Upload 8 files
Browse files- config.json +29 -0
- generation_config.json +8 -0
- model.safetensors +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +44 -0
- trainer_state.json +2112 -0
config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "l3utterfly/tinyllama-1.1b-layla-v4",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 2048,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 5632,
|
| 14 |
+
"max_position_embeddings": 2048,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
+
"model_type": "llama",
|
| 17 |
+
"num_attention_heads": 32,
|
| 18 |
+
"num_hidden_layers": 22,
|
| 19 |
+
"num_key_value_heads": 4,
|
| 20 |
+
"pretraining_tp": 1,
|
| 21 |
+
"rms_norm_eps": 1e-05,
|
| 22 |
+
"rope_scaling": null,
|
| 23 |
+
"rope_theta": 10000.0,
|
| 24 |
+
"tie_word_embeddings": false,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.42.0.dev0",
|
| 27 |
+
"use_cache": false,
|
| 28 |
+
"vocab_size": 32000
|
| 29 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"max_length": 2048,
|
| 6 |
+
"pad_token_id": 0,
|
| 7 |
+
"transformers_version": "4.42.0.dev0"
|
| 8 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d638aa21c24b68510635d50fd9af56aed5707fc75056728a1fe6837d5d5062c5
|
| 3 |
+
size 2200119664
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<unk>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"clean_up_tokenization_spaces": false,
|
| 33 |
+
"eos_token": "</s>",
|
| 34 |
+
"legacy": false,
|
| 35 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 36 |
+
"pad_token": "</s>",
|
| 37 |
+
"padding_side": "right",
|
| 38 |
+
"sp_model_kwargs": {},
|
| 39 |
+
"spaces_between_special_tokens": false,
|
| 40 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 41 |
+
"unk_token": "<unk>",
|
| 42 |
+
"use_default_system_prompt": false,
|
| 43 |
+
"use_fast": true
|
| 44 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2112 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5959,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0033562678301728476,
|
| 13 |
+
"grad_norm": 3.027224540710449,
|
| 14 |
+
"learning_rate": 4.984896794764222e-05,
|
| 15 |
+
"loss": 2.5909,
|
| 16 |
+
"step": 20
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.006712535660345695,
|
| 20 |
+
"grad_norm": 2.6114368438720703,
|
| 21 |
+
"learning_rate": 4.968115455613358e-05,
|
| 22 |
+
"loss": 2.367,
|
| 23 |
+
"step": 40
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010068803490518544,
|
| 27 |
+
"grad_norm": 18.757535934448242,
|
| 28 |
+
"learning_rate": 4.951334116462494e-05,
|
| 29 |
+
"loss": 2.286,
|
| 30 |
+
"step": 60
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01342507132069139,
|
| 34 |
+
"grad_norm": 2.032654047012329,
|
| 35 |
+
"learning_rate": 4.9345527773116296e-05,
|
| 36 |
+
"loss": 2.2961,
|
| 37 |
+
"step": 80
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01678133915086424,
|
| 41 |
+
"grad_norm": 1.8827602863311768,
|
| 42 |
+
"learning_rate": 4.9177714381607655e-05,
|
| 43 |
+
"loss": 2.2622,
|
| 44 |
+
"step": 100
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.020137606981037087,
|
| 48 |
+
"grad_norm": 1.8459181785583496,
|
| 49 |
+
"learning_rate": 4.9009900990099014e-05,
|
| 50 |
+
"loss": 2.2569,
|
| 51 |
+
"step": 120
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.023493874811209934,
|
| 55 |
+
"grad_norm": 1.9432493448257446,
|
| 56 |
+
"learning_rate": 4.8842087598590366e-05,
|
| 57 |
+
"loss": 2.2069,
|
| 58 |
+
"step": 140
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02685014264138278,
|
| 62 |
+
"grad_norm": 1.9693349599838257,
|
| 63 |
+
"learning_rate": 4.8674274207081725e-05,
|
| 64 |
+
"loss": 2.2437,
|
| 65 |
+
"step": 160
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03020641047155563,
|
| 69 |
+
"grad_norm": 1.8286640644073486,
|
| 70 |
+
"learning_rate": 4.8506460815573084e-05,
|
| 71 |
+
"loss": 2.1793,
|
| 72 |
+
"step": 180
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03356267830172848,
|
| 76 |
+
"grad_norm": 1.883164644241333,
|
| 77 |
+
"learning_rate": 4.833864742406444e-05,
|
| 78 |
+
"loss": 2.2013,
|
| 79 |
+
"step": 200
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.036918946131901324,
|
| 83 |
+
"grad_norm": 10.042224884033203,
|
| 84 |
+
"learning_rate": 4.817922470213123e-05,
|
| 85 |
+
"loss": 2.2325,
|
| 86 |
+
"step": 220
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.040275213962074174,
|
| 90 |
+
"grad_norm": 1.909852147102356,
|
| 91 |
+
"learning_rate": 4.801141131062259e-05,
|
| 92 |
+
"loss": 2.2372,
|
| 93 |
+
"step": 240
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.043631481792247025,
|
| 97 |
+
"grad_norm": 1.8228334188461304,
|
| 98 |
+
"learning_rate": 4.784359791911395e-05,
|
| 99 |
+
"loss": 2.236,
|
| 100 |
+
"step": 260
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04698774962241987,
|
| 104 |
+
"grad_norm": 1.8686975240707397,
|
| 105 |
+
"learning_rate": 4.767578452760531e-05,
|
| 106 |
+
"loss": 2.1993,
|
| 107 |
+
"step": 280
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05034401745259272,
|
| 111 |
+
"grad_norm": 1.6455596685409546,
|
| 112 |
+
"learning_rate": 4.750797113609666e-05,
|
| 113 |
+
"loss": 2.2244,
|
| 114 |
+
"step": 300
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05370028528276556,
|
| 118 |
+
"grad_norm": 1.5980720520019531,
|
| 119 |
+
"learning_rate": 4.734015774458802e-05,
|
| 120 |
+
"loss": 2.1412,
|
| 121 |
+
"step": 320
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.05705655311293841,
|
| 125 |
+
"grad_norm": 1.7958396673202515,
|
| 126 |
+
"learning_rate": 4.717234435307938e-05,
|
| 127 |
+
"loss": 2.1516,
|
| 128 |
+
"step": 340
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06041282094311126,
|
| 132 |
+
"grad_norm": 1.6902369260787964,
|
| 133 |
+
"learning_rate": 4.700453096157074e-05,
|
| 134 |
+
"loss": 2.1842,
|
| 135 |
+
"step": 360
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.0637690887732841,
|
| 139 |
+
"grad_norm": 1.7754089832305908,
|
| 140 |
+
"learning_rate": 4.6836717570062096e-05,
|
| 141 |
+
"loss": 2.1591,
|
| 142 |
+
"step": 380
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.06712535660345696,
|
| 146 |
+
"grad_norm": 1.718556523323059,
|
| 147 |
+
"learning_rate": 4.6668904178553455e-05,
|
| 148 |
+
"loss": 2.155,
|
| 149 |
+
"step": 400
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0704816244336298,
|
| 153 |
+
"grad_norm": 1.6255844831466675,
|
| 154 |
+
"learning_rate": 4.650109078704481e-05,
|
| 155 |
+
"loss": 2.2235,
|
| 156 |
+
"step": 420
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07383789226380265,
|
| 160 |
+
"grad_norm": 1.6990654468536377,
|
| 161 |
+
"learning_rate": 4.6333277395536166e-05,
|
| 162 |
+
"loss": 2.1942,
|
| 163 |
+
"step": 440
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.0771941600939755,
|
| 167 |
+
"grad_norm": 1.7159547805786133,
|
| 168 |
+
"learning_rate": 4.6165464004027525e-05,
|
| 169 |
+
"loss": 2.2133,
|
| 170 |
+
"step": 460
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08055042792414835,
|
| 174 |
+
"grad_norm": 1.8792229890823364,
|
| 175 |
+
"learning_rate": 4.5997650612518884e-05,
|
| 176 |
+
"loss": 2.1534,
|
| 177 |
+
"step": 480
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08390669575432119,
|
| 181 |
+
"grad_norm": 1.4530068635940552,
|
| 182 |
+
"learning_rate": 4.582983722101024e-05,
|
| 183 |
+
"loss": 2.1365,
|
| 184 |
+
"step": 500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.08726296358449405,
|
| 188 |
+
"grad_norm": 1.5878769159317017,
|
| 189 |
+
"learning_rate": 4.56620238295016e-05,
|
| 190 |
+
"loss": 2.1513,
|
| 191 |
+
"step": 520
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.09061923141466689,
|
| 195 |
+
"grad_norm": 1.6161599159240723,
|
| 196 |
+
"learning_rate": 4.5494210437992954e-05,
|
| 197 |
+
"loss": 2.1434,
|
| 198 |
+
"step": 540
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.09397549924483974,
|
| 202 |
+
"grad_norm": 1.5275565385818481,
|
| 203 |
+
"learning_rate": 4.5326397046484306e-05,
|
| 204 |
+
"loss": 2.1995,
|
| 205 |
+
"step": 560
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.09733176707501258,
|
| 209 |
+
"grad_norm": 1.3771929740905762,
|
| 210 |
+
"learning_rate": 4.5158583654975665e-05,
|
| 211 |
+
"loss": 2.1546,
|
| 212 |
+
"step": 580
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.10068803490518544,
|
| 216 |
+
"grad_norm": 1.630591630935669,
|
| 217 |
+
"learning_rate": 4.4990770263467024e-05,
|
| 218 |
+
"loss": 2.1317,
|
| 219 |
+
"step": 600
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.10404430273535828,
|
