changjiakawhi commited on
Commit
1261afc
·
verified ·
1 Parent(s): 64d7d5f

Model save

Browse files
.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 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen2.5-1.5B-Instruct
3
+ library_name: transformers
4
+ model_name: Qwen2.5-1.5B-Open-R1-Distill
5
+ tags:
6
+ - generated_from_trainer
7
+ - trl
8
+ - sft
9
+ licence: license
10
+ ---
11
+
12
+ # Model Card for Qwen2.5-1.5B-Open-R1-Distill
13
+
14
+ This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct).
15
+ It has been trained using [TRL](https://github.com/huggingface/trl).
16
+
17
+ ## Quick start
18
+
19
+ ```python
20
+ from transformers import pipeline
21
+
22
+ 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?"
23
+ generator = pipeline("text-generation", model="changjiakawhi/Qwen2.5-1.5B-Open-R1-Distill", device="cuda")
24
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
25
+ print(output["generated_text"])
26
+ ```
27
+
28
+ ## Training procedure
29
+
30
+ [<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/13187091323-hunan-normal-university/huggingface/runs/dw34sw95)
31
+
32
+
33
+ This model was trained with SFT.
34
+
35
+ ### Framework versions
36
+
37
+ - TRL: 0.16.0.dev0
38
+ - Transformers: 4.49.0.dev0
39
+ - Pytorch: 2.5.1
40
+ - Datasets: 3.2.0
41
+ - Tokenizers: 0.21.0
42
+
43
+ ## Citations
44
+
45
+
46
+
47
+ Cite TRL as:
48
+
49
+ ```bibtex
50
+ @misc{vonwerra2022trl,
51
+ title = {{TRL: Transformer Reinforcement Learning}},
52
+ 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},
53
+ year = 2020,
54
+ journal = {GitHub repository},
55
+ publisher = {GitHub},
56
+ howpublished = {\url{https://github.com/huggingface/trl}}
57
+ }
58
+ ```
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 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "total_flos": 76966677970944.0,
3
+ "train_loss": 0.759784622734163,
4
+ "train_runtime": 8483.4218,
5
+ "train_samples": 16610,
6
+ "train_samples_per_second": 2.549,
7
+ "train_steps_per_second": 0.159
8
+ }
config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151645,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1536,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 8960,
13
+ "max_position_embeddings": 32768,
14
+ "max_window_layers": 21,
15
+ "model_type": "qwen2",
16
+ "num_attention_heads": 12,
17
+ "num_hidden_layers": 28,
18
+ "num_key_value_heads": 2,
19
+ "rms_norm_eps": 1e-06,
20
+ "rope_scaling": null,
21
+ "rope_theta": 1000000.0,
22
+ "sliding_window": null,
23
+ "tie_word_embeddings": true,
24
+ "torch_dtype": "bfloat16",
25
+ "transformers_version": "4.49.0.dev0",
26
+ "use_cache": false,
27
+ "use_sliding_window": false,
28
+ "vocab_size": 151936
29
+ }
generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "repetition_penalty": 1.1,
10
+ "temperature": 0.7,
11
+ "top_k": 20,
12
+ "top_p": 0.8,
13
+ "transformers_version": "4.49.0.dev0"
14
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a10f28b8b9f9afbe4e7f4df8260127fa7dab4d847d71f770a1be4ab4b88c7bf
