Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- added_tokens.json +28 -0
- chat_template.jinja +85 -0
- config.json +38 -0
- generation_config.json +6 -0
- latest +1 -0
- merges.txt +0 -0
- model-00001-of-00013.safetensors +3 -0
- model-00002-of-00013.safetensors +3 -0
- model-00003-of-00013.safetensors +3 -0
- model-00004-of-00013.safetensors +3 -0
- model-00005-of-00013.safetensors +3 -0
- model-00006-of-00013.safetensors +3 -0
- model-00007-of-00013.safetensors +3 -0
- model-00008-of-00013.safetensors +3 -0
- model-00009-of-00013.safetensors +3 -0
- model-00010-of-00013.safetensors +3 -0
- model-00011-of-00013.safetensors +3 -0
- model-00012-of-00013.safetensors +3 -0
- model-00013-of-00013.safetensors +3 -0
- model.safetensors.index.json +0 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- rng_state_2.pth +3 -0
- rng_state_3.pth +3 -0
- rng_state_4.pth +3 -0
- rng_state_5.pth +3 -0
- rng_state_6.pth +3 -0
- rng_state_7.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +31 -0
- tokenizer.json +3 -0
- tokenizer_config.json +240 -0
- trainer_state.json +3331 -0
- training_args.bin +3 -0
- vocab.json +0 -0
- zero_to_fp32.py +760 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
added_tokens.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|box_end|>": 151649,
|
| 9 |
+
"<|box_start|>": 151648,
|
| 10 |
+
"<|endoftext|>": 151643,
|
| 11 |
+
"<|file_sep|>": 151664,
|
| 12 |
+
"<|fim_middle|>": 151660,
|
| 13 |
+
"<|fim_pad|>": 151662,
|
| 14 |
+
"<|fim_prefix|>": 151659,
|
| 15 |
+
"<|fim_suffix|>": 151661,
|
| 16 |
+
"<|im_end|>": 151645,
|
| 17 |
+
"<|im_start|>": 151644,
|
| 18 |
+
"<|image_pad|>": 151655,
|
| 19 |
+
"<|object_ref_end|>": 151647,
|
| 20 |
+
"<|object_ref_start|>": 151646,
|
| 21 |
+
"<|quad_end|>": 151651,
|
| 22 |
+
"<|quad_start|>": 151650,
|
| 23 |
+
"<|repo_name|>": 151663,
|
| 24 |
+
"<|video_pad|>": 151656,
|
| 25 |
+
"<|vision_end|>": 151653,
|
| 26 |
+
"<|vision_pad|>": 151654,
|
| 27 |
+
"<|vision_start|>": 151652
|
| 28 |
+
}
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
| 18 |
+
{%- for message in messages[::-1] %}
|
| 19 |
+
{%- set index = (messages|length - 1) - loop.index0 %}
|
| 20 |
+
{%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
| 21 |
+
{%- set ns.multi_step_tool = false %}
|
| 22 |
+
{%- set ns.last_query_index = index %}
|
| 23 |
+
{%- endif %}
|
| 24 |
+
{%- endfor %}
|
| 25 |
+
{%- for message in messages %}
|
| 26 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 27 |
+
{{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
|
| 28 |
+
{%- elif message.role == "assistant" %}
|
| 29 |
+
{%- set content = message.content %}
|
| 30 |
+
{%- set reasoning_content = '' %}
|
| 31 |
+
{%- if message.reasoning_content is defined and message.reasoning_content is not none %}
|
| 32 |
+
{%- set reasoning_content = message.reasoning_content %}
|
| 33 |
+
{%- else %}
|
| 34 |
+
{%- if '</think>' in message.content %}
|
| 35 |
+
{%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
|
| 36 |
+
{%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
| 37 |
+
{%- endif %}
|
| 38 |
+
{%- endif %}
|
| 39 |
+
{%- if loop.index0 > ns.last_query_index %}
|
| 40 |
+
{%- if loop.last or (not loop.last and reasoning_content) %}
|
| 41 |
+
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
| 42 |
+
{%- else %}
|
| 43 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 44 |
+
{%- endif %}
|
| 45 |
+
{%- else %}
|
| 46 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 47 |
+
{%- endif %}
|
| 48 |
+
{%- if message.tool_calls %}
|
| 49 |
+
{%- for tool_call in message.tool_calls %}
|
| 50 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 51 |
+
{{- '\n' }}
|
| 52 |
+
{%- endif %}
|
| 53 |
+
{%- if tool_call.function %}
|
| 54 |
+
{%- set tool_call = tool_call.function %}
|
| 55 |
+
{%- endif %}
|
| 56 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 57 |
+
{{- tool_call.name }}
|
| 58 |
+
{{- '", "arguments": ' }}
|
| 59 |
+
{%- if tool_call.arguments is string %}
|
| 60 |
+
{{- tool_call.arguments }}
|
| 61 |
+
{%- else %}
|
| 62 |
+
{{- tool_call.arguments | tojson }}
|
| 63 |
+
{%- endif %}
|
| 64 |
+
{{- '}\n</tool_call>' }}
|
| 65 |
+
{%- endfor %}
|
| 66 |
+
{%- endif %}
|
| 67 |
+
{{- '<|im_end|>\n' }}
|
| 68 |
+
{%- elif message.role == "tool" %}
|
| 69 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 70 |
+
{{- '<|im_start|>user' }}
|
| 71 |
+
{%- endif %}
|
| 72 |
+
{{- '\n<tool_response>\n' }}
|
| 73 |
+
{{- message.content }}
|
| 74 |
+
{{- '\n</tool_response>' }}
|
| 75 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 76 |
+
{{- '<|im_end|>\n' }}
|
| 77 |
+
{%- endif %}
|
| 78 |
+
{%- endif %}
|
| 79 |
+
{%- endfor %}
|
| 80 |
+
{%- if add_generation_prompt %}
|
| 81 |
+
{{- '<|im_start|>assistant\n' }}
|
| 82 |
+
{%- if enable_thinking is defined and enable_thinking is false %}
|
| 83 |
+
{{- '<think>\n\n</think>\n\n' }}
|
| 84 |
+
{%- endif %}
|
| 85 |
+
{%- endif %}
|
config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3MoeForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"decoder_sparse_step": 1,
|
| 9 |
+
"eos_token_id": 151643,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 6144,
|
| 15 |
+
"max_position_embeddings": 32768,
|
| 16 |
+
"max_window_layers": 48,
|
| 17 |
+
"mlp_only_layers": [],
|
| 18 |
+
"model_type": "qwen3_moe",
|
| 19 |
+
"moe_intermediate_size": 768,
|
| 20 |
+
"norm_topk_prob": true,
|
| 21 |
+
"num_attention_heads": 32,
|
| 22 |
+
"num_experts": 128,
|
| 23 |
+
"num_experts_per_tok": 8,
|
| 24 |
+
"num_hidden_layers": 48,
|
| 25 |
+
"num_key_value_heads": 4,
|
| 26 |
+
"output_router_logits": false,
|
| 27 |
+
"rms_norm_eps": 1e-06,
|
| 28 |
+
"rope_scaling": null,
|
| 29 |
+
"rope_theta": 1000000.0,
|
| 30 |
+
"router_aux_loss_coef": 0.001,
|
| 31 |
+
"sliding_window": null,
|
| 32 |
+
"tie_word_embeddings": false,
|
| 33 |
+
"torch_dtype": "bfloat16",
|
| 34 |
+
"transformers_version": "4.52.0.dev0",
|
| 35 |
+
"use_cache": false,
|
| 36 |
+
"use_sliding_window": false,
|
| 37 |
+
"vocab_size": 151936
|
| 38 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.52.0.dev0"
|
| 6 |
+
}
|
latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step471
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model-00001-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d6dd0e9e0d970fa731b3654aac0a751ea5553847cbd38ee026fb4832e11ce85
|
| 3 |
+
size 4997184968
|
model-00002-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7d5ecc695ea8db7b3bf064294aea69bd708ad1ac2d9099e4f9cffcf6315c0ff
|
| 3 |
+
size 4997741608
|
model-00003-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34a8e6c80b3ade9467334937ac440ec062acf81b7e9ce2df517b846a2ffe7315
|
| 3 |
+
size 4997742208
|
model-00004-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7ca07acd8642d9e339f0f314ebe68041e2d020fa268c9ddc772f9fd4ed4dce4
|
| 3 |
+
size 4997743184
|
model-00005-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea64f8ab9b7693152b34823aef4cf96ff9446a098445d265640c68921d0f2394
|
| 3 |
+
size 4997743184
|
model-00006-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47301ad2f8eed1fabea833464e60eba5139ea901c6762c19806c387affecebb4
|
| 3 |
+
size 4997743184
|
model-00007-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1924eac7d3d4406ea9db05a928d5e1b1e9019ba59ad3bd4f1c7a881dfa5b672
|
| 3 |
+
size 4997743184
|
model-00008-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c126320c256e859d3751f4ea9ffdba4e35aeb2ca634bf93ff6397efca1aa5a2a
|
| 3 |
+
size 4997743184
|
model-00009-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a3d8a52642bd489d789fd57f301d6b949990c484a3aab554e159f1f3178dad0
|
| 3 |
+
size 4997743184
|
model-00010-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:acb067ac414b9f44711954f993f8c25bd106b69c670771292bc43175997faf67
|
| 3 |
+
size 4997743184
|
model-00011-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8cb40602a392f522ed881a7f11626ae2ca8b1f7e0a3f49773b997a28d784dcc3
|
| 3 |
+
size 4997743184
|
model-00012-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:078db1494ebc9e7bb1879753e40e87e4f658b7c1d175de120084b11c4212331e
|
| 3 |
+
size 4997743184
|
model-00013-of-00013.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ed37f0e7f808003cdfb1a8d0c6eb940c86cf24223e96c600715364060920783e
|
| 3 |
+
size 1094220288
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:478b41e9f26d338fd8f896e08cad1adab7c423b61f1b45754113bc78d256a3f9
|
| 3 |
+
size 16389
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce29a8767a7d907dd24987aa2c3e654d4317f3042fbc13b5b72cadb46d43311a
|
| 3 |
+
size 16389
|
rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61a48db011646b4e9a867bf12f4a233cad5dfbfe309686f8996c250196d3783a
|
| 3 |
+
size 16389
|
rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b9562ee822472a4f01dcd6349ab3d1ef42a48915fe3b92e843a0c37db53c8421
|
| 3 |
+
size 16389
|
rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7d2767d83c3bf27f12db022b0632e2c4f8c164274ba75e380cf18f9d5f21819
|
| 3 |
+
size 16389
|
rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:76816358d4e5db8149d60d55234db658d67a13c0c1ce05d7404cf7125a676a5c
|
| 3 |
+
size 16389
|
rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1562e7520c977d178183d641f70abcf3f57da2489938756cfbebf9b6e6c1a9fd
|
| 3 |
+
size 16389
|
rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6b6cabaed045c5398cd1b732f7ec48bd363f3b43cd24e0e70e641a42bd00c28
|
| 3 |
+
size 16389
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7cae8d2350c49a3f0279a941084d1a2be75cc39e0e78ec5e91a5d52713cac1f1
|
| 3 |
+
size 1465
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"eos_token": {
|
| 18 |
+
"content": "<|im_end|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
"pad_token": {
|
| 25 |
+
"content": "<|endoftext|>",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": false,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
}
|
| 31 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
|
| 3 |
+
size 11422654
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,240 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
},
|
| 181 |
+
"151665": {
|
| 182 |
+
"content": "<tool_response>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": false
|
| 188 |
+
},
|
| 189 |
+
"151666": {
|
| 190 |
+
"content": "</tool_response>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": false
|
| 196 |
+
},
|
| 197 |
+
"151667": {
|
| 198 |
+
"content": "<think>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": false
|
| 204 |
+
},
|
| 205 |
+
"151668": {
|
| 206 |
+
"content": "</think>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": false
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|im_start|>",
|
| 216 |
+
"<|im_end|>",
|
| 217 |
+
"<|object_ref_start|>",
|
| 218 |
+
"<|object_ref_end|>",
|
| 219 |
+
"<|box_start|>",
|
| 220 |
+
"<|box_end|>",
|
| 221 |
+
"<|quad_start|>",
|
| 222 |
+
"<|quad_end|>",
|
| 223 |
+
"<|vision_start|>",
|
| 224 |
+
"<|vision_end|>",
|
| 225 |
+
"<|vision_pad|>",
|
| 226 |
+
"<|image_pad|>",
|
| 227 |
+
"<|video_pad|>"
|
| 228 |
+
],
|
| 229 |
+
"bos_token": null,
|
| 230 |
+
"clean_up_tokenization_spaces": false,
|
| 231 |
+
"eos_token": "<|im_end|>",
|
| 232 |
+
"errors": "replace",
|
| 233 |
+
"extra_special_tokens": {},
|
| 234 |
+
"model_max_length": 131072,
|
| 235 |
+
"pad_token": "<|endoftext|>",
|
| 236 |
+
"padding_side": "right",
|
| 237 |
+
"split_special_tokens": false,
|
| 238 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 239 |
+
"unk_token": null
|
| 240 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,3331 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.60288,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 471,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00128,
|
| 14 |
+
"grad_norm": 2.8870697066158506,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 0.8422,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00256,
|
| 21 |
+
"grad_norm": 2.88484389829891,
|
| 22 |
+
"learning_rate": 6.329113924050633e-07,
|
| 23 |
+
"loss": 0.8541,
|
| 24 |
+
"step": 2
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00384,
|
| 28 |
+
"grad_norm": 2.858151965789657,
|
| 29 |
+
"learning_rate": 1.2658227848101265e-06,
|
| 30 |
+
"loss": 0.8376,
|
| 31 |
+
"step": 3
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00512,
|
| 35 |
+
"grad_norm": 2.759628117182127,
|
| 36 |
+
"learning_rate": 1.8987341772151901e-06,
|
| 37 |
+
"loss": 0.8334,
|
| 38 |
+
"step": 4
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0064,
|
| 42 |
+
"grad_norm": 2.796990062811218,
|
| 43 |
+
"learning_rate": 2.531645569620253e-06,
|
| 44 |
+
"loss": 0.8256,
|
| 45 |
+
"step": 5
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00768,
|
| 49 |
+
"grad_norm": 2.5779298795445023,
|
| 50 |
+
"learning_rate": 3.1645569620253167e-06,
|
| 51 |
+
"loss": 0.8301,
|
| 52 |
+
"step": 6
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.00896,
|
| 56 |
+
"grad_norm": 2.182261607936066,
|
| 57 |
+
"learning_rate": 3.7974683544303802e-06,
|
| 58 |
+
"loss": 0.8156,
|
| 59 |
+
"step": 7
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.01024,
|
| 63 |
+
"grad_norm": 1.9615896152651355,
|
| 64 |
+
"learning_rate": 4.430379746835443e-06,
|
| 65 |
+
"loss": 0.7982,
|
| 66 |
+
"step": 8
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.01152,
|
| 70 |
+
"grad_norm": 1.452541644948315,
|
| 71 |
+
"learning_rate": 5.063291139240506e-06,
|
| 72 |
+
"loss": 0.7819,
|
| 73 |
+
"step": 9
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0128,
|
| 77 |
+
"grad_norm": 1.4723286808630864,
|
| 78 |
+
"learning_rate": 5.69620253164557e-06,
|
| 79 |
+
"loss": 0.7906,
|
| 80 |
+
"step": 10
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.01408,
|
| 84 |
+
"grad_norm": 1.3529636617858944,
|
| 85 |
+
"learning_rate": 6.329113924050633e-06,
|
| 86 |
+
"loss": 0.7724,
|
| 87 |
+
"step": 11
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.01536,
|
| 91 |
+
"grad_norm": 1.960737179905222,
|
| 92 |
+
"learning_rate": 6.9620253164556965e-06,
|
| 93 |
+
"loss": 0.7495,
|
| 94 |
+
"step": 12
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.01664,
|
| 98 |
+
"grad_norm": 2.2349101055406337,
|
| 99 |
+
"learning_rate": 7.5949367088607605e-06,
|
| 100 |
+
"loss": 0.7581,
|
| 101 |
+
"step": 13
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.01792,
|
| 105 |
+
"grad_norm": 2.0897577150322477,
|
| 106 |
+
"learning_rate": 8.227848101265822e-06,
|
| 107 |
+
"loss": 0.7404,
|
| 108 |
+
"step": 14
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0192,
|
| 112 |
+
"grad_norm": 1.8227218322635887,
|
| 113 |
+
"learning_rate": 8.860759493670886e-06,
|
| 114 |
+
"loss": 0.7382,
|
| 115 |
+
"step": 15
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.02048,
|
| 119 |
+
"grad_norm": 1.2099951464458898,
|
| 120 |
+
"learning_rate": 9.49367088607595e-06,
|
| 121 |
+
"loss": 0.7231,
|
| 122 |
+
"step": 16
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.02176,
|
| 126 |
+
"grad_norm": 1.2177037129914572,
|
| 127 |
+
"learning_rate": 1.0126582278481012e-05,
|
| 128 |
+
"loss": 0.7259,
|
| 129 |
+
"step": 17
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.02304,
|
| 133 |
+
"grad_norm": 1.1031346132830708,
|
| 134 |
+
"learning_rate": 1.0759493670886076e-05,
|
| 135 |
+
"loss": 0.7059,
|
| 136 |
+
"step": 18
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.02432,
|
| 140 |
+
"grad_norm": 0.9194779600801882,
|
| 141 |
+
"learning_rate": 1.139240506329114e-05,
|
| 142 |
+
"loss": 0.7137,
|
| 143 |
+
"step": 19
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0256,
|
| 147 |
+
"grad_norm": 0.8679468005972053,
|
| 148 |
+
"learning_rate": 1.2025316455696203e-05,
|
| 149 |
+
"loss": 0.7036,
|
| 150 |
+
"step": 20
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.02688,
|
| 154 |
+
"grad_norm": 0.7227287276969042,
|
| 155 |
+
"learning_rate": 1.2658227848101267e-05,
|
| 156 |
+
"loss": 0.696,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.02816,
|
| 161 |
+
"grad_norm": 0.7425882516811844,
|
| 162 |
+
"learning_rate": 1.3291139240506329e-05,
|
| 163 |
+
"loss": 0.6888,
|
| 164 |
+
"step": 22
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.02944,
|
| 168 |
+
"grad_norm": 0.7093793012252196,
|
| 169 |
+
"learning_rate": 1.3924050632911393e-05,
|
| 170 |
+
"loss": 0.6791,
|
| 171 |
+
"step": 23
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.03072,
|
| 175 |
+
"grad_norm": 0.6018215463147907,
|
| 176 |
+
"learning_rate": 1.4556962025316457e-05,
|
| 177 |
+
"loss": 0.6783,
|
| 178 |
+
"step": 24
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.032,
|
| 182 |
+
"grad_norm": 0.5846346732378257,
|
| 183 |
+
"learning_rate": 1.5189873417721521e-05,
|
| 184 |
+
"loss": 0.6811,
|
| 185 |
+
"step": 25
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.03328,
|
| 189 |
+
"grad_norm": 0.5855419788452784,
|
| 190 |
+
"learning_rate": 1.5822784810126583e-05,
|
| 191 |
+
"loss": 0.683,
|
| 192 |
+
"step": 26
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.03456,
|
| 196 |
+
"grad_norm": 0.5096689891724868,
|
| 197 |
+
"learning_rate": 1.6455696202531644e-05,
|
| 198 |
+
"loss": 0.6589,
|
| 199 |
+
"step": 27
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.03584,
|
| 203 |
+
"grad_norm": 0.4871170504081146,
|
| 204 |
+
"learning_rate": 1.7088607594936708e-05,
|
| 205 |
+
"loss": 0.6582,
|
| 206 |
+
"step": 28
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.03712,
|
| 210 |
+
"grad_norm": 0.4949600697144217,
|
| 211 |
+
"learning_rate": 1.7721518987341772e-05,
|
| 212 |
+
"loss": 0.669,
|
| 213 |
+
"step": 29
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0384,
|
| 217 |
+
"grad_norm": 0.5082926031630941,
