Hugging Face's logo

tech4humans
/
Qwen2.5-VL-3B-Instruct-unsloth-r8-rslora

Transformers
Safetensors
Generated from Trainer
unsloth
trl
sft
Model card Files
xet
Community
samuellimabraz commited on
Commit
6bb72fb
·
verified ·
1 Parent(s): 319dc5b

End of training

Browse files
Files changed (16) hide show
  1. .gitattributes +1 -0
  2. README.md +59 -0
  3. adapter_config.json +29 -0
  4. adapter_model.safetensors +3 -0
  5. added_tokens.json +24 -0
  6. all_results.json +11 -0
  7. chat_template.json +3 -0
  8. merges.txt +0 -0
  9. preprocessor_config.json +29 -0
  10. special_tokens_map.json +31 -0
  11. tokenizer.json +3 -0
  12. tokenizer_config.json +210 -0
  13. train_results.json +11 -0
  14. trainer_state.json +2547 -0
  15. training_args.bin +3 -0
  16. vocab.json +0 -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
README.md ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: unsloth/Qwen2.5-VL-3B-Instruct
3
+ library_name: transformers
4
+ model_name: Qwen2.5-VL-3B-Instruct-unsloth-r8-rslora
5
+ tags:
6
+ - generated_from_trainer
7
+ - unsloth
8
+ - trl
9
+ - sft
10
+ licence: license
11
+ ---
12
+
13
+ # Model Card for Qwen2.5-VL-3B-Instruct-unsloth-r8-rslora
14
+
15
+ This model is a fine-tuned version of [unsloth/Qwen2.5-VL-3B-Instruct](https://huggingface.co/unsloth/Qwen2.5-VL-3B-Instruct).
16
+ It has been trained using [TRL](https://github.com/huggingface/trl).
17
+
18
+ ## Quick start
19
+
20
+ ```python
21
+ from transformers import pipeline
22
+
23
+ question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
24
+ generator = pipeline("text-generation", model="tech4humans/Qwen2.5-VL-3B-Instruct-unsloth-r8-rslora", device="cuda")
25
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
26
+ print(output["generated_text"])
27
+ ```
28
+
29
+ ## Training procedure
30
+
31
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/samuel-lima-tech4humans/ocr-finetuning-v2/runs/4lkl7nx9)
32
+
33
+
34
+ This model was trained with SFT.
35
+
36
+ ### Framework versions
37
+
38
+ - TRL: 0.15.1
39
+ - Transformers: 4.49.0.dev0
40
+ - Pytorch: 2.6.0
41
+ - Datasets: 3.3.1
42
+ - Tokenizers: 0.21.0
43
+
44
+ ## Citations
45
+
46
+
47
+
48
+ Cite TRL as:
49
+
50
+ ```bibtex
51
+ @misc{vonwerra2022trl,
52
+ title = {{TRL: Transformer Reinforcement Learning}},
53
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
54
+ year = 2020,
55
+ journal = {GitHub repository},
56
+ publisher = {GitHub},
57
+ howpublished = {\url{https://github.com/huggingface/trl}}
58
+ }
59
+ ```
adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "unsloth/Qwen2.5-VL-3B-Instruct",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 8,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.05,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 8,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": "(?:.*?(?:vision|image|visual|patch|language|text).*?(?:self_attn|attention|attn).*?(?:qkv|proj|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj).*?)|(?:\\bmodel\\.layers\\.[\\d]{1,}\\.(?:self_attn|attention|attn)\\.(?:(?:qkv|proj|gate_proj|up_proj|down_proj|q_proj|k_proj|v_proj|o_proj)))",
26
+ "task_type": "CAUSAL_LM",
27
+ "use_dora": false,
28
+ "use_rslora": false
29
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32f1e017de52d4d12c51a01da250b1039f540b66dac148704e4bad20fd26ab65
3
+ size 22664248
added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
all_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "all_params": 3760275456,
3
+ "memory_footprint": 7531865424,
4
+ "total_flos": 4.419854408240333e+16,
5
+ "train_loss": 0.537436925212387,
6
+ "train_runtime": 3691.1633,
7
+ "train_samples_per_second": 0.863,
8
+ "train_steps_per_second": 0.107,
9
+ "trainable_params": 5652480,
10
+ "trainable_params_percent": 0.15032090244827
11
+ }
chat_template.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "chat_template": "{% set image_count = namespace(value=0) %}{% set video_count = namespace(value=0) %}{% for message in messages %}{% if loop.first and message['role'] != 'system' %}<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n{% endif %}<|im_start|>{{ message['role'] }}\n{% if message['content'] is string %}{{ message['content'] }}<|im_end|>\n{% else %}{% for content in message['content'] %}{% if content['type'] == 'image' or 'image' in content or 'image_url' in content %}{% set image_count.value = image_count.value + 1 %}{% if add_vision_id %}Picture {{ image_count.value }}: {% endif %}<|vision_start|><|image_pad|><|vision_end|>{% elif content['type'] == 'video' or 'video' in content %}{% set video_count.value = video_count.value + 1 %}{% if add_vision_id %}Video {{ video_count.value }}: {% endif %}<|vision_start|><|video_pad|><|vision_end|>{% elif 'text' in content %}{{ content['text'] }}{% endif %}{% endfor %}<|im_end|>\n{% endif %}{% endfor %}{% if add_generation_prompt %}<|im_start|>assistant\n{% endif %}"
3
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": true,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.48145466,
8
+ 0.4578275,
9
+ 0.40821073
10
+ ],
11
+ "image_processor_type": "Qwen2_5_VLImageProcessor",
12
+ "image_std": [
13
+ 0.26862954,
14
+ 0.26130258,
15
+ 0.27577711
16
+ ],
17
+ "max_pixels": 12845056,
18
+ "merge_size": 2,
19
+ "min_pixels": 3136,
20
+ "patch_size": 14,
21
+ "processor_class": "Qwen2_5_VLProcessor",
22
+ "resample": 3,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "longest_edge": 12845056,
26
+ "shortest_edge": 3136
27
+ },
28
+ "temporal_patch_size": 2
29
+ }
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": "<|vision_pad|>",
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 sha256:02e2c18eaf9a0b64d16f11f381664b17389bd92bcf13130ba6f754c95b078300
3
+ size 11422065
tokenizer_config.json ADDED
@@ -0,0 +1,210 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# 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>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\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\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "extra_special_tokens": {},
203
+ "model_max_length": 131072,
204
+ "pad_token": "<|vision_pad|>",
205
+ "padding_side": "right",
206
+ "processor_class": "Qwen2_5_VLProcessor",
207
+ "split_special_tokens": false,
208
+ "tokenizer_class": "Qwen2Tokenizer",
209
+ "unk_token": null
210
+ }
train_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "all_params": 3760275456,
3
+ "memory_footprint": 7531865424,
4
+ "total_flos": 4.419854408240333e+16,
5
+ "train_loss": 0.537436925212387,
6
+ "train_runtime": 3691.1633,
7
+ "train_samples_per_second": 0.863,
8
+ "train_steps_per_second": 0.107,
9
+ "trainable_params": 5652480,
10
+ "trainable_params_percent": 0.15032090244827
11
+ }
trainer_state.json ADDED
@@ -0,0 +1,2547 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.0942394882440567,
3
+ "best_model_checkpoint": "/content/train/Qwen2.5-VL-3B-Instruct-unsloth-r8-rslora/checkpoint-320",
4
+ "epoch": 2.4218455743879472,
5
+ "eval_steps": 10,
6
+ "global_step": 320,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.007532956685499058,
13
+ "grad_norm": 1.6337342262268066,
14
+ "learning_rate": 1e-05,
15
+ "loss": 2.8339,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.015065913370998116,
20
+ "grad_norm": 1.7995260953903198,
21
+ "learning_rate": 2e-05,
22
+ "loss": 3.1612,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.022598870056497175,
27
+ "grad_norm": 1.8331536054611206,
28
+ "learning_rate": 3e-05,
29
+ "loss": 2.9529,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.030131826741996232,
34
+ "grad_norm": 1.8774369955062866,
35
+ "learning_rate": 4e-05,
36
+ "loss": 3.1371,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.03766478342749529,
41
+ "grad_norm": 2.1215269565582275,
42
+ "learning_rate": 5e-05,
43
+ "loss": 3.4216,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.04519774011299435,
48
+ "grad_norm": 1.7083489894866943,
49
+ "learning_rate": 6e-05,
50
+ "loss": 2.5018,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.05273069679849341,
55
+ "grad_norm": 1.9553663730621338,
56
+ "learning_rate": 7e-05,
57
+ "loss": 2.8832,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.060263653483992465,
62
+ "grad_norm": 2.085784673690796,
63
+ "learning_rate": 8e-05,
64
+ "loss": 2.9757,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.06779661016949153,
69
+ "grad_norm": 1.949724793434143,
70
+ "learning_rate": 9e-05,
71
+ "loss": 2.7228,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.07532956685499058,
76
+ "grad_norm": 2.1280415058135986,
77
+ "learning_rate": 0.0001,
78
+ "loss": 2.9347,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.07532956685499058,
83
+ "eval_loss": 2.670034646987915,
84
+ "eval_runtime": 54.8057,
85
+ "eval_samples_per_second": 1.368,
86
+ "eval_steps_per_second": 0.693,
87
+ "step": 10
88
+ },
89
+ {
90
+ "epoch": 0.08286252354048965,
91
+ "grad_norm": 2.418717861175537,
92
+ "learning_rate": 9.999834399079165e-05,
93
+ "loss": 3.0651,
94
+ "step": 11
95
+ },
96
+ {
97
+ "epoch": 0.0903954802259887,
98
+ "grad_norm": 1.9917370080947876,
99