3
+ size 3087467144
special_tokens_map.json ADDED
@@ -0,0 +1,25 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "<|im_end|>"
25
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
tokenizer_config.json ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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 Qwen, created by Alibaba Cloud. 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 Qwen, created by Alibaba Cloud. You 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": "<|im_end|>",
205
+ "split_special_tokens": false,
206
+ "tokenizer_class": "Qwen2Tokenizer",
207
+ "unk_token": null
208
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "total_flos": 76966677970944.0,
3
+ "train_loss": 0.759784622734163,
4
+ "train_runtime": 8483.4218,
5
+ "train_samples": 16610,
6
+ "train_samples_per_second": 2.549,
7
+ "train_steps_per_second": 0.159
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,2320 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9996300406955235,
5
+ "eval_steps": 100,
6
+ "global_step": 1351,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.003699593044765076,
13
+ "grad_norm": 2.757993258887922,
14
+ "learning_rate": 7.352941176470589e-07,
15
+ "loss": 1.1067,
16
+ "mean_token_accuracy": 0.708824381857028,
17
+ "step": 5
18
+ },
19
+ {
20
+ "epoch": 0.007399186089530152,
21
+ "grad_norm": 2.5907330458452984,
22
+ "learning_rate": 1.4705882352941177e-06,
23
+ "loss": 1.0679,
24
+ "mean_token_accuracy": 0.7169299986512495,
25
+ "step": 10
26
+ },
27
+ {
28
+ "epoch": 0.011098779134295227,
29
+ "grad_norm": 2.241392532504099,
30
+ "learning_rate": 2.2058823529411767e-06,
31
+ "loss": 1.1238,
32
+ "mean_token_accuracy": 0.6992935634786113,
33
+ "step": 15
34
+ },
35
+ {
36
+ "epoch": 0.014798372179060304,
37
+ "grad_norm": 1.5956862014648674,
38
+ "learning_rate": 2.9411764705882355e-06,
39
+ "loss": 1.0062,
40
+ "mean_token_accuracy": 0.7288372261973354,
41
+ "step": 20
42
+ },
43
+ {
44
+ "epoch": 0.01849796522382538,
45
+ "grad_norm": 1.6119131199982493,
46
+ "learning_rate": 3.6764705882352946e-06,
47
+ "loss": 1.0517,
48
+ "mean_token_accuracy": 0.7142909651691335,
49
+ "step": 25
50
+ },
51
+ {
52
+ "epoch": 0.022197558268590455,
53
+ "grad_norm": 1.139657389593639,
54
+ "learning_rate": 4.411764705882353e-06,
55
+ "loss": 0.9593,
56
+ "mean_token_accuracy": 0.7346081897149975,
57
+ "step": 30
58
+ },
59
+ {
60
+ "epoch": 0.02589715131335553,
61
+ "grad_norm": 1.2790776332632598,
62
+ "learning_rate": 5.147058823529411e-06,
63
+ "loss": 0.9805,
64
+ "mean_token_accuracy": 0.72720137806358,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.029596744358120607,
69
+ "grad_norm": 1.104655854611052,
70
+ "learning_rate": 5.882352941176471e-06,
71
+ "loss": 0.9114,
72
+ "mean_token_accuracy": 0.7425328465386071,
73
+ "step": 40
74
+ },
75
+ {
76
+ "epoch": 0.033296337402885685,
77
+ "grad_norm": 0.9408582472471563,
78
+ "learning_rate": 6.61764705882353e-06,
79
+ "loss": 0.8885,
80
+ "mean_token_accuracy": 0.7472382565404383,