|
| 218 |
+
"learning_rate": 1.8354430379746836e-05,
|
| 219 |
+
"loss": 0.666,
|
| 220 |
+
"step": 30
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.03968,
|
| 224 |
+
"grad_norm": 0.49381475380567175,
|
| 225 |
+
"learning_rate": 1.89873417721519e-05,
|
| 226 |
+
"loss": 0.6556,
|
| 227 |
+
"step": 31
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.04096,
|
| 231 |
+
"grad_norm": 0.4265624784331274,
|
| 232 |
+
"learning_rate": 1.962025316455696e-05,
|
| 233 |
+
"loss": 0.646,
|
| 234 |
+
"step": 32
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.04224,
|
| 238 |
+
"grad_norm": 0.39190416547723717,
|
| 239 |
+
"learning_rate": 2.0253164556962025e-05,
|
| 240 |
+
"loss": 0.6473,
|
| 241 |
+
"step": 33
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.04352,
|
| 245 |
+
"grad_norm": 0.4631353399929371,
|
| 246 |
+
"learning_rate": 2.088607594936709e-05,
|
| 247 |
+
"loss": 0.6441,
|
| 248 |
+
"step": 34
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0448,
|
| 252 |
+
"grad_norm": 0.3928335126997034,
|
| 253 |
+
"learning_rate": 2.1518987341772153e-05,
|
| 254 |
+
"loss": 0.6352,
|
| 255 |
+
"step": 35
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.04608,
|
| 259 |
+
"grad_norm": 0.36295027582313966,
|
| 260 |
+
"learning_rate": 2.2151898734177217e-05,
|
| 261 |
+
"loss": 0.6333,
|
| 262 |
+
"step": 36
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.04736,
|
| 266 |
+
"grad_norm": 0.35026852064181846,
|
| 267 |
+
"learning_rate": 2.278481012658228e-05,
|
| 268 |
+
"loss": 0.6399,
|
| 269 |
+
"step": 37
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.04864,
|
| 273 |
+
"grad_norm": 0.39778614916835536,
|
| 274 |
+
"learning_rate": 2.341772151898734e-05,
|
| 275 |
+
"loss": 0.6298,
|
| 276 |
+
"step": 38
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.04992,
|
| 280 |
+
"grad_norm": 0.33278348666417684,
|
| 281 |
+
"learning_rate": 2.4050632911392405e-05,
|
| 282 |
+
"loss": 0.6301,
|
| 283 |
+
"step": 39
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0512,
|
| 287 |
+
"grad_norm": 0.31444068712551376,
|
| 288 |
+
"learning_rate": 2.468354430379747e-05,
|
| 289 |
+
"loss": 0.6263,
|
| 290 |
+
"step": 40
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.05248,
|
| 294 |
+
"grad_norm": 0.36059728676958264,
|
| 295 |
+
"learning_rate": 2.5316455696202533e-05,
|
| 296 |
+
"loss": 0.6458,
|
| 297 |
+
"step": 41
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.05376,
|
| 301 |
+
"grad_norm": 0.3916144552301749,
|
| 302 |
+
"learning_rate": 2.5949367088607597e-05,
|
| 303 |
+
"loss": 0.6331,
|
| 304 |
+
"step": 42
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.05504,
|
| 308 |
+
"grad_norm": 0.32338566356420756,
|
| 309 |
+
"learning_rate": 2.6582278481012658e-05,
|
| 310 |
+
"loss": 0.6332,
|
| 311 |
+
"step": 43
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.05632,
|
| 315 |
+
"grad_norm": 0.33704233729853356,
|
| 316 |
+
"learning_rate": 2.7215189873417722e-05,
|
| 317 |
+
"loss": 0.6348,
|
| 318 |
+
"step": 44
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0576,
|
| 322 |
+
"grad_norm": 0.36015399213900634,
|
| 323 |
+
"learning_rate": 2.7848101265822786e-05,
|
| 324 |
+
"loss": 0.6392,
|
| 325 |
+
"step": 45
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.05888,
|
| 329 |
+
"grad_norm": 0.31471331803021757,
|
| 330 |
+
"learning_rate": 2.848101265822785e-05,
|
| 331 |
+
"loss": 0.6272,
|
| 332 |
+
"step": 46
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.06016,
|
| 336 |
+
"grad_norm": 0.3225170654156012,
|
| 337 |
+
"learning_rate": 2.9113924050632914e-05,
|
| 338 |
+
"loss": 0.635,
|
| 339 |
+
"step": 47
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.06144,
|
| 343 |
+
"grad_norm": 0.3064473735810606,
|
| 344 |
+
"learning_rate": 2.9746835443037974e-05,
|
| 345 |
+
"loss": 0.6284,
|
| 346 |
+
"step": 48
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.06272,
|
| 350 |
+
"grad_norm": 0.3038289969291092,
|
| 351 |
+
"learning_rate": 3.0379746835443042e-05,
|
| 352 |
+
"loss": 0.6149,
|
| 353 |
+
"step": 49
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.064,
|
| 357 |
+
"grad_norm": 0.3226803690164346,
|
| 358 |
+
"learning_rate": 3.10126582278481e-05,
|
| 359 |
+
"loss": 0.626,
|
| 360 |
+
"step": 50
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.06528,
|
| 364 |
+
"grad_norm": 0.3096398144524693,
|
| 365 |
+
"learning_rate": 3.1645569620253167e-05,
|
| 366 |
+
"loss": 0.621,
|
| 367 |
+
"step": 51
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.06656,
|
| 371 |
+
"grad_norm": 0.2754757429130796,
|
| 372 |
+
"learning_rate": 3.227848101265823e-05,
|
| 373 |
+
"loss": 0.6185,
|
| 374 |
+
"step": 52
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.06784,
|
| 378 |
+
"grad_norm": 0.3262507218160328,
|
| 379 |
+
"learning_rate": 3.291139240506329e-05,
|
| 380 |
+
"loss": 0.6171,
|
| 381 |
+
"step": 53
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.06912,
|
| 385 |
+
"grad_norm": 0.34971068352090656,
|
| 386 |
+
"learning_rate": 3.354430379746836e-05,
|
| 387 |
+
"loss": 0.616,
|
| 388 |
+
"step": 54
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.0704,
|
| 392 |
+
"grad_norm": 0.2841621281043231,
|
| 393 |
+
"learning_rate": 3.4177215189873416e-05,
|
| 394 |
+
"loss": 0.5995,
|
| 395 |
+
"step": 55
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.07168,
|
| 399 |
+
"grad_norm": 0.4003223636484448,
|
| 400 |
+
"learning_rate": 3.4810126582278487e-05,
|
| 401 |
+
"loss": 0.6169,
|
| 402 |
+
"step": 56
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.07296,
|
| 406 |
+
"grad_norm": 0.31868860231705426,
|
| 407 |
+
"learning_rate": 3.5443037974683544e-05,
|
| 408 |
+
"loss": 0.6077,
|
| 409 |
+
"step": 57
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.07424,
|
| 413 |
+
"grad_norm": 0.3960425782005289,
|
| 414 |
+
"learning_rate": 3.607594936708861e-05,
|
| 415 |
+
"loss": 0.6164,
|
| 416 |
+
"step": 58
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.07552,
|
| 420 |
+
"grad_norm": 0.363865574596696,
|
| 421 |
+
"learning_rate": 3.670886075949367e-05,
|
| 422 |
+
"loss": 0.6118,
|
| 423 |
+
"step": 59
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.0768,
|
| 427 |
+
"grad_norm": 0.33961478774466697,
|
| 428 |
+
"learning_rate": 3.7341772151898736e-05,
|
| 429 |
+
"loss": 0.6137,
|
| 430 |
+
"step": 60
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.07808,
|
| 434 |
+
"grad_norm": 0.4212164741206082,
|
| 435 |
+
"learning_rate": 3.79746835443038e-05,
|
| 436 |
+
"loss": 0.6275,
|
| 437 |
+
"step": 61
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.07936,
|
| 441 |
+
"grad_norm": 0.29878729710395663,
|
| 442 |
+
"learning_rate": 3.8607594936708864e-05,
|
| 443 |
+
"loss": 0.6084,
|
| 444 |
+
"step": 62
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.08064,
|
| 448 |
+
"grad_norm": 0.36745026817379894,
|
| 449 |
+
"learning_rate": 3.924050632911392e-05,
|
| 450 |
+
"loss": 0.607,
|
| 451 |
+
"step": 63
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.08192,
|
| 455 |
+
"grad_norm": 0.38983571508393644,
|
| 456 |
+
"learning_rate": 3.987341772151899e-05,
|
| 457 |
+
"loss": 0.6176,
|
| 458 |
+
"step": 64
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0832,
|
| 462 |
+
"grad_norm": 0.37337392917475115,
|
| 463 |
+
"learning_rate": 4.050632911392405e-05,
|
| 464 |
+
"loss": 0.6184,
|
| 465 |
+
"step": 65
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.08448,
|
| 469 |
+
"grad_norm": 0.3668068115925863,
|
| 470 |
+
"learning_rate": 4.113924050632912e-05,
|
| 471 |
+
"loss": 0.6194,
|
| 472 |
+
"step": 66
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.08576,
|
| 476 |
+
"grad_norm": 0.36138503055306903,
|
| 477 |
+
"learning_rate": 4.177215189873418e-05,
|
| 478 |
+
"loss": 0.6077,
|
| 479 |
+
"step": 67
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.08704,
|
| 483 |
+
"grad_norm": 0.43361127462043814,
|
| 484 |
+
"learning_rate": 4.240506329113924e-05,
|
| 485 |
+
"loss": 0.6147,
|
| 486 |
+
"step": 68
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.08832,
|
| 490 |
+
"grad_norm": 0.33520423726109644,
|
| 491 |
+
"learning_rate": 4.3037974683544305e-05,
|
| 492 |
+
"loss": 0.6118,
|
| 493 |
+
"step": 69
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.0896,
|
| 497 |
+
"grad_norm": 0.4381154362148859,
|
| 498 |
+
"learning_rate": 4.367088607594937e-05,
|
| 499 |
+
"loss": 0.6031,
|
| 500 |
+
"step": 70
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.09088,
|
| 504 |
+
"grad_norm": 0.3717345864324632,
|
| 505 |
+
"learning_rate": 4.430379746835443e-05,
|
| 506 |
+
"loss": 0.6031,
|
| 507 |
+
"step": 71
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.09216,
|
| 511 |
+
"grad_norm": 0.4861728465398392,
|
| 512 |
+
"learning_rate": 4.49367088607595e-05,
|
| 513 |
+
"loss": 0.6006,
|
| 514 |
+
"step": 72
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.09344,
|
| 518 |
+
"grad_norm": 0.3264992939190504,
|
| 519 |
+
"learning_rate": 4.556962025316456e-05,
|
| 520 |
+
"loss": 0.6151,
|
| 521 |
+
"step": 73
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.09472,
|
| 525 |
+
"grad_norm": 0.4319794925001871,
|
| 526 |
+
"learning_rate": 4.6202531645569625e-05,
|
| 527 |
+
"loss": 0.6058,
|
| 528 |
+
"step": 74
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.096,
|
| 532 |
+
"grad_norm": 0.4616345840492333,
|
| 533 |
+
"learning_rate": 4.683544303797468e-05,
|
| 534 |
+
"loss": 0.5967,
|
| 535 |
+
"step": 75
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.09728,
|
| 539 |
+
"grad_norm": 0.4405721152587957,
|
| 540 |
+
"learning_rate": 4.7468354430379746e-05,
|
| 541 |
+
"loss": 0.6002,
|
| 542 |
+
"step": 76
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.09856,
|
| 546 |
+
"grad_norm": 0.5122605377853799,
|
| 547 |
+
"learning_rate": 4.810126582278481e-05,
|
| 548 |
+
"loss": 0.6076,
|
| 549 |
+
"step": 77
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.09984,
|
| 553 |
+
"grad_norm": 0.45313870340097556,
|
| 554 |
+
"learning_rate": 4.8734177215189874e-05,
|
| 555 |
+
"loss": 0.6074,
|
| 556 |
+
"step": 78
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.10112,
|
| 560 |
+
"grad_norm": 0.4340044755876676,
|
| 561 |
+
"learning_rate": 4.936708860759494e-05,
|
| 562 |
+
"loss": 0.606,
|
| 563 |
+
"step": 79
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.1024,
|
| 567 |
+
"grad_norm": 0.4987172862476422,
|
| 568 |
+
"learning_rate": 5e-05,
|
| 569 |
+
"loss": 0.6158,
|
| 570 |
+
"step": 80
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.10368,
|
| 574 |
+
"grad_norm": 0.6226880208665108,
|
| 575 |
+
"learning_rate": 4.999974965737065e-05,
|
| 576 |
+
"loss": 0.621,
|
| 577 |
+
"step": 81
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.10496,
|
| 581 |
+
"grad_norm": 0.5448293131914782,
|
| 582 |
+
"learning_rate": 4.999899863449631e-05,
|
| 583 |
+
"loss": 0.6014,
|
| 584 |
+
"step": 82
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.10624,
|
| 588 |
+
"grad_norm": 0.3427022601926917,
|
| 589 |
+
"learning_rate": 4.999774694641803e-05,
|
| 590 |
+
"loss": 0.6198,
|
| 591 |
+
"step": 83
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.10752,
|
| 595 |
+
"grad_norm": 0.5005152113593655,
|
| 596 |
+
"learning_rate": 4.999599461820387e-05,
|
| 597 |
+
"loss": 0.6054,
|
| 598 |
+
"step": 84
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.1088,
|
| 602 |
+
"grad_norm": 0.5702968806820528,
|
| 603 |
+
"learning_rate": 4.999374168494844e-05,
|
| 604 |
+
"loss": 0.6069,
|
| 605 |
+
"step": 85
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.11008,
|
| 609 |
+
"grad_norm": 0.4671310661706222,
|
| 610 |
+
"learning_rate": 4.999098819177214e-05,
|
| 611 |
+
"loss": 0.6017,
|
| 612 |
+
"step": 86
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.11136,
|
| 616 |
+
"grad_norm": 0.46081768174689064,
|
| 617 |
+
"learning_rate": 4.9987734193820324e-05,
|
| 618 |
+
"loss": 0.5988,
|
| 619 |
+
"step": 87
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.11264,
|
| 623 |
+
"grad_norm": 0.5448729856183013,
|
| 624 |
+
"learning_rate": 4.9983979756262136e-05,
|
| 625 |
+
"loss": 0.6181,
|
| 626 |
+
"step": 88
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.11392,
|
| 630 |
+
"grad_norm": 0.5095775592779056,
|
| 631 |
+
"learning_rate": 4.9979724954289244e-05,
|
| 632 |
+
"loss": 0.608,
|
| 633 |
+
"step": 89
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.1152,
|
| 637 |
+
"grad_norm": 0.41119162739543413,
|
| 638 |
+
"learning_rate": 4.997496987311431e-05,
|
| 639 |
+
"loss": 0.5979,
|
| 640 |
+
"step": 90
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.11648,
|
| 644 |
+
"grad_norm": 0.45501958535738946,
|
| 645 |
+
"learning_rate": 4.996971460796929e-05,
|
| 646 |
+
"loss": 0.6019,
|
| 647 |
+
"step": 91
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.11776,
|
| 651 |
+
"grad_norm": 0.4287172104360816,
|
| 652 |
+
"learning_rate": 4.9963959264103544e-05,
|
| 653 |
+
"loss": 0.5955,
|
| 654 |
+
"step": 92
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.11904,
|
| 658 |
+
"grad_norm": 0.409872269342458,
|
| 659 |
+
"learning_rate": 4.995770395678171e-05,
|
| 660 |
+
"loss": 0.5927,
|
| 661 |
+
"step": 93
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.12032,
|
| 665 |
+
"grad_norm": 0.4304173966206036,
|
| 666 |
+
"learning_rate": 4.995094881128138e-05,
|
| 667 |
+
"loss": 0.5967,
|
| 668 |
+
"step": 94
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.1216,
|
| 672 |
+
"grad_norm": 0.4229799776298517,
|
| 673 |
+
"learning_rate": 4.994369396289063e-05,
|
| 674 |
+
"loss": 0.6084,
|
| 675 |
+
"step": 95
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.12288,
|
| 679 |
+
"grad_norm": 0.4509596954971553,
|
| 680 |
+
"learning_rate": 4.9935939556905295e-05,
|
| 681 |
+
"loss": 0.6134,
|
| 682 |
+
"step": 96
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.12416,
|
| 686 |
+
"grad_norm": 0.467661146414229,
|
| 687 |
+
"learning_rate": 4.992768574862603e-05,
|
| 688 |
+
"loss": 0.5986,
|
| 689 |
+
"step": 97
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.12544,
|
| 693 |
+
"grad_norm": 0.42432875998240194,
|
| 694 |
+
"learning_rate": 4.9918932703355256e-05,
|
| 695 |
+
"loss": 0.6028,
|
| 696 |
+
"step": 98
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.12672,
|
| 700 |
+
"grad_norm": 0.43479377184835605,
|
| 701 |
+
"learning_rate": 4.990968059639379e-05,
|
| 702 |
+
"loss": 0.5942,
|
| 703 |
+
"step": 99
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.128,
|
| 707 |
+
"grad_norm": 0.3680676685801686,
|
| 708 |
+
"learning_rate": 4.989992961303738e-05,
|
| 709 |
+
"loss": 0.5994,
|
| 710 |
+
"step": 100
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.12928,
|
| 714 |
+
"grad_norm": 0.3956815409903461,
|
| 715 |
+
"learning_rate": 4.9889679948572974e-05,
|
| 716 |
+
"loss": 0.5871,
|
| 717 |
+
"step": 101
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.13056,
|
| 721 |
+
"grad_norm": 0.34354949934586104,
|
| 722 |
+
"learning_rate": 4.98789318082748e-05,
|
| 723 |
+
"loss": 0.5873,
|
| 724 |
+
"step": 102
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.13184,
|
| 728 |
+
"grad_norm": 0.3608260963951222,
|
| 729 |
+
"learning_rate": 4.986768540740028e-05,
|
| 730 |
+
"loss": 0.5883,
|
| 731 |
+
"step": 103
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.13312,
|
| 735 |
+
"grad_norm": 0.3937004101078116,
|
| 736 |
+
"learning_rate": 4.98559409711857e-05,
|
| 737 |
+
"loss": 0.6029,
|
| 738 |
+
"step": 104
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.1344,
|
| 742 |
+
"grad_norm": 0.3401718481532899,
|
| 743 |
+
"learning_rate": 4.9843698734841705e-05,
|
| 744 |
+
"loss": 0.5983,
|
| 745 |
+
"step": 105
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.13568,
|
| 749 |
+
"grad_norm": 0.4371868869288284,
|
| 750 |
+
"learning_rate": 4.983095894354858e-05,
|
| 751 |
+
"loss": 0.5866,
|
| 752 |
+
"step": 106
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.13696,
|
| 756 |
+
"grad_norm": 0.3722813571279646,
|
| 757 |
+
"learning_rate": 4.981772185245135e-05,
|
| 758 |
+
"loss": 0.5954,
|
| 759 |
+
"step": 107
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.13824,
|
| 763 |
+
"grad_norm": 0.36493596395606354,
|
| 764 |
+
"learning_rate": 4.980398772665468e-05,
|
| 765 |
+
"loss": 0.5806,
|
| 766 |
+
"step": 108
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.13952,
|
| 770 |
+
"grad_norm": 0.43678937522389644,
|
| 771 |
+
"learning_rate": 4.9789756841217546e-05,
|
| 772 |
+
"loss": 0.595,
|
| 773 |
+
"step": 109
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.1408,
|
| 777 |
+
"grad_norm": 0.34968596729530604,
|
| 778 |
+
"learning_rate": 4.977502948114772e-05,
|
| 779 |
+
"loss": 0.5999,
|
| 780 |
+
"step": 110
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.14208,
|
| 784 |
+
"grad_norm": 0.4035249077012057,
|
| 785 |
+
"learning_rate": 4.9759805941396075e-05,
|
| 786 |
+
"loss": 0.582,
|
| 787 |
+
"step": 111
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.14336,
|
| 791 |
+
"grad_norm": 0.3396387531525401,
|
| 792 |
+
"learning_rate": 4.974408652685072e-05,
|
| 793 |
+
"loss": 0.5912,
|
| 794 |
+
"step": 112
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.14464,
|
| 798 |
+
"grad_norm": 0.3888124435581031,
|
| 799 |
+
"learning_rate": 4.9727871552330794e-05,
|
| 800 |
+
"loss": 0.5994,
|
| 801 |
+
"step": 113
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.14592,
|
| 805 |
+
"grad_norm": 0.3487289265208422,
|
| 806 |
+
"learning_rate": 4.971116134258025e-05,
|
| 807 |
+
"loss": 0.598,
|
| 808 |
+
"step": 114
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.1472,
|