+ "learning_rate": 9.99933760728612e-05,
100
+ "loss": 2.4549,
101
+ "step": 12
102
+ },
103
+ {
104
+ "epoch": 0.09792843691148775,
105
+ "grad_norm": 1.8027087450027466,
106
+ "learning_rate": 9.99850965752854e-05,
107
+ "loss": 2.2586,
108
+ "step": 13
109
+ },
110
+ {
111
+ "epoch": 0.10546139359698682,
112
+ "grad_norm": 1.6262425184249878,
113
+ "learning_rate": 9.997350604650123e-05,
114
+ "loss": 2.0871,
115
+ "step": 14
116
+ },
117
+ {
118
+ "epoch": 0.11299435028248588,
119
+ "grad_norm": 1.5385862588882446,
120
+ "learning_rate": 9.995860525426954e-05,
121
+ "loss": 2.0329,
122
+ "step": 15
123
+ },
124
+ {
125
+ "epoch": 0.12052730696798493,
126
+ "grad_norm": 1.1733174324035645,
127
+ "learning_rate": 9.994039518562432e-05,
128
+ "loss": 1.6894,
129
+ "step": 16
130
+ },
131
+ {
132
+ "epoch": 0.128060263653484,
133
+ "grad_norm": 1.1716901063919067,
134
+ "learning_rate": 9.991887704680724e-05,
135
+ "loss": 1.8195,
136
+ "step": 17
137
+ },
138
+ {
139
+ "epoch": 0.13559322033898305,
140
+ "grad_norm": 1.3725332021713257,
141
+ "learning_rate": 9.989405226318772e-05,
142
+ "loss": 1.9845,
143
+ "step": 18
144
+ },
145
+ {
146
+ "epoch": 0.1431261770244821,
147
+ "grad_norm": 0.8536051511764526,
148
+ "learning_rate": 9.986592247916858e-05,
149
+ "loss": 1.5969,
150
+ "step": 19
151
+ },
152
+ {
153
+ "epoch": 0.15065913370998116,
154
+ "grad_norm": 0.6615162491798401,
155
+ "learning_rate": 9.983448955807708e-05,
156
+ "loss": 1.6763,
157
+ "step": 20
158
+ },
159
+ {
160
+ "epoch": 0.15065913370998116,
161
+ "eval_loss": 1.6168410778045654,
162
+ "eval_runtime": 43.9067,
163
+ "eval_samples_per_second": 1.708,
164
+ "eval_steps_per_second": 0.865,
165
+ "step": 20
166
+ },
167
+ {
168
+ "epoch": 0.15819209039548024,
169
+ "grad_norm": 0.601297914981842,
170
+ "learning_rate": 9.979975558204147e-05,
171
+ "loss": 1.6742,
172
+ "step": 21
173
+ },
174
+ {
175
+ "epoch": 0.1657250470809793,
176
+ "grad_norm": 0.4877667725086212,
177
+ "learning_rate": 9.976172285185314e-05,
178
+ "loss": 1.6194,
179
+ "step": 22
180
+ },
181
+ {
182
+ "epoch": 0.17325800376647835,
183
+ "grad_norm": 0.4728071391582489,
184
+ "learning_rate": 9.972039388681413e-05,
185
+ "loss": 1.6243,
186
+ "step": 23
187
+ },
188
+ {
189
+ "epoch": 0.1807909604519774,
190
+ "grad_norm": 0.37902191281318665,
191
+ "learning_rate": 9.967577142457032e-05,
192
+ "loss": 1.4569,
193
+ "step": 24
194
+ },
195
+ {
196
+ "epoch": 0.18832391713747645,
197
+ "grad_norm": 0.45830923318862915,
198
+ "learning_rate": 9.962785842093003e-05,
199
+ "loss": 1.6214,
200
+ "step": 25
201
+ },
202
+ {
203
+ "epoch": 0.1958568738229755,
204
+ "grad_norm": 0.39010411500930786,
205
+ "learning_rate": 9.957665804966829e-05,
206
+ "loss": 1.345,
207
+ "step": 26
208
+ },
209
+ {
210
+ "epoch": 0.2033898305084746,
211
+ "grad_norm": 0.5287721753120422,
212
+ "learning_rate": 9.952217370231653e-05,
213
+ "loss": 1.6019,
214
+ "step": 27
215
+ },
216
+ {
217
+ "epoch": 0.21092278719397364,
218
+ "grad_norm": 0.4444209635257721,
219
+ "learning_rate": 9.946440898793801e-05,
220
+ "loss": 1.4566,
221
+ "step": 28
222
+ },
223
+ {
224
+ "epoch": 0.2184557438794727,
225
+ "grad_norm": 0.4856286346912384,
226
+ "learning_rate": 9.940336773288865e-05,
227
+ "loss": 1.5087,
228
+ "step": 29
229
+ },
230
+ {
231
+ "epoch": 0.22598870056497175,
232
+ "grad_norm": 0.4327958822250366,
233
+ "learning_rate": 9.933905398056372e-05,
234
+ "loss": 1.2924,
235
+ "step": 30
236
+ },
237
+ {
238
+ "epoch": 0.22598870056497175,
239
+ "eval_loss": 1.4212392568588257,
240
+ "eval_runtime": 44.0456,
241
+ "eval_samples_per_second": 1.703,
242
+ "eval_steps_per_second": 0.863,
243
+ "step": 30
244
+ },
245
+ {
246
+ "epoch": 0.2335216572504708,
247
+ "grad_norm": 0.49079710245132446,
248
+ "learning_rate": 9.92714719911298e-05,
249
+ "loss": 1.3501,
250
+ "step": 31
251
+ },
252
+ {
253
+ "epoch": 0.24105461393596986,
254
+ "grad_norm": 0.8166887760162354,
255
+ "learning_rate": 9.920062624124282e-05,
256
+ "loss": 1.4564,
257
+ "step": 32
258
+ },
259
+ {
260
+ "epoch": 0.24858757062146894,
261
+ "grad_norm": 0.6213160157203674,
262
+ "learning_rate": 9.912652142375132e-05,
263
+ "loss": 1.5757,
264
+ "step": 33
265
+ },
266
+ {
267
+ "epoch": 0.256120527306968,
268
+ "grad_norm": 0.49864110350608826,
269
+ "learning_rate": 9.904916244738571e-05,
270
+ "loss": 1.3778,
271
+ "step": 34
272
+ },
273
+ {
274
+ "epoch": 0.263653483992467,
275
+ "grad_norm": 0.5232111215591431,
276
+ "learning_rate": 9.896855443643308e-05,
277
+ "loss": 1.3466,
278
+ "step": 35
279
+ },
280
+ {
281
+ "epoch": 0.2711864406779661,
282
+ "grad_norm": 0.4295203983783722,
283
+ "learning_rate": 9.888470273039775e-05,
284
+ "loss": 1.3928,
285
+ "step": 36
286
+ },
287
+ {
288
+ "epoch": 0.2787193973634652,
289
+ "grad_norm": 0.3830944895744324,
290
+ "learning_rate": 9.879761288364766e-05,
291
+ "loss": 1.4017,
292
+ "step": 37
293
+ },
294
+ {
295
+ "epoch": 0.2862523540489642,
296
+ "grad_norm": 0.3481033146381378,
297
+ "learning_rate": 9.870729066504629e-05,
298
+ "loss": 1.3344,
299
+ "step": 38
300
+ },
301
+ {
302
+ "epoch": 0.2937853107344633,
303
+ "grad_norm": 0.28261324763298035,
304
+ "learning_rate": 9.861374205757068e-05,
305
+ "loss": 1.2005,
306
+ "step": 39
307
+ },
308
+ {
309
+ "epoch": 0.3013182674199623,
310
+ "grad_norm": 0.31103572249412537,
311
+ "learning_rate": 9.851697325791505e-05,
312
+ "loss": 1.2026,
313
+ "step": 40
314
+ },
315
+ {
316
+ "epoch": 0.3013182674199623,
317
+ "eval_loss": 1.2615383863449097,
318
+ "eval_runtime": 43.98,
319
+ "eval_samples_per_second": 1.705,
320
+ "eval_steps_per_second": 0.864,
321
+ "step": 40
322
+ },
323
+ {
324
+ "epoch": 0.3088512241054614,
325
+ "grad_norm": 0.41986966133117676,
326
+ "learning_rate": 9.841699067608033e-05,
327
+ "loss": 1.2891,
328
+ "step": 41
329
+ },
330
+ {
331
+ "epoch": 0.3163841807909605,
332
+ "grad_norm": 0.3304821848869324,
333
+ "learning_rate": 9.831380093494957e-05,
334
+ "loss": 1.1725,
335
+ "step": 42
336
+ },
337
+ {
338
+ "epoch": 0.3239171374764595,
339
+ "grad_norm": 0.38536179065704346,
340
+ "learning_rate": 9.820741086984924e-05,
341
+ "loss": 1.2084,
342
+ "step": 43
343
+ },
344
+ {
345
+ "epoch": 0.3314500941619586,
346
+ "grad_norm": 0.3935360908508301,
347
+ "learning_rate": 9.809782752809644e-05,
348
+ "loss": 1.2919,
349
+ "step": 44
350
+ },
351
+ {
352
+ "epoch": 0.3389830508474576,
353
+ "grad_norm": 0.3400743305683136,
354
+ "learning_rate": 9.798505816853208e-05,
355
+ "loss": 1.2405,
356
+ "step": 45
357
+ },
358
+ {
359
+ "epoch": 0.3465160075329567,
360
+ "grad_norm": 0.3844086825847626,
361
+ "learning_rate": 9.786911026104007e-05,
362
+ "loss": 1.2489,
363
+ "step": 46
364
+ },
365
+ {
366
+ "epoch": 0.3540489642184557,
367
+ "grad_norm": 0.3005635738372803,
368
+ "learning_rate": 9.774999148605251e-05,
369
+ "loss": 1.1564,
370
+ "step": 47
371
+ },
372
+ {
373
+ "epoch": 0.3615819209039548,
374
+ "grad_norm": 0.2923331558704376,
375
+ "learning_rate": 9.762770973404094e-05,
376
+ "loss": 1.1403,
377
+ "step": 48
378
+ },
379
+ {
380
+ "epoch": 0.3691148775894539,
381
+ "grad_norm": 0.30060315132141113,
382
+ "learning_rate": 9.750227310499366e-05,
383
+ "loss": 1.1172,
384
+ "step": 49
385
+ },
386
+ {
387
+ "epoch": 0.3766478342749529,
388
+ "grad_norm": 0.2783910036087036,
389
+ "learning_rate": 9.737368990787916e-05,
390
+ "loss": 1.0288,
391
+ "step": 50
392
+ },
393
+ {
394
+ "epoch": 0.3766478342749529,
395
+ "eval_loss": 1.122483491897583,
396
+ "eval_runtime": 43.9499,
397
+ "eval_samples_per_second": 1.706,
398
+ "eval_steps_per_second": 0.865,
399
+ "step": 50
400
+ },
401
+ {
402
+ "epoch": 0.384180790960452,
403
+ "grad_norm": 0.2709607779979706,
404
+ "learning_rate": 9.72419686600958e-05,
405
+ "loss": 1.0827,
406
+ "step": 51
407
+ },
408
+ {
409
+ "epoch": 0.391713747645951,
410
+ "grad_norm": 0.27541661262512207,
411
+ "learning_rate": 9.710711808690754e-05,
412
+ "loss": 1.172,
413
+ "step": 52
414
+ },
415
+ {
416
+ "epoch": 0.3992467043314501,
417
+ "grad_norm": 0.25708749890327454,
418
+ "learning_rate": 9.696914712086603e-05,
419
+ "loss": 1.058,
420
+ "step": 53
421
+ },
422
+ {
423
+ "epoch": 0.4067796610169492,
424
+ "grad_norm": 0.29954883456230164,
425
+ "learning_rate": 9.682806490121885e-05,
426
+ "loss": 1.0781,
427
+ "step": 54
428
+ },
429
+ {
430
+ "epoch": 0.4143126177024482,
431
+ "grad_norm": 0.28139016032218933,
432
+ "learning_rate": 9.668388077330421e-05,
433
+ "loss": 1.0746,
434
+ "step": 55
435
+ },
436
+ {
437
+ "epoch": 0.4218455743879473,
438
+ "grad_norm": 0.2823486328125,
439
+ "learning_rate": 9.653660428793188e-05,
440
+ "loss": 1.052,
441
+ "step": 56
442
+ },
443
+ {
444
+ "epoch": 0.4293785310734463,
445
+ "grad_norm": 0.28060609102249146,
446
+ "learning_rate": 9.638624520075046e-05,
447
+ "loss": 1.0088,
448
+ "step": 57
449
+ },
450
+ {
451
+ "epoch": 0.4369114877589454,
452
+ "grad_norm": 0.26586028933525085,
453
+ "learning_rate": 9.623281347160127e-05,
454
+ "loss": 0.9935,
455
+ "step": 58
456
+ },
457
+ {
458
+ "epoch": 0.4444444444444444,
459
+ "grad_norm": 0.2685753405094147,
460
+ "learning_rate": 9.607631926385859e-05,