81
+ "step": 45
82
+ },
83
+ {
84
+ "epoch": 0.03699593044765076,
85
+ "grad_norm": 1.0580888719684283,
86
+ "learning_rate": 7.352941176470589e-06,
87
+ "loss": 0.8658,
88
+ "mean_token_accuracy": 0.7493579770562304,
89
+ "step": 50
90
+ },
91
+ {
92
+ "epoch": 0.040695523492415835,
93
+ "grad_norm": 0.8832977290583615,
94
+ "learning_rate": 8.088235294117648e-06,
95
+ "loss": 0.841,
96
+ "mean_token_accuracy": 0.7568992237585881,
97
+ "step": 55
98
+ },
99
+ {
100
+ "epoch": 0.04439511653718091,
101
+ "grad_norm": 0.9301447232199826,
102
+ "learning_rate": 8.823529411764707e-06,
103
+ "loss": 0.8677,
104
+ "mean_token_accuracy": 0.7502923381848673,
105
+ "step": 60
106
+ },
107
+ {
108
+ "epoch": 0.048094709581945984,
109
+ "grad_norm": 0.856501238425618,
110
+ "learning_rate": 9.558823529411766e-06,
111
+ "loss": 0.878,
112
+ "mean_token_accuracy": 0.7445654933582921,
113
+ "step": 65
114
+ },
115
+ {
116
+ "epoch": 0.05179430262671106,
117
+ "grad_norm": 1.0586539300870963,
118
+ "learning_rate": 1.0294117647058823e-05,
119
+ "loss": 0.8825,
120
+ "mean_token_accuracy": 0.7438754593389819,
121
+ "step": 70
122
+ },
123
+ {
124
+ "epoch": 0.05549389567147614,
125
+ "grad_norm": 1.1122367826346538,
126
+ "learning_rate": 1.1029411764705885e-05,
127
+ "loss": 0.8775,
128
+ "mean_token_accuracy": 0.7447408372141248,
129
+ "step": 75
130
+ },
131
+ {
132
+ "epoch": 0.059193488716241215,
133
+ "grad_norm": 1.0632613858567348,
134
+ "learning_rate": 1.1764705882352942e-05,
135
+ "loss": 0.8347,
136
+ "mean_token_accuracy": 0.7567175732049909,
137
+ "step": 80
138
+ },
139
+ {
140
+ "epoch": 0.06289308176100629,
141
+ "grad_norm": 0.9476656757993303,
142
+ "learning_rate": 1.25e-05,
143
+ "loss": 0.8305,
144
+ "mean_token_accuracy": 0.7566193958141736,
145
+ "step": 85
146
+ },
147
+ {
148
+ "epoch": 0.06659267480577137,
149
+ "grad_norm": 1.1083079276477652,
150
+ "learning_rate": 1.323529411764706e-05,
151
+ "loss": 0.8466,
152
+ "mean_token_accuracy": 0.7519149864344391,
153
+ "step": 90
154
+ },
155
+ {
156
+ "epoch": 0.07029226785053644,
157
+ "grad_norm": 0.9768809343076362,
158
+ "learning_rate": 1.3970588235294118e-05,
159
+ "loss": 0.7894,
160
+ "mean_token_accuracy": 0.768089470723049,
161
+ "step": 95
162
+ },
163
+ {
164
+ "epoch": 0.07399186089530152,
165
+ "grad_norm": 1.0897420744707864,
166
+ "learning_rate": 1.4705882352941179e-05,
167
+ "loss": 0.842,
168
+ "mean_token_accuracy": 0.7519371481531969,
169
+ "step": 100
170
+ },
171
+ {
172
+ "epoch": 0.07399186089530152,
173
+ "eval_loss": 0.8507319688796997,
174
+ "eval_mean_token_accuracy": 0.7484970684902305,
175
+ "eval_runtime": 13.1139,
176
+ "eval_samples_per_second": 9.837,
177
+ "eval_steps_per_second": 2.516,
178
+ "step": 100
179
+ },
180
+ {
181
+ "epoch": 0.07769145394006659,
182
+ "grad_norm": 1.0530600966285744,
183
+ "learning_rate": 1.5441176470588237e-05,
184
+ "loss": 0.8609,
185
+ "mean_token_accuracy": 0.7470814372661423,
186
+ "step": 105
187
+ },
188
+ {
189
+ "epoch": 0.08139104698483167,