| 812 |
+
"grad_norm": 0.34084258932596606,
|
| 813 |
+
"learning_rate": 4.969395623226133e-05,
|
| 814 |
+
"loss": 0.5965,
|
| 815 |
+
"step": 115
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.14848,
|
| 819 |
+
"grad_norm": 0.33211872605390524,
|
| 820 |
+
"learning_rate": 4.967625656594782e-05,
|
| 821 |
+
"loss": 0.5984,
|
| 822 |
+
"step": 116
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.14976,
|
| 826 |
+
"grad_norm": 0.31055192632357626,
|
| 827 |
+
"learning_rate": 4.9658062698118213e-05,
|
| 828 |
+
"loss": 0.593,
|
| 829 |
+
"step": 117
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.15104,
|
| 833 |
+
"grad_norm": 0.35790400007793166,
|
| 834 |
+
"learning_rate": 4.963937499314857e-05,
|
| 835 |
+
"loss": 0.6035,
|
| 836 |
+
"step": 118
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.15232,
|
| 840 |
+
"grad_norm": 0.31118450185510343,
|
| 841 |
+
"learning_rate": 4.962019382530521e-05,
|
| 842 |
+
"loss": 0.5811,
|
| 843 |
+
"step": 119
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.1536,
|
| 847 |
+
"grad_norm": 0.3326176465041298,
|
| 848 |
+
"learning_rate": 4.960051957873725e-05,
|
| 849 |
+
"loss": 0.581,
|
| 850 |
+
"step": 120
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.15488,
|
| 854 |
+
"grad_norm": 0.30210249377153575,
|
| 855 |
+
"learning_rate": 4.958035264746893e-05,
|
| 856 |
+
"loss": 0.5837,
|
| 857 |
+
"step": 121
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.15616,
|
| 861 |
+
"grad_norm": 0.3480385124671555,
|
| 862 |
+
"learning_rate": 4.955969343539162e-05,
|
| 863 |
+
"loss": 0.5768,
|
| 864 |
+
"step": 122
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.15744,
|
| 868 |
+
"grad_norm": 0.3003392569743352,
|
| 869 |
+
"learning_rate": 4.9538542356255866e-05,
|
| 870 |
+
"loss": 0.5938,
|
| 871 |
+
"step": 123
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.15872,
|
| 875 |
+
"grad_norm": 0.32082565179488104,
|
| 876 |
+
"learning_rate": 4.9516899833663e-05,
|
| 877 |
+
"loss": 0.5948,
|
| 878 |
+
"step": 124
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.16,
|
| 882 |
+
"grad_norm": 0.3564349708048278,
|
| 883 |
+
"learning_rate": 4.949476630105669e-05,
|
| 884 |
+
"loss": 0.595,
|
| 885 |
+
"step": 125
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.16128,
|
| 889 |
+
"grad_norm": 0.32049541972124757,
|
| 890 |
+
"learning_rate": 4.94721422017143e-05,
|
| 891 |
+
"loss": 0.5838,
|
| 892 |
+
"step": 126
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.16256,
|
| 896 |
+
"grad_norm": 0.3317680882353993,
|
| 897 |
+
"learning_rate": 4.944902798873794e-05,
|
| 898 |
+
"loss": 0.5952,
|
| 899 |
+
"step": 127
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.16384,
|
| 903 |
+
"grad_norm": 0.3381465061198974,
|
| 904 |
+
"learning_rate": 4.942542412504543e-05,
|
| 905 |
+
"loss": 0.6004,
|
| 906 |
+
"step": 128
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.16512,
|
| 910 |
+
"grad_norm": 0.38351657127693595,
|
| 911 |
+
"learning_rate": 4.940133108336105e-05,
|
| 912 |
+
"loss": 0.6014,
|
| 913 |
+
"step": 129
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.1664,
|
| 917 |
+
"grad_norm": 0.3276142738951724,
|
| 918 |
+
"learning_rate": 4.9376749346206006e-05,
|
| 919 |
+
"loss": 0.5853,
|
| 920 |
+
"step": 130
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.16768,
|
| 924 |
+
"grad_norm": 0.37146400882939534,
|
| 925 |
+
"learning_rate": 4.935167940588887e-05,
|
| 926 |
+
"loss": 0.5995,
|
| 927 |
+
"step": 131
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.16896,
|
| 931 |
+
"grad_norm": 0.32804274509201087,
|
| 932 |
+
"learning_rate": 4.9326121764495596e-05,
|
| 933 |
+
"loss": 0.5955,
|
| 934 |
+
"step": 132
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.17024,
|
| 938 |
+
"grad_norm": 0.3344845806030499,
|
| 939 |
+
"learning_rate": 4.9300076933879574e-05,
|
| 940 |
+
"loss": 0.5818,
|
| 941 |
+
"step": 133
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.17152,
|
| 945 |
+
"grad_norm": 0.3479572078392269,
|
| 946 |
+
"learning_rate": 4.92735454356513e-05,
|
| 947 |
+
"loss": 0.5941,
|
| 948 |
+
"step": 134
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.1728,
|
| 952 |
+
"grad_norm": 0.34868252062960353,
|
| 953 |
+
"learning_rate": 4.924652780116799e-05,
|
| 954 |
+
"loss": 0.5898,
|
| 955 |
+
"step": 135
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.17408,
|
| 959 |
+
"grad_norm": 0.35674279058993497,
|
| 960 |
+
"learning_rate": 4.921902457152289e-05,
|
| 961 |
+
"loss": 0.5899,
|
| 962 |
+
"step": 136
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.17536,
|
| 966 |
+
"grad_norm": 0.3672614416380493,
|
| 967 |
+
"learning_rate": 4.9191036297534454e-05,
|
| 968 |
+
"loss": 0.585,
|
| 969 |
+
"step": 137
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.17664,
|
| 973 |
+
"grad_norm": 0.4039478601084677,
|
| 974 |
+
"learning_rate": 4.916256353973535e-05,
|
| 975 |
+
"loss": 0.5994,
|
| 976 |
+
"step": 138
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.17792,
|
| 980 |
+
"grad_norm": 0.3428958061155067,
|
| 981 |
+
"learning_rate": 4.913360686836117e-05,
|
| 982 |
+
"loss": 0.575,
|
| 983 |
+
"step": 139
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.1792,
|
| 987 |
+
"grad_norm": 0.4024960602256603,
|
| 988 |
+
"learning_rate": 4.910416686333906e-05,
|
| 989 |
+
"loss": 0.5913,
|
| 990 |
+
"step": 140
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.18048,
|
| 994 |
+
"grad_norm": 0.31040065034832104,
|
| 995 |
+
"learning_rate": 4.907424411427608e-05,
|
| 996 |
+
"loss": 0.5761,
|
| 997 |
+
"step": 141
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.18176,
|
| 1001 |
+
"grad_norm": 0.359237099401051,
|
| 1002 |
+
"learning_rate": 4.90438392204474e-05,
|
| 1003 |
+
"loss": 0.5885,
|
| 1004 |
+
"step": 142
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.18304,
|
| 1008 |
+
"grad_norm": 0.3357545415879296,
|
| 1009 |
+
"learning_rate": 4.901295279078431e-05,
|
| 1010 |
+
"loss": 0.5907,
|
| 1011 |
+
"step": 143
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.18432,
|
| 1015 |
+
"grad_norm": 0.2846403022642179,
|
| 1016 |
+
"learning_rate": 4.898158544386201e-05,
|
| 1017 |
+
"loss": 0.5886,
|
| 1018 |
+
"step": 144
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.1856,
|
| 1022 |
+
"grad_norm": 0.3636245125193307,
|
| 1023 |
+
"learning_rate": 4.894973780788722e-05,
|
| 1024 |
+
"loss": 0.5816,
|
| 1025 |
+
"step": 145
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.18688,
|
| 1029 |
+
"grad_norm": 0.25440894793562924,
|
| 1030 |
+
"learning_rate": 4.8917410520685635e-05,
|
| 1031 |
+
"loss": 0.576,
|
| 1032 |
+
"step": 146
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.18816,
|
| 1036 |
+
"grad_norm": 0.3380189678855273,
|
| 1037 |
+
"learning_rate": 4.888460422968908e-05,
|
| 1038 |
+
"loss": 0.5931,
|
| 1039 |
+
"step": 147
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.18944,
|
| 1043 |
+
"grad_norm": 0.3096794617975588,
|
| 1044 |
+
"learning_rate": 4.885131959192262e-05,
|
| 1045 |
+
"loss": 0.5829,
|
| 1046 |
+
"step": 148
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.19072,
|
| 1050 |
+
"grad_norm": 0.280174710159943,
|
| 1051 |
+
"learning_rate": 4.881755727399134e-05,
|
| 1052 |
+
"loss": 0.5794,
|
| 1053 |
+
"step": 149
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.192,
|
| 1057 |
+
"grad_norm": 0.31769340776297994,
|
| 1058 |
+
"learning_rate": 4.878331795206705e-05,
|
| 1059 |
+
"loss": 0.5729,
|
| 1060 |
+
"step": 150
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.19328,
|
| 1064 |
+
"grad_norm": 0.31671973855902796,
|
| 1065 |
+
"learning_rate": 4.8748602311874694e-05,
|
| 1066 |
+
"loss": 0.5905,
|
| 1067 |
+
"step": 151
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.19456,
|
| 1071 |
+
"grad_norm": 0.32614211009906474,
|
| 1072 |
+
"learning_rate": 4.8713411048678635e-05,
|
| 1073 |
+
"loss": 0.5855,
|
| 1074 |
+
"step": 152
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.19584,
|
| 1078 |
+
"grad_norm": 0.29921149443441614,
|
| 1079 |
+
"learning_rate": 4.8677744867268764e-05,
|
| 1080 |
+
"loss": 0.5779,
|
| 1081 |
+
"step": 153
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.19712,
|
| 1085 |
+
"grad_norm": 0.3558339409344647,
|
| 1086 |
+
"learning_rate": 4.8641604481946314e-05,
|
| 1087 |
+
"loss": 0.5892,
|
| 1088 |
+
"step": 154
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.1984,
|
| 1092 |
+
"grad_norm": 0.285079025062,
|
| 1093 |
+
"learning_rate": 4.8604990616509616e-05,
|
| 1094 |
+
"loss": 0.5912,
|
| 1095 |
+
"step": 155
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.19968,
|
| 1099 |
+
"grad_norm": 0.32189736402098207,
|
| 1100 |
+
"learning_rate": 4.856790400423958e-05,
|
| 1101 |
+
"loss": 0.5881,
|
| 1102 |
+
"step": 156
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.20096,
|
| 1106 |
+
"grad_norm": 0.3293153125716864,
|
| 1107 |
+
"learning_rate": 4.8530345387885004e-05,
|
| 1108 |
+
"loss": 0.5679,
|
| 1109 |
+
"step": 157
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.20224,
|
| 1113 |
+
"grad_norm": 0.3020107126026594,
|
| 1114 |
+
"learning_rate": 4.849231551964771e-05,
|
| 1115 |
+
"loss": 0.5892,
|
| 1116 |
+
"step": 158
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.20352,
|
| 1120 |
+
"grad_norm": 0.37859431920215386,
|
| 1121 |
+
"learning_rate": 4.845381516116748e-05,
|
| 1122 |
+
"loss": 0.5752,
|
| 1123 |
+
"step": 159
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.2048,
|
| 1127 |
+
"grad_norm": 0.3310159661704287,
|
| 1128 |
+
"learning_rate": 4.841484508350679e-05,
|
| 1129 |
+
"loss": 0.5746,
|
| 1130 |
+
"step": 160
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.20608,
|
| 1134 |
+
"grad_norm": 0.29255683283621947,
|
| 1135 |
+
"learning_rate": 4.837540606713538e-05,
|
| 1136 |
+
"loss": 0.5822,
|
| 1137 |
+
"step": 161
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.20736,
|
| 1141 |
+
"grad_norm": 0.35293526712063267,
|
| 1142 |
+
"learning_rate": 4.83354989019146e-05,
|
| 1143 |
+
"loss": 0.5872,
|
| 1144 |
+
"step": 162
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.20864,
|
| 1148 |
+
"grad_norm": 0.31406471157688987,
|
| 1149 |
+
"learning_rate": 4.829512438708163e-05,
|
| 1150 |
+
"loss": 0.5837,
|
| 1151 |
+
"step": 163
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.20992,
|
| 1155 |
+
"grad_norm": 0.3746042872826804,
|
| 1156 |
+
"learning_rate": 4.8254283331233464e-05,
|
| 1157 |
+
"loss": 0.5795,
|
| 1158 |
+
"step": 164
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.2112,
|
| 1162 |
+
"grad_norm": 0.36631879902112746,
|
| 1163 |
+
"learning_rate": 4.821297655231066e-05,
|
| 1164 |
+
"loss": 0.5917,
|
| 1165 |
+
"step": 165
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.21248,
|
| 1169 |
+
"grad_norm": 0.31527709590813463,
|
| 1170 |
+
"learning_rate": 4.817120487758104e-05,
|
| 1171 |
+
"loss": 0.583,
|
| 1172 |
+
"step": 166
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.21376,
|
| 1176 |
+
"grad_norm": 0.3311399262333245,
|
| 1177 |
+
"learning_rate": 4.812896914362309e-05,
|
| 1178 |
+
"loss": 0.5806,
|
| 1179 |
+
"step": 167
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.21504,
|
| 1183 |
+
"grad_norm": 0.2797000752887155,
|
| 1184 |
+
"learning_rate": 4.808627019630917e-05,
|
| 1185 |
+
"loss": 0.5818,
|
| 1186 |
+
"step": 168
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.21632,
|
| 1190 |
+
"grad_norm": 0.34099518073149715,
|
| 1191 |
+
"learning_rate": 4.804310889078861e-05,
|
| 1192 |
+
"loss": 0.5827,
|
| 1193 |
+
"step": 169
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.2176,
|
| 1197 |
+
"grad_norm": 0.2858567724381781,
|
| 1198 |
+
"learning_rate": 4.799948609147061e-05,
|
| 1199 |
+
"loss": 0.574,
|
| 1200 |
+
"step": 170
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.21888,
|
| 1204 |
+
"grad_norm": 0.3714847802025429,
|
| 1205 |
+
"learning_rate": 4.7955402672006854e-05,
|
| 1206 |
+
"loss": 0.568,
|
| 1207 |
+
"step": 171
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.22016,
|
| 1211 |
+
"grad_norm": 0.3629775316062746,
|
| 1212 |
+
"learning_rate": 4.791085951527408e-05,
|
| 1213 |
+
"loss": 0.5908,
|
| 1214 |
+
"step": 172
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.22144,
|
| 1218 |
+
"grad_norm": 0.3308553993376432,
|
| 1219 |
+
"learning_rate": 4.786585751335637e-05,
|
| 1220 |
+
"loss": 0.5822,
|
| 1221 |
+
"step": 173
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.22272,
|
| 1225 |
+
"grad_norm": 0.3583796747287126,
|
| 1226 |
+
"learning_rate": 4.782039756752727e-05,
|
| 1227 |
+
"loss": 0.5849,
|
| 1228 |
+
"step": 174
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.224,
|
| 1232 |
+
"grad_norm": 0.3200425393735649,
|
| 1233 |
+
"learning_rate": 4.777448058823179e-05,
|
| 1234 |
+
"loss": 0.5975,
|
| 1235 |
+
"step": 175
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.22528,
|
| 1239 |
+
"grad_norm": 0.292626050602806,
|
| 1240 |
+
"learning_rate": 4.77281074950681e-05,
|
| 1241 |
+
"loss": 0.5858,
|
| 1242 |
+
"step": 176
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.22656,
|
| 1246 |
+
"grad_norm": 0.35199899958158143,
|
| 1247 |
+
"learning_rate": 4.768127921676916e-05,
|
| 1248 |
+
"loss": 0.5956,
|
| 1249 |
+
"step": 177
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.22784,
|
| 1253 |
+
"grad_norm": 0.3049078150389153,
|
| 1254 |
+
"learning_rate": 4.763399669118414e-05,
|
| 1255 |
+
"loss": 0.5818,
|
| 1256 |
+
"step": 178
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.22912,
|
| 1260 |
+
"grad_norm": 0.3171759989327936,
|
| 1261 |
+
"learning_rate": 4.758626086525956e-05,
|
| 1262 |
+
"loss": 0.5788,
|
| 1263 |
+
"step": 179
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.2304,
|
| 1267 |
+
"grad_norm": 0.37616755806228974,
|
| 1268 |
+
"learning_rate": 4.753807269502041e-05,
|
| 1269 |
+
"loss": 0.5764,
|
| 1270 |
+
"step": 180
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.23168,
|
| 1274 |
+
"grad_norm": 0.3111683206800814,
|
| 1275 |
+
"learning_rate": 4.748943314555093e-05,
|
| 1276 |
+
"loss": 0.5904,
|
| 1277 |
+
"step": 181
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.23296,
|
| 1281 |
+
"grad_norm": 0.33324357602957505,
|
| 1282 |
+
"learning_rate": 4.744034319097535e-05,
|
| 1283 |
+
"loss": 0.5837,
|
| 1284 |
+
"step": 182
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.23424,
|
| 1288 |
+
"grad_norm": 0.3033759415187109,
|
| 1289 |
+
"learning_rate": 4.739080381443834e-05,
|
| 1290 |
+
"loss": 0.5872,
|
| 1291 |
+
"step": 183
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.23552,
|
| 1295 |
+
"grad_norm": 0.31221903432674414,
|
| 1296 |
+
"learning_rate": 4.734081600808531e-05,
|
| 1297 |
+
"loss": 0.5788,
|
| 1298 |
+
"step": 184
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.2368,
|
| 1302 |
+
"grad_norm": 0.3521014284242888,
|
| 1303 |
+
"learning_rate": 4.7290380773042575e-05,
|
| 1304 |
+
"loss": 0.5876,
|
| 1305 |
+
"step": 185
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.23808,
|
| 1309 |
+
"grad_norm": 0.310657917751192,
|
| 1310 |
+
"learning_rate": 4.723949911939728e-05,
|
| 1311 |
+
"loss": 0.5799,
|
| 1312 |
+
"step": 186
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.23936,
|
| 1316 |
+
"grad_norm": 0.3571637318567018,
|
| 1317 |
+
"learning_rate": 4.718817206617718e-05,
|
| 1318 |
+
"loss": 0.5768,
|
| 1319 |
+
"step": 187
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.24064,
|
| 1323 |
+
"grad_norm": 0.3019122458668047,
|
| 1324 |
+
"learning_rate": 4.713640064133025e-05,
|
| 1325 |
+
"loss": 0.569,
|
| 1326 |
+
"step": 188
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.24192,
|
| 1330 |
+
"grad_norm": 0.3867595871931151,
|
| 1331 |
+
"learning_rate": 4.7084185881704037e-05,
|
| 1332 |
+
"loss": 0.5807,
|
| 1333 |
+
"step": 189
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.2432,
|
| 1337 |
+
"grad_norm": 0.31887858233709804,
|
| 1338 |
+
"learning_rate": 4.7031528833024976e-05,
|
| 1339 |
+
"loss": 0.5853,
|
| 1340 |
+
"step": 190
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.24448,
|
| 1344 |
+
"grad_norm": 0.33241965371410676,
|
| 1345 |
+
"learning_rate": 4.697843054987737e-05,
|
| 1346 |
+
"loss": 0.5843,
|
| 1347 |
+
"step": 191
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.24576,
|
| 1351 |
+
"grad_norm": 0.3186271240585135,
|
| 1352 |
+
"learning_rate": 4.692489209568234e-05,
|
| 1353 |
+
"loss": 0.58,
|
| 1354 |
+
"step": 192
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.24704,
|
| 1358 |
+
"grad_norm": 0.27257060138401595,
|
| 1359 |
+
"learning_rate": 4.687091454267646e-05,
|
| 1360 |
+
"loss": 0.5782,
|
| 1361 |
+
"step": 193
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.24832,
|
| 1365 |
+
"grad_norm": 0.3068678257607851,
|
| 1366 |
+
"learning_rate": 4.681649897189036e-05,
|
| 1367 |
+
"loss": 0.5715,
|
| 1368 |
+
"step": 194
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.2496,
|
| 1372 |
+
"grad_norm": 0.2895553434845913,
|
| 1373 |
+
"learning_rate": 4.6761646473126985e-05,
|
| 1374 |
+
"loss": 0.5864,
|
| 1375 |
+
"step": 195
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.25088,
|
| 1379 |
+
"grad_norm": 0.3320177973067205,
|
| 1380 |
+
"learning_rate": 4.670635814493984e-05,
|
| 1381 |
+
"loss": 0.5895,
|
| 1382 |
+
"step": 196
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.25216,
|
| 1386 |
+
"grad_norm": 0.28847585573804785,
|
| 1387 |
+
"learning_rate": 4.665063509461097e-05,
|
| 1388 |
+
"loss": 0.5906,
|
| 1389 |
+
"step": 197
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.25344,
|
| 1393 |
+
"grad_norm": 0.36087996937741346,
|
| 1394 |
+
"learning_rate": 4.6594478438128757e-05,