461
+ "loss": 1.002,
462
+ "step": 59
463
+ },
464
+ {
465
+ "epoch": 0.4519774011299435,
466
+ "grad_norm": 0.24896016716957092,
467
+ "learning_rate": 9.591677294375636e-05,
468
+ "loss": 0.9303,
469
+ "step": 60
470
+ },
471
+ {
472
+ "epoch": 0.4519774011299435,
473
+ "eval_loss": 0.9904586672782898,
474
+ "eval_runtime": 43.9768,
475
+ "eval_samples_per_second": 1.705,
476
+ "eval_steps_per_second": 0.864,
477
+ "step": 60
478
+ },
479
+ {
480
+ "epoch": 0.4595103578154426,
481
+ "grad_norm": 0.2723137140274048,
482
+ "learning_rate": 9.575418507970161e-05,
483
+ "loss": 1.0173,
484
+ "step": 61
485
+ },
486
+ {
487
+ "epoch": 0.4670433145009416,
488
+ "grad_norm": 0.28442397713661194,
489
+ "learning_rate": 9.558856644157432e-05,
490
+ "loss": 0.9589,
491
+ "step": 62
492
+ },
493
+ {
494
+ "epoch": 0.4745762711864407,
495
+ "grad_norm": 0.25184300541877747,
496
+ "learning_rate": 9.541992800001409e-05,
497
+ "loss": 0.9013,
498
+ "step": 63
499
+ },
500
+ {
501
+ "epoch": 0.4821092278719397,
502
+ "grad_norm": 0.2699225842952728,
503
+ "learning_rate": 9.52482809256934e-05,
504
+ "loss": 0.9436,
505
+ "step": 64
506
+ },
507
+ {
508
+ "epoch": 0.4896421845574388,
509
+ "grad_norm": 0.24603509902954102,
510
+ "learning_rate": 9.507363658857768e-05,
511
+ "loss": 0.907,
512
+ "step": 65
513
+ },
514
+ {
515
+ "epoch": 0.4971751412429379,
516
+ "grad_norm": 0.24069125950336456,
517
+ "learning_rate": 9.489600655717217e-05,
518
+ "loss": 0.9147,
519
+ "step": 66
520
+ },
521
+ {
522
+ "epoch": 0.504708097928437,
523
+ "grad_norm": 0.2531817555427551,
524
+ "learning_rate": 9.471540259775554e-05,
525
+ "loss": 0.8429,
526
+ "step": 67
527
+ },
528
+ {
529
+ "epoch": 0.512241054613936,
530
+ "grad_norm": 0.29864469170570374,
531
+ "learning_rate": 9.453183667360062e-05,
532
+ "loss": 0.9324,
533
+ "step": 68
534
+ },
535
+ {
536
+ "epoch": 0.519774011299435,
537
+ "grad_norm": 0.28371143341064453,
538
+ "learning_rate": 9.43453209441818e-05,
539
+ "loss": 0.8185,
540
+ "step": 69
541
+ },
542
+ {
543
+ "epoch": 0.527306967984934,
544
+ "grad_norm": 0.3073464035987854,
545
+ "learning_rate": 9.415586776436973e-05,
546
+ "loss": 0.9043,
547
+ "step": 70
548
+ },
549
+ {
550
+ "epoch": 0.527306967984934,
551
+ "eval_loss": 0.8400442600250244,
552
+ "eval_runtime": 43.8471,
553
+ "eval_samples_per_second": 1.71,
554
+ "eval_steps_per_second": 0.867,
555
+ "step": 70
556
+ },
557
+ {
558
+ "epoch": 0.5348399246704332,
559
+ "grad_norm": 0.31865832209587097,
560
+ "learning_rate": 9.396348968361281e-05,
561
+ "loss": 0.8984,
562
+ "step": 71
563
+ },
564
+ {
565
+ "epoch": 0.5423728813559322,
566
+ "grad_norm": 0.30962345004081726,
567
+ "learning_rate": 9.376819944510598e-05,
568
+ "loss": 0.8379,
569
+ "step": 72
570
+ },
571
+ {
572
+ "epoch": 0.5499058380414312,
573
+ "grad_norm": 0.40249139070510864,
574
+ "learning_rate": 9.357000998494656e-05,
575
+ "loss": 0.8553,
576
+ "step": 73
577
+ },
578
+ {
579
+ "epoch": 0.5574387947269304,
580
+ "grad_norm": 0.33023273944854736,
581
+ "learning_rate": 9.336893443127738e-05,
582
+ "loss": 0.762,
583
+ "step": 74
584
+ },
585
+ {
586
+ "epoch": 0.5649717514124294,
587
+ "grad_norm": 0.2891211211681366,
588
+ "learning_rate": 9.31649861034172e-05,
589
+ "loss": 0.7443,
590
+ "step": 75
591
+ },
592
+ {
593
+ "epoch": 0.5725047080979284,
594
+ "grad_norm": 0.33249810338020325,
595
+ "learning_rate": 9.295817851097837e-05,
596
+ "loss": 0.789,
597
+ "step": 76
598
+ },
599
+ {
600
+ "epoch": 0.5800376647834274,
601
+ "grad_norm": 0.32798218727111816,
602
+ "learning_rate": 9.274852535297198e-05,
603
+ "loss": 0.7544,
604
+ "step": 77
605
+ },
606
+ {
607
+ "epoch": 0.5875706214689266,
608
+ "grad_norm": 0.38781678676605225,
609
+ "learning_rate": 9.253604051690046e-05,
610
+ "loss": 0.691,
611
+ "step": 78
612
+ },
613
+ {
614
+ "epoch": 0.5951035781544256,
615
+ "grad_norm": 0.3911118805408478,
616
+ "learning_rate": 9.232073807783759e-05,
617
+ "loss": 0.6637,
618
+ "step": 79
619
+ },
620
+ {
621
+ "epoch": 0.6026365348399246,
622
+ "grad_norm": 0.31271758675575256,
623
+ "learning_rate": 9.210263229749626e-05,
624
+ "loss": 0.6649,
625
+ "step": 80
626
+ },
627
+ {
628
+ "epoch": 0.6026365348399246,
629
+ "eval_loss": 0.657122015953064,
630
+ "eval_runtime": 43.9006,
631
+ "eval_samples_per_second": 1.708,
632
+ "eval_steps_per_second": 0.866,
633
+ "step": 80
634
+ },
635
+ {
636
+ "epoch": 0.6101694915254238,
637
+ "grad_norm": 0.3603556454181671,
638
+ "learning_rate": 9.188173762328367e-05,
639
+ "loss": 0.6501,
640
+ "step": 81
641
+ },
642
+ {
643
+ "epoch": 0.6177024482109228,
644
+ "grad_norm": 0.39841267466545105,
645
+ "learning_rate": 9.165806868734444e-05,
646
+ "loss": 0.6572,
647
+ "step": 82
648
+ },
649
+ {
650
+ "epoch": 0.6252354048964218,
651
+ "grad_norm": 0.34062060713768005,
652
+ "learning_rate": 9.143164030559122e-05,
653
+ "loss": 0.6706,
654
+ "step": 83
655
+ },
656
+ {
657
+ "epoch": 0.632768361581921,
658
+ "grad_norm": 0.31477442383766174,
659
+ "learning_rate": 9.120246747672347e-05,
660
+ "loss": 0.629,
661
+ "step": 84
662
+ },
663
+ {
664
+ "epoch": 0.64030131826742,
665
+ "grad_norm": 0.28587597608566284,
666
+ "learning_rate": 9.097056538123376e-05,
667
+ "loss": 0.5698,
668
+ "step": 85
669
+ },
670
+ {
671
+ "epoch": 0.647834274952919,
672
+ "grad_norm": 0.2795222997665405,
673
+ "learning_rate": 9.073594938040231e-05,
674
+ "loss": 0.573,
675
+ "step": 86
676
+ },
677
+ {
678
+ "epoch": 0.655367231638418,
679
+ "grad_norm": 0.23265816271305084,
680
+ "learning_rate": 9.049863501527947e-05,
681
+ "loss": 0.626,
682
+ "step": 87
683
+ },
684
+ {
685
+ "epoch": 0.6629001883239172,
686
+ "grad_norm": 0.26728200912475586,
687
+ "learning_rate": 9.025863800565613e-05,
688
+ "loss": 0.6103,
689
+ "step": 88
690
+ },
691
+ {
692
+ "epoch": 0.6704331450094162,
693
+ "grad_norm": 0.24294917285442352,
694
+ "learning_rate": 9.001597424902267e-05,
695
+ "loss": 0.5109,
696
+ "step": 89
697
+ },
698
+ {
699
+ "epoch": 0.6779661016949152,
700
+ "grad_norm": 0.27384141087532043,
701
+ "learning_rate": 8.977065981951566e-05,
702
+ "loss": 0.5266,
703
+ "step": 90
704
+ },
705
+ {
706
+ "epoch": 0.6779661016949152,
707
+ "eval_loss": 0.5219850540161133,
708
+ "eval_runtime": 43.852,
709
+ "eval_samples_per_second": 1.71,
710
+ "eval_steps_per_second": 0.867,
711
+ "step": 90
712
+ },
713
+ {
714
+ "epoch": 0.6854990583804144,
715
+ "grad_norm": 0.2532135248184204,
716
+ "learning_rate": 8.952271096685332e-05,
717
+ "loss": 0.5027,
718
+ "step": 91
719
+ },
720
+ {
721
+ "epoch": 0.6930320150659134,
722
+ "grad_norm": 0.25869038701057434,
723
+ "learning_rate": 8.927214411525895e-05,
724
+ "loss": 0.5846,
725
+ "step": 92
726
+ },
727
+ {
728
+ "epoch": 0.7005649717514124,
729
+ "grad_norm": 0.25413447618484497,
730
+ "learning_rate": 8.90189758623731e-05,
731
+ "loss": 0.5171,
732
+ "step": 93
733
+ },
734
+ {
735
+ "epoch": 0.7080979284369114,
736
+ "grad_norm": 0.2756996154785156,
737
+ "learning_rate": 8.876322297815405e-05,
738
+ "loss": 0.5104,
739
+ "step": 94
740
+ },
741
+ {
742
+ "epoch": 0.7156308851224106,
743
+ "grad_norm": 0.25177237391471863,
744
+ "learning_rate": 8.850490240376711e-05,
745
+ "loss": 0.4921,
746
+ "step": 95
747
+ },
748
+ {
749
+ "epoch": 0.7231638418079096,
750
+ "grad_norm": 0.22342444956302643,
751
+ "learning_rate": 8.824403125046225e-05,
752
+ "loss": 0.45,
753
+ "step": 96
754
+ },
755
+ {
756
+ "epoch": 0.7306967984934086,
757
+ "grad_norm": 0.2055438607931137,
758
+ "learning_rate": 8.798062679844077e-05,
759
+ "loss": 0.4527,
760
+ "step": 97
761
+ },
762
+ {
763
+ "epoch": 0.7382297551789078,
764
+ "grad_norm": 0.21503853797912598,
765
+ "learning_rate": 8.771470649571056e-05,
766
+ "loss": 0.5282,
767
+ "step": 98
768
+ },
769
+ {
770
+ "epoch": 0.7457627118644068,
771
+ "grad_norm": 0.1799071580171585,
772
+ "learning_rate": 8.744628795693047e-05,
773
+ "loss": 0.5086,
774
+ "step": 99
775
+ },
776
+ {
777
+ "epoch": 0.7532956685499058,
778
+ "grad_norm": 0.20049640536308289,
779
+ "learning_rate": 8.717538896224332e-05,
780
+ "loss": 0.5187,
781
+ "step": 100
782
+ },
783
+ {
784
+ "epoch": 0.7532956685499058,
785
+ "eval_loss": 0.457345575094223,
786
+ "eval_runtime": 43.8906,
787
+ "eval_samples_per_second": 1.709,
788
+ "eval_steps_per_second": 0.866,
789
+ "step": 100
790
+ },
791
+ {
792
+ "epoch": 0.7608286252354048,
793
+ "grad_norm": 0.19504646956920624,
794
+ "learning_rate": 8.690202745609835e-05,
795
+ "loss": 0.3967,
796
+ "step": 101
797
+ },
798
+ {
799
+ "epoch": 0.768361581920904,
800
+ "grad_norm": 0.18335066735744476,
801
+ "learning_rate": 8.662622154606237e-05,
802
+ "loss": 0.4931,
803
+ "step": 102
804
+ },
805
+ {
806
+ "epoch": 0.775894538606403,
807
+ "grad_norm": 0.2755972743034363,
808
+ "learning_rate": 8.634798950162048e-05,
809
+ "loss": 0.5392,
810
+ "step": 103
811
+ },
812
+ {
813
+ "epoch": 0.783427495291902,
814
+ "grad_norm": 0.18094103038311005,
815
+ "learning_rate": 8.606734975296578e-05,
816
+ "loss": 0.4579,
817
+ "step": 104
818
+ },
819
+ {
820
+ "epoch": 0.7909604519774012,
821
+ "grad_norm": 0.20168878138065338,