| 223 |
+
"grad_norm": 1.5199378728866577,
|
| 224 |
+
"learning_rate": 4.482295687195838e-05,
|
| 225 |
+
"loss": 2.116,
|
| 226 |
+
"step": 620
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.10740057056553112,
|
| 230 |
+
"grad_norm": 1.7431048154830933,
|
| 231 |
+
"learning_rate": 4.465514348044974e-05,
|
| 232 |
+
"loss": 2.1379,
|
| 233 |
+
"step": 640
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.11075683839570398,
|
| 237 |
+
"grad_norm": 1.6243093013763428,
|
| 238 |
+
"learning_rate": 4.44873300889411e-05,
|
| 239 |
+
"loss": 2.1478,
|
| 240 |
+
"step": 660
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.11411310622587682,
|
| 244 |
+
"grad_norm": 1.4836941957473755,
|
| 245 |
+
"learning_rate": 4.431951669743245e-05,
|
| 246 |
+
"loss": 2.1522,
|
| 247 |
+
"step": 680
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.11746937405604967,
|
| 251 |
+
"grad_norm": 1.5157350301742554,
|
| 252 |
+
"learning_rate": 4.415170330592381e-05,
|
| 253 |
+
"loss": 2.104,
|
| 254 |
+
"step": 700
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.12082564188622252,
|
| 258 |
+
"grad_norm": 1.3875478506088257,
|
| 259 |
+
"learning_rate": 4.398388991441517e-05,
|
| 260 |
+
"loss": 2.1495,
|
| 261 |
+
"step": 720
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.12418190971639537,
|
| 265 |
+
"grad_norm": 1.4270449876785278,
|
| 266 |
+
"learning_rate": 4.381607652290653e-05,
|
| 267 |
+
"loss": 2.1275,
|
| 268 |
+
"step": 740
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1275381775465682,
|
| 272 |
+
"grad_norm": 1.522594690322876,
|
| 273 |
+
"learning_rate": 4.364826313139789e-05,
|
| 274 |
+
"loss": 2.1439,
|
| 275 |
+
"step": 760
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.13089444537674105,
|
| 279 |
+
"grad_norm": 1.3996539115905762,
|
| 280 |
+
"learning_rate": 4.348044973988925e-05,
|
| 281 |
+
"loss": 2.1384,
|
| 282 |
+
"step": 780
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.13425071320691392,
|
| 286 |
+
"grad_norm": 1.5052591562271118,
|
| 287 |
+
"learning_rate": 4.33126363483806e-05,
|
| 288 |
+
"loss": 2.1005,
|
| 289 |
+
"step": 800
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.13760698103708677,
|
| 293 |
+
"grad_norm": 1.4939976930618286,
|
| 294 |
+
"learning_rate": 4.314482295687196e-05,
|
| 295 |
+
"loss": 2.078,
|
| 296 |
+
"step": 820
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1409632488672596,
|
| 300 |
+
"grad_norm": 1.5155856609344482,
|
| 301 |
+
"learning_rate": 4.297700956536332e-05,
|
| 302 |
+
"loss": 2.0766,
|
| 303 |
+
"step": 840
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.14431951669743245,
|
| 307 |
+
"grad_norm": 1.5032541751861572,
|
| 308 |
+
"learning_rate": 4.280919617385468e-05,
|
| 309 |
+
"loss": 2.1383,
|
| 310 |
+
"step": 860
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1476757845276053,
|
| 314 |
+
"grad_norm": 1.5815625190734863,
|
| 315 |
+
"learning_rate": 4.2641382782346036e-05,
|
| 316 |
+
"loss": 2.138,
|
| 317 |
+
"step": 880
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.15103205235777814,
|
| 321 |
+
"grad_norm": 1.5529719591140747,
|
| 322 |
+
"learning_rate": 4.2473569390837395e-05,
|
| 323 |
+
"loss": 2.1252,
|
| 324 |
+
"step": 900
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.154388320187951,
|
| 328 |
+
"grad_norm": 1.5002542734146118,
|
| 329 |
+
"learning_rate": 4.230575599932875e-05,
|
| 330 |
+
"loss": 2.1158,
|
| 331 |
+
"step": 920
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.15774458801812385,
|
| 335 |
+
"grad_norm": 1.4248254299163818,
|
| 336 |
+
"learning_rate": 4.2137942607820106e-05,
|
| 337 |
+
"loss": 2.1441,
|
| 338 |
+
"step": 940
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1611008558482967,
|
| 342 |
+
"grad_norm": 1.3779915571212769,
|
| 343 |
+
"learning_rate": 4.1970129216311465e-05,
|
| 344 |
+
"loss": 2.1105,
|
| 345 |
+
"step": 960
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.16445712367846954,
|
| 349 |
+
"grad_norm": 1.439987301826477,
|
| 350 |
+
"learning_rate": 4.1802315824802824e-05,
|
| 351 |
+
"loss": 2.0988,
|
| 352 |
+
"step": 980
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.16781339150864238,
|
| 356 |
+
"grad_norm": 1.3819688558578491,
|
| 357 |
+
"learning_rate": 4.163450243329418e-05,
|
| 358 |
+
"loss": 2.0594,
|
| 359 |
+
"step": 1000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.17116965933881523,
|
| 363 |
+
"grad_norm": 1.4762790203094482,
|
| 364 |
+
"learning_rate": 4.146668904178554e-05,
|
| 365 |
+
"loss": 2.1448,
|
| 366 |
+
"step": 1020
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1745259271689881,
|
| 370 |
+
"grad_norm": 1.3910346031188965,
|
| 371 |
+
"learning_rate": 4.1298875650276894e-05,
|
| 372 |
+
"loss": 2.1189,
|
| 373 |
+
"step": 1040
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.17788219499916094,
|
| 377 |
+
"grad_norm": 1.3728772401809692,
|
| 378 |
+
"learning_rate": 4.1131062258768246e-05,
|
| 379 |
+
"loss": 2.1025,
|
| 380 |
+
"step": 1060
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.18123846282933379,
|
| 384 |
+
"grad_norm": 1.466235637664795,
|
| 385 |
+
"learning_rate": 4.0963248867259605e-05,
|
| 386 |
+
"loss": 2.1163,
|
| 387 |
+
"step": 1080
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.18459473065950663,
|
| 391 |
+
"grad_norm": 1.418556809425354,
|
| 392 |
+
"learning_rate": 4.0795435475750964e-05,
|
| 393 |
+
"loss": 2.11,
|
| 394 |
+
"step": 1100
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.18795099848967947,
|
| 398 |
+
"grad_norm": 1.3588992357254028,
|
| 399 |
+
"learning_rate": 4.062762208424232e-05,
|
| 400 |
+
"loss": 2.1174,
|
| 401 |
+
"step": 1120
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.19130726631985231,
|
| 405 |
+
"grad_norm": 1.490565538406372,
|
| 406 |
+
"learning_rate": 4.045980869273368e-05,
|
| 407 |
+
"loss": 2.1176,
|
| 408 |
+
"step": 1140
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.19466353415002516,
|
| 412 |
+
"grad_norm": 1.4399744272232056,
|
| 413 |
+
"learning_rate": 4.029199530122504e-05,
|
| 414 |
+
"loss": 2.1222,
|
| 415 |
+
"step": 1160
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.19801980198019803,
|
| 419 |
+
"grad_norm": 1.379418134689331,
|
| 420 |
+
"learning_rate": 4.012418190971639e-05,
|
| 421 |
+
"loss": 2.1003,
|
| 422 |
+
"step": 1180
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.20137606981037087,
|
| 426 |
+
"grad_norm": 1.4052506685256958,
|
| 427 |
+
"learning_rate": 3.995636851820775e-05,
|
| 428 |
+
"loss": 2.0951,
|
| 429 |
+
"step": 1200
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.20473233764054372,
|
| 433 |
+
"grad_norm": 1.3870118856430054,
|
| 434 |
+
"learning_rate": 3.978855512669911e-05,
|
| 435 |
+
"loss": 2.1201,
|
| 436 |
+
"step": 1220
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.20808860547071656,
|
| 440 |
+
"grad_norm": 1.334796667098999,
|
| 441 |
+
"learning_rate": 3.962074173519047e-05,
|
| 442 |
+
"loss": 2.1252,
|
| 443 |
+
"step": 1240
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2114448733008894,
|
| 447 |
+
"grad_norm": 1.2433503866195679,
|
| 448 |
+
"learning_rate": 3.945292834368183e-05,
|
| 449 |
+
"loss": 2.1048,
|
| 450 |
+
"step": 1260
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.21480114113106225,
|
| 454 |
+
"grad_norm": 1.422275185585022,
|
| 455 |
+
"learning_rate": 3.928511495217319e-05,
|
| 456 |
+
"loss": 2.0829,
|
| 457 |
+
"step": 1280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.21815740896123512,
|
| 461 |
+
"grad_norm": 1.429117202758789,
|
| 462 |
+
"learning_rate": 3.911730156066454e-05,
|
| 463 |
+
"loss": 2.0545,
|
| 464 |
+
"step": 1300
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.22151367679140796,
|
| 468 |
+
"grad_norm": 1.40293550491333,
|
| 469 |
+
"learning_rate": 3.89494881691559e-05,
|
| 470 |
+
"loss": 2.0176,
|
| 471 |
+
"step": 1320
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2248699446215808,
|
| 475 |
+
"grad_norm": 1.2488166093826294,
|
| 476 |
+
"learning_rate": 3.878167477764726e-05,
|
| 477 |
+
"loss": 2.0846,
|
| 478 |
+
"step": 1340
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.22822621245175365,
|
| 482 |
+
"grad_norm": 1.2881702184677124,
|
| 483 |
+
"learning_rate": 3.861386138613862e-05,
|
| 484 |
+
"loss": 2.069,
|
| 485 |
+
"step": 1360
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2315824802819265,
|
| 489 |
+
"grad_norm": 1.3793479204177856,
|
| 490 |
+
"learning_rate": 3.8446047994629976e-05,
|
| 491 |
+
"loss": 2.0808,
|
| 492 |
+
"step": 1380
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.23493874811209933,