|
| 1395 |
+
"loss": 0.5827,
|
| 1396 |
+
"step": 198
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.25472,
|
| 1400 |
+
"grad_norm": 0.2999930376743691,
|
| 1401 |
+
"learning_rate": 4.653788930016562e-05,
|
| 1402 |
+
"loss": 0.5661,
|
| 1403 |
+
"step": 199
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.256,
|
| 1407 |
+
"grad_norm": 0.3488298254178776,
|
| 1408 |
+
"learning_rate": 4.6480868814055424e-05,
|
| 1409 |
+
"loss": 0.5793,
|
| 1410 |
+
"step": 200
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.25728,
|
| 1414 |
+
"grad_norm": 0.3489687054350795,
|
| 1415 |
+
"learning_rate": 4.6423418121770855e-05,
|
| 1416 |
+
"loss": 0.5804,
|
| 1417 |
+
"step": 201
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.25856,
|
| 1421 |
+
"grad_norm": 0.32040065196445794,
|
| 1422 |
+
"learning_rate": 4.636553837390051e-05,
|
| 1423 |
+
"loss": 0.5885,
|
| 1424 |
+
"step": 202
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.25984,
|
| 1428 |
+
"grad_norm": 0.3211301436932052,
|
| 1429 |
+
"learning_rate": 4.630723072962584e-05,
|
| 1430 |
+
"loss": 0.5712,
|
| 1431 |
+
"step": 203
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.26112,
|
| 1435 |
+
"grad_norm": 0.359318175552222,
|
| 1436 |
+
"learning_rate": 4.6248496356697966e-05,
|
| 1437 |
+
"loss": 0.5708,
|
| 1438 |
+
"step": 204
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.2624,
|
| 1442 |
+
"grad_norm": 0.28862834349726524,
|
| 1443 |
+
"learning_rate": 4.618933643141428e-05,
|
| 1444 |
+
"loss": 0.5761,
|
| 1445 |
+
"step": 205
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.26368,
|
| 1449 |
+
"grad_norm": 0.37248048616519797,
|
| 1450 |
+
"learning_rate": 4.6129752138594874e-05,
|
| 1451 |
+
"loss": 0.5853,
|
| 1452 |
+
"step": 206
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.26496,
|
| 1456 |
+
"grad_norm": 0.26505735554799875,
|
| 1457 |
+
"learning_rate": 4.6069744671558835e-05,
|
| 1458 |
+
"loss": 0.5711,
|
| 1459 |
+
"step": 207
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.26624,
|
| 1463 |
+
"grad_norm": 0.4313554228926337,
|
| 1464 |
+
"learning_rate": 4.6009315232100324e-05,
|
| 1465 |
+
"loss": 0.5805,
|
| 1466 |
+
"step": 208
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.26752,
|
| 1470 |
+
"grad_norm": 0.25367772845462405,
|
| 1471 |
+
"learning_rate": 4.5948465030464536e-05,
|
| 1472 |
+
"loss": 0.5804,
|
| 1473 |
+
"step": 209
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.2688,
|
| 1477 |
+
"grad_norm": 0.4241960582103298,
|
| 1478 |
+
"learning_rate": 4.588719528532342e-05,
|
| 1479 |
+
"loss": 0.5781,
|
| 1480 |
+
"step": 210
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.27008,
|
| 1484 |
+
"grad_norm": 0.26959374978777695,
|
| 1485 |
+
"learning_rate": 4.58255072237513e-05,
|
| 1486 |
+
"loss": 0.5872,
|
| 1487 |
+
"step": 211
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.27136,
|
| 1491 |
+
"grad_norm": 0.32686893212462864,
|
| 1492 |
+
"learning_rate": 4.5763402081200294e-05,
|
| 1493 |
+
"loss": 0.5865,
|
| 1494 |
+
"step": 212
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.27264,
|
| 1498 |
+
"grad_norm": 0.23434165675561655,
|
| 1499 |
+
"learning_rate": 4.570088110147559e-05,
|
| 1500 |
+
"loss": 0.5605,
|
| 1501 |
+
"step": 213
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.27392,
|
| 1505 |
+
"grad_norm": 0.29221024109189686,
|
| 1506 |
+
"learning_rate": 4.56379455367105e-05,
|
| 1507 |
+
"loss": 0.5798,
|
| 1508 |
+
"step": 214
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.2752,
|
| 1512 |
+
"grad_norm": 0.24948843663318818,
|
| 1513 |
+
"learning_rate": 4.557459664734141e-05,
|
| 1514 |
+
"loss": 0.5647,
|
| 1515 |
+
"step": 215
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.27648,
|
| 1519 |
+
"grad_norm": 0.2697506211061829,
|
| 1520 |
+
"learning_rate": 4.551083570208252e-05,
|
| 1521 |
+
"loss": 0.574,
|
| 1522 |
+
"step": 216
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.27776,
|
| 1526 |
+
"grad_norm": 0.26982678855787623,
|
| 1527 |
+
"learning_rate": 4.544666397790043e-05,
|
| 1528 |
+
"loss": 0.5769,
|
| 1529 |
+
"step": 217
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.27904,
|
| 1533 |
+
"grad_norm": 0.26312816652400706,
|
| 1534 |
+
"learning_rate": 4.538208275998861e-05,
|
| 1535 |
+
"loss": 0.5794,
|
| 1536 |
+
"step": 218
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.28032,
|
| 1540 |
+
"grad_norm": 0.23961265146671684,
|
| 1541 |
+
"learning_rate": 4.531709334174161e-05,
|
| 1542 |
+
"loss": 0.5713,
|
| 1543 |
+
"step": 219
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.2816,
|
| 1547 |
+
"grad_norm": 0.2521656675623654,
|
| 1548 |
+
"learning_rate": 4.5251697024729165e-05,
|
| 1549 |
+
"loss": 0.5799,
|
| 1550 |
+
"step": 220
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.28288,
|
| 1554 |
+
"grad_norm": 0.2554557690482637,
|
| 1555 |
+
"learning_rate": 4.518589511867017e-05,
|
| 1556 |
+
"loss": 0.5612,
|
| 1557 |
+
"step": 221
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.28416,
|
| 1561 |
+
"grad_norm": 0.2575994504688328,
|
| 1562 |
+
"learning_rate": 4.511968894140639e-05,
|
| 1563 |
+
"loss": 0.5646,
|
| 1564 |
+
"step": 222
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.28544,
|
| 1568 |
+
"grad_norm": 0.24148588167926954,
|
| 1569 |
+
"learning_rate": 4.50530798188761e-05,
|
| 1570 |
+
"loss": 0.5752,
|
| 1571 |
+
"step": 223
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.28672,
|
| 1575 |
+
"grad_norm": 0.24795970441926052,
|
| 1576 |
+
"learning_rate": 4.498606908508754e-05,
|
| 1577 |
+
"loss": 0.5651,
|
| 1578 |
+
"step": 224
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.288,
|
| 1582 |
+
"grad_norm": 0.2594616590711217,
|
| 1583 |
+
"learning_rate": 4.491865808209215e-05,
|
| 1584 |
+
"loss": 0.5677,
|
| 1585 |
+
"step": 225
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.28928,
|
| 1589 |
+
"grad_norm": 0.25674541102342535,
|
| 1590 |
+
"learning_rate": 4.485084815995778e-05,
|
| 1591 |
+
"loss": 0.5612,
|
| 1592 |
+
"step": 226
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.29056,
|
| 1596 |
+
"grad_norm": 0.24985240413836018,
|
| 1597 |
+
"learning_rate": 4.478264067674155e-05,
|
| 1598 |
+
"loss": 0.5699,
|
| 1599 |
+
"step": 227
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.29184,
|
| 1603 |
+
"grad_norm": 0.2565786888027258,
|
| 1604 |
+
"learning_rate": 4.471403699846272e-05,
|
| 1605 |
+
"loss": 0.5649,
|
| 1606 |
+
"step": 228
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.29312,
|
| 1610 |
+
"grad_norm": 0.24372178698061564,
|
| 1611 |
+
"learning_rate": 4.4645038499075296e-05,
|
| 1612 |
+
"loss": 0.5727,
|
| 1613 |
+
"step": 229
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.2944,
|
| 1617 |
+
"grad_norm": 0.26957885657754116,
|
| 1618 |
+
"learning_rate": 4.457564656044056e-05,
|
| 1619 |
+
"loss": 0.5673,
|
| 1620 |
+
"step": 230
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.29568,
|
| 1624 |
+
"grad_norm": 0.2607171265747611,
|
| 1625 |
+
"learning_rate": 4.4505862572299315e-05,
|
| 1626 |
+
"loss": 0.5745,
|
| 1627 |
+
"step": 231
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.29696,
|
| 1631 |
+
"grad_norm": 0.24863876648291447,
|
| 1632 |
+
"learning_rate": 4.443568793224415e-05,
|
| 1633 |
+
"loss": 0.5681,
|
| 1634 |
+
"step": 232
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.29824,
|
| 1638 |
+
"grad_norm": 0.263062932590105,
|
| 1639 |
+
"learning_rate": 4.436512404569136e-05,
|
| 1640 |
+
"loss": 0.5644,
|
| 1641 |
+
"step": 233
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.29952,
|
| 1645 |
+
"grad_norm": 0.2414635376635183,
|
| 1646 |
+
"learning_rate": 4.429417232585288e-05,
|
| 1647 |
+
"loss": 0.5701,
|
| 1648 |
+
"step": 234
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.3008,
|
| 1652 |
+
"grad_norm": 0.25798244731348213,
|
| 1653 |
+
"learning_rate": 4.422283419370789e-05,
|
| 1654 |
+
"loss": 0.5749,
|
| 1655 |
+
"step": 235
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.30208,
|
| 1659 |
+
"grad_norm": 0.2705236848186794,
|
| 1660 |
+
"learning_rate": 4.415111107797445e-05,
|
| 1661 |
+
"loss": 0.5722,
|
| 1662 |
+
"step": 236
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.30336,
|
| 1666 |
+
"grad_norm": 0.2532310162226822,
|
| 1667 |
+
"learning_rate": 4.407900441508084e-05,
|
| 1668 |
+
"loss": 0.5613,
|
| 1669 |
+
"step": 237
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.30464,
|
| 1673 |
+
"grad_norm": 0.27015507001673206,
|
| 1674 |
+
"learning_rate": 4.400651564913676e-05,
|
| 1675 |
+
"loss": 0.5764,
|
| 1676 |
+
"step": 238
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.30592,
|
| 1680 |
+
"grad_norm": 0.2600356410106743,
|
| 1681 |
+
"learning_rate": 4.3933646231904504e-05,
|
| 1682 |
+
"loss": 0.5621,
|
| 1683 |
+
"step": 239
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.3072,
|
| 1687 |
+
"grad_norm": 0.2947154143151425,
|
| 1688 |
+
"learning_rate": 4.3860397622769756e-05,
|
| 1689 |
+
"loss": 0.5776,
|
| 1690 |
+
"step": 240
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.30848,
|
| 1694 |
+
"grad_norm": 0.278006374370773,
|
| 1695 |
+
"learning_rate": 4.37867712887125e-05,
|
| 1696 |
+
"loss": 0.5779,
|
| 1697 |
+
"step": 241
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.30976,
|
| 1701 |
+
"grad_norm": 0.2737057808752138,
|
| 1702 |
+
"learning_rate": 4.371276870427753e-05,
|
| 1703 |
+
"loss": 0.5691,
|
| 1704 |
+
"step": 242
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.31104,
|
| 1708 |
+
"grad_norm": 0.2779579791922396,
|
| 1709 |
+
"learning_rate": 4.363839135154497e-05,
|
| 1710 |
+
"loss": 0.5561,
|
| 1711 |
+
"step": 243
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.31232,
|
| 1715 |
+
"grad_norm": 0.2900125824723869,
|
| 1716 |
+
"learning_rate": 4.356364072010059e-05,
|
| 1717 |
+
"loss": 0.5862,
|
| 1718 |
+
"step": 244
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.3136,
|
| 1722 |
+
"grad_norm": 0.2864520071161508,
|
| 1723 |
+
"learning_rate": 4.348851830700593e-05,
|
| 1724 |
+
"loss": 0.5622,
|
| 1725 |
+
"step": 245
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.31488,
|
| 1729 |
+
"grad_norm": 0.26378798903912654,
|
| 1730 |
+
"learning_rate": 4.3413025616768424e-05,
|
| 1731 |
+
"loss": 0.5709,
|
| 1732 |
+
"step": 246
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.31616,
|
| 1736 |
+
"grad_norm": 0.3060187732302713,
|
| 1737 |
+
"learning_rate": 4.333716416131115e-05,
|
| 1738 |
+
"loss": 0.5669,
|
| 1739 |
+
"step": 247
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.31744,
|
| 1743 |
+
"grad_norm": 0.29432171614697805,
|
| 1744 |
+
"learning_rate": 4.3260935459942584e-05,
|
| 1745 |
+
"loss": 0.5682,
|
| 1746 |
+
"step": 248
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.31872,
|
| 1750 |
+
"grad_norm": 0.26017905775079875,
|
| 1751 |
+
"learning_rate": 4.318434103932622e-05,
|
| 1752 |
+
"loss": 0.5885,
|
| 1753 |
+
"step": 249
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.32,
|
| 1757 |
+
"grad_norm": 0.31505146670137313,
|
| 1758 |
+
"learning_rate": 4.310738243344996e-05,
|
| 1759 |
+
"loss": 0.5761,
|
| 1760 |
+
"step": 250
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.32128,
|
| 1764 |
+
"grad_norm": 0.28372810809515076,
|
| 1765 |
+
"learning_rate": 4.303006118359537e-05,
|
| 1766 |
+
"loss": 0.5592,
|
| 1767 |
+
"step": 251
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.32256,
|
| 1771 |
+
"grad_norm": 0.2651991449577939,
|
| 1772 |
+
"learning_rate": 4.295237883830685e-05,
|
| 1773 |
+
"loss": 0.5697,
|
| 1774 |
+
"step": 252
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.32384,
|
| 1778 |
+
"grad_norm": 0.24913000845932673,
|
| 1779 |
+
"learning_rate": 4.2874336953360616e-05,
|
| 1780 |
+
"loss": 0.5776,
|
| 1781 |
+
"step": 253
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.32512,
|
| 1785 |
+
"grad_norm": 0.2430027683011546,
|
| 1786 |
+
"learning_rate": 4.2795937091733515e-05,
|
| 1787 |
+
"loss": 0.5677,
|
| 1788 |
+
"step": 254
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.3264,
|
| 1792 |
+
"grad_norm": 0.2568626552616791,
|
| 1793 |
+
"learning_rate": 4.271718082357175e-05,
|
| 1794 |
+
"loss": 0.582,
|
| 1795 |
+
"step": 255
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.32768,
|
| 1799 |
+
"grad_norm": 0.23741828351399039,
|
| 1800 |
+
"learning_rate": 4.2638069726159424e-05,
|
| 1801 |
+
"loss": 0.5609,
|
| 1802 |
+
"step": 256
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.32896,
|
| 1806 |
+
"grad_norm": 0.23355776394833105,
|
| 1807 |
+
"learning_rate": 4.255860538388694e-05,
|
| 1808 |
+
"loss": 0.572,
|
| 1809 |
+
"step": 257
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.33024,
|
| 1813 |
+
"grad_norm": 0.28500006430456526,
|
| 1814 |
+
"learning_rate": 4.247878938821929e-05,
|
| 1815 |
+
"loss": 0.5783,
|
| 1816 |
+
"step": 258
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.33152,
|
| 1820 |
+
"grad_norm": 0.23832612238075213,
|
| 1821 |
+
"learning_rate": 4.2398623337664176e-05,
|
| 1822 |
+
"loss": 0.5591,
|
| 1823 |
+
"step": 259
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.3328,
|
| 1827 |
+
"grad_norm": 0.3027829669410571,
|
| 1828 |
+
"learning_rate": 4.231810883773999e-05,
|
| 1829 |
+
"loss": 0.5723,
|
| 1830 |
+
"step": 260
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.33408,
|
| 1834 |
+
"grad_norm": 0.24742812416416485,
|
| 1835 |
+
"learning_rate": 4.223724750094366e-05,
|
| 1836 |
+
"loss": 0.562,
|
| 1837 |
+
"step": 261
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.33536,
|
| 1841 |
+
"grad_norm": 0.281426189909417,
|
| 1842 |
+
"learning_rate": 4.215604094671835e-05,
|
| 1843 |
+
"loss": 0.5711,
|
| 1844 |
+
"step": 262
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.33664,
|
| 1848 |
+
"grad_norm": 0.3464883424573956,
|
| 1849 |
+
"learning_rate": 4.207449080142104e-05,
|
| 1850 |
+
"loss": 0.56,
|
| 1851 |
+
"step": 263
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.33792,
|
| 1855 |
+
"grad_norm": 0.23536130124235047,
|
| 1856 |
+
"learning_rate": 4.199259869828998e-05,
|
| 1857 |
+
"loss": 0.5714,
|
| 1858 |
+
"step": 264
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.3392,
|
| 1862 |
+
"grad_norm": 0.32901239458674364,
|
| 1863 |
+
"learning_rate": 4.191036627741191e-05,
|
| 1864 |
+
"loss": 0.5656,
|
| 1865 |
+
"step": 265
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.34048,
|
| 1869 |
+
"grad_norm": 0.2513197034442368,
|
| 1870 |
+
"learning_rate": 4.182779518568926e-05,
|
| 1871 |
+
"loss": 0.5549,
|
| 1872 |
+
"step": 266
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.34176,
|
| 1876 |
+
"grad_norm": 0.2956698427587275,
|
| 1877 |
+
"learning_rate": 4.174488707680717e-05,
|
| 1878 |
+
"loss": 0.565,
|
| 1879 |
+
"step": 267
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.34304,
|
| 1883 |
+
"grad_norm": 0.2767183631207283,
|
| 1884 |
+
"learning_rate": 4.1661643611200366e-05,
|
| 1885 |
+
"loss": 0.5751,
|
| 1886 |
+
"step": 268
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.34432,
|
| 1890 |
+
"grad_norm": 0.24523313623813994,
|
| 1891 |
+
"learning_rate": 4.157806645601988e-05,
|
| 1892 |
+
"loss": 0.5742,
|
| 1893 |
+
"step": 269
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.3456,
|
| 1897 |
+
"grad_norm": 0.2908460538863046,
|
| 1898 |
+
"learning_rate": 4.149415728509971e-05,
|
| 1899 |
+
"loss": 0.5593,
|
| 1900 |
+
"step": 270
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.34688,
|
| 1904 |
+
"grad_norm": 0.4424923362280487,
|
| 1905 |
+
"learning_rate": 4.140991777892324e-05,
|
| 1906 |
+
"loss": 0.5814,
|
| 1907 |
+
"step": 271
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.34816,
|
| 1911 |
+
"grad_norm": 0.2575686169084144,
|
| 1912 |
+
"learning_rate": 4.132534962458962e-05,
|
| 1913 |
+
"loss": 0.5678,
|
| 1914 |
+
"step": 272
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.34944,
|
| 1918 |
+
"grad_norm": 0.2629805233702073,
|
| 1919 |
+
"learning_rate": 4.124045451578001e-05,
|
| 1920 |
+
"loss": 0.5629,
|
| 1921 |
+
"step": 273
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.35072,
|
| 1925 |
+
"grad_norm": 0.22898476571255702,
|
| 1926 |
+
"learning_rate": 4.115523415272358e-05,
|
| 1927 |
+
"loss": 0.569,
|
| 1928 |
+
"step": 274
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.352,
|
| 1932 |
+
"grad_norm": 0.2443223155892946,
|
| 1933 |
+
"learning_rate": 4.1069690242163484e-05,
|
| 1934 |
+
"loss": 0.5624,
|
| 1935 |
+
"step": 275
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.35328,
|
| 1939 |
+
"grad_norm": 0.24146954716830435,
|
| 1940 |
+
"learning_rate": 4.0983824497322755e-05,
|
| 1941 |
+
"loss": 0.5783,
|
| 1942 |
+
"step": 276
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.35456,
|
| 1946 |
+
"grad_norm": 0.23532353314194146,
|
| 1947 |
+
"learning_rate": 4.0897638637869874e-05,
|
| 1948 |
+
"loss": 0.5846,
|
| 1949 |
+
"step": 277
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.35584,
|
| 1953 |
+
"grad_norm": 0.21666728547483338,
|
| 1954 |
+
"learning_rate": 4.0811134389884433e-05,
|
| 1955 |
+
"loss": 0.5788,
|
| 1956 |
+
"step": 278
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.35712,
|
| 1960 |
+
"grad_norm": 0.2296647805474214,
|
| 1961 |
+
"learning_rate": 4.07243134858225e-05,
|
| 1962 |
+
"loss": 0.5679,
|
| 1963 |
+
"step": 279
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.3584,
|
| 1967 |
+
"grad_norm": 0.23340342966367908,
|
| 1968 |
+
"learning_rate": 4.063717766448194e-05,
|
| 1969 |
+
"loss": 0.5645,
|
| 1970 |
+
"step": 280
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.35968,
|
| 1974 |
+
"grad_norm": 0.23757239204061076,