190
+ "grad_norm": 0.9420676582335842,
191
+ "learning_rate": 1.6176470588235296e-05,
192
+ "loss": 0.8056,
193
+ "mean_token_accuracy": 0.7604963244668818,
194
+ "step": 110
195
+ },
196
+ {
197
+ "epoch": 0.08509064002959675,
198
+ "grad_norm": 0.868613480082479,
199
+ "learning_rate": 1.6911764705882355e-05,
200
+ "loss": 0.8432,
201
+ "mean_token_accuracy": 0.7519378614781397,
202
+ "step": 115
203
+ },
204
+ {
205
+ "epoch": 0.08879023307436182,
206
+ "grad_norm": 1.0518122686708429,
207
+ "learning_rate": 1.7647058823529414e-05,
208
+ "loss": 0.8237,
209
+ "mean_token_accuracy": 0.7588873826848553,
210
+ "step": 120
211
+ },
212
+ {
213
+ "epoch": 0.0924898261191269,
214
+ "grad_norm": 1.135357117838434,
215
+ "learning_rate": 1.8382352941176472e-05,
216
+ "loss": 0.8417,
217
+ "mean_token_accuracy": 0.7510501482017762,
218
+ "step": 125
219
+ },
220
+ {
221
+ "epoch": 0.09618941916389197,
222
+ "grad_norm": 1.0511646764690201,
223
+ "learning_rate": 1.911764705882353e-05,
224
+ "loss": 0.8024,
225
+ "mean_token_accuracy": 0.7630132146126162,
226
+ "step": 130
227
+ },
228
+ {
229
+ "epoch": 0.09988901220865705,
230
+ "grad_norm": 1.003659422839662,
231
+ "learning_rate": 1.985294117647059e-05,
232
+ "loss": 0.7846,
233
+ "mean_token_accuracy": 0.7684428405908383,
234
+ "step": 135
235
+ },
236
+ {
237
+ "epoch": 0.10358860525342212,
238
+ "grad_norm": 0.9783718356092802,
239
+ "learning_rate": 1.9999465148392906e-05,
240
+ "loss": 0.7978,
241
+ "mean_token_accuracy": 0.7638017765978411,
242
+ "step": 140
243
+ },
244
+ {
245
+ "epoch": 0.1072881982981872,
246
+ "grad_norm": 0.9325940574831129,
247
+ "learning_rate": 1.999729241179462e-05,
248
+ "loss": 0.7899,
249
+ "mean_token_accuracy": 0.7657915220582932,
250
+ "step": 145
251
+ },
252
+ {
253
+ "epoch": 0.11098779134295228,
254
+ "grad_norm": 0.9587245678809391,
255
+ "learning_rate": 1.999344872485215e-05,
256
+ "loss": 0.7935,
257
+ "mean_token_accuracy": 0.7643045615369954,
258
+ "step": 150
259
+ },
260
+ {
261
+ "epoch": 0.11468738438771735,
262
+ "grad_norm": 1.2026644648319857,
263
+ "learning_rate": 1.9987934730000457e-05,
264
+ "loss": 0.7922,
265
+ "mean_token_accuracy": 0.7667913563744123,
266
+ "step": 155
267
+ },
268
+ {
269
+ "epoch": 0.11838697743248243,
270
+ "grad_norm": 1.0527655093884445,
271
+ "learning_rate": 1.998075134885022e-05,
272
+ "loss": 0.8036,
273
+ "mean_token_accuracy": 0.7613009885033224,
274
+ "step": 160
275
+ },
276
+ {
277
+ "epoch": 0.1220865704772475,
278
+ "grad_norm": 1.2386325448592015,
279
+ "learning_rate": 1.9971899782033853e-05,
280
+ "loss": 0.8013,
281
+ "mean_token_accuracy": 0.7620870662678945,
282
+ "step": 165
283
+ },
284
+ {
285
+ "epoch": 0.12578616352201258,
286
+ "grad_norm": 1.127408462312219,
287
+ "learning_rate": 1.9961381509004785e-05,
288
+ "loss": 0.808,
289
+ "mean_token_accuracy": 0.7593721598276708,
290
+ "step": 170
291
+ },
292
+ {
293
+ "epoch": 0.12948575656677766,
294
+ "grad_norm": 0.9438483145827001,
295
+ "learning_rate": 1.9949198287790215e-05,