|
| 496 |
+
"grad_norm": 1.3845285177230835,
|
| 497 |
+
"learning_rate": 3.8278234603121335e-05,
|
| 498 |
+
"loss": 2.1008,
|
| 499 |
+
"step": 1400
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2382950159422722,
|
| 503 |
+
"grad_norm": 1.4609571695327759,
|
| 504 |
+
"learning_rate": 3.811042121161269e-05,
|
| 505 |
+
"loss": 2.085,
|
| 506 |
+
"step": 1420
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.24165128377244505,
|
| 510 |
+
"grad_norm": 1.3259775638580322,
|
| 511 |
+
"learning_rate": 3.7942607820104046e-05,
|
| 512 |
+
"loss": 2.0631,
|
| 513 |
+
"step": 1440
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2450075516026179,
|
| 517 |
+
"grad_norm": 1.248612880706787,
|
| 518 |
+
"learning_rate": 3.7774794428595405e-05,
|
| 519 |
+
"loss": 2.0908,
|
| 520 |
+
"step": 1460
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.24836381943279073,
|
| 524 |
+
"grad_norm": 1.4042353630065918,
|
| 525 |
+
"learning_rate": 3.7606981037086764e-05,
|
| 526 |
+
"loss": 2.0705,
|
| 527 |
+
"step": 1480
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2517200872629636,
|
| 531 |
+
"grad_norm": 1.2640959024429321,
|
| 532 |
+
"learning_rate": 3.743916764557812e-05,
|
| 533 |
+
"loss": 2.0626,
|
| 534 |
+
"step": 1500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2550763550931364,
|
| 538 |
+
"grad_norm": 1.355383038520813,
|
| 539 |
+
"learning_rate": 3.7271354254069475e-05,
|
| 540 |
+
"loss": 2.1025,
|
| 541 |
+
"step": 1520
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.25843262292330926,
|
| 545 |
+
"grad_norm": 1.2184921503067017,
|
| 546 |
+
"learning_rate": 3.7103540862560834e-05,
|
| 547 |
+
"loss": 2.0187,
|
| 548 |
+
"step": 1540
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2617888907534821,
|
| 552 |
+
"grad_norm": 1.2813303470611572,
|
| 553 |
+
"learning_rate": 3.6935727471052186e-05,
|
| 554 |
+
"loss": 2.087,
|
| 555 |
+
"step": 1560
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.26514515858365495,
|
| 559 |
+
"grad_norm": 1.3968662023544312,
|
| 560 |
+
"learning_rate": 3.6767914079543545e-05,
|
| 561 |
+
"loss": 2.0643,
|
| 562 |
+
"step": 1580
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.26850142641382785,
|
| 566 |
+
"grad_norm": 1.337203025817871,
|
| 567 |
+
"learning_rate": 3.6600100688034904e-05,
|
| 568 |
+
"loss": 2.0377,
|
| 569 |
+
"step": 1600
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2718576942440007,
|
| 573 |
+
"grad_norm": 1.3844518661499023,
|
| 574 |
+
"learning_rate": 3.643228729652626e-05,
|
| 575 |
+
"loss": 2.0807,
|
| 576 |
+
"step": 1620
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.27521396207417353,
|
| 580 |
+
"grad_norm": 1.787477731704712,
|
| 581 |
+
"learning_rate": 3.626447390501762e-05,
|
| 582 |
+
"loss": 2.0775,
|
| 583 |
+
"step": 1640
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.2785702299043464,
|
| 587 |
+
"grad_norm": 1.3247798681259155,
|
| 588 |
+
"learning_rate": 3.609666051350898e-05,
|
| 589 |
+
"loss": 2.0996,
|
| 590 |
+
"step": 1660
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2819264977345192,
|
| 594 |
+
"grad_norm": 1.2748069763183594,
|
| 595 |
+
"learning_rate": 3.592884712200033e-05,
|
| 596 |
+
"loss": 2.0737,
|
| 597 |
+
"step": 1680
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.28528276556469206,
|
| 601 |
+
"grad_norm": 1.346238374710083,
|
| 602 |
+
"learning_rate": 3.576103373049169e-05,
|
| 603 |
+
"loss": 2.1076,
|
| 604 |
+
"step": 1700
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2886390333948649,
|
| 608 |
+
"grad_norm": 1.3545295000076294,
|
| 609 |
+
"learning_rate": 3.559322033898305e-05,
|
| 610 |
+
"loss": 2.0663,
|
| 611 |
+
"step": 1720
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.29199530122503775,
|
| 615 |
+
"grad_norm": 1.3100897073745728,
|
| 616 |
+
"learning_rate": 3.542540694747441e-05,
|
| 617 |
+
"loss": 2.0872,
|
| 618 |
+
"step": 1740
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2953515690552106,
|
| 622 |
+
"grad_norm": 1.3519947528839111,
|
| 623 |
+
"learning_rate": 3.525759355596577e-05,
|
| 624 |
+
"loss": 2.0269,
|
| 625 |
+
"step": 1760
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.29870783688538344,
|
| 629 |
+
"grad_norm": 1.2966337203979492,
|
| 630 |
+
"learning_rate": 3.508978016445713e-05,
|
| 631 |
+
"loss": 2.0784,
|
| 632 |
+
"step": 1780
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3020641047155563,
|
| 636 |
+
"grad_norm": 1.3702917098999023,
|
| 637 |
+
"learning_rate": 3.492196677294848e-05,
|
| 638 |
+
"loss": 2.0906,
|
| 639 |
+
"step": 1800
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3054203725457291,
|
| 643 |
+
"grad_norm": 1.338409185409546,
|
| 644 |
+
"learning_rate": 3.475415338143984e-05,
|
| 645 |
+
"loss": 2.0899,
|
| 646 |
+
"step": 1820
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.308776640375902,
|
| 650 |
+
"grad_norm": 1.3070242404937744,
|
| 651 |
+
"learning_rate": 3.45863399899312e-05,
|
| 652 |
+
"loss": 2.0514,
|
| 653 |
+
"step": 1840
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.31213290820607487,
|
| 657 |
+
"grad_norm": 1.2753440141677856,
|
| 658 |
+
"learning_rate": 3.441852659842256e-05,
|
| 659 |
+
"loss": 2.0809,
|
| 660 |
+
"step": 1860
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3154891760362477,
|
| 664 |
+
"grad_norm": 1.3125340938568115,
|
| 665 |
+
"learning_rate": 3.4250713206913916e-05,
|
| 666 |
+
"loss": 2.0372,
|
| 667 |
+
"step": 1880
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.31884544386642055,
|
| 671 |
+
"grad_norm": 1.3050023317337036,
|
| 672 |
+
"learning_rate": 3.4082899815405275e-05,
|
| 673 |
+
"loss": 2.072,
|
| 674 |
+
"step": 1900
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3222017116965934,
|
| 678 |
+
"grad_norm": 1.4283207654953003,
|
| 679 |
+
"learning_rate": 3.391508642389663e-05,
|
| 680 |
+
"loss": 2.0489,
|
| 681 |
+
"step": 1920
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.32555797952676624,
|
| 685 |
+
"grad_norm": 1.4320636987686157,
|
| 686 |
+
"learning_rate": 3.3747273032387986e-05,
|
| 687 |
+
"loss": 2.1004,
|
| 688 |
+
"step": 1940
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3289142473569391,
|
| 692 |
+
"grad_norm": 1.3432669639587402,
|
| 693 |
+
"learning_rate": 3.3579459640879345e-05,
|
| 694 |
+
"loss": 2.0837,
|
| 695 |
+
"step": 1960
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3322705151871119,
|
| 699 |
+
"grad_norm": 1.22812819480896,
|
| 700 |
+
"learning_rate": 3.3411646249370704e-05,
|
| 701 |
+
"loss": 2.0651,
|
| 702 |
+
"step": 1980
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.33562678301728477,
|
| 706 |
+
"grad_norm": 1.3818988800048828,
|
| 707 |
+
"learning_rate": 3.324383285786206e-05,
|
| 708 |
+
"loss": 2.0446,
|
| 709 |
+
"step": 2000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3389830508474576,
|
| 713 |
+
"grad_norm": 1.2890549898147583,
|
| 714 |
+
"learning_rate": 3.3076019466353415e-05,
|
| 715 |
+
"loss": 2.0326,
|
| 716 |
+
"step": 2020
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.34233931867763046,
|
| 720 |
+
"grad_norm": 1.341260552406311,
|
| 721 |
+
"learning_rate": 3.2908206074844774e-05,
|
| 722 |
+
"loss": 2.0733,
|
| 723 |
+
"step": 2040
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3456955865078033,
|
| 727 |
+
"grad_norm": 1.2970625162124634,
|
| 728 |
+
"learning_rate": 3.274039268333613e-05,
|
| 729 |
+
"loss": 2.0749,
|
| 730 |
+
"step": 2060
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3490518543379762,
|
| 734 |
+
"grad_norm": 1.3300013542175293,
|
| 735 |
+
"learning_rate": 3.2572579291827485e-05,
|
| 736 |
+
"loss": 2.0283,
|
| 737 |
+
"step": 2080
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.35240812216814904,
|
| 741 |
+
"grad_norm": 1.372938871383667,
|
| 742 |
+
"learning_rate": 3.2404765900318844e-05,
|
| 743 |
+
"loss": 2.0703,
|
| 744 |
+
"step": 2100
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3557643899983219,
|
| 748 |
+
"grad_norm": 1.355000615119934,
|
| 749 |
+
"learning_rate": 3.22369525088102e-05,
|
| 750 |
+
"loss": 2.0641,
|
| 751 |
+
"step": 2120
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3591206578284947,
|
| 755 |
+
"grad_norm": 1.3288871049880981,
|
| 756 |
+
"learning_rate": 3.206913911730156e-05,
|
| 757 |
+
"loss": 2.0169,
|
| 758 |
+
"step": 2140
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.36247692565866757,
|
| 762 |
+
"grad_norm": 1.2701308727264404,
|
| 763 |
+
"learning_rate": 3.190132572579292e-05,
|
| 764 |
+
"loss": 2.1025,
|
| 765 |
+
"step": 2160
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3658331934888404,