822
+ "learning_rate": 8.578432088977859e-05,
823
+ "loss": 0.3906,
824
+ "step": 105
825
+ },
826
+ {
827
+ "epoch": 0.7984934086629002,
828
+ "grad_norm": 0.20075590908527374,
829
+ "learning_rate": 8.549892165999505e-05,
830
+ "loss": 0.4453,
831
+ "step": 106
832
+ },
833
+ {
834
+ "epoch": 0.8060263653483992,
835
+ "grad_norm": 0.19419971108436584,
836
+ "learning_rate": 8.521117096856528e-05,
837
+ "loss": 0.4881,
838
+ "step": 107
839
+ },
840
+ {
841
+ "epoch": 0.8135593220338984,
842
+ "grad_norm": 0.25679534673690796,
843
+ "learning_rate": 8.492108787620105e-05,
844
+ "loss": 0.4528,
845
+ "step": 108
846
+ },
847
+ {
848
+ "epoch": 0.8210922787193974,
849
+ "grad_norm": 0.17266082763671875,
850
+ "learning_rate": 8.462869159811327e-05,
851
+ "loss": 0.4072,
852
+ "step": 109
853
+ },
854
+ {
855
+ "epoch": 0.8286252354048964,
856
+ "grad_norm": 0.20196932554244995,
857
+ "learning_rate": 8.433400150273906e-05,
858
+ "loss": 0.4506,
859
+ "step": 110
860
+ },
861
+ {
862
+ "epoch": 0.8286252354048964,
863
+ "eval_loss": 0.4128032326698303,
864
+ "eval_runtime": 43.8756,
865
+ "eval_samples_per_second": 1.709,
866
+ "eval_steps_per_second": 0.866,
867
+ "step": 110
868
+ },
869
+ {
870
+ "epoch": 0.8361581920903954,
871
+ "grad_norm": 0.18674571812152863,
872
+ "learning_rate": 8.403703711045892e-05,
873
+ "loss": 0.4425,
874
+ "step": 111
875
+ },
876
+ {
877
+ "epoch": 0.8436911487758946,
878
+ "grad_norm": 0.19765950739383698,
879
+ "learning_rate": 8.373781809230355e-05,
880
+ "loss": 0.4913,
881
+ "step": 112
882
+ },
883
+ {
884
+ "epoch": 0.8512241054613936,
885
+ "grad_norm": 0.157321497797966,
886
+ "learning_rate": 8.343636426865096e-05,
887
+ "loss": 0.4163,
888
+ "step": 113
889
+ },
890
+ {
891
+ "epoch": 0.8587570621468926,
892
+ "grad_norm": 0.20251426100730896,
893
+ "learning_rate": 8.313269560791342e-05,
894
+ "loss": 0.3659,
895
+ "step": 114
896
+ },
897
+ {
898
+ "epoch": 0.8662900188323918,
899
+ "grad_norm": 0.16248001158237457,
900
+ "learning_rate": 8.28268322252149e-05,
901
+ "loss": 0.4309,
902
+ "step": 115
903
+ },
904
+ {
905
+ "epoch": 0.8738229755178908,
906
+ "grad_norm": 0.16724838316440582,
907
+ "learning_rate": 8.251879438105854e-05,
908
+ "loss": 0.4167,
909
+ "step": 116
910
+ },
911
+ {
912
+ "epoch": 0.8813559322033898,
913
+ "grad_norm": 0.20615121722221375,
914
+ "learning_rate": 8.220860247998456e-05,
915
+ "loss": 0.3559,
916
+ "step": 117
917
+ },
918
+ {
919
+ "epoch": 0.8888888888888888,
920
+ "grad_norm": 0.2065141499042511,
921
+ "learning_rate": 8.189627706921877e-05,
922
+ "loss": 0.3626,
923
+ "step": 118
924
+ },
925
+ {
926
+ "epoch": 0.896421845574388,
927
+ "grad_norm": 0.17293284833431244,
928
+ "learning_rate": 8.15818388373114e-05,
929
+ "loss": 0.4325,
930
+ "step": 119
931
+ },
932
+ {
933
+ "epoch": 0.903954802259887,
934
+ "grad_norm": 0.19183604419231415,
935
+ "learning_rate": 8.126530861276677e-05,
936
+ "loss": 0.3387,
937
+ "step": 120
938
+ },
939
+ {
940
+ "epoch": 0.903954802259887,
941
+ "eval_loss": 0.38071778416633606,
942
+ "eval_runtime": 43.8754,
943
+ "eval_samples_per_second": 1.709,
944
+ "eval_steps_per_second": 0.866,
945
+ "step": 120
946
+ },
947
+ {
948
+ "epoch": 0.911487758945386,
949
+ "grad_norm": 0.1738622784614563,
950
+ "learning_rate": 8.094670736266353e-05,
951
+ "loss": 0.4149,
952
+ "step": 121
953
+ },
954
+ {
955
+ "epoch": 0.9190207156308852,
956
+ "grad_norm": 0.23153111338615417,
957
+ "learning_rate": 8.062605619126584e-05,
958
+ "loss": 0.4434,
959
+ "step": 122
960
+ },
961
+ {
962
+ "epoch": 0.9265536723163842,
963
+ "grad_norm": 0.26979267597198486,
964
+ "learning_rate": 8.030337633862542e-05,
965
+ "loss": 0.3639,
966
+ "step": 123
967
+ },
968
+ {
969
+ "epoch": 0.9340866290018832,
970
+ "grad_norm": 0.19837673008441925,
971
+ "learning_rate": 7.997868917917453e-05,
972
+ "loss": 0.3336,
973
+ "step": 124
974
+ },
975
+ {
976
+ "epoch": 0.9416195856873822,
977
+ "grad_norm": 0.18979325890541077,
978
+ "learning_rate": 7.965201622031021e-05,
979
+ "loss": 0.3715,
980
+ "step": 125
981
+ },
982
+ {
983
+ "epoch": 0.9491525423728814,
984
+ "grad_norm": 0.20212630927562714,
985
+ "learning_rate": 7.932337910096961e-05,
986
+ "loss": 0.333,
987
+ "step": 126
988
+ },
989
+ {
990
+ "epoch": 0.9566854990583804,
991
+ "grad_norm": 0.22705507278442383,
992
+ "learning_rate": 7.899279959019654e-05,
993
+ "loss": 0.3456,
994
+ "step": 127
995
+ },
996
+ {
997
+ "epoch": 0.9642184557438794,
998
+ "grad_norm": 0.24515411257743835,
999
+ "learning_rate": 7.866029958569956e-05,
1000
+ "loss": 0.3336,
1001
+ "step": 128
1002
+ },
1003
+ {
1004
+ "epoch": 0.9717514124293786,
1005
+ "grad_norm": 0.2214023470878601,
1006
+ "learning_rate": 7.832590111240145e-05,
1007
+ "loss": 0.3951,
1008
+ "step": 129
1009
+ },
1010
+ {
1011
+ "epoch": 0.9792843691148776,
1012
+ "grad_norm": 0.20626534521579742,
1013
+ "learning_rate": 7.798962632098024e-05,
1014
+ "loss": 0.3636,
1015
+ "step": 130
1016
+ },
1017
+ {
1018
+ "epoch": 0.9792843691148776,
1019
+ "eval_loss": 0.3497096598148346,
1020
+ "eval_runtime": 43.8538,
1021
+ "eval_samples_per_second": 1.71,
1022
+ "eval_steps_per_second": 0.867,
1023
+ "step": 130
1024
+ },
1025
+ {
1026
+ "epoch": 0.9868173258003766,
1027
+ "grad_norm": 0.21280796825885773,
1028
+ "learning_rate": 7.765149748640197e-05,
1029
+ "loss": 0.2931,
1030
+ "step": 131
1031
+ },
1032
+ {
1033
+ "epoch": 0.9943502824858758,
1034
+ "grad_norm": 0.18561464548110962,
1035
+ "learning_rate": 7.73115370064452e-05,
1036
+ "loss": 0.3363,
1037
+ "step": 132
1038
+ },
1039
+ {
1040
+ "epoch": 1.0075329566854991,
1041
+ "grad_norm": 0.3493395447731018,
1042
+ "learning_rate": 7.696976740021733e-05,
1043
+ "loss": 0.7106,
1044
+ "step": 133
1045
+ },
1046
+ {
1047
+ "epoch": 1.015065913370998,
1048
+ "grad_norm": 0.24458995461463928,
1049
+ "learning_rate": 7.6626211306663e-05,
1050
+ "loss": 0.2806,
1051
+ "step": 134
1052
+ },
1053
+ {
1054
+ "epoch": 1.0225988700564972,
1055
+ "grad_norm": 0.2562405467033386,
1056
+ "learning_rate": 7.628089148306434e-05,
1057
+ "loss": 0.3048,
1058
+ "step": 135
1059
+ },
1060
+ {
1061
+ "epoch": 1.0301318267419963,
1062
+ "grad_norm": 0.22578994929790497,
1063
+ "learning_rate": 7.59338308035337e-05,
1064
+ "loss": 0.3142,
1065
+ "step": 136
1066
+ },
1067
+ {
1068
+ "epoch": 1.0376647834274952,
1069
+ "grad_norm": 0.20230619609355927,
1070
+ "learning_rate": 7.558505225749827e-05,
1071
+ "loss": 0.3104,
1072
+ "step": 137
1073
+ },
1074
+ {
1075
+ "epoch": 1.0451977401129944,
1076
+ "grad_norm": 0.22089356184005737,
1077
+ "learning_rate": 7.523457894817745e-05,
1078
+ "loss": 0.2626,
1079
+ "step": 138
1080
+ },
1081
+ {
1082
+ "epoch": 1.0527306967984935,
1083
+ "grad_norm": 0.23944571614265442,
1084
+ "learning_rate": 7.488243409105233e-05,
1085
+ "loss": 0.2984,
1086
+ "step": 139
1087
+ },
1088
+ {
1089
+ "epoch": 1.0602636534839924,
1090
+ "grad_norm": 0.2213144749403,
1091
+ "learning_rate": 7.452864101232798e-05,
1092
+ "loss": 0.3714,
1093
+ "step": 140
1094
+ },
1095
+ {
1096
+ "epoch": 1.0602636534839924,
1097
+ "eval_loss": 0.31596410274505615,
1098
+ "eval_runtime": 43.9549,
1099
+ "eval_samples_per_second": 1.706,
1100
+ "eval_steps_per_second": 0.865,
1101
+ "step": 140
1102
+ },
1103
+ {
1104
+ "epoch": 1.0677966101694916,
1105
+ "grad_norm": 0.2673434019088745,
1106
+ "learning_rate": 7.417322314738822e-05,
1107
+ "loss": 0.3455,
1108
+ "step": 141
1109
+ },
1110
+ {
1111
+ "epoch": 1.0753295668549905,
1112
+ "grad_norm": 0.25528621673583984,
1113
+ "learning_rate": 7.381620403924333e-05,
1114
+ "loss": 0.311,
1115
+ "step": 142
1116
+ },
1117
+ {
1118
+ "epoch": 1.0828625235404896,
1119
+ "grad_norm": 0.3240588903427124,
1120
+ "learning_rate": 7.345760733697055e-05,
1121
+ "loss": 0.3363,
1122
+ "step": 143
1123
+ },
1124
+ {
1125
+ "epoch": 1.0903954802259888,
1126
+ "grad_norm": 0.31120482087135315,
1127
+ "learning_rate": 7.30974567941475e-05,
1128
+ "loss": 0.3529,
1129
+ "step": 144
1130
+ },
1131
+ {
1132
+ "epoch": 1.0979284369114877,
1133
+ "grad_norm": 0.2683728039264679,
1134
+ "learning_rate": 7.273577626727884e-05,
1135
+ "loss": 0.2926,
1136
+ "step": 145
1137
+ },
1138
+ {
1139
+ "epoch": 1.1054613935969868,
1140
+ "grad_norm": 0.2490655928850174,
1141
+ "learning_rate": 7.237258971421587e-05,
1142
+ "loss": 0.3029,
1143
+ "step": 146
1144
+ },
1145
+ {
1146
+ "epoch": 1.112994350282486,
1147
+ "grad_norm": 0.2618696093559265,
1148
+ "learning_rate": 7.20079211925696e-05,
1149
+ "loss": 0.278,
1150
+ "step": 147
1151
+ },
1152
+ {
1153
+ "epoch": 1.1205273069679849,
1154
+ "grad_norm": 0.25481289625167847,
1155
+ "learning_rate": 7.164179485811727e-05,
1156
+ "loss": 0.2936,
1157
+ "step": 148
1158
+ },
1159
+ {
1160
+ "epoch": 1.128060263653484,
1161
+ "grad_norm": 0.36896899342536926,
1162
+ "learning_rate": 7.127423496320212e-05,
1163
+ "loss": 0.2876,
1164
+ "step": 149
1165
+ },
1166
+ {
1167
+ "epoch": 1.1355932203389831,
1168
+ "grad_norm": 0.3275793194770813,
1169
+ "learning_rate": 7.090526585512696e-05,
1170
+ "loss": 0.3279,
1171
+ "step": 150
1172
+ },
1173
+ {
1174
+ "epoch": 1.1355932203389831,