|
| 1975 |
+
"learning_rate": 4.05497286709676e-05,
|
| 1976 |
+
"loss": 0.5745,
|
| 1977 |
+
"step": 281
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.36096,
|
| 1981 |
+
"grad_norm": 0.24284616004111326,
|
| 1982 |
+
"learning_rate": 4.0461968256656376e-05,
|
| 1983 |
+
"loss": 0.5659,
|
| 1984 |
+
"step": 282
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.36224,
|
| 1988 |
+
"grad_norm": 0.2317609526061923,
|
| 1989 |
+
"learning_rate": 4.037389817916208e-05,
|
| 1990 |
+
"loss": 0.5657,
|
| 1991 |
+
"step": 283
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.36352,
|
| 1995 |
+
"grad_norm": 0.2779219267379158,
|
| 1996 |
+
"learning_rate": 4.028552020230031e-05,
|
| 1997 |
+
"loss": 0.5612,
|
| 1998 |
+
"step": 284
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.3648,
|
| 2002 |
+
"grad_norm": 0.2568685752720875,
|
| 2003 |
+
"learning_rate": 4.019683609605305e-05,
|
| 2004 |
+
"loss": 0.5618,
|
| 2005 |
+
"step": 285
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.36608,
|
| 2009 |
+
"grad_norm": 0.24357977645804754,
|
| 2010 |
+
"learning_rate": 4.010784763653331e-05,
|
| 2011 |
+
"loss": 0.5532,
|
| 2012 |
+
"step": 286
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.36736,
|
| 2016 |
+
"grad_norm": 0.2854087705449231,
|
| 2017 |
+
"learning_rate": 4.001855660594948e-05,
|
| 2018 |
+
"loss": 0.5598,
|
| 2019 |
+
"step": 287
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.36864,
|
| 2023 |
+
"grad_norm": 0.27153778848296095,
|
| 2024 |
+
"learning_rate": 3.9928964792569655e-05,
|
| 2025 |
+
"loss": 0.5661,
|
| 2026 |
+
"step": 288
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.36992,
|
| 2030 |
+
"grad_norm": 0.3013570035330345,
|
| 2031 |
+
"learning_rate": 3.983907399068587e-05,
|
| 2032 |
+
"loss": 0.576,
|
| 2033 |
+
"step": 289
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.3712,
|
| 2037 |
+
"grad_norm": 0.23579225525517492,
|
| 2038 |
+
"learning_rate": 3.974888600057808e-05,
|
| 2039 |
+
"loss": 0.5685,
|
| 2040 |
+
"step": 290
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.37248,
|
| 2044 |
+
"grad_norm": 0.29344672148557993,
|
| 2045 |
+
"learning_rate": 3.965840262847818e-05,
|
| 2046 |
+
"loss": 0.5652,
|
| 2047 |
+
"step": 291
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.37376,
|
| 2051 |
+
"grad_norm": 0.2771114973156099,
|
| 2052 |
+
"learning_rate": 3.956762568653378e-05,
|
| 2053 |
+
"loss": 0.5615,
|
| 2054 |
+
"step": 292
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.37504,
|
| 2058 |
+
"grad_norm": 0.2475727777258234,
|
| 2059 |
+
"learning_rate": 3.947655699277197e-05,
|
| 2060 |
+
"loss": 0.5622,
|
| 2061 |
+
"step": 293
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.37632,
|
| 2065 |
+
"grad_norm": 0.3283313482469374,
|
| 2066 |
+
"learning_rate": 3.9385198371062845e-05,
|
| 2067 |
+
"loss": 0.5688,
|
| 2068 |
+
"step": 294
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.3776,
|
| 2072 |
+
"grad_norm": 0.2567128249078116,
|
| 2073 |
+
"learning_rate": 3.929355165108299e-05,
|
| 2074 |
+
"loss": 0.5685,
|
| 2075 |
+
"step": 295
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.37888,
|
| 2079 |
+
"grad_norm": 0.33868790476694477,
|
| 2080 |
+
"learning_rate": 3.920161866827889e-05,
|
| 2081 |
+
"loss": 0.5603,
|
| 2082 |
+
"step": 296
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.38016,
|
| 2086 |
+
"grad_norm": 0.23159577247265578,
|
| 2087 |
+
"learning_rate": 3.910940126383013e-05,
|
| 2088 |
+
"loss": 0.5614,
|
| 2089 |
+
"step": 297
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.38144,
|
| 2093 |
+
"grad_norm": 0.261651084842406,
|
| 2094 |
+
"learning_rate": 3.9016901284612474e-05,
|
| 2095 |
+
"loss": 0.5634,
|
| 2096 |
+
"step": 298
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.38272,
|
| 2100 |
+
"grad_norm": 0.25918417083311507,
|
| 2101 |
+
"learning_rate": 3.8924120583160985e-05,
|
| 2102 |
+
"loss": 0.564,
|
| 2103 |
+
"step": 299
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.384,
|
| 2107 |
+
"grad_norm": 0.24530554345595257,
|
| 2108 |
+
"learning_rate": 3.883106101763285e-05,
|
| 2109 |
+
"loss": 0.5751,
|
| 2110 |
+
"step": 300
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.38528,
|
| 2114 |
+
"grad_norm": 0.25269762220161,
|
| 2115 |
+
"learning_rate": 3.873772445177015e-05,
|
| 2116 |
+
"loss": 0.5666,
|
| 2117 |
+
"step": 301
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.38656,
|
| 2121 |
+
"grad_norm": 0.2686509936643206,
|
| 2122 |
+
"learning_rate": 3.8644112754862614e-05,
|
| 2123 |
+
"loss": 0.5787,
|
| 2124 |
+
"step": 302
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.38784,
|
| 2128 |
+
"grad_norm": 0.2306788708136305,
|
| 2129 |
+
"learning_rate": 3.85502278017101e-05,
|
| 2130 |
+
"loss": 0.5639,
|
| 2131 |
+
"step": 303
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.38912,
|
| 2135 |
+
"grad_norm": 0.25921338779127046,
|
| 2136 |
+
"learning_rate": 3.84560714725851e-05,
|
| 2137 |
+
"loss": 0.5611,
|
| 2138 |
+
"step": 304
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.3904,
|
| 2142 |
+
"grad_norm": 0.2875441041443726,
|
| 2143 |
+
"learning_rate": 3.8361645653195026e-05,
|
| 2144 |
+
"loss": 0.5641,
|
| 2145 |
+
"step": 305
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.39168,
|
| 2149 |
+
"grad_norm": 0.24679004673798802,
|
| 2150 |
+
"learning_rate": 3.8266952234644545e-05,
|
| 2151 |
+
"loss": 0.5696,
|
| 2152 |
+
"step": 306
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.39296,
|
| 2156 |
+
"grad_norm": 0.32137055545965876,
|
| 2157 |
+
"learning_rate": 3.817199311339759e-05,
|
| 2158 |
+
"loss": 0.5642,
|
| 2159 |
+
"step": 307
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.39424,
|
| 2163 |
+
"grad_norm": 0.25511061251478745,
|
| 2164 |
+
"learning_rate": 3.807677019123944e-05,
|
| 2165 |
+
"loss": 0.5535,
|
| 2166 |
+
"step": 308
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.39552,
|
| 2170 |
+
"grad_norm": 0.28918335928655264,
|
| 2171 |
+
"learning_rate": 3.798128537523865e-05,
|
| 2172 |
+
"loss": 0.5628,
|
| 2173 |
+
"step": 309
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.3968,
|
| 2177 |
+
"grad_norm": 0.24654149918381577,
|
| 2178 |
+
"learning_rate": 3.7885540577708804e-05,
|
| 2179 |
+
"loss": 0.5693,
|
| 2180 |
+
"step": 310
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.39808,
|
| 2184 |
+
"grad_norm": 0.29445546859209903,
|
| 2185 |
+
"learning_rate": 3.7789537716170256e-05,
|
| 2186 |
+
"loss": 0.5603,
|
| 2187 |
+
"step": 311
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.39936,
|
| 2191 |
+
"grad_norm": 0.2546352881022481,
|
| 2192 |
+
"learning_rate": 3.76932787133117e-05,
|
| 2193 |
+
"loss": 0.5605,
|
| 2194 |
+
"step": 312
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.40064,
|
| 2198 |
+
"grad_norm": 0.2694209869990075,
|
| 2199 |
+
"learning_rate": 3.759676549695168e-05,
|
| 2200 |
+
"loss": 0.5631,
|
| 2201 |
+
"step": 313
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.40192,
|
| 2205 |
+
"grad_norm": 0.25374147038521105,
|
| 2206 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 2207 |
+
"loss": 0.5737,
|
| 2208 |
+
"step": 314
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.4032,
|
| 2212 |
+
"grad_norm": 0.2692940305682822,
|
| 2213 |
+
"learning_rate": 3.740298416041898e-05,
|
| 2214 |
+
"loss": 0.5592,
|
| 2215 |
+
"step": 315
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.40448,
|
| 2219 |
+
"grad_norm": 0.27292894144562746,
|
| 2220 |
+
"learning_rate": 3.730571992118462e-05,
|
| 2221 |
+
"loss": 0.56,
|
| 2222 |
+
"step": 316
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.40576,
|
| 2226 |
+
"grad_norm": 0.2831296008226712,
|
| 2227 |
+
"learning_rate": 3.720820923024778e-05,
|
| 2228 |
+
"loss": 0.556,
|
| 2229 |
+
"step": 317
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.40704,
|
| 2233 |
+
"grad_norm": 0.24842153386433,
|
| 2234 |
+
"learning_rate": 3.711045404049507e-05,
|
| 2235 |
+
"loss": 0.5615,
|
| 2236 |
+
"step": 318
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.40832,
|
| 2240 |
+
"grad_norm": 0.25137316651759817,
|
| 2241 |
+
"learning_rate": 3.701245630970979e-05,
|
| 2242 |
+
"loss": 0.5519,
|
| 2243 |
+
"step": 319
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.4096,
|
| 2247 |
+
"grad_norm": 0.256842976111285,
|
| 2248 |
+
"learning_rate": 3.69142180005327e-05,
|
| 2249 |
+
"loss": 0.559,
|
| 2250 |
+
"step": 320
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.41088,
|
| 2254 |
+
"grad_norm": 0.23856265603702867,
|
| 2255 |
+
"learning_rate": 3.681574108042274e-05,
|
| 2256 |
+
"loss": 0.5513,
|
| 2257 |
+
"step": 321
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.41216,
|
| 2261 |
+
"grad_norm": 0.2636505871798519,
|
| 2262 |
+
"learning_rate": 3.6717027521617595e-05,
|
| 2263 |
+
"loss": 0.5553,
|
| 2264 |
+
"step": 322
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.41344,
|
| 2268 |
+
"grad_norm": 0.23126506975873334,
|
| 2269 |
+
"learning_rate": 3.6618079301094216e-05,
|
| 2270 |
+
"loss": 0.5535,
|
| 2271 |
+
"step": 323
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.41472,
|
| 2275 |
+
"grad_norm": 0.2765627545708339,
|
| 2276 |
+
"learning_rate": 3.6518898400529214e-05,
|
| 2277 |
+
"loss": 0.5727,
|
| 2278 |
+
"step": 324
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.416,
|
| 2282 |
+
"grad_norm": 0.25065312946741247,
|
| 2283 |
+
"learning_rate": 3.6419486806259194e-05,
|
| 2284 |
+
"loss": 0.5642,
|
| 2285 |
+
"step": 325
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.41728,
|
| 2289 |
+
"grad_norm": 0.24801934318992586,
|
| 2290 |
+
"learning_rate": 3.631984650924094e-05,
|
| 2291 |
+
"loss": 0.5489,
|
| 2292 |
+
"step": 326
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.41856,
|
| 2296 |
+
"grad_norm": 0.2505101313201257,
|
| 2297 |
+
"learning_rate": 3.621997950501156e-05,
|
| 2298 |
+
"loss": 0.5575,
|
| 2299 |
+
"step": 327
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.41984,
|
| 2303 |
+
"grad_norm": 0.22423685749771013,
|
| 2304 |
+
"learning_rate": 3.611988779364853e-05,
|
| 2305 |
+
"loss": 0.5468,
|
| 2306 |
+
"step": 328
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.42112,
|
| 2310 |
+
"grad_norm": 0.2613413831880175,
|
| 2311 |
+
"learning_rate": 3.6019573379729643e-05,
|
| 2312 |
+
"loss": 0.5742,
|
| 2313 |
+
"step": 329
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.4224,
|
| 2317 |
+
"grad_norm": 0.23270975616366224,
|
| 2318 |
+
"learning_rate": 3.591903827229282e-05,
|
| 2319 |
+
"loss": 0.567,
|
| 2320 |
+
"step": 330
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.42368,
|
| 2324 |
+
"grad_norm": 0.22747265941762546,
|
| 2325 |
+
"learning_rate": 3.5818284484795904e-05,
|
| 2326 |
+
"loss": 0.5521,
|
| 2327 |
+
"step": 331
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.42496,
|
| 2331 |
+
"grad_norm": 0.21305915578050127,
|
| 2332 |
+
"learning_rate": 3.5717314035076355e-05,
|
| 2333 |
+
"loss": 0.5507,
|
| 2334 |
+
"step": 332
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.42624,
|
| 2338 |
+
"grad_norm": 0.22564541882328742,
|
| 2339 |
+
"learning_rate": 3.56161289453108e-05,
|
| 2340 |
+
"loss": 0.5612,
|
| 2341 |
+
"step": 333
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.42752,
|
| 2345 |
+
"grad_norm": 0.24927047862972154,
|
| 2346 |
+
"learning_rate": 3.5514731241974544e-05,
|
| 2347 |
+
"loss": 0.5511,
|
| 2348 |
+
"step": 334
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.4288,
|
| 2352 |
+
"grad_norm": 0.23391865199996986,
|
| 2353 |
+
"learning_rate": 3.5413122955801005e-05,
|
| 2354 |
+
"loss": 0.5638,
|
| 2355 |
+
"step": 335
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.43008,
|
| 2359 |
+
"grad_norm": 0.2538820004383474,
|
| 2360 |
+
"learning_rate": 3.5311306121741015e-05,
|
| 2361 |
+
"loss": 0.5629,
|
| 2362 |
+
"step": 336
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.43136,
|
| 2366 |
+
"grad_norm": 0.23071027016549392,
|
| 2367 |
+
"learning_rate": 3.5209282778922106e-05,
|
| 2368 |
+
"loss": 0.5524,
|
| 2369 |
+
"step": 337
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.43264,
|
| 2373 |
+
"grad_norm": 0.2538902369228176,
|
| 2374 |
+
"learning_rate": 3.510705497060762e-05,
|
| 2375 |
+
"loss": 0.5589,
|
| 2376 |
+
"step": 338
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.43392,
|
| 2380 |
+
"grad_norm": 0.2396787654237747,
|
| 2381 |
+
"learning_rate": 3.500462474415584e-05,
|
| 2382 |
+
"loss": 0.5583,
|
| 2383 |
+
"step": 339
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.4352,
|
| 2387 |
+
"grad_norm": 0.24498876339943568,
|
| 2388 |
+
"learning_rate": 3.490199415097892e-05,
|
| 2389 |
+
"loss": 0.5586,
|
| 2390 |
+
"step": 340
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.43648,
|
| 2394 |
+
"grad_norm": 0.22429707447384664,
|
| 2395 |
+
"learning_rate": 3.479916524650188e-05,
|
| 2396 |
+
"loss": 0.5591,
|
| 2397 |
+
"step": 341
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.43776,
|
| 2401 |
+
"grad_norm": 0.2488615974495946,
|
| 2402 |
+
"learning_rate": 3.4696140090121376e-05,
|
| 2403 |
+
"loss": 0.5716,
|
| 2404 |
+
"step": 342
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.43904,
|
| 2408 |
+
"grad_norm": 0.24601068839661924,
|
| 2409 |
+
"learning_rate": 3.459292074516449e-05,
|
| 2410 |
+
"loss": 0.572,
|
| 2411 |
+
"step": 343
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.44032,
|
| 2415 |
+
"grad_norm": 0.24772654051656678,
|
| 2416 |
+
"learning_rate": 3.4489509278847414e-05,
|
| 2417 |
+
"loss": 0.5627,
|
| 2418 |
+
"step": 344
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.4416,
|
| 2422 |
+
"grad_norm": 0.24320980786790716,
|
| 2423 |
+
"learning_rate": 3.4385907762234e-05,
|
| 2424 |
+
"loss": 0.556,
|
| 2425 |
+
"step": 345
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.44288,
|
| 2429 |
+
"grad_norm": 0.2899880667030433,
|
| 2430 |
+
"learning_rate": 3.428211827019434e-05,
|
| 2431 |
+
"loss": 0.5625,
|
| 2432 |
+
"step": 346
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.44416,
|
| 2436 |
+
"grad_norm": 0.24951619342805165,
|
| 2437 |
+
"learning_rate": 3.417814288136319e-05,
|
| 2438 |
+
"loss": 0.5606,
|
| 2439 |
+
"step": 347
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.44544,
|
| 2443 |
+
"grad_norm": 0.29645948538693856,
|
| 2444 |
+
"learning_rate": 3.407398367809832e-05,
|
| 2445 |
+
"loss": 0.5696,
|
| 2446 |
+
"step": 348
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.44672,
|
| 2450 |
+
"grad_norm": 0.2628573817303425,
|
| 2451 |
+
"learning_rate": 3.3969642746438836e-05,
|
| 2452 |
+
"loss": 0.5548,
|
| 2453 |
+
"step": 349
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.448,
|
| 2457 |
+
"grad_norm": 0.25156314937857505,
|
| 2458 |
+
"learning_rate": 3.386512217606339e-05,
|
| 2459 |
+
"loss": 0.5584,
|
| 2460 |
+
"step": 350
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.44928,
|
| 2464 |
+
"grad_norm": 0.24416044960818195,
|
| 2465 |
+
"learning_rate": 3.3760424060248344e-05,
|
| 2466 |
+
"loss": 0.5501,
|
| 2467 |
+
"step": 351
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.45056,
|
| 2471 |
+
"grad_norm": 0.2508322157210064,
|
| 2472 |
+
"learning_rate": 3.365555049582582e-05,
|
| 2473 |
+
"loss": 0.5575,
|
| 2474 |
+
"step": 352
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.45184,
|
| 2478 |
+
"grad_norm": 0.23612772269109591,
|
| 2479 |
+
"learning_rate": 3.355050358314172e-05,
|
| 2480 |
+
"loss": 0.5682,
|
| 2481 |
+
"step": 353
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.45312,
|
| 2485 |
+
"grad_norm": 0.24541798896943803,
|
| 2486 |
+
"learning_rate": 3.3445285426013685e-05,
|
| 2487 |
+
"loss": 0.551,
|
| 2488 |
+
"step": 354
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.4544,
|
| 2492 |
+
"grad_norm": 0.2234016182436274,
|
| 2493 |
+
"learning_rate": 3.3339898131688914e-05,
|
| 2494 |
+
"loss": 0.5591,
|
| 2495 |
+
"step": 355
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.45568,
|
| 2499 |
+
"grad_norm": 0.24455388410655837,
|
| 2500 |
+
"learning_rate": 3.323434381080199e-05,
|
| 2501 |
+
"loss": 0.5542,
|
| 2502 |
+
"step": 356
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.45696,
|
| 2506 |
+
"grad_norm": 0.2447461284103842,
|
| 2507 |
+
"learning_rate": 3.312862457733263e-05,
|
| 2508 |
+
"loss": 0.5689,
|
| 2509 |
+
"step": 357
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.45824,
|
| 2513 |
+
"grad_norm": 0.2611063437561967,
|
| 2514 |
+
"learning_rate": 3.302274254856329e-05,
|
| 2515 |
+
"loss": 0.5564,
|
| 2516 |
+
"step": 358
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.45952,
|
| 2520 |
+
"grad_norm": 0.2609646712338443,
|
| 2521 |
+
"learning_rate": 3.2916699845036816e-05,
|
| 2522 |
+
"loss": 0.5609,
|
| 2523 |
+
"step": 359
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.4608,
|
| 2527 |
+
"grad_norm": 0.2668094661495938,
|
| 2528 |
+
"learning_rate": 3.281049859051394e-05,
|
| 2529 |
+
"loss": 0.55,
|
| 2530 |
+
"step": 360
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.46208,
|
| 2534 |
+
"grad_norm": 0.23869480950176758,
|
| 2535 |
+
"learning_rate": 3.270414091193077e-05,
|
| 2536 |
+
"loss": 0.5613,
|
| 2537 |
+
"step": 361
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.46336,
|
| 2541 |
+
"grad_norm": 0.24021036759397824,
|
| 2542 |
+
"learning_rate": 3.2597628939356175e-05,
|
| 2543 |
+
"loss": 0.5431,
|
| 2544 |
+
"step": 362
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.46464,
|
| 2548 |
+
"grad_norm": 0.22964918725760114,
|
| 2549 |
+
"learning_rate": 3.2490964805949145e-05,
|
| 2550 |
+
"loss": 0.5568,
|
| 2551 |
+
"step": 363
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.46592,