|
| 769 |
+
"grad_norm": 1.3620569705963135,
|
| 770 |
+
"learning_rate": 3.173351233428428e-05,
|
| 771 |
+
"loss": 2.0621,
|
| 772 |
+
"step": 2180
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.36918946131901326,
|
| 776 |
+
"grad_norm": 1.1960408687591553,
|
| 777 |
+
"learning_rate": 3.156569894277563e-05,
|
| 778 |
+
"loss": 2.0452,
|
| 779 |
+
"step": 2200
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3725457291491861,
|
| 783 |
+
"grad_norm": 1.3412033319473267,
|
| 784 |
+
"learning_rate": 3.139788555126699e-05,
|
| 785 |
+
"loss": 2.0655,
|
| 786 |
+
"step": 2220
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.37590199697935894,
|
| 790 |
+
"grad_norm": 1.3558080196380615,
|
| 791 |
+
"learning_rate": 3.123007215975835e-05,
|
| 792 |
+
"loss": 2.0277,
|
| 793 |
+
"step": 2240
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.3792582648095318,
|
| 797 |
+
"grad_norm": 1.2877964973449707,
|
| 798 |
+
"learning_rate": 3.106225876824971e-05,
|
| 799 |
+
"loss": 2.0384,
|
| 800 |
+
"step": 2260
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.38261453263970463,
|
| 804 |
+
"grad_norm": 1.3294323682785034,
|
| 805 |
+
"learning_rate": 3.089444537674107e-05,
|
| 806 |
+
"loss": 1.9997,
|
| 807 |
+
"step": 2280
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.3859708004698775,
|
| 811 |
+
"grad_norm": 1.2727768421173096,
|
| 812 |
+
"learning_rate": 3.072663198523243e-05,
|
| 813 |
+
"loss": 1.9849,
|
| 814 |
+
"step": 2300
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3893270683000503,
|
| 818 |
+
"grad_norm": 1.2134991884231567,
|
| 819 |
+
"learning_rate": 3.055881859372378e-05,
|
| 820 |
+
"loss": 2.0725,
|
| 821 |
+
"step": 2320
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3926833361302232,
|
| 825 |
+
"grad_norm": 1.2667819261550903,
|
| 826 |
+
"learning_rate": 3.0391005202215138e-05,
|
| 827 |
+
"loss": 2.0429,
|
| 828 |
+
"step": 2340
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.39603960396039606,
|
| 832 |
+
"grad_norm": 1.3160877227783203,
|
| 833 |
+
"learning_rate": 3.0223191810706497e-05,
|
| 834 |
+
"loss": 2.0154,
|
| 835 |
+
"step": 2360
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3993958717905689,
|
| 839 |
+
"grad_norm": 1.2551796436309814,
|
| 840 |
+
"learning_rate": 3.0055378419197856e-05,
|
| 841 |
+
"loss": 2.0383,
|
| 842 |
+
"step": 2380
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.40275213962074174,
|
| 846 |
+
"grad_norm": 1.3122833967208862,
|
| 847 |
+
"learning_rate": 2.988756502768921e-05,
|
| 848 |
+
"loss": 2.0263,
|
| 849 |
+
"step": 2400
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4061084074509146,
|
| 853 |
+
"grad_norm": 1.269384503364563,
|
| 854 |
+
"learning_rate": 2.971975163618057e-05,
|
| 855 |
+
"loss": 2.1241,
|
| 856 |
+
"step": 2420
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.40946467528108743,
|
| 860 |
+
"grad_norm": 1.3302291631698608,
|
| 861 |
+
"learning_rate": 2.955193824467193e-05,
|
| 862 |
+
"loss": 2.0616,
|
| 863 |
+
"step": 2440
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4128209431112603,
|
| 867 |
+
"grad_norm": 1.2562564611434937,
|
| 868 |
+
"learning_rate": 2.9384124853163285e-05,
|
| 869 |
+
"loss": 2.0613,
|
| 870 |
+
"step": 2460
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4161772109414331,
|
| 874 |
+
"grad_norm": 1.2373179197311401,
|
| 875 |
+
"learning_rate": 2.9216311461654644e-05,
|
| 876 |
+
"loss": 2.0202,
|
| 877 |
+
"step": 2480
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.41953347877160596,
|
| 881 |
+
"grad_norm": 1.3266092538833618,
|
| 882 |
+
"learning_rate": 2.9048498070145996e-05,
|
| 883 |
+
"loss": 2.0174,
|
| 884 |
+
"step": 2500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4228897466017788,
|
| 888 |
+
"grad_norm": 1.27751624584198,
|
| 889 |
+
"learning_rate": 2.8880684678637355e-05,
|
| 890 |
+
"loss": 2.0549,
|
| 891 |
+
"step": 2520
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.42624601443195165,
|
| 895 |
+
"grad_norm": 1.2350589036941528,
|
| 896 |
+
"learning_rate": 2.871287128712871e-05,
|
| 897 |
+
"loss": 2.0126,
|
| 898 |
+
"step": 2540
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.4296022822621245,
|
| 902 |
+
"grad_norm": 1.3619853258132935,
|
| 903 |
+
"learning_rate": 2.854505789562007e-05,
|
| 904 |
+
"loss": 1.9725,
|
| 905 |
+
"step": 2560
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4329585500922974,
|
| 909 |
+
"grad_norm": 1.354687213897705,
|
| 910 |
+
"learning_rate": 2.837724450411143e-05,
|
| 911 |
+
"loss": 2.0492,
|
| 912 |
+
"step": 2580
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.43631481792247023,
|
| 916 |
+
"grad_norm": 1.3418916463851929,
|
| 917 |
+
"learning_rate": 2.8209431112602784e-05,
|
| 918 |
+
"loss": 1.976,
|
| 919 |
+
"step": 2600
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4396710857526431,
|
| 923 |
+
"grad_norm": 1.232704520225525,
|
| 924 |
+
"learning_rate": 2.8041617721094143e-05,
|
| 925 |
+
"loss": 2.0035,
|
| 926 |
+
"step": 2620
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4430273535828159,
|
| 930 |
+
"grad_norm": 1.3459244966506958,
|
| 931 |
+
"learning_rate": 2.7873804329585502e-05,
|
| 932 |
+
"loss": 2.0612,
|
| 933 |
+
"step": 2640
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.44638362141298876,
|
| 937 |
+
"grad_norm": 1.2673430442810059,
|
| 938 |
+
"learning_rate": 2.7705990938076858e-05,
|
| 939 |
+
"loss": 2.0134,
|
| 940 |
+
"step": 2660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4497398892431616,
|
| 944 |
+
"grad_norm": 1.229880928993225,
|
| 945 |
+
"learning_rate": 2.7538177546568216e-05,
|
| 946 |
+
"loss": 2.0324,
|
| 947 |
+
"step": 2680
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.45309615707333445,
|
| 951 |
+
"grad_norm": 1.3053526878356934,
|
| 952 |
+
"learning_rate": 2.7370364155059575e-05,
|
| 953 |
+
"loss": 2.0031,
|
| 954 |
+
"step": 2700
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4564524249035073,
|
| 958 |
+
"grad_norm": 1.3416264057159424,
|
| 959 |
+
"learning_rate": 2.720255076355093e-05,
|
| 960 |
+
"loss": 2.0513,
|
| 961 |
+
"step": 2720
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.45980869273368014,
|
| 965 |
+
"grad_norm": 1.3494229316711426,
|
| 966 |
+
"learning_rate": 2.703473737204229e-05,
|
| 967 |
+
"loss": 2.0527,
|
| 968 |
+
"step": 2740
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.463164960563853,
|
| 972 |
+
"grad_norm": 1.2861367464065552,
|
| 973 |
+
"learning_rate": 2.686692398053365e-05,
|
| 974 |
+
"loss": 2.0327,
|
| 975 |
+
"step": 2760
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.4665212283940258,
|
| 979 |
+
"grad_norm": 1.260968565940857,
|
| 980 |
+
"learning_rate": 2.6699110589025004e-05,
|
| 981 |
+
"loss": 2.0399,
|
| 982 |
+
"step": 2780
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.46987749622419867,
|
| 986 |
+
"grad_norm": 1.4496228694915771,
|
| 987 |
+
"learning_rate": 2.6531297197516363e-05,
|
| 988 |
+
"loss": 2.0532,
|
| 989 |
+
"step": 2800
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.47323376405437156,
|
| 993 |
+
"grad_norm": 1.2266364097595215,
|
| 994 |
+
"learning_rate": 2.6363483806007722e-05,
|
| 995 |
+
"loss": 2.0196,
|
| 996 |
+
"step": 2820
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4765900318845444,
|
| 1000 |
+
"grad_norm": 1.4289458990097046,
|
| 1001 |
+
"learning_rate": 2.6195670414499078e-05,
|
| 1002 |
+
"loss": 2.0379,
|
| 1003 |
+
"step": 2840
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.47994629971471725,
|
| 1007 |
+
"grad_norm": 1.3267526626586914,
|
| 1008 |
+
"learning_rate": 2.6027857022990437e-05,
|
| 1009 |
+
"loss": 2.0068,
|
| 1010 |
+
"step": 2860
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.4833025675448901,
|
| 1014 |
+
"grad_norm": 1.308477520942688,
|
| 1015 |
+
"learning_rate": 2.5860043631481796e-05,
|
| 1016 |
+
"loss": 2.0204,
|
| 1017 |
+
"step": 2880
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.48665883537506294,
|
| 1021 |
+
"grad_norm": 1.2613425254821777,
|
| 1022 |
+
"learning_rate": 2.569223023997315e-05,
|
| 1023 |
+
"loss": 2.0319,
|
| 1024 |
+
"step": 2900
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.4900151032052358,
|
| 1028 |
+
"grad_norm": 1.2851049900054932,
|
| 1029 |
+
"learning_rate": 2.552441684846451e-05,
|
| 1030 |
+
"loss": 2.044,
|
| 1031 |
+
"step": 2920
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.4933713710354086,
|
| 1035 |
+
"grad_norm": 1.2507027387619019,
|
| 1036 |
+
"learning_rate": 2.535660345695587e-05,
|
| 1037 |
+
"loss": 1.9964,
|
| 1038 |