1175
+ "eval_loss": 0.2823197543621063,
1176
+ "eval_runtime": 43.9363,
1177
+ "eval_samples_per_second": 1.707,
1178
+ "eval_steps_per_second": 0.865,
1179
+ "step": 150
1180
+ },
1181
+ {
1182
+ "epoch": 1.143126177024482,
1183
+ "grad_norm": 0.263179749250412,
1184
+ "learning_rate": 7.053491197454142e-05,
1185
+ "loss": 0.2342,
1186
+ "step": 151
1187
+ },
1188
+ {
1189
+ "epoch": 1.1506591337099812,
1190
+ "grad_norm": 0.2754746377468109,
1191
+ "learning_rate": 7.016319785382296e-05,
1192
+ "loss": 0.3234,
1193
+ "step": 152
1194
+ },
1195
+ {
1196
+ "epoch": 1.1581920903954803,
1197
+ "grad_norm": 0.29197824001312256,
1198
+ "learning_rate": 6.979014811545189e-05,
1199
+ "loss": 0.2458,
1200
+ "step": 153
1201
+ },
1202
+ {
1203
+ "epoch": 1.1657250470809792,
1204
+ "grad_norm": 0.45912623405456543,
1205
+ "learning_rate": 6.941578747038023e-05,
1206
+ "loss": 0.2357,
1207
+ "step": 154
1208
+ },
1209
+ {
1210
+ "epoch": 1.1732580037664784,
1211
+ "grad_norm": 0.36572179198265076,
1212
+ "learning_rate": 6.904014071639503e-05,
1213
+ "loss": 0.283,
1214
+ "step": 155
1215
+ },
1216
+ {
1217
+ "epoch": 1.1807909604519775,
1218
+ "grad_norm": 0.29460588097572327,
1219
+ "learning_rate": 6.866323273647563e-05,
1220
+ "loss": 0.2653,
1221
+ "step": 156
1222
+ },
1223
+ {
1224
+ "epoch": 1.1883239171374764,
1225
+ "grad_norm": 0.2520604729652405,
1226
+ "learning_rate": 6.828508849714546e-05,
1227
+ "loss": 0.3155,
1228
+ "step": 157
1229
+ },
1230
+ {
1231
+ "epoch": 1.1958568738229756,
1232
+ "grad_norm": 0.2615698575973511,
1233
+ "learning_rate": 6.79057330468182e-05,
1234
+ "loss": 0.1852,
1235
+ "step": 158
1236
+ },
1237
+ {
1238
+ "epoch": 1.2033898305084745,
1239
+ "grad_norm": 0.30008143186569214,
1240
+ "learning_rate": 6.752519151413861e-05,
1241
+ "loss": 0.2543,
1242
+ "step": 159
1243
+ },
1244
+ {
1245
+ "epoch": 1.2109227871939736,
1246
+ "grad_norm": 0.3347761929035187,
1247
+ "learning_rate": 6.7143489106318e-05,
1248
+ "loss": 0.3602,
1249
+ "step": 160
1250
+ },
1251
+ {
1252
+ "epoch": 1.2109227871939736,
1253
+ "eval_loss": 0.25164613127708435,
1254
+ "eval_runtime": 44.1497,
1255
+ "eval_samples_per_second": 1.699,
1256
+ "eval_steps_per_second": 0.861,
1257
+ "step": 160
1258
+ },
1259
+ {
1260
+ "epoch": 1.2184557438794728,
1261
+ "grad_norm": 0.3353792726993561,
1262
+ "learning_rate": 6.676065110746444e-05,
1263
+ "loss": 0.2489,
1264
+ "step": 161
1265
+ },
1266
+ {
1267
+ "epoch": 1.2259887005649717,
1268
+ "grad_norm": 0.32569241523742676,
1269
+ "learning_rate": 6.637670287690799e-05,
1270
+ "loss": 0.194,
1271
+ "step": 162
1272
+ },
1273
+ {
1274
+ "epoch": 1.2335216572504708,
1275
+ "grad_norm": 0.336117684841156,
1276
+ "learning_rate": 6.599166984752087e-05,
1277
+ "loss": 0.1816,
1278
+ "step": 163
1279
+ },
1280
+ {
1281
+ "epoch": 1.24105461393597,
1282
+ "grad_norm": 0.37371379137039185,
1283
+ "learning_rate": 6.560557752403277e-05,
1284
+ "loss": 0.2807,
1285
+ "step": 164
1286
+ },
1287
+ {
1288
+ "epoch": 1.2485875706214689,
1289
+ "grad_norm": 0.2796350419521332,
1290
+ "learning_rate": 6.52184514813414e-05,
1291
+ "loss": 0.2149,
1292
+ "step": 165
1293
+ },
1294
+ {
1295
+ "epoch": 1.256120527306968,
1296
+ "grad_norm": 0.4764145314693451,
1297
+ "learning_rate": 6.483031736281843e-05,
1298
+ "loss": 0.1903,
1299
+ "step": 166
1300
+ },
1301
+ {
1302
+ "epoch": 1.2636534839924671,
1303
+ "grad_norm": 0.36564409732818604,
1304
+ "learning_rate": 6.444120087861081e-05,
1305
+ "loss": 0.2638,
1306
+ "step": 167
1307
+ },
1308
+ {
1309
+ "epoch": 1.271186440677966,
1310
+ "grad_norm": 0.34570133686065674,
1311
+ "learning_rate": 6.40511278039378e-05,
1312
+ "loss": 0.2568,
1313
+ "step": 168
1314
+ },
1315
+ {
1316
+ "epoch": 1.2787193973634652,
1317
+ "grad_norm": 0.3726440966129303,
1318
+ "learning_rate": 6.366012397738355e-05,
1319
+ "loss": 0.2396,
1320
+ "step": 169
1321
+ },
1322
+ {
1323
+ "epoch": 1.286252354048964,
1324
+ "grad_norm": 0.22987572848796844,
1325
+ "learning_rate": 6.326821529918553e-05,
1326
+ "loss": 0.2168,
1327
+ "step": 170
1328
+ },
1329
+ {
1330
+ "epoch": 1.286252354048964,
1331
+ "eval_loss": 0.22047367691993713,
1332
+ "eval_runtime": 43.9518,
1333
+ "eval_samples_per_second": 1.706,
1334
+ "eval_steps_per_second": 0.865,
1335
+ "step": 170
1336
+ },
1337
+ {
1338
+ "epoch": 1.2937853107344632,
1339
+ "grad_norm": 0.38589489459991455,
1340
+ "learning_rate": 6.287542772951897e-05,
1341
+ "loss": 0.2671,
1342
+ "step": 171
1343
+ },
1344
+ {
1345
+ "epoch": 1.3013182674199624,
1346
+ "grad_norm": 0.26084208488464355,
1347
+ "learning_rate": 6.248178728677711e-05,
1348
+ "loss": 0.2628,
1349
+ "step": 172
1350
+ },
1351
+ {
1352
+ "epoch": 1.3088512241054615,
1353
+ "grad_norm": 0.3167526423931122,
1354
+ "learning_rate": 6.208732004584791e-05,
1355
+ "loss": 0.2128,
1356
+ "step": 173
1357
+ },
1358
+ {
1359
+ "epoch": 1.3163841807909604,
1360
+ "grad_norm": 0.3979116976261139,
1361
+ "learning_rate": 6.16920521363867e-05,
1362
+ "loss": 0.217,
1363
+ "step": 174
1364
+ },
1365
+ {
1366
+ "epoch": 1.3239171374764596,
1367
+ "grad_norm": 0.3168756663799286,
1368
+ "learning_rate": 6.129600974108538e-05,
1369
+ "loss": 0.2284,
1370
+ "step": 175
1371
+ },
1372
+ {
1373
+ "epoch": 1.3314500941619585,
1374
+ "grad_norm": 0.30765026807785034,
1375
+ "learning_rate": 6.089921909393812e-05,
1376
+ "loss": 0.2698,
1377
+ "step": 176
1378
+ },
1379
+ {
1380
+ "epoch": 1.3389830508474576,
1381
+ "grad_norm": 0.3305312991142273,
1382
+ "learning_rate": 6.050170647850351e-05,
1383
+ "loss": 0.257,
1384
+ "step": 177
1385
+ },
1386
+ {
1387
+ "epoch": 1.3465160075329567,
1388
+ "grad_norm": 0.6542951464653015,
1389
+ "learning_rate": 6.0103498226163603e-05,
1390
+ "loss": 0.2666,
1391
+ "step": 178
1392
+ },
1393
+ {
1394
+ "epoch": 1.3540489642184557,
1395
+ "grad_norm": 0.3485718071460724,
1396
+ "learning_rate": 5.970462071437973e-05,
1397
+ "loss": 0.1579,
1398
+ "step": 179
1399
+ },
1400
+ {
1401
+ "epoch": 1.3615819209039548,
1402
+ "grad_norm": 0.4073808789253235,
1403
+ "learning_rate": 5.93051003649452e-05,
1404
+ "loss": 0.1643,
1405
+ "step": 180
1406
+ },
1407
+ {
1408
+ "epoch": 1.3615819209039548,
1409
+ "eval_loss": 0.19270579516887665,
1410
+ "eval_runtime": 44.023,
1411
+ "eval_samples_per_second": 1.704,
1412
+ "eval_steps_per_second": 0.863,
1413
+ "step": 180
1414
+ },
1415
+ {
1416
+ "epoch": 1.369114877589454,
1417
+ "grad_norm": 0.37884190678596497,
1418
+ "learning_rate": 5.890496364223509e-05,
1419
+ "loss": 0.2015,
1420
+ "step": 181
1421
+ },
1422
+ {
1423
+ "epoch": 1.3766478342749529,
1424
+ "grad_norm": 0.2823837101459503,
1425
+ "learning_rate": 5.850423705145334e-05,
1426
+ "loss": 0.1774,
1427
+ "step": 182
1428
+ },
1429
+ {
1430
+ "epoch": 1.384180790960452,
1431
+ "grad_norm": 0.33302009105682373,
1432
+ "learning_rate": 5.8102947136876876e-05,
1433
+ "loss": 0.2202,
1434
+ "step": 183
1435
+ },
1436
+ {
1437
+ "epoch": 1.3917137476459511,
1438
+ "grad_norm": 0.30890604853630066,
1439
+ "learning_rate": 5.770112048009747e-05,
1440
+ "loss": 0.1372,
1441
+ "step": 184
1442
+ },
1443
+ {
1444
+ "epoch": 1.39924670433145,
1445
+ "grad_norm": 0.3033400774002075,
1446
+ "learning_rate": 5.7298783698260874e-05,
1447
+ "loss": 0.1857,
1448
+ "step": 185
1449
+ },
1450
+ {
1451
+ "epoch": 1.4067796610169492,
1452
+ "grad_norm": 0.2779131829738617,
1453
+ "learning_rate": 5.68959634423037e-05,
1454
+ "loss": 0.187,
1455
+ "step": 186
1456
+ },
1457
+ {
1458
+ "epoch": 1.414312617702448,
1459
+ "grad_norm": 0.485245943069458,
1460
+ "learning_rate": 5.64926863951881e-05,
1461
+ "loss": 0.2437,
1462
+ "step": 187
1463
+ },
1464
+ {
1465
+ "epoch": 1.4218455743879472,
1466
+ "grad_norm": 0.37200334668159485,
1467
+ "learning_rate": 5.60889792701342e-05,
1468
+ "loss": 0.1923,
1469
+ "step": 188
1470
+ },
1471
+ {
1472
+ "epoch": 1.4293785310734464,
1473
+ "grad_norm": 0.32306909561157227,
1474
+ "learning_rate": 5.568486880885068e-05,
1475
+ "loss": 0.1713,
1476
+ "step": 189
1477
+ },
1478
+ {
1479
+ "epoch": 1.4369114877589455,
1480
+ "grad_norm": 0.3483230471611023,
1481
+ "learning_rate": 5.52803817797633e-05,
1482
+ "loss": 0.1336,
1483
+ "step": 190
1484
+ },
1485
+ {
1486
+ "epoch": 1.4369114877589455,
1487
+ "eval_loss": 0.17033681273460388,
1488
+ "eval_runtime": 43.9231,
1489
+ "eval_samples_per_second": 1.708,
1490
+ "eval_steps_per_second": 0.865,
1491
+ "step": 190
1492
+ },
1493
+ {
1494
+ "epoch": 1.4444444444444444,
1495
+ "grad_norm": 0.3841436505317688,
1496
+ "learning_rate": 5.487554497624189e-05,
1497
+ "loss": 0.1797,
1498
+ "step": 191
1499
+ },
1500
+ {
1501
+ "epoch": 1.4519774011299436,
1502
+ "grad_norm": 0.3441992998123169,
1503
+ "learning_rate": 5.4470385214825416e-05,
1504
+ "loss": 0.2176,
1505
+ "step": 192
1506
+ },
1507
+ {
1508
+ "epoch": 1.4595103578154425,
1509
+ "grad_norm": 0.5087612867355347,
1510
+ "learning_rate": 5.406492933344571e-05,
1511
+ "loss": 0.2663,
1512
+ "step": 193
1513
+ },
1514
+ {
1515
+ "epoch": 1.4670433145009416,
1516
+ "grad_norm": 0.3248097598552704,
1517
+ "learning_rate": 5.365920418964973e-05,
1518
+ "loss": 0.1739,
1519