296
+ "loss": 0.7861,
297
+ "mean_token_accuracy": 0.7660641312412623,
298
+ "step": 175
299
+ },
300
+ {
301
+ "epoch": 0.13318534961154274,
302
+ "grad_norm": 0.9318614898436913,
303
+ "learning_rate": 1.9935352154697257e-05,
304
+ "loss": 0.8038,
305
+ "mean_token_accuracy": 0.7602947954939976,
306
+ "step": 180
307
+ },
308
+ {
309
+ "epoch": 0.1368849426563078,
310
+ "grad_norm": 0.8876177342352581,
311
+ "learning_rate": 1.9919845423972603e-05,
312
+ "loss": 0.757,
313
+ "mean_token_accuracy": 0.7742074952233244,
314
+ "step": 185
315
+ },
316
+ {
317
+ "epoch": 0.14058453570107288,
318
+ "grad_norm": 1.005618517605514,
319
+ "learning_rate": 1.9902680687415704e-05,
320
+ "loss": 0.7985,
321
+ "mean_token_accuracy": 0.7622990063822126,
322
+ "step": 190
323
+ },
324
+ {
325
+ "epoch": 0.14428412874583796,
326
+ "grad_norm": 1.0909612602188234,
327
+ "learning_rate": 1.9883860813945596e-05,
328
+ "loss": 0.7997,
329
+ "mean_token_accuracy": 0.7611327796064032,
330
+ "step": 195
331
+ },
332
+ {
333
+ "epoch": 0.14798372179060304,
334
+ "grad_norm": 0.9223546224550383,
335
+ "learning_rate": 1.986338894912137e-05,
336
+ "loss": 0.7888,
337
+ "mean_token_accuracy": 0.7650975947505124,
338
+ "step": 200
339
+ },
340
+ {
341
+ "epoch": 0.14798372179060304,
342
+ "eval_loss": 0.8212586641311646,
343
+ "eval_mean_token_accuracy": 0.753984452665533,
344
+ "eval_runtime": 12.914,
345
+ "eval_samples_per_second": 9.989,
346
+ "eval_steps_per_second": 2.555,
347
+ "step": 200
348
+ },
349
+ {
350
+ "epoch": 0.15168331483536812,
351
+ "grad_norm": 0.9461631972455284,
352
+ "learning_rate": 1.9841268514616434e-05,
353
+ "loss": 0.8206,
354
+ "mean_token_accuracy": 0.7560038239175771,
355
+ "step": 205
356
+ },
357
+ {
358
+ "epoch": 0.15538290788013318,
359
+ "grad_norm": 1.0128847648211983,
360
+ "learning_rate": 1.9817503207646606e-05,
361
+ "loss": 0.7827,
362
+ "mean_token_accuracy": 0.7663386260842018,
363
+ "step": 210
364
+ },
365
+ {
366
+ "epoch": 0.15908250092489826,
367
+ "grad_norm": 1.0613216235343779,
368
+ "learning_rate": 1.979209700035216e-05,
369
+ "loss": 0.8097,
370
+ "mean_token_accuracy": 0.7595550837031688,
371
+ "step": 215
372
+ },
373
+ {
374
+ "epoch": 0.16278209396966334,
375
+ "grad_norm": 1.0811772085526155,
376
+ "learning_rate": 1.976505413913393e-05,
377
+ "loss": 0.7771,
378
+ "mean_token_accuracy": 0.7685113508318921,
379
+ "step": 220
380
+ },
381
+ {
382
+ "epoch": 0.16648168701442842,
383
+ "grad_norm": 1.0032217781891688,
384
+ "learning_rate": 1.9736379143943565e-05,
385
+ "loss": 0.7788,
386
+ "mean_token_accuracy": 0.767374709683548,
387
+ "step": 225
388
+ },
389
+ {
390
+ "epoch": 0.1701812800591935,
391
+ "grad_norm": 1.0033846900829124,
392
+ "learning_rate": 1.9706076807528044e-05,
393
+ "loss": 0.7792,
394
+ "mean_token_accuracy": 0.7671742181015657,
395
+ "step": 230
396
+ },
397
+ {
398
+ "epoch": 0.17388087310395856,
399
+ "grad_norm": 1.016761717770083,
400
+ "learning_rate": 1.967415219462864e-05,
401