+
"step": 2940
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.49672763886558147,
|
| 1042 |
+
"grad_norm": 1.3747198581695557,
|
| 1043 |
+
"learning_rate": 2.5188790065447225e-05,
|
| 1044 |
+
"loss": 1.9995,
|
| 1045 |
+
"step": 2960
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5000839066957543,
|
| 1049 |
+
"grad_norm": 1.2839736938476562,
|
| 1050 |
+
"learning_rate": 2.5020976673938584e-05,
|
| 1051 |
+
"loss": 2.0137,
|
| 1052 |
+
"step": 2980
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5034401745259272,
|
| 1056 |
+
"grad_norm": 1.428585410118103,
|
| 1057 |
+
"learning_rate": 2.485316328242994e-05,
|
| 1058 |
+
"loss": 2.0149,
|
| 1059 |
+
"step": 3000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5067964423561,
|
| 1063 |
+
"grad_norm": 1.3017884492874146,
|
| 1064 |
+
"learning_rate": 2.46853498909213e-05,
|
| 1065 |
+
"loss": 2.0023,
|
| 1066 |
+
"step": 3020
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5101527101862728,
|
| 1070 |
+
"grad_norm": 1.2209047079086304,
|
| 1071 |
+
"learning_rate": 2.4517536499412654e-05,
|
| 1072 |
+
"loss": 2.0092,
|
| 1073 |
+
"step": 3040
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5135089780164457,
|
| 1077 |
+
"grad_norm": 1.2542091608047485,
|
| 1078 |
+
"learning_rate": 2.4349723107904013e-05,
|
| 1079 |
+
"loss": 2.0137,
|
| 1080 |
+
"step": 3060
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5168652458466185,
|
| 1084 |
+
"grad_norm": 1.1981834173202515,
|
| 1085 |
+
"learning_rate": 2.4181909716395372e-05,
|
| 1086 |
+
"loss": 2.0013,
|
| 1087 |
+
"step": 3080
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5202215136767914,
|
| 1091 |
+
"grad_norm": 1.3641618490219116,
|
| 1092 |
+
"learning_rate": 2.4014096324886727e-05,
|
| 1093 |
+
"loss": 2.0412,
|
| 1094 |
+
"step": 3100
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5235777815069642,
|
| 1098 |
+
"grad_norm": 1.2675738334655762,
|
| 1099 |
+
"learning_rate": 2.3846282933378083e-05,
|
| 1100 |
+
"loss": 2.0088,
|
| 1101 |
+
"step": 3120
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5269340493371371,
|
| 1105 |
+
"grad_norm": 1.2499597072601318,
|
| 1106 |
+
"learning_rate": 2.3678469541869442e-05,
|
| 1107 |
+
"loss": 1.9926,
|
| 1108 |
+
"step": 3140
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5302903171673099,
|
| 1112 |
+
"grad_norm": 1.260705828666687,
|
| 1113 |
+
"learning_rate": 2.3510656150360798e-05,
|
| 1114 |
+
"loss": 1.9775,
|
| 1115 |
+
"step": 3160
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5336465849974829,
|
| 1119 |
+
"grad_norm": 1.2815592288970947,
|
| 1120 |
+
"learning_rate": 2.3342842758852156e-05,
|
| 1121 |
+
"loss": 1.9964,
|
| 1122 |
+
"step": 3180
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5370028528276557,
|
| 1126 |
+
"grad_norm": 1.191841721534729,
|
| 1127 |
+
"learning_rate": 2.3175029367343515e-05,
|
| 1128 |
+
"loss": 1.9858,
|
| 1129 |
+
"step": 3200
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5403591206578285,
|
| 1133 |
+
"grad_norm": 1.2903523445129395,
|
| 1134 |
+
"learning_rate": 2.300721597583487e-05,
|
| 1135 |
+
"loss": 2.0101,
|
| 1136 |
+
"step": 3220
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5437153884880014,
|
| 1140 |
+
"grad_norm": 1.257596731185913,
|
| 1141 |
+
"learning_rate": 2.283940258432623e-05,
|
| 1142 |
+
"loss": 1.9916,
|
| 1143 |
+
"step": 3240
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5470716563181742,
|
| 1147 |
+
"grad_norm": 1.2740247249603271,
|
| 1148 |
+
"learning_rate": 2.267158919281759e-05,
|
| 1149 |
+
"loss": 1.9724,
|
| 1150 |
+
"step": 3260
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5504279241483471,
|
| 1154 |
+
"grad_norm": 1.2834815979003906,
|
| 1155 |
+
"learning_rate": 2.2503775801308944e-05,
|
| 1156 |
+
"loss": 2.0144,
|
| 1157 |
+
"step": 3280
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5537841919785199,
|
| 1161 |
+
"grad_norm": 1.2792032957077026,
|
| 1162 |
+
"learning_rate": 2.2335962409800303e-05,
|
| 1163 |
+
"loss": 2.0241,
|
| 1164 |
+
"step": 3300
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5571404598086928,
|
| 1168 |
+
"grad_norm": 1.3966797590255737,
|
| 1169 |
+
"learning_rate": 2.2168149018291662e-05,
|
| 1170 |
+
"loss": 2.0202,
|
| 1171 |
+
"step": 3320
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5604967276388656,
|
| 1175 |
+
"grad_norm": 1.2389239072799683,
|
| 1176 |
+
"learning_rate": 2.2000335626783018e-05,
|
| 1177 |
+
"loss": 2.0221,
|
| 1178 |
+
"step": 3340
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5638529954690384,
|
| 1182 |
+
"grad_norm": 1.2616690397262573,
|
| 1183 |
+
"learning_rate": 2.1832522235274374e-05,
|
| 1184 |
+
"loss": 2.0266,
|
| 1185 |
+
"step": 3360
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5672092632992113,
|
| 1189 |
+
"grad_norm": 1.2490557432174683,
|
| 1190 |
+
"learning_rate": 2.1664708843765732e-05,
|
| 1191 |
+
"loss": 2.011,
|
| 1192 |
+
"step": 3380
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5705655311293841,
|
| 1196 |
+
"grad_norm": 1.2576720714569092,
|
| 1197 |
+
"learning_rate": 2.149689545225709e-05,
|
| 1198 |
+
"loss": 2.0368,
|
| 1199 |
+
"step": 3400
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.573921798959557,
|
| 1203 |
+
"grad_norm": 1.2379933595657349,
|
| 1204 |
+
"learning_rate": 2.1329082060748447e-05,
|
| 1205 |
+
"loss": 1.9964,
|
| 1206 |
+
"step": 3420
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5772780667897298,
|
| 1210 |
+
"grad_norm": 1.259509801864624,
|
| 1211 |
+
"learning_rate": 2.1161268669239806e-05,
|
| 1212 |
+
"loss": 1.9827,
|
| 1213 |
+
"step": 3440
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5806343346199027,
|
| 1217 |
+
"grad_norm": 1.2831110954284668,
|
| 1218 |
+
"learning_rate": 2.0993455277731165e-05,
|
| 1219 |
+
"loss": 2.0438,
|
| 1220 |
+
"step": 3460
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5839906024500755,
|
| 1224 |
+
"grad_norm": 1.331018090248108,
|
| 1225 |
+
"learning_rate": 2.082564188622252e-05,
|
| 1226 |
+
"loss": 2.0088,
|
| 1227 |
+
"step": 3480
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5873468702802483,
|
| 1231 |
+
"grad_norm": 1.2574256658554077,
|
| 1232 |
+
"learning_rate": 2.065782849471388e-05,
|
| 1233 |
+
"loss": 2.0474,
|
| 1234 |
+
"step": 3500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5907031381104212,
|
| 1238 |
+
"grad_norm": 1.283894658088684,
|
| 1239 |
+
"learning_rate": 2.049001510320524e-05,
|
| 1240 |
+
"loss": 2.0543,
|
| 1241 |
+
"step": 3520
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.594059405940594,
|
| 1245 |
+
"grad_norm": 1.2675976753234863,
|
| 1246 |
+
"learning_rate": 2.0322201711696594e-05,
|
| 1247 |
+
"loss": 2.0182,
|
| 1248 |
+
"step": 3540
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5974156737707669,
|
| 1252 |
+
"grad_norm": 1.2579065561294556,
|
| 1253 |
+
"learning_rate": 2.0154388320187953e-05,
|
| 1254 |
+
"loss": 1.9765,
|
| 1255 |
+
"step": 3560
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6007719416009397,
|
| 1259 |
+
"grad_norm": 1.2290009260177612,
|
| 1260 |
+
"learning_rate": 1.9986574928679312e-05,
|
| 1261 |
+
"loss": 1.9856,
|
| 1262 |
+
"step": 3580
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6041282094311126,
|
| 1266 |
+
"grad_norm": 1.3220741748809814,
|
| 1267 |
+
"learning_rate": 1.9818761537170667e-05,
|
| 1268 |
+
"loss": 1.9789,
|
| 1269 |
+
"step": 3600
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6074844772612854,
|
| 1273 |
+
"grad_norm": 1.3620175123214722,
|
| 1274 |
+
"learning_rate": 1.9650948145662023e-05,
|
| 1275 |
+
"loss": 1.973,
|
| 1276 |
+
"step": 3620
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6108407450914582,
|
| 1280 |
+
"grad_norm": 1.3105921745300293,
|
| 1281 |
+
"learning_rate": 1.9483134754153382e-05,
|
| 1282 |
+
"loss": 2.0177,
|
| 1283 |
+
"step": 3640
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6141970129216311,
|
| 1287 |
+
"grad_norm": 1.303818941116333,
|
| 1288 |
+
"learning_rate": 1.931532136264474e-05,
|
| 1289 |
+
"loss": 2.0125,
|
| 1290 |
+
"step": 3660
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.617553280751804,
|
| 1294 |
+
"grad_norm": 1.2042673826217651,
|
| 1295 |
+
"learning_rate": 1.9147507971136096e-05,
|
| 1296 |
+
"loss": 2.0166,
|
| 1297 |
+
"step": 3680
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6209095485819769,
|
| 1301 |
+
"grad_norm": 1.2778286933898926,
|
| 1302 |
+
"learning_rate": 1.8979694579627455e-05,
|
| 1303 |
+
"loss": 1.9743,
|
| 1304 |
+
"step": 3700
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6242658164121497,