+ "step": 194
1520
+ },
1521
+ {
1522
+ "epoch": 1.4745762711864407,
1523
+ "grad_norm": 0.4247044622898102,
1524
+ "learning_rate": 5.3253236658820396e-05,
1525
+ "loss": 0.1454,
1526
+ "step": 195
1527
+ },
1528
+ {
1529
+ "epoch": 1.4821092278719397,
1530
+ "grad_norm": 0.2898694574832916,
1531
+ "learning_rate": 5.28470536323965e-05,
1532
+ "loss": 0.1609,
1533
+ "step": 196
1534
+ },
1535
+ {
1536
+ "epoch": 1.4896421845574388,
1537
+ "grad_norm": 0.41251513361930847,
1538
+ "learning_rate": 5.244068201609133e-05,
1539
+ "loss": 0.143,
1540
+ "step": 197
1541
+ },
1542
+ {
1543
+ "epoch": 1.497175141242938,
1544
+ "grad_norm": 0.525326132774353,
1545
+ "learning_rate": 5.2034148728110424e-05,
1546
+ "loss": 0.1938,
1547
+ "step": 198
1548
+ },
1549
+ {
1550
+ "epoch": 1.5047080979284368,
1551
+ "grad_norm": 0.36119720339775085,
1552
+ "learning_rate": 5.162748069736851e-05,
1553
+ "loss": 0.1518,
1554
+ "step": 199
1555
+ },
1556
+ {
1557
+ "epoch": 1.512241054613936,
1558
+ "grad_norm": 0.38599205017089844,
1559
+ "learning_rate": 5.1220704861705774e-05,
1560
+ "loss": 0.1198,
1561
+ "step": 200
1562
+ },
1563
+ {
1564
+ "epoch": 1.512241054613936,
1565
+ "eval_loss": 0.15222826600074768,
1566
+ "eval_runtime": 44.0589,
1567
+ "eval_samples_per_second": 1.702,
1568
+ "eval_steps_per_second": 0.862,
1569
+ "step": 200
1570
+ },
1571
+ {
1572
+ "epoch": 1.5197740112994351,
1573
+ "grad_norm": 0.36092710494995117,
1574
+ "learning_rate": 5.081384816610336e-05,
1575
+ "loss": 0.1485,
1576
+ "step": 201
1577
+ },
1578
+ {
1579
+ "epoch": 1.527306967984934,
1580
+ "grad_norm": 0.3418121039867401,
1581
+ "learning_rate": 5.0406937560898646e-05,
1582
+ "loss": 0.1368,
1583
+ "step": 202
1584
+ },
1585
+ {
1586
+ "epoch": 1.5348399246704332,
1587
+ "grad_norm": 0.39115968346595764,
1588
+ "learning_rate": 5e-05,
1589
+ "loss": 0.1843,
1590
+ "step": 203
1591
+ },
1592
+ {
1593
+ "epoch": 1.542372881355932,
1594
+ "grad_norm": 0.4022577106952667,
1595
+ "learning_rate": 4.9593062439101365e-05,
1596
+ "loss": 0.128,
1597
+ "step": 204
1598
+ },
1599
+ {
1600
+ "epoch": 1.5499058380414312,
1601
+ "grad_norm": 0.3265657126903534,
1602
+ "learning_rate": 4.918615183389665e-05,
1603
+ "loss": 0.1365,
1604
+ "step": 205
1605
+ },
1606
+ {
1607
+ "epoch": 1.5574387947269304,
1608
+ "grad_norm": 0.32065853476524353,
1609
+ "learning_rate": 4.877929513829424e-05,
1610
+ "loss": 0.1495,
1611
+ "step": 206
1612
+ },
1613
+ {
1614
+ "epoch": 1.5649717514124295,
1615
+ "grad_norm": 0.2755489647388458,
1616
+ "learning_rate": 4.8372519302631486e-05,
1617
+ "loss": 0.1105,
1618
+ "step": 207
1619
+ },
1620
+ {
1621
+ "epoch": 1.5725047080979284,
1622
+ "grad_norm": 0.5003211498260498,
1623
+ "learning_rate": 4.796585127188958e-05,
1624
+ "loss": 0.1284,
1625
+ "step": 208
1626
+ },
1627
+ {
1628
+ "epoch": 1.5800376647834273,
1629
+ "grad_norm": 0.43929675221443176,
1630
+ "learning_rate": 4.755931798390867e-05,
1631
+ "loss": 0.1889,
1632
+ "step": 209
1633
+ },
1634
+ {
1635
+ "epoch": 1.5875706214689265,
1636
+ "grad_norm": 0.2825930118560791,
1637
+ "learning_rate": 4.715294636760352e-05,
1638
+ "loss": 0.1364,
1639
+ "step": 210
1640
+ },
1641
+ {
1642
+ "epoch": 1.5875706214689265,
1643
+ "eval_loss": 0.13791552186012268,
1644
+ "eval_runtime": 43.9041,
1645
+ "eval_samples_per_second": 1.708,
1646
+ "eval_steps_per_second": 0.866,
1647
+ "step": 210
1648
+ },
1649
+ {
1650
+ "epoch": 1.5951035781544256,
1651
+ "grad_norm": 0.3584269881248474,
1652
+ "learning_rate": 4.674676334117962e-05,
1653
+ "loss": 0.1261,
1654
+ "step": 211
1655
+ },
1656
+ {
1657
+ "epoch": 1.6026365348399247,
1658
+ "grad_norm": 0.3463616967201233,
1659
+ "learning_rate": 4.634079581035029e-05,
1660
+ "loss": 0.185,
1661
+ "step": 212
1662
+ },
1663
+ {
1664
+ "epoch": 1.6101694915254239,
1665
+ "grad_norm": 0.3442526161670685,
1666
+ "learning_rate": 4.59350706665543e-05,
1667
+ "loss": 0.104,
1668
+ "step": 213
1669
+ },
1670
+ {
1671
+ "epoch": 1.6177024482109228,
1672
+ "grad_norm": 0.3831660747528076,
1673
+ "learning_rate": 4.55296147851746e-05,
1674
+ "loss": 0.1294,
1675
+ "step": 214
1676
+ },
1677
+ {
1678
+ "epoch": 1.6252354048964217,
1679
+ "grad_norm": 0.3194776475429535,
1680
+ "learning_rate": 4.512445502375813e-05,
1681
+ "loss": 0.1068,
1682
+ "step": 215
1683
+ },
1684
+ {
1685
+ "epoch": 1.6327683615819208,
1686
+ "grad_norm": 0.3435104787349701,
1687
+ "learning_rate": 4.471961822023671e-05,
1688
+ "loss": 0.0864,
1689
+ "step": 216
1690
+ },
1691
+ {
1692
+ "epoch": 1.64030131826742,
1693
+ "grad_norm": 0.2775423526763916,
1694
+ "learning_rate": 4.431513119114934e-05,
1695
+ "loss": 0.1858,
1696
+ "step": 217
1697
+ },
1698
+ {
1699
+ "epoch": 1.6478342749529191,
1700
+ "grad_norm": 0.3337323069572449,
1701
+ "learning_rate": 4.391102072986581e-05,
1702
+ "loss": 0.1946,
1703
+ "step": 218
1704
+ },
1705
+ {
1706
+ "epoch": 1.655367231638418,
1707
+ "grad_norm": 0.32516106963157654,
1708
+ "learning_rate": 4.350731360481191e-05,
1709
+ "loss": 0.1014,
1710
+ "step": 219
1711
+ },
1712
+ {
1713
+ "epoch": 1.6629001883239172,
1714
+ "grad_norm": 0.2920524775981903,
1715
+ "learning_rate": 4.3104036557696295e-05,
1716
+ "loss": 0.2689,
1717
+ "step": 220
1718
+ },
1719
+ {
1720
+ "epoch": 1.6629001883239172,
1721
+ "eval_loss": 0.12488219887018204,
1722
+ "eval_runtime": 43.9386,
1723
+ "eval_samples_per_second": 1.707,
1724
+ "eval_steps_per_second": 0.865,
1725
+ "step": 220
1726
+ },
1727
+ {
1728
+ "epoch": 1.670433145009416,
1729
+ "grad_norm": 0.4062795042991638,
1730
+ "learning_rate": 4.270121630173913e-05,
1731
+ "loss": 0.1495,
1732
+ "step": 221
1733
+ },
1734
+ {
1735
+ "epoch": 1.6779661016949152,
1736
+ "grad_norm": 0.30497950315475464,
1737
+ "learning_rate": 4.229887951990255e-05,
1738
+ "loss": 0.1369,
1739
+ "step": 222
1740
+ },
1741
+ {
1742
+ "epoch": 1.6854990583804144,
1743
+ "grad_norm": 0.26471954584121704,
1744
+ "learning_rate": 4.189705286312314e-05,
1745
+ "loss": 0.0776,
1746
+ "step": 223
1747
+ },
1748
+ {
1749
+ "epoch": 1.6930320150659135,
1750
+ "grad_norm": 0.22838452458381653,
1751
+ "learning_rate": 4.149576294854668e-05,
1752
+ "loss": 0.0942,
1753
+ "step": 224
1754
+ },
1755
+ {
1756
+ "epoch": 1.7005649717514124,
1757
+ "grad_norm": 0.2828744649887085,
1758
+ "learning_rate": 4.1095036357764915e-05,
1759
+ "loss": 0.0787,
1760
+ "step": 225
1761
+ },
1762
+ {
1763
+ "epoch": 1.7080979284369113,
1764
+ "grad_norm": 0.3124982714653015,
1765
+ "learning_rate": 4.069489963505482e-05,
1766
+ "loss": 0.13,
1767
+ "step": 226
1768
+ },
1769
+ {
1770
+ "epoch": 1.7156308851224105,
1771
+ "grad_norm": 0.3018610179424286,
1772
+ "learning_rate": 4.029537928562028e-05,
1773
+ "loss": 0.1263,
1774
+ "step": 227
1775
+ },
1776
+ {
1777
+ "epoch": 1.7231638418079096,
1778
+ "grad_norm": 0.28386422991752625,
1779
+ "learning_rate": 3.98965017738364e-05,
1780
+ "loss": 0.089,
1781
+ "step": 228
1782
+ },
1783
+ {
1784
+ "epoch": 1.7306967984934087,
1785
+ "grad_norm": 0.24454158544540405,
1786
+ "learning_rate": 3.9498293521496503e-05,
1787
+ "loss": 0.1525,
1788
+ "step": 229
1789
+ },
1790
+ {
1791
+ "epoch": 1.7382297551789079,
1792
+ "grad_norm": 0.29519349336624146,
1793
+ "learning_rate": 3.9100780906061896e-05,
1794
+ "loss": 0.1383,
1795
+ "step": 230
1796
+ },
1797
+ {
1798
+ "epoch": 1.7382297551789079,
1799
+ "eval_loss": 0.11655016243457794,
1800
+ "eval_runtime": 43.924,
1801
+ "eval_samples_per_second": 1.707,
1802
+ "eval_steps_per_second": 0.865,
1803
+ "step": 230
1804
+ },
1805
+ {
1806
+ "epoch": 1.7457627118644068,
1807
+ "grad_norm": 0.2894473075866699,
1808
+ "learning_rate": 3.8703990258914614e-05,
1809
+ "loss": 0.1709,
1810
+ "step": 231
1811
+ },
1812
+ {
1813
+ "epoch": 1.7532956685499057,
1814
+ "grad_norm": 0.31395775079727173,
1815
+ "learning_rate": 3.83079478636133e-05,
1816
+ "loss": 0.0944,
1817
+ "step": 232
1818
+ },
1819
+ {
1820
+ "epoch": 1.7608286252354048,
1821
+ "grad_norm": 0.2754004895687103,
1822
+ "learning_rate": 3.791267995415208e-05,
1823
+ "loss": 0.1075,
1824
+ "step": 233
1825
+ },
1826
+ {
1827
+ "epoch": 1.768361581920904,
1828
+ "grad_norm": 0.27213406562805176,
1829
+ "learning_rate": 3.7518212713222906e-05,
1830
+ "loss": 0.1264,
1831
+ "step": 234
1832
+ },
1833
+ {
1834
+ "epoch": 1.7758945386064031,
1835
+ "grad_norm": 0.44370901584625244,
1836
+ "learning_rate": 3.7124572270481056e-05,
1837
+ "loss": 0.1066,
1838
+ "step": 235
1839
+ },
1840
+ {
1841
+ "epoch": 1.783427495291902,
1842
+ "grad_norm": 0.27574431896209717,
1843
+ "learning_rate": 3.673178470081448e-05,
1844
+ "loss": 0.1138,
1845
+ "step": 236
1846
+ },
1847
+ {
1848
+ "epoch": 1.7909604519774012,
1849
+ "grad_norm": 0.25503936409950256,
1850
+ "learning_rate": 3.633987602261647e-05,
1851
+ "loss": 0.0913,
1852
+ "step": 237
1853
+ },
1854
+ {
1855
+ "epoch": 1.7984934086629,
1856
+ "grad_norm": 0.33877211809158325,
1857
+ "learning_rate": 3.594887219606221e-05,
1858
+ "loss": 0.1084,
1859
+ "step": 238
1860
+ },
1861
+ {
1862
+ "epoch": 1.8060263653483992,
1863
+ "grad_norm": 0.3899424374103546,
1864
+ "learning_rate": 3.55587991213892e-05,
1865