|
| 2555 |
+
"grad_norm": 0.22304779121360874,
|
| 2556 |
+
"learning_rate": 3.238415064791603e-05,
|
| 2557 |
+
"loss": 0.561,
|
| 2558 |
+
"step": 364
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.4672,
|
| 2562 |
+
"grad_norm": 0.23921210910709523,
|
| 2563 |
+
"learning_rate": 3.227718860446782e-05,
|
| 2564 |
+
"loss": 0.5457,
|
| 2565 |
+
"step": 365
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.46848,
|
| 2569 |
+
"grad_norm": 0.2209452744046842,
|
| 2570 |
+
"learning_rate": 3.217008081777726e-05,
|
| 2571 |
+
"loss": 0.5543,
|
| 2572 |
+
"step": 366
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.46976,
|
| 2576 |
+
"grad_norm": 0.24033224570485728,
|
| 2577 |
+
"learning_rate": 3.206282943293593e-05,
|
| 2578 |
+
"loss": 0.5491,
|
| 2579 |
+
"step": 367
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.47104,
|
| 2583 |
+
"grad_norm": 0.23799116455210376,
|
| 2584 |
+
"learning_rate": 3.195543659791132e-05,
|
| 2585 |
+
"loss": 0.5473,
|
| 2586 |
+
"step": 368
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.47232,
|
| 2590 |
+
"grad_norm": 0.22503904067880973,
|
| 2591 |
+
"learning_rate": 3.1847904463503816e-05,
|
| 2592 |
+
"loss": 0.5633,
|
| 2593 |
+
"step": 369
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.4736,
|
| 2597 |
+
"grad_norm": 0.22994992202231845,
|
| 2598 |
+
"learning_rate": 3.17402351833036e-05,
|
| 2599 |
+
"loss": 0.56,
|
| 2600 |
+
"step": 370
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.47488,
|
| 2604 |
+
"grad_norm": 0.2454490715349298,
|
| 2605 |
+
"learning_rate": 3.163243091364752e-05,
|
| 2606 |
+
"loss": 0.5502,
|
| 2607 |
+
"step": 371
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.47616,
|
| 2611 |
+
"grad_norm": 0.2567184232746229,
|
| 2612 |
+
"learning_rate": 3.152449381357593e-05,
|
| 2613 |
+
"loss": 0.5606,
|
| 2614 |
+
"step": 372
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.47744,
|
| 2618 |
+
"grad_norm": 0.22104709877502515,
|
| 2619 |
+
"learning_rate": 3.141642604478942e-05,
|
| 2620 |
+
"loss": 0.5452,
|
| 2621 |
+
"step": 373
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.47872,
|
| 2625 |
+
"grad_norm": 0.254791340469147,
|
| 2626 |
+
"learning_rate": 3.130822977160554e-05,
|
| 2627 |
+
"loss": 0.5575,
|
| 2628 |
+
"step": 374
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.48,
|
| 2632 |
+
"grad_norm": 0.2172815217297066,
|
| 2633 |
+
"learning_rate": 3.119990716091546e-05,
|
| 2634 |
+
"loss": 0.5592,
|
| 2635 |
+
"step": 375
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.48128,
|
| 2639 |
+
"grad_norm": 0.23931030435414483,
|
| 2640 |
+
"learning_rate": 3.109146038214055e-05,
|
| 2641 |
+
"loss": 0.5473,
|
| 2642 |
+
"step": 376
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.48256,
|
| 2646 |
+
"grad_norm": 0.2290268854716801,
|
| 2647 |
+
"learning_rate": 3.098289160718895e-05,
|
| 2648 |
+
"loss": 0.5645,
|
| 2649 |
+
"step": 377
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.48384,
|
| 2653 |
+
"grad_norm": 0.23603315462362215,
|
| 2654 |
+
"learning_rate": 3.087420301041206e-05,
|
| 2655 |
+
"loss": 0.5408,
|
| 2656 |
+
"step": 378
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.48512,
|
| 2660 |
+
"grad_norm": 0.20930260887273594,
|
| 2661 |
+
"learning_rate": 3.076539676856101e-05,
|
| 2662 |
+
"loss": 0.5497,
|
| 2663 |
+
"step": 379
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.4864,
|
| 2667 |
+
"grad_norm": 0.20115083009307477,
|
| 2668 |
+
"learning_rate": 3.065647506074306e-05,
|
| 2669 |
+
"loss": 0.565,
|
| 2670 |
+
"step": 380
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.48768,
|
| 2674 |
+
"grad_norm": 0.24482468983205832,
|
| 2675 |
+
"learning_rate": 3.054744006837794e-05,
|
| 2676 |
+
"loss": 0.5682,
|
| 2677 |
+
"step": 381
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.48896,
|
| 2681 |
+
"grad_norm": 0.21118406075167703,
|
| 2682 |
+
"learning_rate": 3.0438293975154186e-05,
|
| 2683 |
+
"loss": 0.5477,
|
| 2684 |
+
"step": 382
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.49024,
|
| 2688 |
+
"grad_norm": 0.23616813058785022,
|
| 2689 |
+
"learning_rate": 3.03290389669854e-05,
|
| 2690 |
+
"loss": 0.5563,
|
| 2691 |
+
"step": 383
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.49152,
|
| 2695 |
+
"grad_norm": 0.21365522272880177,
|
| 2696 |
+
"learning_rate": 3.021967723196647e-05,
|
| 2697 |
+
"loss": 0.5507,
|
| 2698 |
+
"step": 384
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.4928,
|
| 2702 |
+
"grad_norm": 0.22727170936570085,
|
| 2703 |
+
"learning_rate": 3.0110210960329725e-05,
|
| 2704 |
+
"loss": 0.5603,
|
| 2705 |
+
"step": 385
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.49408,
|
| 2709 |
+
"grad_norm": 0.20731397878172675,
|
| 2710 |
+
"learning_rate": 3.0000642344401113e-05,
|
| 2711 |
+
"loss": 0.5563,
|
| 2712 |
+
"step": 386
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.49536,
|
| 2716 |
+
"grad_norm": 0.21475557099958575,
|
| 2717 |
+
"learning_rate": 2.9890973578556268e-05,
|
| 2718 |
+
"loss": 0.552,
|
| 2719 |
+
"step": 387
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 0.49664,
|
| 2723 |
+
"grad_norm": 0.26877099196744664,
|
| 2724 |
+
"learning_rate": 2.978120685917656e-05,
|
| 2725 |
+
"loss": 0.5409,
|
| 2726 |
+
"step": 388
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.49792,
|
| 2730 |
+
"grad_norm": 0.21789486858704812,
|
| 2731 |
+
"learning_rate": 2.9671344384605127e-05,
|
| 2732 |
+
"loss": 0.5486,
|
| 2733 |
+
"step": 389
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 0.4992,
|
| 2737 |
+
"grad_norm": 0.23650494006372377,
|
| 2738 |
+
"learning_rate": 2.956138835510282e-05,
|
| 2739 |
+
"loss": 0.5454,
|
| 2740 |
+
"step": 390
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.50048,
|
| 2744 |
+
"grad_norm": 0.22781069793383732,
|
| 2745 |
+
"learning_rate": 2.945134097280417e-05,
|
| 2746 |
+
"loss": 0.5615,
|
| 2747 |
+
"step": 391
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.50176,
|
| 2751 |
+
"grad_norm": 0.2365389525743412,
|
| 2752 |
+
"learning_rate": 2.9341204441673266e-05,
|
| 2753 |
+
"loss": 0.5544,
|
| 2754 |
+
"step": 392
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 0.50304,
|
| 2758 |
+
"grad_norm": 0.21603949855780968,
|
| 2759 |
+
"learning_rate": 2.9230980967459593e-05,
|
| 2760 |
+
"loss": 0.569,
|
| 2761 |
+
"step": 393
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 0.50432,
|
| 2765 |
+
"grad_norm": 0.2037144247370564,
|
| 2766 |
+
"learning_rate": 2.9120672757653916e-05,
|
| 2767 |
+
"loss": 0.5497,
|
| 2768 |
+
"step": 394
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.5056,
|
| 2772 |
+
"grad_norm": 0.23323915671730888,
|
| 2773 |
+
"learning_rate": 2.9010282021444008e-05,
|
| 2774 |
+
"loss": 0.5583,
|
| 2775 |
+
"step": 395
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.50688,
|
| 2779 |
+
"grad_norm": 0.21232243650279164,
|
| 2780 |
+
"learning_rate": 2.8899810969670448e-05,
|
| 2781 |
+
"loss": 0.5632,
|
| 2782 |
+
"step": 396
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 0.50816,
|
| 2786 |
+
"grad_norm": 0.23946451797290258,
|
| 2787 |
+
"learning_rate": 2.8789261814782316e-05,
|
| 2788 |
+
"loss": 0.5593,
|
| 2789 |
+
"step": 397
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 0.50944,
|
| 2793 |
+
"grad_norm": 0.2277720908380823,
|
| 2794 |
+
"learning_rate": 2.8678636770792906e-05,
|
| 2795 |
+
"loss": 0.5557,
|
| 2796 |
+
"step": 398
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 0.51072,
|
| 2800 |
+
"grad_norm": 0.2121131449498051,
|
| 2801 |
+
"learning_rate": 2.856793805323536e-05,
|
| 2802 |
+
"loss": 0.5494,
|
| 2803 |
+
"step": 399
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 0.512,
|
| 2807 |
+
"grad_norm": 0.2257692000346662,
|
| 2808 |
+
"learning_rate": 2.845716787911833e-05,
|
| 2809 |
+
"loss": 0.5669,
|
| 2810 |
+
"step": 400
|
| 2811 |
+
},
|
| 2812 |
+
{
|
| 2813 |
+
"epoch": 0.51328,
|
| 2814 |
+
"grad_norm": 0.23206067761113425,
|
| 2815 |
+
"learning_rate": 2.8346328466881545e-05,
|
| 2816 |
+
"loss": 0.5587,
|
| 2817 |
+
"step": 401
|
| 2818 |
+
},
|
| 2819 |
+
{
|
| 2820 |
+
"epoch": 0.51456,
|
| 2821 |
+
"grad_norm": 0.22901702627476864,
|
| 2822 |
+
"learning_rate": 2.8235422036351382e-05,
|
| 2823 |
+
"loss": 0.5719,
|
| 2824 |
+
"step": 402
|
| 2825 |
+
},
|
| 2826 |
+
{
|
| 2827 |
+
"epoch": 0.51584,
|
| 2828 |
+
"grad_norm": 0.2284208784515934,
|
| 2829 |
+
"learning_rate": 2.812445080869646e-05,
|
| 2830 |
+
"loss": 0.5608,
|
| 2831 |
+
"step": 403
|
| 2832 |
+
},
|
| 2833 |
+
{
|
| 2834 |
+
"epoch": 0.51712,
|
| 2835 |
+
"grad_norm": 0.21851283140724692,
|
| 2836 |
+
"learning_rate": 2.8013417006383076e-05,
|
| 2837 |
+
"loss": 0.5567,
|
| 2838 |
+
"step": 404
|
| 2839 |
+
},
|
| 2840 |
+
{
|
| 2841 |
+
"epoch": 0.5184,
|
| 2842 |
+
"grad_norm": 0.21275077693373082,
|
| 2843 |
+
"learning_rate": 2.7902322853130757e-05,
|
| 2844 |
+
"loss": 0.545,
|
| 2845 |
+
"step": 405
|
| 2846 |
+
},
|
| 2847 |
+
{
|
| 2848 |
+
"epoch": 0.51968,
|
| 2849 |
+
"grad_norm": 0.2251800050901655,
|
| 2850 |
+
"learning_rate": 2.77911705738677e-05,
|
| 2851 |
+
"loss": 0.555,
|
| 2852 |
+
"step": 406
|
| 2853 |
+
},
|
| 2854 |
+
{
|
| 2855 |
+
"epoch": 0.52096,
|
| 2856 |
+
"grad_norm": 0.21241829251347838,
|
| 2857 |
+
"learning_rate": 2.7679962394686198e-05,
|
| 2858 |
+
"loss": 0.5581,
|
| 2859 |
+
"step": 407
|
| 2860 |
+
},
|
| 2861 |
+
{
|
| 2862 |
+
"epoch": 0.52224,
|
| 2863 |
+
"grad_norm": 0.21703951180685088,
|
| 2864 |
+
"learning_rate": 2.756870054279811e-05,
|
| 2865 |
+
"loss": 0.5488,
|
| 2866 |
+
"step": 408
|
| 2867 |
+
},
|
| 2868 |
+
{
|
| 2869 |
+
"epoch": 0.52352,
|
| 2870 |
+
"grad_norm": 0.22069571037639818,
|
| 2871 |
+
"learning_rate": 2.745738724649018e-05,
|
| 2872 |
+
"loss": 0.5535,
|
| 2873 |
+
"step": 409
|
| 2874 |
+
},
|
| 2875 |
+
{
|
| 2876 |
+
"epoch": 0.5248,
|
| 2877 |
+
"grad_norm": 0.25932488146409416,
|
| 2878 |
+
"learning_rate": 2.7346024735079486e-05,
|
| 2879 |
+
"loss": 0.5583,
|
| 2880 |
+
"step": 410
|
| 2881 |
+
},
|
| 2882 |
+
{
|
| 2883 |
+
"epoch": 0.52608,
|
| 2884 |
+
"grad_norm": 0.20727956981757625,
|
| 2885 |
+
"learning_rate": 2.7234615238868732e-05,
|
| 2886 |
+
"loss": 0.5632,
|
| 2887 |
+
"step": 411
|
| 2888 |
+
},
|
| 2889 |
+
{
|
| 2890 |
+
"epoch": 0.52736,
|
| 2891 |
+
"grad_norm": 0.24066945021398153,
|
| 2892 |
+
"learning_rate": 2.712316098910162e-05,
|
| 2893 |
+
"loss": 0.563,
|
| 2894 |
+
"step": 412
|
| 2895 |
+
},
|
| 2896 |
+
{
|
| 2897 |
+
"epoch": 0.52864,
|
| 2898 |
+
"grad_norm": 0.25008559453154683,
|
| 2899 |
+
"learning_rate": 2.7011664217918154e-05,
|
| 2900 |
+
"loss": 0.5488,
|
| 2901 |
+
"step": 413
|
| 2902 |
+
},
|
| 2903 |
+
{
|
| 2904 |
+
"epoch": 0.52992,
|
| 2905 |
+
"grad_norm": 0.23013581555834695,
|
| 2906 |
+
"learning_rate": 2.6900127158309903e-05,
|
| 2907 |
+
"loss": 0.5633,
|
| 2908 |
+
"step": 414
|
| 2909 |
+
},
|
| 2910 |
+
{
|
| 2911 |
+
"epoch": 0.5312,
|
| 2912 |
+
"grad_norm": 0.22224479213901044,
|
| 2913 |
+
"learning_rate": 2.6788552044075344e-05,
|
| 2914 |
+
"loss": 0.549,
|
| 2915 |
+
"step": 415
|
| 2916 |
+
},
|
| 2917 |
+
{
|
| 2918 |
+
"epoch": 0.53248,
|
| 2919 |
+
"grad_norm": 0.2654591019527864,
|
| 2920 |
+
"learning_rate": 2.667694110977506e-05,
|
| 2921 |
+
"loss": 0.5574,
|
| 2922 |
+
"step": 416
|
| 2923 |
+
},
|
| 2924 |
+
{
|
| 2925 |
+
"epoch": 0.53376,
|
| 2926 |
+
"grad_norm": 0.19545471307044301,
|
| 2927 |
+
"learning_rate": 2.656529659068705e-05,
|
| 2928 |
+
"loss": 0.5488,
|
| 2929 |
+
"step": 417
|
| 2930 |
+
},
|
| 2931 |
+
{
|
| 2932 |
+
"epoch": 0.53504,
|
| 2933 |
+
"grad_norm": 0.20254148474738726,
|
| 2934 |
+
"learning_rate": 2.6453620722761896e-05,
|
| 2935 |
+
"loss": 0.5638,
|
| 2936 |
+
"step": 418
|
| 2937 |
+
},
|
| 2938 |
+
{
|
| 2939 |
+
"epoch": 0.53632,
|
| 2940 |
+
"grad_norm": 0.22685700497041825,
|
| 2941 |
+
"learning_rate": 2.6341915742578037e-05,
|
| 2942 |
+
"loss": 0.5414,
|
| 2943 |
+
"step": 419
|
| 2944 |
+
},
|
| 2945 |
+
{
|
| 2946 |
+
"epoch": 0.5376,
|
| 2947 |
+
"grad_norm": 0.20942540533409634,
|
| 2948 |
+
"learning_rate": 2.6230183887296955e-05,
|
| 2949 |
+
"loss": 0.5565,
|
| 2950 |
+
"step": 420
|
| 2951 |
+
},
|
| 2952 |
+
{
|
| 2953 |
+
"epoch": 0.53888,
|
| 2954 |
+
"grad_norm": 0.21464639120298495,
|
| 2955 |
+
"learning_rate": 2.6118427394618357e-05,
|
| 2956 |
+
"loss": 0.5561,
|
| 2957 |
+
"step": 421
|
| 2958 |
+
},
|
| 2959 |
+
{
|
| 2960 |
+
"epoch": 0.54016,
|
| 2961 |
+
"grad_norm": 0.22217802407368148,
|
| 2962 |
+
"learning_rate": 2.600664850273538e-05,
|
| 2963 |
+
"loss": 0.5533,
|
| 2964 |
+
"step": 422
|
| 2965 |
+
},
|
| 2966 |
+
{
|
| 2967 |
+
"epoch": 0.54144,
|
| 2968 |
+
"grad_norm": 0.18275865926888668,
|
| 2969 |
+
"learning_rate": 2.5894849450289764e-05,
|
| 2970 |
+
"loss": 0.5488,
|
| 2971 |
+
"step": 423
|
| 2972 |
+
},
|
| 2973 |
+
{
|
| 2974 |
+
"epoch": 0.54272,
|
| 2975 |
+
"grad_norm": 0.2167014210634381,
|
| 2976 |
+
"learning_rate": 2.5783032476327007e-05,
|
| 2977 |
+
"loss": 0.5461,
|
| 2978 |
+
"step": 424
|
| 2979 |
+
},
|
| 2980 |
+
{
|
| 2981 |
+
"epoch": 0.544,
|
| 2982 |
+
"grad_norm": 0.2034252786956083,
|
| 2983 |
+
"learning_rate": 2.5671199820251534e-05,
|
| 2984 |
+
"loss": 0.5509,
|
| 2985 |
+
"step": 425
|
| 2986 |
+
},
|
| 2987 |
+
{
|
| 2988 |
+
"epoch": 0.54528,
|
| 2989 |
+
"grad_norm": 0.18569564751312728,
|
| 2990 |
+
"learning_rate": 2.5559353721781832e-05,
|
| 2991 |
+
"loss": 0.5478,
|
| 2992 |
+
"step": 426
|
| 2993 |
+
},
|
| 2994 |
+
{
|
| 2995 |
+
"epoch": 0.54656,
|
| 2996 |
+
"grad_norm": 0.20022896045042796,
|
| 2997 |
+
"learning_rate": 2.544749642090561e-05,
|
| 2998 |
+
"loss": 0.5477,
|
| 2999 |
+
"step": 427
|
| 3000 |
+
},
|
| 3001 |
+
{
|
| 3002 |
+
"epoch": 0.54784,
|
| 3003 |
+
"grad_norm": 0.21037283126869577,
|
| 3004 |
+
"learning_rate": 2.5335630157834937e-05,
|
| 3005 |
+
"loss": 0.5539,
|
| 3006 |
+
"step": 428
|
| 3007 |
+
},
|
| 3008 |
+
{
|
| 3009 |
+
"epoch": 0.54912,
|
| 3010 |
+
"grad_norm": 0.19972461705925887,
|
| 3011 |
+
"learning_rate": 2.522375717296137e-05,
|
| 3012 |
+
"loss": 0.5569,
|
| 3013 |
+
"step": 429
|
| 3014 |
+
},
|
| 3015 |
+
{
|
| 3016 |
+
"epoch": 0.5504,
|
| 3017 |
+
"grad_norm": 0.19005063735053182,
|
| 3018 |
+
"learning_rate": 2.5111879706811087e-05,
|
| 3019 |
+
"loss": 0.5457,
|
| 3020 |
+
"step": 430
|
| 3021 |
+
},
|
| 3022 |
+
{
|
| 3023 |
+
"epoch": 0.55168,
|
| 3024 |
+
"grad_norm": 0.21356314651647584,
|
| 3025 |
+
"learning_rate": 2.5e-05,
|
| 3026 |
+
"loss": 0.5459,
|
| 3027 |
+
"step": 431
|
| 3028 |
+
},
|
| 3029 |
+
{
|
| 3030 |
+
"epoch": 0.55296,
|
| 3031 |
+
"grad_norm": 0.20945885149378243,
|
| 3032 |
+
"learning_rate": 2.4888120293188916e-05,
|
| 3033 |
+
"loss": 0.5492,
|
| 3034 |
+
"step": 432
|
| 3035 |
+
},
|
| 3036 |
+
{
|
| 3037 |
+
"epoch": 0.55424,
|
| 3038 |
+
"grad_norm": 0.2230850015397349,
|
| 3039 |
+
"learning_rate": 2.4776242827038636e-05,
|
| 3040 |
+
"loss": 0.5545,
|
| 3041 |
+
"step": 433
|
| 3042 |
+
},
|
| 3043 |
+
{
|
| 3044 |
+
"epoch": 0.55552,
|
| 3045 |
+
"grad_norm": 0.23589759701529497,
|
| 3046 |
+
"learning_rate": 2.4664369842165068e-05,
|
| 3047 |
+
"loss": 0.5412,
|
| 3048 |
+
"step": 434
|
| 3049 |
+
},
|
| 3050 |
+
{
|
| 3051 |
+
"epoch": 0.5568,
|
| 3052 |
+
"grad_norm": 0.22581333235197817,
|
| 3053 |
+
"learning_rate": 2.4552503579094397e-05,
|
| 3054 |
+
"loss": 0.5435,
|
| 3055 |
+
"step": 435
|
| 3056 |
+
},
|
| 3057 |
+
{
|
| 3058 |
+
"epoch": 0.55808,
|
| 3059 |
+
"grad_norm": 0.24919774658733163,
|
| 3060 |
+
"learning_rate": 2.4440646278218177e-05,
|
| 3061 |
+
"loss": 0.545,
|
| 3062 |
+
"step": 436
|
| 3063 |
+
},
|
| 3064 |
+
{
|
| 3065 |
+
"epoch": 0.55936,
|
| 3066 |
+
"grad_norm": 0.2190295213590674,
|
| 3067 |
+
"learning_rate": 2.4328800179748475e-05,
|
| 3068 |
+
"loss": 0.5408,
|
| 3069 |
+
"step": 437
|
| 3070 |
+
},
|
| 3071 |
+
{
|
| 3072 |
+
"epoch": 0.56064,
|
| 3073 |
+
"grad_norm": 0.2628486805976341,
|
| 3074 |
+
"learning_rate": 2.4216967523673e-05,
|
| 3075 |
+
"loss": 0.5655,
|
| 3076 |
+
"step": 438
|
| 3077 |
+
},
|
| 3078 |
+
{
|
| 3079 |
+
"epoch": 0.56192,
|
| 3080 |
+
"grad_norm": 0.25252172819470975,
|
| 3081 |
+
"learning_rate": 2.4105150549710238e-05,
|
| 3082 |
+
"loss": 0.5552,
|
| 3083 |
+
"step": 439
|
| 3084 |
+
},
|
| 3085 |
+
{
|
| 3086 |
+
"epoch": 0.5632,
|
| 3087 |
+
"grad_norm": 0.20229952269202467,
|
| 3088 |
+
"learning_rate": 2.399335149726463e-05,
|
| 3089 |
+
"loss": 0.5429,
|
| 3090 |
+
"step": 440
|
| 3091 |
+
},
|
| 3092 |
+
{
|
| 3093 |
+
"epoch": 0.56448,
|
| 3094 |
+
"grad_norm": 0.24469450227466819,
|
| 3095 |
+
"learning_rate": 2.388157260538165e-05,
|
| 3096 |
+
"loss": 0.5582,
|
| 3097 |
+
"step": 441
|
| 3098 |
+
},
|
| 3099 |
+
{
|
| 3100 |
+
"epoch": 0.56576,
|
| 3101 |
+
"grad_norm": 0.21164210405032619,
|
| 3102 |
+
"learning_rate": 2.3769816112703047e-05,
|
| 3103 |
+
"loss": 0.5506,
|
| 3104 |
+
"step": 442
|
| 3105 |
+
},
|
| 3106 |
+
{
|
| 3107 |
+
"epoch": 0.56704,
|
| 3108 |
+
"grad_norm": 0.22950726772051805,
|
| 3109 |
+
"learning_rate": 2.365808425742196e-05,
|
| 3110 |
+
"loss": 0.5497,
|
| 3111 |
+
"step": 443
|
| 3112 |
+
},
|
| 3113 |
+
{
|
| 3114 |
+
"epoch": 0.56832,
|
| 3115 |
+
"grad_norm": 0.23336024035401096,
|
| 3116 |
+
"learning_rate": 2.3546379277238107e-05,
|
| 3117 |
+
"loss": 0.5415,
|
| 3118 |
+
"step": 444
|
| 3119 |
+
},
|
| 3120 |
+
{
|
| 3121 |
+
"epoch": 0.5696,
|
| 3122 |
+
"grad_norm": 0.19765357154592897,
|
| 3123 |
+
"learning_rate": 2.3434703409312954e-05,
|
| 3124 |
+
"loss": 0.5423,
|
| 3125 |
+
"step": 445
|
| 3126 |
+
},
|
| 3127 |
+
{
|
| 3128 |
+
"epoch": 0.57088,
|
| 3129 |
+
"grad_norm": 0.20108815980408287,
|
| 3130 |
+