|
| 1308 |
+
"grad_norm": 1.3580365180969238,
|
| 1309 |
+
"learning_rate": 1.8811881188118814e-05,
|
| 1310 |
+
"loss": 2.0165,
|
| 1311 |
+
"step": 3720
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6276220842423226,
|
| 1315 |
+
"grad_norm": 1.2921738624572754,
|
| 1316 |
+
"learning_rate": 1.864406779661017e-05,
|
| 1317 |
+
"loss": 1.96,
|
| 1318 |
+
"step": 3740
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6309783520724954,
|
| 1322 |
+
"grad_norm": 1.347383737564087,
|
| 1323 |
+
"learning_rate": 1.847625440510153e-05,
|
| 1324 |
+
"loss": 1.9895,
|
| 1325 |
+
"step": 3760
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6343346199026683,
|
| 1329 |
+
"grad_norm": 1.2058087587356567,
|
| 1330 |
+
"learning_rate": 1.8308441013592888e-05,
|
| 1331 |
+
"loss": 1.9735,
|
| 1332 |
+
"step": 3780
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6376908877328411,
|
| 1336 |
+
"grad_norm": 1.3013418912887573,
|
| 1337 |
+
"learning_rate": 1.8140627622084243e-05,
|
| 1338 |
+
"loss": 1.96,
|
| 1339 |
+
"step": 3800
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.641047155563014,
|
| 1343 |
+
"grad_norm": 1.2196553945541382,
|
| 1344 |
+
"learning_rate": 1.7972814230575602e-05,
|
| 1345 |
+
"loss": 1.9808,
|
| 1346 |
+
"step": 3820
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6444034233931868,
|
| 1350 |
+
"grad_norm": 1.2661124467849731,
|
| 1351 |
+
"learning_rate": 1.7805000839066958e-05,
|
| 1352 |
+
"loss": 2.0022,
|
| 1353 |
+
"step": 3840
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6477596912233596,
|
| 1357 |
+
"grad_norm": 1.3183765411376953,
|
| 1358 |
+
"learning_rate": 1.7637187447558313e-05,
|
| 1359 |
+
"loss": 1.9696,
|
| 1360 |
+
"step": 3860
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6511159590535325,
|
| 1364 |
+
"grad_norm": 1.2164355516433716,
|
| 1365 |
+
"learning_rate": 1.7469374056049672e-05,
|
| 1366 |
+
"loss": 1.9387,
|
| 1367 |
+
"step": 3880
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6544722268837053,
|
| 1371 |
+
"grad_norm": 1.2561441659927368,
|
| 1372 |
+
"learning_rate": 1.730156066454103e-05,
|
| 1373 |
+
"loss": 2.0252,
|
| 1374 |
+
"step": 3900
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6578284947138782,
|
| 1378 |
+
"grad_norm": 1.240861177444458,
|
| 1379 |
+
"learning_rate": 1.7133747273032387e-05,
|
| 1380 |
+
"loss": 1.9737,
|
| 1381 |
+
"step": 3920
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.661184762544051,
|
| 1385 |
+
"grad_norm": 1.2271203994750977,
|
| 1386 |
+
"learning_rate": 1.6965933881523746e-05,
|
| 1387 |
+
"loss": 1.991,
|
| 1388 |
+
"step": 3940
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.6645410303742239,
|
| 1392 |
+
"grad_norm": 1.224165678024292,
|
| 1393 |
+
"learning_rate": 1.6798120490015105e-05,
|
| 1394 |
+
"loss": 1.9771,
|
| 1395 |
+
"step": 3960
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.6678972982043967,
|
| 1399 |
+
"grad_norm": 1.222956657409668,
|
| 1400 |
+
"learning_rate": 1.663030709850646e-05,
|
| 1401 |
+
"loss": 1.9659,
|
| 1402 |
+
"step": 3980
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6712535660345695,
|
| 1406 |
+
"grad_norm": 1.2539464235305786,
|
| 1407 |
+
"learning_rate": 1.646249370699782e-05,
|
| 1408 |
+
"loss": 1.9811,
|
| 1409 |
+
"step": 4000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6746098338647424,
|
| 1413 |
+
"grad_norm": 1.272801160812378,
|
| 1414 |
+
"learning_rate": 1.629468031548918e-05,
|
| 1415 |
+
"loss": 2.0171,
|
| 1416 |
+
"step": 4020
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6779661016949152,
|
| 1420 |
+
"grad_norm": 1.2539173364639282,
|
| 1421 |
+
"learning_rate": 1.6126866923980534e-05,
|
| 1422 |
+
"loss": 1.953,
|
| 1423 |
+
"step": 4040
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6813223695250881,
|
| 1427 |
+
"grad_norm": 1.2918198108673096,
|
| 1428 |
+
"learning_rate": 1.5959053532471893e-05,
|
| 1429 |
+
"loss": 1.9742,
|
| 1430 |
+
"step": 4060
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6846786373552609,
|
| 1434 |
+
"grad_norm": 1.3085472583770752,
|
| 1435 |
+
"learning_rate": 1.5791240140963252e-05,
|
| 1436 |
+
"loss": 1.9693,
|
| 1437 |
+
"step": 4080
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6880349051854338,
|
| 1441 |
+
"grad_norm": 1.2754298448562622,
|
| 1442 |
+
"learning_rate": 1.5623426749454607e-05,
|
| 1443 |
+
"loss": 1.9819,
|
| 1444 |
+
"step": 4100
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6913911730156066,
|
| 1448 |
+
"grad_norm": 1.3267842531204224,
|
| 1449 |
+
"learning_rate": 1.5455613357945963e-05,
|
| 1450 |
+
"loss": 1.9549,
|
| 1451 |
+
"step": 4120
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6947474408457794,
|
| 1455 |
+
"grad_norm": 1.1842241287231445,
|
| 1456 |
+
"learning_rate": 1.5287799966437322e-05,
|
| 1457 |
+
"loss": 1.9816,
|
| 1458 |
+
"step": 4140
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6981037086759524,
|
| 1462 |
+
"grad_norm": 1.2266135215759277,
|
| 1463 |
+
"learning_rate": 1.511998657492868e-05,
|
| 1464 |
+
"loss": 1.956,
|
| 1465 |
+
"step": 4160
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7014599765061252,
|
| 1469 |
+
"grad_norm": 1.2384594678878784,
|
| 1470 |
+
"learning_rate": 1.4952173183420038e-05,
|
| 1471 |
+
"loss": 1.9761,
|
| 1472 |
+
"step": 4180
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7048162443362981,
|
| 1476 |
+
"grad_norm": 1.2703733444213867,
|
| 1477 |
+
"learning_rate": 1.4784359791911395e-05,
|
| 1478 |
+
"loss": 1.9602,
|
| 1479 |
+
"step": 4200
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7081725121664709,
|
| 1483 |
+
"grad_norm": 1.3347383737564087,
|
| 1484 |
+
"learning_rate": 1.4616546400402753e-05,
|
| 1485 |
+
"loss": 1.9863,
|
| 1486 |
+
"step": 4220
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7115287799966438,
|
| 1490 |
+
"grad_norm": 1.2556087970733643,
|
| 1491 |
+
"learning_rate": 1.4448733008894112e-05,
|
| 1492 |
+
"loss": 1.9937,
|
| 1493 |
+
"step": 4240
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7148850478268166,
|
| 1497 |
+
"grad_norm": 1.2546402215957642,
|
| 1498 |
+
"learning_rate": 1.4280919617385469e-05,
|
| 1499 |
+
"loss": 1.9428,
|
| 1500 |
+
"step": 4260
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7182413156569895,
|
| 1504 |
+
"grad_norm": 1.20867121219635,
|
| 1505 |
+
"learning_rate": 1.4113106225876826e-05,
|
| 1506 |
+
"loss": 1.9702,
|
| 1507 |
+
"step": 4280
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7215975834871623,
|
| 1511 |
+
"grad_norm": 1.2617672681808472,
|
| 1512 |
+
"learning_rate": 1.3945292834368185e-05,
|
| 1513 |
+
"loss": 1.9688,
|
| 1514 |
+
"step": 4300
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7249538513173351,
|
| 1518 |
+
"grad_norm": 1.306674599647522,
|
| 1519 |
+
"learning_rate": 1.3777479442859542e-05,
|
| 1520 |
+
"loss": 1.9497,
|
| 1521 |
+
"step": 4320
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.728310119147508,
|
| 1525 |
+
"grad_norm": 1.3713186979293823,
|
| 1526 |
+
"learning_rate": 1.3609666051350898e-05,
|
| 1527 |
+
"loss": 2.0179,
|
| 1528 |
+
"step": 4340
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7316663869776808,
|
| 1532 |
+
"grad_norm": 1.1903387308120728,
|
| 1533 |
+
"learning_rate": 1.3441852659842255e-05,
|
| 1534 |
+
"loss": 1.9861,
|
| 1535 |
+
"step": 4360
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7350226548078537,
|
| 1539 |
+
"grad_norm": 1.2360427379608154,
|
| 1540 |
+
"learning_rate": 1.3274039268333612e-05,
|
| 1541 |
+
"loss": 1.9742,
|
| 1542 |
+
"step": 4380
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7383789226380265,
|
| 1546 |
+
"grad_norm": 1.1744493246078491,
|
| 1547 |
+
"learning_rate": 1.3106225876824971e-05,
|
| 1548 |
+
"loss": 2.0021,
|
| 1549 |
+
"step": 4400
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7417351904681994,
|
| 1553 |
+
"grad_norm": 1.25652015209198,
|
| 1554 |
+
"learning_rate": 1.2938412485316329e-05,
|
| 1555 |
+
"loss": 2.0034,
|
| 1556 |
+
"step": 4420
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7450914582983722,
|
| 1560 |
+
"grad_norm": 1.2795733213424683,
|
| 1561 |
+
"learning_rate": 1.2770599093807686e-05,
|
| 1562 |
+
"loss": 2.0015,
|
| 1563 |
+
"step": 4440
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.748447726128545,
|
| 1567 |
+
"grad_norm": 1.2552342414855957,
|
| 1568 |
+
"learning_rate": 1.2602785702299045e-05,
|
| 1569 |
+
"loss": 1.9703,
|
| 1570 |
+
"step": 4460
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7518039939587179,
|
| 1574 |
+
"grad_norm": 1.196215271949768,