+ "loss": 0.203,
1866
+ "step": 239
1867
+ },
1868
+ {
1869
+ "epoch": 1.8135593220338984,
1870
+ "grad_norm": 0.3390370011329651,
1871
+ "learning_rate": 3.516968263718159e-05,
1872
+ "loss": 0.1145,
1873
+ "step": 240
1874
+ },
1875
+ {
1876
+ "epoch": 1.8135593220338984,
1877
+ "eval_loss": 0.11021808534860611,
1878
+ "eval_runtime": 43.9451,
1879
+ "eval_samples_per_second": 1.707,
1880
+ "eval_steps_per_second": 0.865,
1881
+ "step": 240
1882
+ },
1883
+ {
1884
+ "epoch": 1.8210922787193975,
1885
+ "grad_norm": 0.33010998368263245,
1886
+ "learning_rate": 3.47815485186586e-05,
1887
+ "loss": 0.0726,
1888
+ "step": 241
1889
+ },
1890
+ {
1891
+ "epoch": 1.8286252354048964,
1892
+ "grad_norm": 0.280018150806427,
1893
+ "learning_rate": 3.439442247596724e-05,
1894
+ "loss": 0.1107,
1895
+ "step": 242
1896
+ },
1897
+ {
1898
+ "epoch": 1.8361581920903953,
1899
+ "grad_norm": 0.2712264657020569,
1900
+ "learning_rate": 3.400833015247913e-05,
1901
+ "loss": 0.0737,
1902
+ "step": 243
1903
+ },
1904
+ {
1905
+ "epoch": 1.8436911487758945,
1906
+ "grad_norm": 0.2368086278438568,
1907
+ "learning_rate": 3.3623297123092006e-05,
1908
+ "loss": 0.1474,
1909
+ "step": 244
1910
+ },
1911
+ {
1912
+ "epoch": 1.8512241054613936,
1913
+ "grad_norm": 0.29197973012924194,
1914
+ "learning_rate": 3.323934889253556e-05,
1915
+ "loss": 0.078,
1916
+ "step": 245
1917
+ },
1918
+ {
1919
+ "epoch": 1.8587570621468927,
1920
+ "grad_norm": 0.2464105784893036,
1921
+ "learning_rate": 3.285651089368202e-05,
1922
+ "loss": 0.1029,
1923
+ "step": 246
1924
+ },
1925
+ {
1926
+ "epoch": 1.8662900188323919,
1927
+ "grad_norm": 0.24727390706539154,
1928
+ "learning_rate": 3.2474808485861397e-05,
1929
+ "loss": 0.0552,
1930
+ "step": 247
1931
+ },
1932
+ {
1933
+ "epoch": 1.8738229755178908,
1934
+ "grad_norm": 0.23096492886543274,
1935
+ "learning_rate": 3.209426695318182e-05,
1936
+ "loss": 0.0818,
1937
+ "step": 248
1938
+ },
1939
+ {
1940
+ "epoch": 1.8813559322033897,
1941
+ "grad_norm": 0.23165109753608704,
1942
+ "learning_rate": 3.171491150285456e-05,
1943
+ "loss": 0.1493,
1944
+ "step": 249
1945
+ },
1946
+ {
1947
+ "epoch": 1.8888888888888888,
1948
+ "grad_norm": 0.3371273875236511,
1949
+ "learning_rate": 3.133676726352438e-05,
1950
+ "loss": 0.1037,
1951
+ "step": 250
1952
+ },
1953
+ {
1954
+ "epoch": 1.8888888888888888,
1955
+ "eval_loss": 0.10490331053733826,
1956
+ "eval_runtime": 43.9307,
1957
+ "eval_samples_per_second": 1.707,
1958
+ "eval_steps_per_second": 0.865,
1959
+ "step": 250
1960
+ },
1961
+ {
1962
+ "epoch": 1.896421845574388,
1963
+ "grad_norm": 0.2399485558271408,
1964
+ "learning_rate": 3.0959859283604984e-05,
1965
+ "loss": 0.0929,
1966
+ "step": 251
1967
+ },
1968
+ {
1969
+ "epoch": 1.9039548022598871,
1970
+ "grad_norm": 0.26773276925086975,
1971
+ "learning_rate": 3.0584212529619775e-05,
1972
+ "loss": 0.0637,
1973
+ "step": 252
1974
+ },
1975
+ {
1976
+ "epoch": 1.911487758945386,
1977
+ "grad_norm": 0.41889140009880066,
1978
+ "learning_rate": 3.0209851884548117e-05,
1979
+ "loss": 0.1263,
1980
+ "step": 253
1981
+ },
1982
+ {
1983
+ "epoch": 1.9190207156308852,
1984
+ "grad_norm": 0.2959991991519928,
1985
+ "learning_rate": 2.9836802146177034e-05,
1986
+ "loss": 0.1339,
1987
+ "step": 254
1988
+ },
1989
+ {
1990
+ "epoch": 1.926553672316384,
1991
+ "grad_norm": 0.24563738703727722,
1992
+ "learning_rate": 2.9465088025458586e-05,
1993
+ "loss": 0.1496,
1994
+ "step": 255
1995
+ },
1996
+ {
1997
+ "epoch": 1.9340866290018832,
1998
+ "grad_norm": 0.24098043143749237,
1999
+ "learning_rate": 2.9094734144873036e-05,
2000
+ "loss": 0.0539,
2001
+ "step": 256
2002
+ },
2003
+ {
2004
+ "epoch": 1.9416195856873824,
2005
+ "grad_norm": 0.3055378198623657,
2006
+ "learning_rate": 2.8725765036797892e-05,
2007
+ "loss": 0.1087,
2008
+ "step": 257
2009
+ },
2010
+ {
2011
+ "epoch": 1.9491525423728815,
2012
+ "grad_norm": 0.27689605951309204,
2013
+ "learning_rate": 2.835820514188273e-05,
2014
+ "loss": 0.1004,
2015
+ "step": 258
2016
+ },
2017
+ {
2018
+ "epoch": 1.9566854990583804,
2019
+ "grad_norm": 0.30273178219795227,
2020
+ "learning_rate": 2.7992078807430422e-05,
2021
+ "loss": 0.1261,
2022
+ "step": 259
2023
+ },
2024
+ {
2025
+ "epoch": 1.9642184557438793,
2026
+ "grad_norm": 0.3399432897567749,
2027
+ "learning_rate": 2.7627410285784163e-05,
2028
+ "loss": 0.1318,
2029
+ "step": 260
2030
+ },
2031
+ {
2032
+ "epoch": 1.9642184557438793,
2033
+ "eval_loss": 0.10169798880815506,
2034
+ "eval_runtime": 43.9845,
2035
+ "eval_samples_per_second": 1.705,
2036
+ "eval_steps_per_second": 0.864,
2037
+ "step": 260
2038
+ },
2039
+ {
2040
+ "epoch": 1.9717514124293785,
2041
+ "grad_norm": 0.2374039739370346,
2042
+ "learning_rate": 2.7264223732721167e-05,
2043
+ "loss": 0.0795,
2044
+ "step": 261
2045
+ },
2046
+ {
2047
+ "epoch": 1.9792843691148776,
2048
+ "grad_norm": 0.2347286343574524,
2049
+ "learning_rate": 2.6902543205852492e-05,
2050
+ "loss": 0.1129,
2051
+ "step": 262
2052
+ },
2053
+ {
2054
+ "epoch": 1.9868173258003767,
2055
+ "grad_norm": 0.2325945943593979,
2056
+ "learning_rate": 2.6542392663029463e-05,
2057
+ "loss": 0.117,
2058
+ "step": 263
2059
+ },
2060
+ {
2061
+ "epoch": 1.9943502824858759,
2062
+ "grad_norm": 0.2451542764902115,
2063
+ "learning_rate": 2.618379596075668e-05,
2064
+ "loss": 0.0742,
2065
+ "step": 264
2066
+ },
2067
+ {
2068
+ "epoch": 2.007532956685499,
2069
+ "grad_norm": 0.4711216390132904,
2070
+ "learning_rate": 2.582677685261179e-05,
2071
+ "loss": 0.2505,
2072
+ "step": 265
2073
+ },
2074
+ {
2075
+ "epoch": 2.0150659133709983,
2076
+ "grad_norm": 0.25324326753616333,
2077
+ "learning_rate": 2.5471358987672017e-05,
2078
+ "loss": 0.0925,
2079
+ "step": 266
2080
+ },
2081
+ {
2082
+ "epoch": 2.022598870056497,
2083
+ "grad_norm": 0.24556219577789307,
2084
+ "learning_rate": 2.511756590894765e-05,
2085
+ "loss": 0.069,
2086
+ "step": 267
2087
+ },
2088
+ {
2089
+ "epoch": 2.030131826741996,
2090
+ "grad_norm": 0.20775027573108673,
2091
+ "learning_rate": 2.476542105182254e-05,
2092
+ "loss": 0.0687,
2093
+ "step": 268
2094
+ },
2095
+ {
2096
+ "epoch": 2.0376647834274952,
2097
+ "grad_norm": 0.2542414367198944,
2098
+ "learning_rate": 2.4414947742501744e-05,
2099
+ "loss": 0.0769,
2100
+ "step": 269
2101
+ },
2102
+ {
2103
+ "epoch": 2.0451977401129944,
2104
+ "grad_norm": 0.26574474573135376,
2105
+ "learning_rate": 2.4066169196466326e-05,
2106
+ "loss": 0.0834,
2107
+ "step": 270
2108
+ },
2109
+ {
2110
+ "epoch": 2.0451977401129944,
2111
+ "eval_loss": 0.0993381068110466,
2112
+ "eval_runtime": 43.9044,
2113
+ "eval_samples_per_second": 1.708,
2114
+ "eval_steps_per_second": 0.866,
2115
+ "step": 270
2116
+ },
2117
+ {
2118
+ "epoch": 2.0527306967984935,
2119
+ "grad_norm": 0.3231644034385681,
2120
+ "learning_rate": 2.3719108516935683e-05,
2121
+ "loss": 0.0774,
2122
+ "step": 271
2123
+ },
2124
+ {
2125
+ "epoch": 2.0602636534839927,
2126
+ "grad_norm": 0.22078053653240204,
2127
+ "learning_rate": 2.3373788693337024e-05,
2128
+ "loss": 0.0884,
2129
+ "step": 272
2130
+ },
2131
+ {
2132
+ "epoch": 2.0677966101694913,
2133
+ "grad_norm": 0.22590897977352142,
2134
+ "learning_rate": 2.303023259978267e-05,
2135
+ "loss": 0.0781,
2136
+ "step": 273
2137
+ },
2138
+ {
2139
+ "epoch": 2.0753295668549905,
2140
+ "grad_norm": 0.23885948956012726,
2141
+ "learning_rate": 2.268846299355481e-05,
2142
+ "loss": 0.0543,
2143
+ "step": 274
2144
+ },
2145
+ {
2146
+ "epoch": 2.0828625235404896,
2147
+ "grad_norm": 0.2120169848203659,
2148
+ "learning_rate": 2.2348502513598035e-05,
2149
+ "loss": 0.0925,
2150
+ "step": 275
2151
+ },
2152
+ {
2153
+ "epoch": 2.0903954802259888,
2154
+ "grad_norm": 0.2760743498802185,
2155
+ "learning_rate": 2.2010373679019776e-05,
2156
+ "loss": 0.1427,
2157
+ "step": 276
2158
+ },
2159
+ {
2160
+ "epoch": 2.097928436911488,
2161
+ "grad_norm": 0.3039568066596985,
2162
+ "learning_rate": 2.167409888759856e-05,
2163
+ "loss": 0.1766,
2164
+ "step": 277
2165
+ },
2166
+ {
2167
+ "epoch": 2.105461393596987,
2168
+ "grad_norm": 0.2834893763065338,
2169
+ "learning_rate": 2.133970041430044e-05,
2170
+ "loss": 0.0816,
2171
+ "step": 278
2172
+ },
2173
+ {
2174
+ "epoch": 2.1129943502824857,
2175
+ "grad_norm": 0.2928575277328491,
2176
+ "learning_rate": 2.1007200409803462e-05,
2177
+ "loss": 0.1076,
2178
+ "step": 279
2179
+ },
2180
+ {
2181
+ "epoch": 2.120527306967985,
2182
+ "grad_norm": 0.2669938802719116,
2183
+ "learning_rate": 2.067662089903039e-05,
2184
+ "loss": 0.0889,
2185
+ "step": 280
2186
+ },
2187
+ {
2188
+ "epoch": 2.120527306967985,
2189
+ "eval_loss": 0.09731351584196091,
2190
+ "eval_runtime": 44.078,
2191
+ "eval_samples_per_second": 1.702,
2192
+ "eval_steps_per_second": 0.862,
2193
+ "step": 280
2194
+ },
2195
+ {
2196
+ "epoch": 2.128060263653484,
2197
+ "grad_norm": 0.28592410683631897,
2198
+ "learning_rate": 2.03479837796898e-05,
2199
+ "loss": 0.0685,
2200
+ "step": 281
2201
+ },
2202
+ {
2203
+ "epoch": 2.135593220338983,
2204
+ "grad_norm": 0.32021915912628174,
2205
+ "learning_rate": 2.002131082082549e-05,
2206
+ "loss": 0.0927,
2207
+ "step": 282
2208
+ },
2209
+ {
2210
+ "epoch": 2.1431261770244823,