"learning_rate": 2.3323058890224938e-05,
|
| 3131 |
+
"loss": 0.5543,
|
| 3132 |
+
"step": 446
|
| 3133 |
+
},
|
| 3134 |
+
{
|
| 3135 |
+
"epoch": 0.57216,
|
| 3136 |
+
"grad_norm": 0.20801363546967222,
|
| 3137 |
+
"learning_rate": 2.321144795592467e-05,
|
| 3138 |
+
"loss": 0.5579,
|
| 3139 |
+
"step": 447
|
| 3140 |
+
},
|
| 3141 |
+
{
|
| 3142 |
+
"epoch": 0.57344,
|
| 3143 |
+
"grad_norm": 0.19116838604162356,
|
| 3144 |
+
"learning_rate": 2.3099872841690103e-05,
|
| 3145 |
+
"loss": 0.5418,
|
| 3146 |
+
"step": 448
|
| 3147 |
+
},
|
| 3148 |
+
{
|
| 3149 |
+
"epoch": 0.57472,
|
| 3150 |
+
"grad_norm": 0.18998557353906567,
|
| 3151 |
+
"learning_rate": 2.2988335782081855e-05,
|
| 3152 |
+
"loss": 0.5575,
|
| 3153 |
+
"step": 449
|
| 3154 |
+
},
|
| 3155 |
+
{
|
| 3156 |
+
"epoch": 0.576,
|
| 3157 |
+
"grad_norm": 0.20390602126403667,
|
| 3158 |
+
"learning_rate": 2.2876839010898377e-05,
|
| 3159 |
+
"loss": 0.5517,
|
| 3160 |
+
"step": 450
|
| 3161 |
+
},
|
| 3162 |
+
{
|
| 3163 |
+
"epoch": 0.57728,
|
| 3164 |
+
"grad_norm": 0.20273804136737242,
|
| 3165 |
+
"learning_rate": 2.2765384761131274e-05,
|
| 3166 |
+
"loss": 0.5512,
|
| 3167 |
+
"step": 451
|
| 3168 |
+
},
|
| 3169 |
+
{
|
| 3170 |
+
"epoch": 0.57856,
|
| 3171 |
+
"grad_norm": 0.2323345427375102,
|
| 3172 |
+
"learning_rate": 2.265397526492052e-05,
|
| 3173 |
+
"loss": 0.5552,
|
| 3174 |
+
"step": 452
|
| 3175 |
+
},
|
| 3176 |
+
{
|
| 3177 |
+
"epoch": 0.57984,
|
| 3178 |
+
"grad_norm": 0.21618001641053508,
|
| 3179 |
+
"learning_rate": 2.2542612753509823e-05,
|
| 3180 |
+
"loss": 0.5497,
|
| 3181 |
+
"step": 453
|
| 3182 |
+
},
|
| 3183 |
+
{
|
| 3184 |
+
"epoch": 0.58112,
|
| 3185 |
+
"grad_norm": 0.2089693691856067,
|
| 3186 |
+
"learning_rate": 2.24312994572019e-05,
|
| 3187 |
+
"loss": 0.5437,
|
| 3188 |
+
"step": 454
|
| 3189 |
+
},
|
| 3190 |
+
{
|
| 3191 |
+
"epoch": 0.5824,
|
| 3192 |
+
"grad_norm": 0.21321327766142864,
|
| 3193 |
+
"learning_rate": 2.2320037605313808e-05,
|
| 3194 |
+
"loss": 0.5457,
|
| 3195 |
+
"step": 455
|
| 3196 |
+
},
|
| 3197 |
+
{
|
| 3198 |
+
"epoch": 0.58368,
|
| 3199 |
+
"grad_norm": 0.1866665167284803,
|
| 3200 |
+
"learning_rate": 2.2208829426132307e-05,
|
| 3201 |
+
"loss": 0.5589,
|
| 3202 |
+
"step": 456
|
| 3203 |
+
},
|
| 3204 |
+
{
|
| 3205 |
+
"epoch": 0.58496,
|
| 3206 |
+
"grad_norm": 0.19549868239805784,
|
| 3207 |
+
"learning_rate": 2.2097677146869242e-05,
|
| 3208 |
+
"loss": 0.5428,
|
| 3209 |
+
"step": 457
|
| 3210 |
+
},
|
| 3211 |
+
{
|
| 3212 |
+
"epoch": 0.58624,
|
| 3213 |
+
"grad_norm": 0.21465688730228774,
|
| 3214 |
+
"learning_rate": 2.1986582993616926e-05,
|
| 3215 |
+
"loss": 0.5409,
|
| 3216 |
+
"step": 458
|
| 3217 |
+
},
|
| 3218 |
+
{
|
| 3219 |
+
"epoch": 0.58752,
|
| 3220 |
+
"grad_norm": 0.19658929521150698,
|
| 3221 |
+
"learning_rate": 2.1875549191303545e-05,
|
| 3222 |
+
"loss": 0.5464,
|
| 3223 |
+
"step": 459
|
| 3224 |
+
},
|
| 3225 |
+
{
|
| 3226 |
+
"epoch": 0.5888,
|
| 3227 |
+
"grad_norm": 0.198475793396338,
|
| 3228 |
+
"learning_rate": 2.1764577963648614e-05,
|
| 3229 |
+
"loss": 0.5468,
|
| 3230 |
+
"step": 460
|
| 3231 |
+
},
|
| 3232 |
+
{
|
| 3233 |
+
"epoch": 0.59008,
|
| 3234 |
+
"grad_norm": 0.21158010098295732,
|
| 3235 |
+
"learning_rate": 2.1653671533118468e-05,
|
| 3236 |
+
"loss": 0.5537,
|
| 3237 |
+
"step": 461
|
| 3238 |
+
},
|
| 3239 |
+
{
|
| 3240 |
+
"epoch": 0.59136,
|
| 3241 |
+
"grad_norm": 0.21320070068715855,
|
| 3242 |
+
"learning_rate": 2.154283212088168e-05,
|
| 3243 |
+
"loss": 0.5486,
|
| 3244 |
+
"step": 462
|
| 3245 |
+
},
|
| 3246 |
+
{
|
| 3247 |
+
"epoch": 0.59264,
|
| 3248 |
+
"grad_norm": 0.2212774837957101,
|
| 3249 |
+
"learning_rate": 2.1432061946764644e-05,
|
| 3250 |
+
"loss": 0.5381,
|
| 3251 |
+
"step": 463
|
| 3252 |
+
},
|
| 3253 |
+
{
|
| 3254 |
+
"epoch": 0.59392,
|
| 3255 |
+
"grad_norm": 0.19109434465910716,
|
| 3256 |
+
"learning_rate": 2.1321363229207096e-05,
|
| 3257 |
+
"loss": 0.5538,
|
| 3258 |
+
"step": 464
|
| 3259 |
+
},
|
| 3260 |
+
{
|
| 3261 |
+
"epoch": 0.5952,
|
| 3262 |
+
"grad_norm": 0.194950360975785,
|
| 3263 |
+
"learning_rate": 2.121073818521769e-05,
|
| 3264 |
+
"loss": 0.5539,
|
| 3265 |
+
"step": 465
|
| 3266 |
+
},
|
| 3267 |
+
{
|
| 3268 |
+
"epoch": 0.59648,
|
| 3269 |
+
"grad_norm": 0.20628175723560369,
|
| 3270 |
+
"learning_rate": 2.1100189030329558e-05,
|
| 3271 |
+
"loss": 0.5414,
|
| 3272 |
+
"step": 466
|
| 3273 |
+
},
|
| 3274 |
+
{
|
| 3275 |
+
"epoch": 0.59776,
|
| 3276 |
+
"grad_norm": 0.19027258987716833,
|
| 3277 |
+
"learning_rate": 2.098971797855599e-05,
|
| 3278 |
+
"loss": 0.5465,
|
| 3279 |
+
"step": 467
|
| 3280 |
+
},
|
| 3281 |
+
{
|
| 3282 |
+
"epoch": 0.59904,
|
| 3283 |
+
"grad_norm": 0.1968353186636271,
|
| 3284 |
+
"learning_rate": 2.0879327242346093e-05,
|
| 3285 |
+
"loss": 0.5429,
|
| 3286 |
+
"step": 468
|
| 3287 |
+
},
|
| 3288 |
+
{
|
| 3289 |
+
"epoch": 0.60032,
|
| 3290 |
+
"grad_norm": 0.19197267253975747,
|
| 3291 |
+
"learning_rate": 2.0769019032540416e-05,
|
| 3292 |
+
"loss": 0.5576,
|
| 3293 |
+
"step": 469
|
| 3294 |
+
},
|
| 3295 |
+
{
|
| 3296 |
+
"epoch": 0.6016,
|
| 3297 |
+
"grad_norm": 0.1989897674210096,
|
| 3298 |
+
"learning_rate": 2.0658795558326743e-05,
|
| 3299 |
+
"loss": 0.5421,
|
| 3300 |
+
"step": 470
|
| 3301 |
+
},
|
| 3302 |
+
{
|
| 3303 |
+
"epoch": 0.60288,
|
| 3304 |
+
"grad_norm": 0.2081969628582826,
|
| 3305 |
+
"learning_rate": 2.054865902719584e-05,
|
| 3306 |
+
"loss": 0.567,
|
| 3307 |
+
"step": 471
|
| 3308 |
+
}
|
| 3309 |
+
],
|
| 3310 |
+
"logging_steps": 1,
|
| 3311 |
+
"max_steps": 781,
|
| 3312 |
+
"num_input_tokens_seen": 0,
|
| 3313 |
+
"num_train_epochs": 1,
|
| 3314 |
+
"save_steps": 157,
|
| 3315 |
+
"stateful_callbacks": {
|
| 3316 |
+
"TrainerControl": {
|
| 3317 |
+
"args": {
|
| 3318 |
+
"should_epoch_stop": false,
|
| 3319 |
+
"should_evaluate": false,
|
| 3320 |
+
"should_log": false,
|
| 3321 |
+
"should_save": true,
|
| 3322 |
+
"should_training_stop": false
|
| 3323 |
+
},
|
| 3324 |
+
"attributes": {}
|
| 3325 |
+
}
|
| 3326 |
+
},
|
| 3327 |
+
"total_flos": 584441230589952.0,
|
| 3328 |
+
"train_batch_size": 2,
|
| 3329 |
+
"trial_name": null,
|
| 3330 |
+
"trial_params": null
|
| 3331 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17d4ae4b6b6214515a5dd9c49ec78773756cbcc25fe748eaf748b8059d9427af
|
| 3 |
+
size 8081
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
zero_to_fp32.py
ADDED
|
@@ -0,0 +1,760 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/usr/bin/env python
|
| 2 |
+
|
| 3 |
+
# Copyright (c) Microsoft Corporation.
|
| 4 |
+
# SPDX-License-Identifier: Apache-2.0
|
| 5 |
+
|
| 6 |
+
# DeepSpeed Team
|
| 7 |
+
|
| 8 |
+
# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
|
| 9 |
+
# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
|
| 10 |
+
# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
|
| 11 |
+
# application.
|
| 12 |
+
#
|
| 13 |
+
# example:
|
| 14 |
+
# python zero_to_fp32.py . output_dir/
|
| 15 |
+
# or
|
| 16 |
+
# python zero_to_fp32.py . output_dir/ --safe_serialization
|
| 17 |
+
|
| 18 |
+
import argparse
|
| 19 |
+
import torch
|
| 20 |
+
import glob
|
| 21 |
+
import math
|
| 22 |
+
import os
|
| 23 |
+
import re
|
| 24 |
+
import gc
|
| 25 |
+
import json
|
| 26 |
+
import numpy as np
|
| 27 |
+
from tqdm import tqdm
|
| 28 |
+
from collections import OrderedDict
|
| 29 |
+
from dataclasses import dataclass
|
| 30 |
+
|
| 31 |
+
# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
|
| 32 |
+
# DeepSpeed data structures it has to be available in the current python environment.
|
| 33 |
+
from deepspeed.utils import logger
|
| 34 |
+
from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
|
| 35 |
+
FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
|
| 36 |
+
FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
|
| 37 |
+
|
| 38 |
+
|
| 39 |
+
@dataclass
|
| 40 |
+
class zero_model_state:
|
| 41 |
+
buffers: dict()
|
| 42 |
+
param_shapes: dict()
|
| 43 |
+
shared_params: list
|
| 44 |
+
ds_version: int
|
| 45 |
+
frozen_param_shapes: dict()
|
| 46 |
+
frozen_param_fragments: dict()
|
| 47 |
+
|
| 48 |
+
|
| 49 |
+
debug = 0
|
| 50 |
+
|
| 51 |
+
# load to cpu
|
| 52 |
+
device = torch.device('cpu')
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
def atoi(text):
|
| 56 |
+
return int(text) if text.isdigit() else text
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
def natural_keys(text):
|
| 60 |
+
'''
|
| 61 |
+
alist.sort(key=natural_keys) sorts in human order
|
| 62 |
+
http://nedbatchelder.com/blog/200712/human_sorting.html
|
| 63 |
+
(See Toothy's implementation in the comments)
|
| 64 |
+
'''
|
| 65 |
+
return [atoi(c) for c in re.split(r'(\d+)', text)]
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
def get_model_state_file(checkpoint_dir, zero_stage):
|
| 69 |
+
if not os.path.isdir(checkpoint_dir):
|
| 70 |
+
raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
|
| 71 |
+
|
| 72 |
+
# there should be only one file
|
| 73 |
+
if zero_stage <= 2:
|
| 74 |
+
file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
|
| 75 |
+
elif zero_stage == 3:
|
| 76 |
+
file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
|
| 77 |
+
|
| 78 |
+
if not os.path.exists(file):
|
| 79 |
+
raise FileNotFoundError(f"can't find model states file at '{file}'")
|
| 80 |
+
|
| 81 |
+
return file
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
def get_checkpoint_files(checkpoint_dir, glob_pattern):
|
| 85 |
+
# XXX: need to test that this simple glob rule works for multi-node setup too
|
| 86 |
+
ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
|
| 87 |
+
|
| 88 |
+
if len(ckpt_files) == 0:
|
| 89 |
+
raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
|
| 90 |
+
|
| 91 |
+
return ckpt_files
|
| 92 |
+
|
| 93 |
+
|
| 94 |
+
def get_optim_files(checkpoint_dir):
|
| 95 |
+
return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
|
| 96 |
+
|
| 97 |
+
|
| 98 |
+
def get_model_state_files(checkpoint_dir):
|
| 99 |
+
return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
|
| 100 |
+
|
| 101 |
+
|
| 102 |
+
def parse_model_states(files):
|
| 103 |
+
zero_model_states = []
|
| 104 |
+
for file in files:
|
| 105 |
+
state_dict = torch.load(file, map_location=device, weights_only=False)
|
| 106 |
+
|
| 107 |
+
if BUFFER_NAMES not in state_dict:
|
| 108 |
+
raise ValueError(f"{file} is not a model state checkpoint")
|
| 109 |
+
buffer_names = state_dict[BUFFER_NAMES]
|
| 110 |
+
if debug:
|
| 111 |
+
print("Found buffers:", buffer_names)
|
| 112 |
+
|
| 113 |
+
# recover just the buffers while restoring them to fp32 if they were saved in fp16
|
| 114 |
+
buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
|
| 115 |
+
param_shapes = state_dict[PARAM_SHAPES]
|
| 116 |
+
|
| 117 |
+
# collect parameters that are included in param_shapes
|
| 118 |
+
param_names = []
|
| 119 |
+
for s in param_shapes:
|
| 120 |
+
for name in s.keys():
|
| 121 |
+
param_names.append(name)
|
| 122 |
+
|
| 123 |
+
# update with frozen parameters
|
| 124 |
+
frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
|
| 125 |
+
if frozen_param_shapes is not None:
|
| 126 |
+
if debug:
|
| 127 |
+
print(f"Found frozen_param_shapes: {frozen_param_shapes}")
|
| 128 |
+
param_names += list(frozen_param_shapes.keys())
|
| 129 |
+
|
| 130 |
+
# handle shared params
|
| 131 |
+
shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
|
| 132 |
+
|
| 133 |
+
ds_version = state_dict.get(DS_VERSION, None)
|
| 134 |
+
|
| 135 |
+
frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
|
| 136 |
+
|
| 137 |
+
z_model_state = zero_model_state(buffers=buffers,
|
| 138 |
+
param_shapes=param_shapes,
|
| 139 |
+
shared_params=shared_params,
|
| 140 |
+
ds_version=ds_version,
|
| 141 |
+
frozen_param_shapes=frozen_param_shapes,
|
| 142 |
+
frozen_param_fragments=frozen_param_fragments)
|
| 143 |
+
zero_model_states.append(z_model_state)
|
| 144 |
+
|
| 145 |
+
return zero_model_states
|
| 146 |
+
|
| 147 |
+
|
| 148 |
+
def parse_optim_states(files, ds_checkpoint_dir):
|
| 149 |
+
total_files = len(files)
|
| 150 |
+
state_dicts = []
|
| 151 |
+
for f in tqdm(files, desc='Loading checkpoint shards'):
|
| 152 |
+
state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
|
| 153 |
+
# immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
|
| 154 |
+
# and also handle the case where it was already removed by another helper script
|
| 155 |
+
state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
|
| 156 |
+
state_dicts.append(state_dict)
|
| 157 |
+
|
| 158 |
+
if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
|
| 159 |
+
raise ValueError(f"{files[0]} is not a zero checkpoint")
|
| 160 |
+
zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
|
| 161 |
+
world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
|
| 162 |
+
|
| 163 |
+
# For ZeRO-2 each param group can have different partition_count as data parallelism for expert
|
| 164 |
+
# parameters can be different from data parallelism for non-expert parameters. So we can just
|
| 165 |
+
# use the max of the partition_count to get the dp world_size.
|
| 166 |
+
|
| 167 |
+
if type(world_size) is list:
|
| 168 |
+
world_size = max(world_size)
|
| 169 |
+
|
| 170 |
+
if world_size != total_files:
|
| 171 |
+
raise ValueError(
|
| 172 |
+
f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
|
| 173 |
+
"Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
|
| 174 |
+
)
|
| 175 |
+
|
| 176 |
+
# the groups are named differently in each stage
|
| 177 |
+
if zero_stage <= 2:
|
| 178 |
+
fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
|
| 179 |
+
elif zero_stage == 3:
|
| 180 |
+
fp32_groups_key = FP32_FLAT_GROUPS
|
| 181 |
+
else:
|
| 182 |
+
raise ValueError(f"unknown zero stage {zero_stage}")
|
| 183 |
+
|
| 184 |
+
fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
|
| 185 |
+
return zero_stage, world_size, fp32_flat_groups
|
| 186 |
+
|
| 187 |
+
|
| 188 |
+
def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
|
| 189 |
+
"""
|
| 190 |
+
Returns fp32 state_dict reconstructed from ds checkpoint
|
| 191 |
+
|
| 192 |
+
Args:
|
| 193 |
+
- ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
|
| 194 |
+
|
| 195 |
+
"""
|
| 196 |
+
print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
|
| 197 |
+
|
| 198 |
+
optim_files = get_optim_files(ds_checkpoint_dir)
|
| 199 |
+
zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
|
| 200 |
+
print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
|
| 201 |
+
|
| 202 |
+
model_files = get_model_state_files(ds_checkpoint_dir)
|
| 203 |
+
|
| 204 |
+
zero_model_states = parse_model_states(model_files)
|
| 205 |
+
print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
|
| 206 |
+
|
| 207 |
+
if zero_stage <= 2:
|
| 208 |
+
return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 209 |
+
exclude_frozen_parameters)
|
| 210 |
+
elif zero_stage == 3:
|
| 211 |
+
return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 212 |
+
exclude_frozen_parameters)
|
| 213 |
+
|
| 214 |
+
|
| 215 |
+
def _zero2_merge_frozen_params(state_dict, zero_model_states):
|
| 216 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 217 |
+
return
|
| 218 |
+
|
| 219 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 220 |
+
frozen_param_fragments = zero_model_states[0].frozen_param_fragments
|
| 221 |
+
|
| 222 |
+
if debug:
|
| 223 |
+
num_elem = sum(s.numel() for s in frozen_param_shapes.values())
|
| 224 |
+
print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 225 |
+
|
| 226 |
+
wanted_params = len(frozen_param_shapes)
|
| 227 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 228 |
+
avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
|
| 229 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 230 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 231 |
+
|
| 232 |
+
total_params = 0
|
| 233 |
+
total_numel = 0
|
| 234 |
+
for name, shape in frozen_param_shapes.items():
|
| 235 |
+
total_params += 1
|
| 236 |
+
unpartitioned_numel = shape.numel()
|
| 237 |
+
total_numel += unpartitioned_numel
|
| 238 |
+
|
| 239 |
+
state_dict[name] = frozen_param_fragments[name]
|
| 240 |
+
|
| 241 |
+
if debug:
|
| 242 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 243 |
+
|
| 244 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 245 |
+
|
| 246 |
+
|
| 247 |
+
def _has_callable(obj, fn):
|
| 248 |
+
attr = getattr(obj, fn, None)
|
| 249 |
+
return callable(attr)
|
| 250 |
+
|
| 251 |
+
|
| 252 |
+
def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 253 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 254 |
+
|
| 255 |
+
# Reconstruction protocol:
|
| 256 |
+
#
|
| 257 |
+
# XXX: document this
|
| 258 |
+
|
| 259 |
+
if debug:
|
| 260 |
+
for i in range(world_size):
|
| 261 |
+
for j in range(len(fp32_flat_groups[0])):
|
| 262 |
+
print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
|
| 263 |
+
|
| 264 |
+
# XXX: memory usage doubles here (zero2)
|
| 265 |
+
num_param_groups = len(fp32_flat_groups[0])
|
| 266 |
+
merged_single_partition_of_fp32_groups = []
|
| 267 |
+
for i in range(num_param_groups):
|
| 268 |
+
merged_partitions = [sd[i] for sd in fp32_flat_groups]
|
| 269 |
+
full_single_fp32_vector = torch.cat(merged_partitions, 0)
|
| 270 |
+
merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
|
| 271 |
+
avail_numel = sum(
|
| 272 |
+
[full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
|
| 273 |
+
|
| 274 |
+
if debug:
|
| 275 |
+
wanted_params = sum([len(shapes) for shapes in param_shapes])
|
| 276 |
+
wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
|
| 277 |
+
# not asserting if there is a mismatch due to possible padding
|
| 278 |
+
print(f"Have {avail_numel} numels to process.")