|
| 1575 |
+
"learning_rate": 1.2434972310790402e-05,
|
| 1576 |
+
"loss": 1.9434,
|
| 1577 |
+
"step": 4480
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.7551602617888907,
|
| 1581 |
+
"grad_norm": 1.2752282619476318,
|
| 1582 |
+
"learning_rate": 1.226715891928176e-05,
|
| 1583 |
+
"loss": 2.0011,
|
| 1584 |
+
"step": 4500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.7585165296190636,
|
| 1588 |
+
"grad_norm": 1.361426591873169,
|
| 1589 |
+
"learning_rate": 1.2099345527773117e-05,
|
| 1590 |
+
"loss": 2.001,
|
| 1591 |
+
"step": 4520
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.7618727974492364,
|
| 1595 |
+
"grad_norm": 1.295401930809021,
|
| 1596 |
+
"learning_rate": 1.1931532136264474e-05,
|
| 1597 |
+
"loss": 1.9785,
|
| 1598 |
+
"step": 4540
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.7652290652794093,
|
| 1602 |
+
"grad_norm": 1.1889725923538208,
|
| 1603 |
+
"learning_rate": 1.1763718744755831e-05,
|
| 1604 |
+
"loss": 2.0203,
|
| 1605 |
+
"step": 4560
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.7685853331095821,
|
| 1609 |
+
"grad_norm": 1.2194198369979858,
|
| 1610 |
+
"learning_rate": 1.159590535324719e-05,
|
| 1611 |
+
"loss": 1.954,
|
| 1612 |
+
"step": 4580
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.771941600939755,
|
| 1616 |
+
"grad_norm": 1.1998612880706787,
|
| 1617 |
+
"learning_rate": 1.1428091961738547e-05,
|
| 1618 |
+
"loss": 1.9129,
|
| 1619 |
+
"step": 4600
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.7752978687699278,
|
| 1623 |
+
"grad_norm": 1.1526966094970703,
|
| 1624 |
+
"learning_rate": 1.1260278570229905e-05,
|
| 1625 |
+
"loss": 1.941,
|
| 1626 |
+
"step": 4620
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7786541366001006,
|
| 1630 |
+
"grad_norm": 1.2833491563796997,
|
| 1631 |
+
"learning_rate": 1.1092465178721262e-05,
|
| 1632 |
+
"loss": 1.9606,
|
| 1633 |
+
"step": 4640
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7820104044302736,
|
| 1637 |
+
"grad_norm": 1.3405060768127441,
|
| 1638 |
+
"learning_rate": 1.0924651787212619e-05,
|
| 1639 |
+
"loss": 1.9828,
|
| 1640 |
+
"step": 4660
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7853666722604464,
|
| 1644 |
+
"grad_norm": 1.2940865755081177,
|
| 1645 |
+
"learning_rate": 1.0756838395703978e-05,
|
| 1646 |
+
"loss": 1.9309,
|
| 1647 |
+
"step": 4680
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7887229400906193,
|
| 1651 |
+
"grad_norm": 1.3541054725646973,
|
| 1652 |
+
"learning_rate": 1.0589025004195335e-05,
|
| 1653 |
+
"loss": 1.9213,
|
| 1654 |
+
"step": 4700
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7920792079207921,
|
| 1658 |
+
"grad_norm": 1.2736974954605103,
|
| 1659 |
+
"learning_rate": 1.0421211612686693e-05,
|
| 1660 |
+
"loss": 1.9526,
|
| 1661 |
+
"step": 4720
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.795435475750965,
|
| 1665 |
+
"grad_norm": 1.2431901693344116,
|
| 1666 |
+
"learning_rate": 1.0253398221178052e-05,
|
| 1667 |
+
"loss": 1.9348,
|
| 1668 |
+
"step": 4740
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7987917435811378,
|
| 1672 |
+
"grad_norm": 1.245354175567627,
|
| 1673 |
+
"learning_rate": 1.0085584829669407e-05,
|
| 1674 |
+
"loss": 1.9294,
|
| 1675 |
+
"step": 4760
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8021480114113106,
|
| 1679 |
+
"grad_norm": 1.2285710573196411,
|
| 1680 |
+
"learning_rate": 9.917771438160766e-06,
|
| 1681 |
+
"loss": 1.9766,
|
| 1682 |
+
"step": 4780
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8055042792414835,
|
| 1686 |
+
"grad_norm": 1.1929337978363037,
|
| 1687 |
+
"learning_rate": 9.749958046652123e-06,
|
| 1688 |
+
"loss": 1.9265,
|
| 1689 |
+
"step": 4800
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8088605470716563,
|
| 1693 |
+
"grad_norm": 1.3306503295898438,
|
| 1694 |
+
"learning_rate": 9.58214465514348e-06,
|
| 1695 |
+
"loss": 1.966,
|
| 1696 |
+
"step": 4820
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8122168149018292,
|
| 1700 |
+
"grad_norm": 1.203856110572815,
|
| 1701 |
+
"learning_rate": 9.41433126363484e-06,
|
| 1702 |
+
"loss": 1.9775,
|
| 1703 |
+
"step": 4840
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.815573082732002,
|
| 1707 |
+
"grad_norm": 1.2907806634902954,
|
| 1708 |
+
"learning_rate": 9.246517872126197e-06,
|
| 1709 |
+
"loss": 1.9544,
|
| 1710 |
+
"step": 4860
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8189293505621749,
|
| 1714 |
+
"grad_norm": 1.36453378200531,
|
| 1715 |
+
"learning_rate": 9.078704480617552e-06,
|
| 1716 |
+
"loss": 1.935,
|
| 1717 |
+
"step": 4880
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8222856183923477,
|
| 1721 |
+
"grad_norm": 1.228378415107727,
|
| 1722 |
+
"learning_rate": 8.910891089108911e-06,
|
| 1723 |
+
"loss": 1.9942,
|
| 1724 |
+
"step": 4900
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8256418862225205,
|
| 1728 |
+
"grad_norm": 1.2804518938064575,
|
| 1729 |
+
"learning_rate": 8.743077697600269e-06,
|
| 1730 |
+
"loss": 1.9865,
|
| 1731 |
+
"step": 4920
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8289981540526934,
|
| 1735 |
+
"grad_norm": 1.2092355489730835,
|
| 1736 |
+
"learning_rate": 8.575264306091626e-06,
|
| 1737 |
+
"loss": 1.9931,
|
| 1738 |
+
"step": 4940
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8323544218828662,
|
| 1742 |
+
"grad_norm": 1.248852252960205,
|
| 1743 |
+
"learning_rate": 8.407450914582985e-06,
|
| 1744 |
+
"loss": 1.9831,
|
| 1745 |
+
"step": 4960
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8357106897130391,
|
| 1749 |
+
"grad_norm": 1.3584181070327759,
|
| 1750 |
+
"learning_rate": 8.239637523074342e-06,
|
| 1751 |
+
"loss": 1.9706,
|
| 1752 |
+
"step": 4980
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8390669575432119,
|
| 1756 |
+
"grad_norm": 1.250568151473999,
|
| 1757 |
+
"learning_rate": 8.0718241315657e-06,
|
| 1758 |
+
"loss": 1.9449,
|
| 1759 |
+
"step": 5000
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8424232253733848,
|
| 1763 |
+
"grad_norm": 1.3448213338851929,
|
| 1764 |
+
"learning_rate": 7.904010740057057e-06,
|
| 1765 |
+
"loss": 1.9221,
|
| 1766 |
+
"step": 5020
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8457794932035576,
|
| 1770 |
+
"grad_norm": 1.2519199848175049,
|
| 1771 |
+
"learning_rate": 7.736197348548414e-06,
|
| 1772 |
+
"loss": 1.9444,
|
| 1773 |
+
"step": 5040
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8491357610337305,
|
| 1777 |
+
"grad_norm": 1.2377424240112305,
|
| 1778 |
+
"learning_rate": 7.568383957039772e-06,
|
| 1779 |
+
"loss": 1.9438,
|
| 1780 |
+
"step": 5060
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8524920288639033,
|
| 1784 |
+
"grad_norm": 1.3130027055740356,
|
| 1785 |
+
"learning_rate": 7.40057056553113e-06,
|
| 1786 |
+
"loss": 1.9791,
|
| 1787 |
+
"step": 5080
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8558482966940761,
|
| 1791 |
+
"grad_norm": 1.2841044664382935,
|
| 1792 |
+
"learning_rate": 7.232757174022488e-06,
|
| 1793 |
+
"loss": 1.9539,
|
| 1794 |
+
"step": 5100
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.859204564524249,
|
| 1798 |
+
"grad_norm": 1.2105425596237183,
|
| 1799 |
+
"learning_rate": 7.064943782513845e-06,
|
| 1800 |
+
"loss": 1.9738,
|
| 1801 |
+
"step": 5120
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8625608323544219,
|
| 1805 |
+
"grad_norm": 1.2560657262802124,
|
| 1806 |
+
"learning_rate": 6.897130391005202e-06,
|
| 1807 |
+
"loss": 1.9486,
|
| 1808 |
+
"step": 5140
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8659171001845948,
|
| 1812 |
+
"grad_norm": 1.2118816375732422,
|
| 1813 |
+
"learning_rate": 6.72931699949656e-06,
|
| 1814 |
+
"loss": 1.9107,
|
| 1815 |
+
"step": 5160
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8692733680147676,
|
| 1819 |
+
"grad_norm": 1.2164198160171509,
|
| 1820 |
+
"learning_rate": 6.561503607987918e-06,
|
| 1821 |
+
"loss": 2.0168,
|
| 1822 |
+
"step": 5180
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8726296358449405,
|
| 1826 |
+
"grad_norm": 1.358729362487793,
|
| 1827 |
+
"learning_rate": 6.393690216479275e-06,
|
| 1828 |
+
"loss": 1.9401,
|
| 1829 |
+
"step": 5200
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8759859036751133,
|
| 1833 |
+
"grad_norm": 1.2905333042144775,
|
| 1834 |
+
"learning_rate": 6.225876824970633e-06,
|
| 1835 |
+
"loss": 1.9235,
|
| 1836 |
+
"step": 5220
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8793421715052862,
|
| 1840 |
+
"grad_norm": 1.2965201139450073,
|
| 1841 |
+
"learning_rate": 6.058063433461991e-06,
|
| 1842 |
+
"loss": 1.9166,
|