2211
+ "grad_norm": 0.2118028998374939,
2212
+ "learning_rate": 1.9696623661374618e-05,
2213
+ "loss": 0.0634,
2214
+ "step": 283
2215
+ },
2216
+ {
2217
+ "epoch": 2.150659133709981,
2218
+ "grad_norm": 0.22075195610523224,
2219
+ "learning_rate": 1.937394380873418e-05,
2220
+ "loss": 0.1464,
2221
+ "step": 284
2222
+ },
2223
+ {
2224
+ "epoch": 2.15819209039548,
2225
+ "grad_norm": 0.24222980439662933,
2226
+ "learning_rate": 1.905329263733649e-05,
2227
+ "loss": 0.1251,
2228
+ "step": 285
2229
+ },
2230
+ {
2231
+ "epoch": 2.1657250470809792,
2232
+ "grad_norm": 0.28131377696990967,
2233
+ "learning_rate": 1.873469138723325e-05,
2234
+ "loss": 0.1184,
2235
+ "step": 286
2236
+ },
2237
+ {
2238
+ "epoch": 2.1732580037664784,
2239
+ "grad_norm": 0.22016428411006927,
2240
+ "learning_rate": 1.8418161162688615e-05,
2241
+ "loss": 0.0399,
2242
+ "step": 287
2243
+ },
2244
+ {
2245
+ "epoch": 2.1807909604519775,
2246
+ "grad_norm": 0.19806838035583496,
2247
+ "learning_rate": 1.8103722930781247e-05,
2248
+ "loss": 0.0756,
2249
+ "step": 288
2250
+ },
2251
+ {
2252
+ "epoch": 2.1883239171374766,
2253
+ "grad_norm": 0.332441121339798,
2254
+ "learning_rate": 1.779139752001545e-05,
2255
+ "loss": 0.0938,
2256
+ "step": 289
2257
+ },
2258
+ {
2259
+ "epoch": 2.1958568738229753,
2260
+ "grad_norm": 0.19058367609977722,
2261
+ "learning_rate": 1.748120561894147e-05,
2262
+ "loss": 0.0643,
2263
+ "step": 290
2264
+ },
2265
+ {
2266
+ "epoch": 2.1958568738229753,
2267
+ "eval_loss": 0.09646200388669968,
2268
+ "eval_runtime": 43.99,
2269
+ "eval_samples_per_second": 1.705,
2270
+ "eval_steps_per_second": 0.864,
2271
+ "step": 290
2272
+ },
2273
+ {
2274
+ "epoch": 2.2033898305084745,
2275
+ "grad_norm": 0.168031245470047,
2276
+ "learning_rate": 1.7173167774785092e-05,
2277
+ "loss": 0.0524,
2278
+ "step": 291
2279
+ },
2280
+ {
2281
+ "epoch": 2.2109227871939736,
2282
+ "grad_norm": 0.23299571871757507,
2283
+ "learning_rate": 1.6867304392086575e-05,
2284
+ "loss": 0.0775,
2285
+ "step": 292
2286
+ },
2287
+ {
2288
+ "epoch": 2.2184557438794728,
2289
+ "grad_norm": 0.258078008890152,
2290
+ "learning_rate": 1.6563635731349057e-05,
2291
+ "loss": 0.0662,
2292
+ "step": 293
2293
+ },
2294
+ {
2295
+ "epoch": 2.225988700564972,
2296
+ "grad_norm": 0.21041317284107208,
2297
+ "learning_rate": 1.6262181907696454e-05,
2298
+ "loss": 0.1229,
2299
+ "step": 294
2300
+ },
2301
+ {
2302
+ "epoch": 2.2335216572504706,
2303
+ "grad_norm": 0.1758544147014618,
2304
+ "learning_rate": 1.5962962889541105e-05,
2305
+ "loss": 0.0487,
2306
+ "step": 295
2307
+ },
2308
+ {
2309
+ "epoch": 2.2410546139359697,
2310
+ "grad_norm": 0.24289867281913757,
2311
+ "learning_rate": 1.5665998497260958e-05,
2312
+ "loss": 0.103,
2313
+ "step": 296
2314
+ },
2315
+ {
2316
+ "epoch": 2.248587570621469,
2317
+ "grad_norm": 0.41178449988365173,
2318
+ "learning_rate": 1.5371308401886757e-05,
2319
+ "loss": 0.1671,
2320
+ "step": 297
2321
+ },
2322
+ {
2323
+ "epoch": 2.256120527306968,
2324
+ "grad_norm": 0.22324170172214508,
2325
+ "learning_rate": 1.5078912123798961e-05,
2326
+ "loss": 0.089,
2327
+ "step": 298
2328
+ },
2329
+ {
2330
+ "epoch": 2.263653483992467,
2331
+ "grad_norm": 0.2477371245622635,
2332
+ "learning_rate": 1.4788829031434732e-05,
2333
+ "loss": 0.0397,
2334
+ "step": 299
2335
+ },
2336
+ {
2337
+ "epoch": 2.2711864406779663,
2338
+ "grad_norm": 0.21119807660579681,
2339
+ "learning_rate": 1.4501078340004953e-05,
2340
+ "loss": 0.0677,
2341
+ "step": 300
2342
+ },
2343
+ {
2344
+ "epoch": 2.2711864406779663,
2345
+ "eval_loss": 0.09540043771266937,
2346
+ "eval_runtime": 43.9302,
2347
+ "eval_samples_per_second": 1.707,
2348
+ "eval_steps_per_second": 0.865,
2349
+ "step": 300
2350
+ },
2351
+ {
2352
+ "epoch": 2.2787193973634654,
2353
+ "grad_norm": 0.24356690049171448,
2354
+ "learning_rate": 1.4215679110221413e-05,
2355
+ "loss": 0.1493,
2356
+ "step": 301
2357
+ },
2358
+ {
2359
+ "epoch": 2.286252354048964,
2360
+ "grad_norm": 0.180653914809227,
2361
+ "learning_rate": 1.3932650247034218e-05,
2362
+ "loss": 0.0703,
2363
+ "step": 302
2364
+ },
2365
+ {
2366
+ "epoch": 2.2937853107344632,
2367
+ "grad_norm": 0.3206869959831238,
2368
+ "learning_rate": 1.3652010498379519e-05,
2369
+ "loss": 0.1167,
2370
+ "step": 303
2371
+ },
2372
+ {
2373
+ "epoch": 2.3013182674199624,
2374
+ "grad_norm": 0.21196141839027405,
2375
+ "learning_rate": 1.337377845393763e-05,
2376
+ "loss": 0.0721,
2377
+ "step": 304
2378
+ },
2379
+ {
2380
+ "epoch": 2.3088512241054615,
2381
+ "grad_norm": 0.34639954566955566,
2382
+ "learning_rate": 1.309797254390167e-05,
2383
+ "loss": 0.1114,
2384
+ "step": 305
2385
+ },
2386
+ {
2387
+ "epoch": 2.3163841807909606,
2388
+ "grad_norm": 0.28393760323524475,
2389
+ "learning_rate": 1.2824611037756684e-05,
2390
+ "loss": 0.0527,
2391
+ "step": 306
2392
+ },
2393
+ {
2394
+ "epoch": 2.3239171374764593,
2395
+ "grad_norm": 0.2436118721961975,
2396
+ "learning_rate": 1.255371204306956e-05,
2397
+ "loss": 0.0825,
2398
+ "step": 307
2399
+ },
2400
+ {
2401
+ "epoch": 2.3314500941619585,
2402
+ "grad_norm": 0.2877277135848999,
2403
+ "learning_rate": 1.2285293504289447e-05,
2404
+ "loss": 0.0721,
2405
+ "step": 308
2406
+ },
2407
+ {
2408
+ "epoch": 2.3389830508474576,
2409
+ "grad_norm": 0.18935468792915344,
2410
+ "learning_rate": 1.2019373201559247e-05,
2411
+ "loss": 0.0826,
2412
+ "step": 309
2413
+ },
2414
+ {
2415
+ "epoch": 2.3465160075329567,
2416
+ "grad_norm": 0.3491658568382263,
2417
+ "learning_rate": 1.1755968749537754e-05,
2418
+ "loss": 0.1784,
2419
+ "step": 310
2420
+ },
2421
+ {
2422
+ "epoch": 2.3465160075329567,
2423
+ "eval_loss": 0.09475459903478622,
2424
+ "eval_runtime": 43.9621,
2425
+ "eval_samples_per_second": 1.706,
2426
+ "eval_steps_per_second": 0.864,
2427
+ "step": 310
2428
+ },
2429
+ {
2430
+ "epoch": 2.354048964218456,
2431
+ "grad_norm": 0.31381261348724365,
2432
+ "learning_rate": 1.1495097596232901e-05,
2433
+ "loss": 0.1266,
2434
+ "step": 311
2435
+ },
2436
+ {
2437
+ "epoch": 2.361581920903955,
2438
+ "grad_norm": 0.23978573083877563,
2439
+ "learning_rate": 1.1236777021845956e-05,
2440
+ "loss": 0.1047,
2441
+ "step": 312
2442
+ },
2443
+ {
2444
+ "epoch": 2.3691148775894537,
2445
+ "grad_norm": 0.24144862592220306,
2446
+ "learning_rate": 1.0981024137626922e-05,
2447
+ "loss": 0.0572,
2448
+ "step": 313
2449
+ },
2450
+ {
2451
+ "epoch": 2.376647834274953,
2452
+ "grad_norm": 0.21653713285923004,
2453
+ "learning_rate": 1.0727855884741056e-05,
2454
+ "loss": 0.0672,
2455
+ "step": 314
2456
+ },
2457
+ {
2458
+ "epoch": 2.384180790960452,
2459
+ "grad_norm": 0.22251500189304352,
2460
+ "learning_rate": 1.0477289033146675e-05,
2461
+ "loss": 0.0598,
2462
+ "step": 315
2463
+ },
2464
+ {
2465
+ "epoch": 2.391713747645951,
2466
+ "grad_norm": 0.297451913356781,
2467
+ "learning_rate": 1.022934018048432e-05,
2468
+ "loss": 0.0518,
2469
+ "step": 316
2470
+ },
2471
+ {
2472
+ "epoch": 2.3992467043314503,
2473
+ "grad_norm": 0.29483604431152344,
2474
+ "learning_rate": 9.984025750977339e-06,
2475
+ "loss": 0.0875,
2476
+ "step": 317
2477
+ },
2478
+ {
2479
+ "epoch": 2.406779661016949,
2480
+ "grad_norm": 0.3049258887767792,
2481
+ "learning_rate": 9.741361994343867e-06,
2482
+ "loss": 0.0843,
2483
+ "step": 318
2484
+ },
2485
+ {
2486
+ "epoch": 2.414312617702448,
2487
+ "grad_norm": 0.21589063107967377,
2488
+ "learning_rate": 9.501364984720557e-06,
2489
+ "loss": 0.0588,
2490
+ "step": 319
2491
+ },
2492
+ {
2493
+ "epoch": 2.4218455743879472,
2494
+ "grad_norm": 0.2357209175825119,
2495
+ "learning_rate": 9.264050619597697e-06,
2496
+ "loss": 0.105,
2497
+ "step": 320
2498
+ },
2499
+ {
2500
+ "epoch": 2.4218455743879472,
2501
+ "eval_loss": 0.0942394882440567,
2502
+ "eval_runtime": 43.9875,
2503
+ "eval_samples_per_second": 1.705,
2504
+ "eval_steps_per_second": 0.864,
2505
+ "step": 320
2506
+ },
2507
+ {
2508
+ "epoch": 2.4218455743879472,
2509
+ "step": 320,
2510
+ "total_flos": 4.419854408240333e+16,
2511
+ "train_loss": 0.537436925212387,
2512
+ "train_runtime": 3691.1633,
2513
+ "train_samples_per_second": 0.863,
2514
+ "train_steps_per_second": 0.107
2515
+ }
2516
+ ],
2517
+ "logging_steps": 1,
2518
+ "max_steps": 396,
2519
+ "num_input_tokens_seen": 0,
2520
+ "num_train_epochs": 3,
2521
+ "save_steps": 500,
2522
+ "stateful_callbacks": {
2523
+ "EarlyStoppingCallback": {
2524
+ "args": {
2525
+ "early_stopping_patience": 4,
2526
+ "early_stopping_threshold": 0.0015
2527
+ },
2528
+ "attributes": {
2529
+ "early_stopping_patience_counter": 4
2530
+ }
2531
+ },
2532
+ "TrainerControl": {
2533
+ "args": {
2534
+ "should_epoch_stop": false,
2535
+ "should_evaluate": false,
2536
+ "should_log": false,
2537
+ "should_save": true,
2538
+ "should_training_stop": true
2539
+ },
2540
+ "attributes": {}
2541
+ }
2542
+ },
2543
+ "total_flos": 4.419854408240333e+16,
2544
+ "train_batch_size": 2,
2545
+ "trial_name": null,
2546
+ "trial_params": null
2547
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c1d7ab402d8665708ee23c722193c44538d1716b7f2b2f4c64c67480baaad8f
3
+ size 5752
vocab.json ADDED
The diff for this file is too large to render. See raw diff