+ "loss": 0.776,
402
+ "mean_token_accuracy": 0.7692263473301846,
403
+ "step": 235
404
+ },
405
+ {
406
+ "epoch": 0.17758046614872364,
407
+ "grad_norm": 1.0390090395504632,
408
+ "learning_rate": 1.9640610641134383e-05,
409
+ "loss": 0.8125,
410
+ "mean_token_accuracy": 0.7578594416116643,
411
+ "step": 240
412
+ },
413
+ {
414
+ "epoch": 0.18128005919348872,
415
+ "grad_norm": 0.9506480558904421,
416
+ "learning_rate": 1.9605457753190224e-05,
417
+ "loss": 0.774,
418
+ "mean_token_accuracy": 0.7669808330485,
419
+ "step": 245
420
+ },
421
+ {
422
+ "epoch": 0.1849796522382538,
423
+ "grad_norm": 0.9074565067919744,
424
+ "learning_rate": 1.9568699406260016e-05,
425
+ "loss": 0.7631,
426
+ "mean_token_accuracy": 0.7712889443688947,
427
+ "step": 250
428
+ },
429
+ {
430
+ "epoch": 0.18867924528301888,
431
+ "grad_norm": 0.870192572015487,
432
+ "learning_rate": 1.953034174414449e-05,
433
+ "loss": 0.8041,
434
+ "mean_token_accuracy": 0.7584920492345467,
435
+ "step": 255
436
+ },
437
+ {
438
+ "epoch": 0.19237883832778394,
439
+ "grad_norm": 0.9436163053750114,
440
+ "learning_rate": 1.9490391177954383e-05,
441
+ "loss": 0.7697,
442
+ "mean_token_accuracy": 0.7687670857457188,
443
+ "step": 260
444
+ },
445
+ {
446
+ "epoch": 0.19607843137254902,
447
+ "grad_norm": 0.8454337167883038,
448
+ "learning_rate": 1.944885438503888e-05,
449
+ "loss": 0.7854,
450
+ "mean_token_accuracy": 0.762450394993266,
451
+ "step": 265
452
+ },
453
+ {
454
+ "epoch": 0.1997780244173141,
455
+ "grad_norm": 0.8388911143997999,
456
+ "learning_rate": 1.9405738307869565e-05,
457
+ "loss": 0.7568,
458
+ "mean_token_accuracy": 0.7739168937821217,
459
+ "step": 270
460
+ },
461
+ {
462
+ "epoch": 0.20347761746207918,
463
+ "grad_norm": 0.8905388426225519,
464
+ "learning_rate": 1.936105015288003e-05,
465
+ "loss": 0.7874,
466
+ "mean_token_accuracy": 0.7657060442327474,
467
+ "step": 275
468
+ },
469
+ {
470
+ "epoch": 0.20717721050684423,
471
+ "grad_norm": 0.9449437240241089,
472
+ "learning_rate": 1.9314797389261426e-05,
473
+ "loss": 0.7964,
474
+ "mean_token_accuracy": 0.7608504929094035,
475
+ "step": 280
476
+ },
477
+ {
478
+ "epoch": 0.21087680355160932,
479
+ "grad_norm": 0.9586909770652294,
480
+ "learning_rate": 1.9266987747714036e-05,
481
+ "loss": 0.7539,
482
+ "mean_token_accuracy": 0.7740995999688999,
483
+ "step": 285
484
+ },
485
+ {
486
+ "epoch": 0.2145763965963744,
487
+ "grad_norm": 0.8457904946858513,
488
+ "learning_rate": 1.9217629219155172e-05,
489
+ "loss": 0.7596,
490
+ "mean_token_accuracy": 0.7711086747415601,
491
+ "step": 290
492
+ },
493
+ {
494
+ "epoch": 0.21827598964113948,
495
+ "grad_norm": 0.9562284538878311,
496
+ "learning_rate": 1.916673005338357e-05,
497
+ "loss": 0.7636,
498
+ "mean_token_accuracy": 0.7705879155908494,
499
+ "step": 295
500
+ },
501
+ {
502
+ "epoch": 0.22197558268590456,
503
+ "grad_norm": 0.9311027437584551,
504
+ "learning_rate": 1.9114298757700508e-05,
505
+ "loss": 0.7833,
506
+ "mean_token_accuracy": 0.7640983829119388,
507
+ "step": 300