| 1843 |
+
"step": 5240
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.882698439335459,
|
| 1847 |
+
"grad_norm": 1.3677794933319092,
|
| 1848 |
+
"learning_rate": 5.890250041953348e-06,
|
| 1849 |
+
"loss": 1.9327,
|
| 1850 |
+
"step": 5260
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8860547071656318,
|
| 1854 |
+
"grad_norm": 1.312015175819397,
|
| 1855 |
+
"learning_rate": 5.722436650444705e-06,
|
| 1856 |
+
"loss": 1.9032,
|
| 1857 |
+
"step": 5280
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8894109749958047,
|
| 1861 |
+
"grad_norm": 1.2666916847229004,
|
| 1862 |
+
"learning_rate": 5.554623258936063e-06,
|
| 1863 |
+
"loss": 1.9014,
|
| 1864 |
+
"step": 5300
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8927672428259775,
|
| 1868 |
+
"grad_norm": 1.2646892070770264,
|
| 1869 |
+
"learning_rate": 5.3868098674274214e-06,
|
| 1870 |
+
"loss": 1.9586,
|
| 1871 |
+
"step": 5320
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8961235106561504,
|
| 1875 |
+
"grad_norm": 1.2514413595199585,
|
| 1876 |
+
"learning_rate": 5.218996475918779e-06,
|
| 1877 |
+
"loss": 1.959,
|
| 1878 |
+
"step": 5340
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8994797784863232,
|
| 1882 |
+
"grad_norm": 1.28076171875,
|
| 1883 |
+
"learning_rate": 5.051183084410136e-06,
|
| 1884 |
+
"loss": 1.9393,
|
| 1885 |
+
"step": 5360
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.902836046316496,
|
| 1889 |
+
"grad_norm": 1.3480361700057983,
|
| 1890 |
+
"learning_rate": 4.883369692901494e-06,
|
| 1891 |
+
"loss": 1.9074,
|
| 1892 |
+
"step": 5380
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9061923141466689,
|
| 1896 |
+
"grad_norm": 1.3494782447814941,
|
| 1897 |
+
"learning_rate": 4.715556301392851e-06,
|
| 1898 |
+
"loss": 1.9162,
|
| 1899 |
+
"step": 5400
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9095485819768417,
|
| 1903 |
+
"grad_norm": 1.3703207969665527,
|
| 1904 |
+
"learning_rate": 4.547742909884209e-06,
|
| 1905 |
+
"loss": 1.9728,
|
| 1906 |
+
"step": 5420
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9129048498070146,
|
| 1910 |
+
"grad_norm": 1.2115719318389893,
|
| 1911 |
+
"learning_rate": 4.379929518375567e-06,
|
| 1912 |
+
"loss": 1.9381,
|
| 1913 |
+
"step": 5440
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.9162611176371874,
|
| 1917 |
+
"grad_norm": 1.2599093914031982,
|
| 1918 |
+
"learning_rate": 4.212116126866924e-06,
|
| 1919 |
+
"loss": 1.9405,
|
| 1920 |
+
"step": 5460
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9196173854673603,
|
| 1924 |
+
"grad_norm": 1.1931716203689575,
|
| 1925 |
+
"learning_rate": 4.044302735358282e-06,
|
| 1926 |
+
"loss": 1.9205,
|
| 1927 |
+
"step": 5480
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.9229736532975331,
|
| 1931 |
+
"grad_norm": 1.2630369663238525,
|
| 1932 |
+
"learning_rate": 3.876489343849639e-06,
|
| 1933 |
+
"loss": 1.9257,
|
| 1934 |
+
"step": 5500
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.926329921127706,
|
| 1938 |
+
"grad_norm": 1.26536226272583,
|
| 1939 |
+
"learning_rate": 3.7086759523409966e-06,
|
| 1940 |
+
"loss": 1.9136,
|
| 1941 |
+
"step": 5520
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9296861889578788,
|
| 1945 |
+
"grad_norm": 1.25338876247406,
|
| 1946 |
+
"learning_rate": 3.5408625608323547e-06,
|
| 1947 |
+
"loss": 1.875,
|
| 1948 |
+
"step": 5540
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9330424567880516,
|
| 1952 |
+
"grad_norm": 1.2348542213439941,
|
| 1953 |
+
"learning_rate": 3.3730491693237124e-06,
|
| 1954 |
+
"loss": 1.959,
|
| 1955 |
+
"step": 5560
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.9363987246182245,
|
| 1959 |
+
"grad_norm": 1.3011400699615479,
|
| 1960 |
+
"learning_rate": 3.2052357778150696e-06,
|
| 1961 |
+
"loss": 1.9571,
|
| 1962 |
+
"step": 5580
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.9397549924483973,
|
| 1966 |
+
"grad_norm": 1.343929409980774,
|
| 1967 |
+
"learning_rate": 3.0374223863064273e-06,
|
| 1968 |
+
"loss": 1.9492,
|
| 1969 |
+
"step": 5600
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.9431112602785703,
|
| 1973 |
+
"grad_norm": 1.281829833984375,
|
| 1974 |
+
"learning_rate": 2.869608994797785e-06,
|
| 1975 |
+
"loss": 1.9429,
|
| 1976 |
+
"step": 5620
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.9464675281087431,
|
| 1980 |
+
"grad_norm": 1.1878955364227295,
|
| 1981 |
+
"learning_rate": 2.7017956032891427e-06,
|
| 1982 |
+
"loss": 1.9288,
|
| 1983 |
+
"step": 5640
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.949823795938916,
|
| 1987 |
+
"grad_norm": 1.2426503896713257,
|
| 1988 |
+
"learning_rate": 2.5339822117805e-06,
|
| 1989 |
+
"loss": 1.9498,
|
| 1990 |
+
"step": 5660
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9531800637690888,
|
| 1994 |
+
"grad_norm": 1.3009203672409058,
|
| 1995 |
+
"learning_rate": 2.366168820271858e-06,
|
| 1996 |
+
"loss": 1.9254,
|
| 1997 |
+
"step": 5680
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9565363315992617,
|
| 2001 |
+
"grad_norm": 1.3365298509597778,
|
| 2002 |
+
"learning_rate": 2.1983554287632153e-06,
|
| 2003 |
+
"loss": 1.9745,
|
| 2004 |
+
"step": 5700
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9598925994294345,
|
| 2008 |
+
"grad_norm": 1.2553895711898804,
|
| 2009 |
+
"learning_rate": 2.030542037254573e-06,
|
| 2010 |
+
"loss": 1.9636,
|
| 2011 |
+
"step": 5720
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9632488672596073,
|
| 2015 |
+
"grad_norm": 1.168750286102295,
|
| 2016 |
+
"learning_rate": 1.8627286457459307e-06,
|
| 2017 |
+
"loss": 1.9333,
|
| 2018 |
+
"step": 5740
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9666051350897802,
|
| 2022 |
+
"grad_norm": 1.3157403469085693,
|
| 2023 |
+
"learning_rate": 1.6949152542372882e-06,
|
| 2024 |
+
"loss": 1.9635,
|
| 2025 |
+
"step": 5760
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.969961402919953,
|
| 2029 |
+
"grad_norm": 1.2229833602905273,
|
| 2030 |
+
"learning_rate": 1.5271018627286458e-06,
|
| 2031 |
+
"loss": 1.9534,
|
| 2032 |
+
"step": 5780
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9733176707501259,
|
| 2036 |
+
"grad_norm": 1.3683537244796753,
|
| 2037 |
+
"learning_rate": 1.3592884712200033e-06,
|
| 2038 |
+
"loss": 1.9916,
|
| 2039 |
+
"step": 5800
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9766739385802987,
|
| 2043 |
+
"grad_norm": 1.1630358695983887,
|
| 2044 |
+
"learning_rate": 1.191475079711361e-06,
|
| 2045 |
+
"loss": 1.9098,
|
| 2046 |
+
"step": 5820
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9800302064104716,
|
| 2050 |
+
"grad_norm": 1.249770164489746,
|
| 2051 |
+
"learning_rate": 1.0236616882027187e-06,
|
| 2052 |
+
"loss": 1.9411,
|
| 2053 |
+
"step": 5840
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9833864742406444,
|
| 2057 |
+
"grad_norm": 1.2611877918243408,
|
| 2058 |
+
"learning_rate": 8.558482966940762e-07,
|
| 2059 |
+
"loss": 1.9093,
|
| 2060 |
+
"step": 5860
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9867427420708172,
|
| 2064 |
+
"grad_norm": 1.2660034894943237,
|
| 2065 |
+
"learning_rate": 6.880349051854338e-07,
|
| 2066 |
+
"loss": 1.9,
|
| 2067 |
+
"step": 5880
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.9900990099009901,
|
| 2071 |
+
"grad_norm": 1.263708233833313,
|
| 2072 |
+
"learning_rate": 5.202215136767914e-07,
|
| 2073 |
+
"loss": 1.9256,
|
| 2074 |
+
"step": 5900
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9934552777311629,
|
| 2078 |
+
"grad_norm": 1.257265329360962,
|
| 2079 |
+
"learning_rate": 3.5240812216814904e-07,
|
| 2080 |
+
"loss": 1.9474,
|
| 2081 |
+
"step": 5920
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9968115455613358,
|
| 2085 |
+
"grad_norm": 1.3897913694381714,
|
| 2086 |
+
"learning_rate": 1.8459473065950665e-07,
|
| 2087 |
+
"loss": 1.9567,
|
| 2088 |
+
"step": 5940
|
| 2089 |
+
}
|
| 2090 |
+
],
|
| 2091 |
+
"logging_steps": 20,
|
| 2092 |
+
"max_steps": 5959,
|
| 2093 |
+
"num_input_tokens_seen": 0,
|
| 2094 |
+
"num_train_epochs": 1,
|
| 2095 |
+
"save_steps": 5959,
|
| 2096 |
+
"stateful_callbacks": {
|
| 2097 |
+
"TrainerControl": {
|
| 2098 |
+
"args": {
|
| 2099 |
+
"should_epoch_stop": false,
|
| 2100 |
+
"should_evaluate": false,
|
| 2101 |
+
"should_log": false,
|
| 2102 |
+
"should_save": true,
|
| 2103 |
+
"should_training_stop": true
|
| 2104 |
+
},
|
| 2105 |
+
"attributes": {}
|
| 2106 |
+
}
|
| 2107 |
+
},
|
| 2108 |
+
"total_flos": 6.81749288803369e+17,
|
| 2109 |
+
"train_batch_size": 18,
|
| 2110 |
+
"trial_name": null,
|
| 2111 |
+
"trial_params": null
|
| 2112 |
+
}
|