|
| 279 |
+
print(f"Need {wanted_numel} numels in {wanted_params} params.")
|
| 280 |
+
|
| 281 |
+
# params
|
| 282 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 283 |
+
# out-of-core computing solution
|
| 284 |
+
total_numel = 0
|
| 285 |
+
total_params = 0
|
| 286 |
+
for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
|
| 287 |
+
offset = 0
|
| 288 |
+
avail_numel = full_single_fp32_vector.numel()
|
| 289 |
+
for name, shape in shapes.items():
|
| 290 |
+
|
| 291 |
+
unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
|
| 292 |
+
total_numel += unpartitioned_numel
|
| 293 |
+
total_params += 1
|
| 294 |
+
|
| 295 |
+
if debug:
|
| 296 |
+
print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
|
| 297 |
+
state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
|
| 298 |
+
offset += unpartitioned_numel
|
| 299 |
+
|
| 300 |
+
# Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
|
| 301 |
+
# avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
|
| 302 |
+
# paddings performed in the code it's almost impossible to predict the exact numbers w/o the
|
| 303 |
+
# live optimizer object, so we are checking that the numbers are within the right range
|
| 304 |
+
align_to = 2 * world_size
|
| 305 |
+
|
| 306 |
+
def zero2_align(x):
|
| 307 |
+
return align_to * math.ceil(x / align_to)
|
| 308 |
+
|
| 309 |
+
if debug:
|
| 310 |
+
print(f"original offset={offset}, avail_numel={avail_numel}")
|
| 311 |
+
|
| 312 |
+
offset = zero2_align(offset)
|
| 313 |
+
avail_numel = zero2_align(avail_numel)
|
| 314 |
+
|
| 315 |
+
if debug:
|
| 316 |
+
print(f"aligned offset={offset}, avail_numel={avail_numel}")
|
| 317 |
+
|
| 318 |
+
# Sanity check
|
| 319 |
+
if offset != avail_numel:
|
| 320 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 321 |
+
|
| 322 |
+
print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
|
| 323 |
+
|
| 324 |
+
|
| 325 |
+
def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 326 |
+
exclude_frozen_parameters):
|
| 327 |
+
state_dict = OrderedDict()
|
| 328 |
+
|
| 329 |
+
# buffers
|
| 330 |
+
buffers = zero_model_states[0].buffers
|
| 331 |
+
state_dict.update(buffers)
|
| 332 |
+
if debug:
|
| 333 |
+
print(f"added {len(buffers)} buffers")
|
| 334 |
+
|
| 335 |
+
if not exclude_frozen_parameters:
|
| 336 |
+
_zero2_merge_frozen_params(state_dict, zero_model_states)
|
| 337 |
+
|
| 338 |
+
_zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 339 |
+
|
| 340 |
+
# recover shared parameters
|
| 341 |
+
for pair in zero_model_states[0].shared_params:
|
| 342 |
+
if pair[1] in state_dict:
|
| 343 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 344 |
+
|
| 345 |
+
return state_dict
|
| 346 |
+
|
| 347 |
+
|
| 348 |
+
def zero3_partitioned_param_info(unpartitioned_numel, world_size):
|
| 349 |
+
remainder = unpartitioned_numel % world_size
|
| 350 |
+
padding_numel = (world_size - remainder) if remainder else 0
|
| 351 |
+
partitioned_numel = math.ceil(unpartitioned_numel / world_size)
|
| 352 |
+
return partitioned_numel, padding_numel
|
| 353 |
+
|
| 354 |
+
|
| 355 |
+
def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
|
| 356 |
+
if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
|
| 357 |
+
return
|
| 358 |
+
|
| 359 |
+
if debug:
|
| 360 |
+
for i in range(world_size):
|
| 361 |
+
num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
|
| 362 |
+
print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
|
| 363 |
+
|
| 364 |
+
frozen_param_shapes = zero_model_states[0].frozen_param_shapes
|
| 365 |
+
wanted_params = len(frozen_param_shapes)
|
| 366 |
+
wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
|
| 367 |
+
avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
|
| 368 |
+
print(f'Frozen params: Have {avail_numel} numels to process.')
|
| 369 |
+
print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
|
| 370 |
+
|
| 371 |
+
total_params = 0
|
| 372 |
+
total_numel = 0
|
| 373 |
+
for name, shape in zero_model_states[0].frozen_param_shapes.items():
|
| 374 |
+
total_params += 1
|
| 375 |
+
unpartitioned_numel = shape.numel()
|
| 376 |
+
total_numel += unpartitioned_numel
|
| 377 |
+
|
| 378 |
+
param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
|
| 379 |
+
state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
|
| 380 |
+
|
| 381 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 382 |
+
|
| 383 |
+
if debug:
|
| 384 |
+
print(
|
| 385 |
+
f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 386 |
+
)
|
| 387 |
+
|
| 388 |
+
print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
|
| 389 |
+
|
| 390 |
+
|
| 391 |
+
class GatheredTensor:
|
| 392 |
+
"""
|
| 393 |
+
A pseudo tensor that collects partitioned weights.
|
| 394 |
+
It is more memory efficient when there are multiple groups.
|
| 395 |
+
"""
|
| 396 |
+
|
| 397 |
+
def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
|
| 398 |
+
self.flat_groups = flat_groups
|
| 399 |
+
self.flat_groups_offset = flat_groups_offset
|
| 400 |
+
self.offset = offset
|
| 401 |
+
self.partitioned_numel = partitioned_numel
|
| 402 |
+
self.shape = shape
|
| 403 |
+
self.dtype = self.flat_groups[0][0].dtype
|
| 404 |
+
|
| 405 |
+
def contiguous(self):
|
| 406 |
+
"""
|
| 407 |
+
Merge partitioned weights from flat_groups into a single tensor.
|
| 408 |
+
"""
|
| 409 |
+
end_idx = self.offset + self.partitioned_numel
|
| 410 |
+
world_size = len(self.flat_groups)
|
| 411 |
+
pad_flat_param_chunks = []
|
| 412 |
+
|
| 413 |
+
for rank_i in range(world_size):
|
| 414 |
+
# for each rank, we need to collect weights from related group/groups
|
| 415 |
+
flat_groups_at_rank_i = self.flat_groups[rank_i]
|
| 416 |
+
start_group_id = None
|
| 417 |
+
end_group_id = None
|
| 418 |
+
for group_id in range(len(self.flat_groups_offset)):
|
| 419 |
+
if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
|
| 420 |
+
start_group_id = group_id
|
| 421 |
+
if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
|
| 422 |
+
end_group_id = group_id
|
| 423 |
+
break
|
| 424 |
+
# collect weights from related group/groups
|
| 425 |
+
for group_id in range(start_group_id, end_group_id + 1):
|
| 426 |
+
flat_tensor = flat_groups_at_rank_i[group_id]
|
| 427 |
+
start_offset = self.offset - self.flat_groups_offset[group_id]
|
| 428 |
+
end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
|
| 429 |
+
pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
|
| 430 |
+
|
| 431 |
+
# collect weights from all ranks
|
| 432 |
+
pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
|
| 433 |
+
param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
|
| 434 |
+
return param
|
| 435 |
+
|
| 436 |
+
|
| 437 |
+
def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
|
| 438 |
+
param_shapes = zero_model_states[0].param_shapes
|
| 439 |
+
avail_numel = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * world_size
|
| 440 |
+
|
| 441 |
+
# Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
|
| 442 |
+
# param, re-consolidating each param, while dealing with padding if any
|
| 443 |
+
|
| 444 |
+
# merge list of dicts, preserving order
|
| 445 |
+
param_shapes = {k: v for d in param_shapes for k, v in d.items()}
|
| 446 |
+
|
| 447 |
+
if debug:
|
| 448 |
+
for i in range(world_size):
|
| 449 |
+
print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
|
| 450 |
+
|
| 451 |
+
wanted_params = len(param_shapes)
|
| 452 |
+
wanted_numel = sum(shape.numel() for shape in param_shapes.values())
|
| 453 |
+
# not asserting if there is a mismatch due to possible padding
|
| 454 |
+
avail_numel = fp32_flat_groups[0].numel() * world_size
|
| 455 |
+
print(f"Trainable params: Have {avail_numel} numels to process.")
|
| 456 |
+
print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
|
| 457 |
+
|
| 458 |
+
# params
|
| 459 |
+
# XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
|
| 460 |
+
# out-of-core computing solution
|
| 461 |
+
offset = 0
|
| 462 |
+
total_numel = 0
|
| 463 |
+
total_params = 0
|
| 464 |
+
flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
|
| 465 |
+
for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
|
| 466 |
+
unpartitioned_numel = shape.numel()
|
| 467 |
+
total_numel += unpartitioned_numel
|
| 468 |
+
total_params += 1
|
| 469 |
+
partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
|
| 470 |
+
|
| 471 |
+
if debug:
|
| 472 |
+
print(
|
| 473 |
+
f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
|
| 474 |
+
)
|
| 475 |
+
|
| 476 |
+
# memory efficient tensor
|
| 477 |
+
tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
|
| 478 |
+
state_dict[name] = tensor
|
| 479 |
+
offset += partitioned_numel
|
| 480 |
+
|
| 481 |
+
offset *= world_size
|
| 482 |
+
|
| 483 |
+
# Sanity check
|
| 484 |
+
if offset != avail_numel:
|
| 485 |
+
raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
|
| 486 |
+
|
| 487 |
+
print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
|
| 488 |
+
|
| 489 |
+
|
| 490 |
+
def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
|
| 491 |
+
exclude_frozen_parameters):
|
| 492 |
+
state_dict = OrderedDict()
|
| 493 |
+
|
| 494 |
+
# buffers
|
| 495 |
+
buffers = zero_model_states[0].buffers
|
| 496 |
+
state_dict.update(buffers)
|
| 497 |
+
if debug:
|
| 498 |
+
print(f"added {len(buffers)} buffers")
|
| 499 |
+
|
| 500 |
+
if not exclude_frozen_parameters:
|
| 501 |
+
_zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
|
| 502 |
+
|
| 503 |
+
_zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
|
| 504 |
+
|
| 505 |
+
# recover shared parameters
|
| 506 |
+
for pair in zero_model_states[0].shared_params:
|
| 507 |
+
if pair[1] in state_dict:
|
| 508 |
+
state_dict[pair[0]] = state_dict[pair[1]]
|
| 509 |
+
|
| 510 |
+
return state_dict
|
| 511 |
+
|
| 512 |
+
|
| 513 |
+
def to_torch_tensor(state_dict, return_empty_tensor=False):
|
| 514 |
+
"""
|
| 515 |
+
Convert state_dict of GatheredTensor to torch tensor
|
| 516 |
+
"""
|
| 517 |
+
torch_state_dict = {}
|
| 518 |
+
converted_tensors = {}
|
| 519 |
+
for name, tensor in state_dict.items():
|
| 520 |
+
tensor_id = id(tensor)
|
| 521 |
+
if tensor_id in converted_tensors: # shared tensors
|
| 522 |
+
shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
|
| 523 |
+
torch_state_dict[name] = shared_tensor
|
| 524 |
+
else:
|
| 525 |
+
converted_tensors[tensor_id] = name
|
| 526 |
+
if return_empty_tensor:
|
| 527 |
+
torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
|
| 528 |
+
else:
|
| 529 |
+
torch_state_dict[name] = tensor.contiguous()
|
| 530 |
+
return torch_state_dict
|
| 531 |
+
|
| 532 |
+
|
| 533 |
+
def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 534 |
+
tag=None,
|
| 535 |
+
exclude_frozen_parameters=False,
|
| 536 |
+
lazy_mode=False):
|
| 537 |
+
"""
|
| 538 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
|
| 539 |
+
``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
|
| 540 |
+
via a model hub.
|
| 541 |
+
|
| 542 |
+
Args:
|
| 543 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder
|
| 544 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
|
| 545 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 546 |
+
- ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
|
| 547 |
+
Convert the pesduo tensor to torch tensor by ``.contiguous()``
|
| 548 |
+
|
| 549 |
+
Returns:
|
| 550 |
+
- pytorch ``state_dict``
|
| 551 |
+
|
| 552 |
+
A typical usage might be ::
|
| 553 |
+
|
| 554 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 555 |
+
# do the training and checkpoint saving
|
| 556 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
|
| 557 |
+
model = model.cpu() # move to cpu
|
| 558 |
+
model.load_state_dict(state_dict)
|
| 559 |
+
# submit to model hub or save the model to share with others
|
| 560 |
+
|
| 561 |
+
In this example the ``model`` will no longer be usable in the deepspeed context of the same
|
| 562 |
+
application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 563 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 564 |
+
|
| 565 |
+
If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
|
| 566 |
+
|
| 567 |
+
Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
|
| 568 |
+
You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
|
| 569 |
+
the checkpoint. Or you can load state_dict in lazy mode ::
|
| 570 |
+
|
| 571 |
+
from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
|
| 572 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
|
| 573 |
+
for name, lazy_tensor in state_dict.item():
|
| 574 |
+
tensor = lazy_tensor.contiguous() # to cpu
|
| 575 |
+
print(name, tensor)
|
| 576 |
+
# del tensor to release memory if it no longer in use
|
| 577 |
+
"""
|
| 578 |
+
if tag is None:
|
| 579 |
+
latest_path = os.path.join(checkpoint_dir, 'latest')
|
| 580 |
+
if os.path.isfile(latest_path):
|
| 581 |
+
with open(latest_path, 'r') as fd:
|
| 582 |
+
tag = fd.read().strip()
|
| 583 |
+
else:
|
| 584 |
+
raise ValueError(f"Unable to find 'latest' file at {latest_path}")
|
| 585 |
+
|
| 586 |
+
ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
|
| 587 |
+
|
| 588 |
+
if not os.path.isdir(ds_checkpoint_dir):
|
| 589 |
+
raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
|
| 590 |
+
|
| 591 |
+
state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
|
| 592 |
+
if lazy_mode:
|
| 593 |
+
return state_dict
|
| 594 |
+
else:
|
| 595 |
+
return to_torch_tensor(state_dict)
|
| 596 |
+
|
| 597 |
+
|
| 598 |
+
def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
|
| 599 |
+
output_dir,
|
| 600 |
+
max_shard_size="5GB",
|
| 601 |
+
safe_serialization=False,
|
| 602 |
+
tag=None,
|
| 603 |
+
exclude_frozen_parameters=False):
|
| 604 |
+
"""
|
| 605 |
+
Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
|
| 606 |
+
loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
|
| 607 |
+
|
| 608 |
+
Args:
|
| 609 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 610 |
+
- ``output_dir``: directory to the pytorch fp32 state_dict output files
|
| 611 |
+
- ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
|
| 612 |
+
- ``safe_serialization``: whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
|
| 613 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 614 |
+
- ``exclude_frozen_parameters``: exclude frozen parameters
|
| 615 |
+
"""
|
| 616 |
+
|
| 617 |
+
# Dependency pre-check
|
| 618 |
+
if safe_serialization:
|
| 619 |
+
try:
|
| 620 |
+
from safetensors.torch import save_file
|
| 621 |
+
except ImportError:
|
| 622 |
+
print('If you want to use `safe_serialization`, please `pip install safetensors`')
|
| 623 |
+
raise
|
| 624 |
+
if max_shard_size is not None:
|
| 625 |
+
try:
|
| 626 |
+
from huggingface_hub import split_torch_state_dict_into_shards
|
| 627 |
+
except ImportError:
|
| 628 |
+
print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
|
| 629 |
+
raise
|
| 630 |
+
|
| 631 |
+
# Convert zero checkpoint to state_dict
|
| 632 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
|
| 633 |
+
tag,
|
| 634 |
+
exclude_frozen_parameters,
|
| 635 |
+
lazy_mode=True)
|
| 636 |
+
|
| 637 |
+
# Shard the model if it is too big.
|
| 638 |
+
weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
|
| 639 |
+
if max_shard_size is not None:
|
| 640 |
+
filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
|
| 641 |
+
# an memory-efficient approach for sharding
|
| 642 |
+
empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
|
| 643 |
+
state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
|
| 644 |
+
filename_pattern=filename_pattern,
|
| 645 |
+
max_shard_size=max_shard_size)
|
| 646 |
+
else:
|
| 647 |
+
from collections import namedtuple
|
| 648 |
+
StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
|
| 649 |
+
state_dict_split = StateDictSplit(is_sharded=False,
|
| 650 |
+
filename_to_tensors={weights_name: list(state_dict.keys())})
|
| 651 |
+
|
| 652 |
+
# Save the model by shard
|
| 653 |
+
os.makedirs(output_dir, exist_ok=True)
|
| 654 |
+
filename_to_tensors = state_dict_split.filename_to_tensors.items()
|
| 655 |
+
for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
|
| 656 |
+
shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
|
| 657 |
+
shard_state_dict = to_torch_tensor(shard_state_dict)
|
| 658 |
+
output_path = os.path.join(output_dir, shard_file)
|
| 659 |
+
if safe_serialization:
|
| 660 |
+
save_file(shard_state_dict, output_path, metadata={"format": "pt"})
|
| 661 |
+
else:
|
| 662 |
+
torch.save(shard_state_dict, output_path)
|
| 663 |
+
# release the memory of current shard
|
| 664 |
+
for tensor_name in list(shard_state_dict.keys()):
|
| 665 |
+
del state_dict[tensor_name]
|
| 666 |
+
del shard_state_dict[tensor_name]
|
| 667 |
+
del shard_state_dict
|
| 668 |
+
gc.collect()
|
| 669 |
+
|
| 670 |
+
# Save index if sharded
|
| 671 |
+
if state_dict_split.is_sharded:
|
| 672 |
+
index = {
|
| 673 |
+
"metadata": state_dict_split.metadata,
|
| 674 |
+
"weight_map": state_dict_split.tensor_to_filename,
|
| 675 |
+
}
|
| 676 |
+
save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
|
| 677 |
+
save_index_file = os.path.join(output_dir, save_index_file)
|
| 678 |
+
with open(save_index_file, "w", encoding="utf-8") as f:
|
| 679 |
+
content = json.dumps(index, indent=2, sort_keys=True) + "\n"
|
| 680 |
+
f.write(content)
|
| 681 |
+
|
| 682 |
+
|
| 683 |
+
def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
|
| 684 |
+
"""
|
| 685 |
+
1. Put the provided model to cpu
|
| 686 |
+
2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
|
| 687 |
+
3. Load it into the provided model
|
| 688 |
+
|
| 689 |
+
Args:
|
| 690 |
+
- ``model``: the model object to update
|
| 691 |
+
- ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
|
| 692 |
+
- ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
|
| 693 |
+
|
| 694 |
+
Returns:
|
| 695 |
+
- ``model`: modified model
|
| 696 |
+
|
| 697 |
+
Make sure you have plenty of CPU memory available before you call this function. If you don't
|
| 698 |
+
have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
|
| 699 |
+
conveniently placed for you in the checkpoint folder.
|
| 700 |
+
|
| 701 |
+
A typical usage might be ::
|
| 702 |
+
|
| 703 |
+
from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
|
| 704 |
+
model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
|
| 705 |
+
# submit to model hub or save the model to share with others
|
| 706 |
+
|
| 707 |
+
Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
|
| 708 |
+
of the same application. i.e. you will need to re-initialize the deepspeed engine, since
|
| 709 |
+
``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
|
| 710 |
+
|
| 711 |
+
"""
|
| 712 |
+
logger.info(f"Extracting fp32 weights")
|
| 713 |
+
state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
|
| 714 |
+
|
| 715 |
+
logger.info(f"Overwriting model with fp32 weights")
|
| 716 |
+
model = model.cpu()
|
| 717 |
+
model.load_state_dict(state_dict, strict=False)
|
| 718 |
+
|
| 719 |
+
return model
|
| 720 |
+
|
| 721 |
+
|
| 722 |
+
if __name__ == "__main__":
|
| 723 |
+
parser = argparse.ArgumentParser()
|
| 724 |
+
parser.add_argument("checkpoint_dir",
|
| 725 |
+
type=str,
|
| 726 |
+
help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
|
| 727 |
+
parser.add_argument("output_dir",
|
| 728 |
+
type=str,
|
| 729 |
+
help="directory to the pytorch fp32 state_dict output files"
|
| 730 |
+
"(e.g. path/checkpoint-12-output/)")
|
| 731 |
+
parser.add_argument(
|
| 732 |
+
"--max_shard_size",
|
| 733 |
+
type=str,
|
| 734 |
+
default="5GB",
|
| 735 |
+
help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
|
| 736 |
+
"lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
|
| 737 |
+
"We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
|
| 738 |
+
"without CPU OOM issues.")
|
| 739 |
+
parser.add_argument(
|
| 740 |
+
"--safe_serialization",
|
| 741 |
+
default=False,
|
| 742 |
+
action='store_true',
|
| 743 |
+
help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
|
| 744 |
+
parser.add_argument("-t",
|
| 745 |
+
"--tag",
|
| 746 |
+
type=str,
|
| 747 |
+
default=None,
|
| 748 |
+
help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
|
| 749 |
+
parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
|
| 750 |
+
parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
|
| 751 |
+
args = parser.parse_args()
|
| 752 |
+
|
| 753 |
+
debug = args.debug
|
| 754 |
+
|
| 755 |
+
convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
|
| 756 |
+
args.output_dir,
|
| 757 |
+
max_shard_size=args.max_shard_size,
|
| 758 |
+
safe_serialization=args.safe_serialization,
|
| 759 |
+
tag=args.tag,
|
| 760 |
+
exclude_frozen_parameters=args.exclude_frozen_parameters)
|