508
+ },
509
+ {
510
+ "epoch": 0.22197558268590456,
511
+ "eval_loss": 0.8046144247055054,
512
+ "eval_mean_token_accuracy": 0.7572653228255071,
513
+ "eval_runtime": 12.9112,
514
+ "eval_samples_per_second": 9.991,
515
+ "eval_steps_per_second": 2.556,
516
+ "step": 300
517
+ },
518
+ {
519
+ "epoch": 0.2256751757306696,
520
+ "grad_norm": 0.981751974465348,
521
+ "learning_rate": 1.9060344095487916e-05,
522
+ "loss": 0.7517,
523
+ "mean_token_accuracy": 0.7743910330942259,
524
+ "step": 305
525
+ },
526
+ {
527
+ "epoch": 0.2293747687754347,
528
+ "grad_norm": 0.9758899701140303,
529
+ "learning_rate": 1.9004875084743624e-05,
530
+ "loss": 0.7744,
531
+ "mean_token_accuracy": 0.7681139265197442,
532
+ "step": 310
533
+ },
534
+ {
535
+ "epoch": 0.23307436182019978,
536
+ "grad_norm": 1.0996867282039497,
537
+ "learning_rate": 1.8947900996574133e-05,
538
+ "loss": 0.784,
539
+ "mean_token_accuracy": 0.7653104608265264,
540
+ "step": 315
541
+ },
542
+ {
543
+ "epoch": 0.23677395486496486,
544
+ "grad_norm": 0.8876749775333296,
545
+ "learning_rate": 1.8889431353645004e-05,
546
+ "loss": 0.7856,
547
+ "mean_token_accuracy": 0.7637695086387486,
548
+ "step": 320
549
+ },
550
+ {
551
+ "epoch": 0.24047354790972994,
552
+ "grad_norm": 0.9254315942079439,
553
+ "learning_rate": 1.8829475928589272e-05,
554
+ "loss": 0.7478,
555
+ "mean_token_accuracy": 0.7747918700572913,
556
+ "step": 325
557
+ },
558
+ {
559
+ "epoch": 0.244173140954495,
560
+ "grad_norm": 0.8975251081081479,
561
+ "learning_rate": 1.8768044742374008e-05,
562
+ "loss": 0.7838,
563
+ "mean_token_accuracy": 0.7662794783326515,
564
+ "step": 330
565
+ },
566
+ {
567
+ "epoch": 0.24787273399926008,
568
+ "grad_norm": 0.9659229415705355,
569
+ "learning_rate": 1.870514806262544e-05,
570
+ "loss": 0.8064,
571
+ "mean_token_accuracy": 0.7575206306307912,
572
+ "step": 335
573
+ },
574
+ {
575
+ "epoch": 0.25157232704402516,
576
+ "grad_norm": 0.8730001205655563,
577
+ "learning_rate": 1.8640796401912805e-05,
578
+ "loss": 0.7786,
579
+ "mean_token_accuracy": 0.7647639047537955,
580
+ "step": 340
581
+ },
582
+ {
583
+ "epoch": 0.25527192008879024,
584
+ "grad_norm": 0.8503523554051183,
585
+ "learning_rate": 1.8575000515991283e-05,
586
+ "loss": 0.7303,
587
+ "mean_token_accuracy": 0.7790401063745367,
588
+ "step": 345
589
+ },
590
+ {
591
+ "epoch": 0.2589715131335553,
592
+ "grad_norm": 0.952888955967417,
593
+ "learning_rate": 1.850777140200427e-05,
594
+ "loss": 0.7481,
595
+ "mean_token_accuracy": 0.7754539961681624,
596
+ "step": 350
597
+ },
598
+ {
599
+ "epoch": 0.2626711061783204,
600
+ "grad_norm": 0.8607187749209038,
601
+ "learning_rate": 1.843912029664531e-05,
602
+ "loss": 0.7628,
603
+ "mean_token_accuracy": 0.7705045859827028,
604
+ "step": 355
605
+ },
606
+ {
607
+ "epoch": 0.2663706992230855,
608
+ "grad_norm": 0.9612681025779204,
609
+ "learning_rate": 1.8369058674280004e-05,
610
+ "loss": 0.7751,
611
+ "mean_token_accuracy": 0.7662138710016833,
612
+ "step": 360
613
+ },
614
+ {
615
+ "epoch": 0.2700702922678505,
616
+ "grad_norm": 0.9382382938099896,
617
+ "learning_rate": 1.8297598245028173e-05,
618
+ "loss": 0.7723,
619
+ "mean_token_accuracy": 0.767681726807316,
620
+ "step": 365
621
+ },
622
+ {
623
+ "epoch": 0.2737698853126156,
624
+ "grad_norm": 0.9997500486677628,
625
+ "learning_rate": 1.8224750952806626e-05,
626
+ "loss": 0.7611,
627
+ "mean_token_accuracy": 0.7707074275165302,
628
+ "step": 370
629
+ },
630
+ {
631
+ "epoch": 0.27746947835738067,
632
+ "grad_norm": 1.0014266441284037,
633
+ "learning_rate": 1.815052897333284e-05,
634
+ "loss": 0.7505,
635
+ "mean_token_accuracy": 0.7740457102243052,
636
+ "step": 375
637
+ },
638
+ {
639
+ "epoch": 0.28116907140214575,
640
+ "grad_norm": 1.1075366477297495,
641
+ "learning_rate": 1.8074944712089925e-05,
642
+ "loss": 0.7641,
643
+ "mean_token_accuracy": 0.7693695886355318,
644
+ "step": 380
645
+ },
646
+ {
647
+ "epoch": 0.28486866444691084,
648
+ "grad_norm": 0.880418326713614,
649
+ "learning_rate": 1.799801080225316e-05,
650
+ "loss": 0.7539,
651
+ "mean_token_accuracy": 0.7710243423063697,
652
+ "step": 385
653
+ },
654
+ {
655
+ "epoch": 0.2885682574916759,
656
+ "grad_norm": 0.9012750826548256,
657
+ "learning_rate": 1.7919740102578482e-05,
658
+ "loss": 0.7781,
659
+ "mean_token_accuracy": 0.7661210900072783,
660
+ "step": 390
661
+ },
662
+ {
663
+ "epoch": 0.292267850536441,
664
+ "grad_norm": 0.9230339938349867,
665
+ "learning_rate": 1.7840145695253258e-05,
666
+ "loss": 0.7708,
667
+ "mean_token_accuracy": 0.7679753783644399,
668
+ "step": 395
669
+ },
670
+ {
671
+ "epoch": 0.2959674435812061,
672
+ "grad_norm": 0.8616748960380592,
673
+ "learning_rate": 1.7759240883709745e-05,
674
+ "loss": 0.7643,
675
+ "mean_token_accuracy": 0.7698643926023239,
676
+ "step": 400
677
+ },
678
+ {
679
+ "epoch": 0.2959674435812061,
680
+ "eval_loss": 0.7939132452011108,
681
+ "eval_mean_token_accuracy": 0.7597238566071922,
682
+ "eval_runtime": 12.8935,
683
+ "eval_samples_per_second": 10.005,
684
+ "eval_steps_per_second": 2.559,
685
+ "step": 400
686
+ },
687
+ {
688
+ "epoch": 0.29966703662597116,
689
+ "grad_norm": 0.8626025158912716,
690
+ "learning_rate": 1.7677039190401538e-05,
691
+ "loss": 0.7945,
692
+ "mean_token_accuracy": 0.7610350956243586,
693
+ "step": 405
694
+ },
695
+ {
696
+ "epoch": 0.30336662967073624,
697
+ "grad_norm": 0.8809551406016749,
698
+ "learning_rate": 1.759355435454342e-05,
699
+ "loss": 0.7554,
700
+ "mean_token_accuracy": 0.7720821829110623,
701
+ "step": 410
702
+ },
703
+ {
704
+ "epoch": 0.30706622271550127,
705
+ "grad_norm": 0.8398343511073675,
706
+ "learning_rate": 1.7508800329814993e-05,
707
+ "loss": 0.758,
708
+ "mean_token_accuracy": 0.7709962708725867,
709
+ "step": 415
710
+ },
711
+ {
712
+ "epoch": 0.31076581576026635,
713
+ "grad_norm": 0.8717214162340853,
714
+ "learning_rate": 1.7422791282028457e-05,
715
+ "loss": 0.7456,
716
+ "mean_token_accuracy": 0.7747603119006603,
717
+ "step": 420
718
+ },
719
+ {
720
+ "epoch": 0.31446540880503143,
721
+ "grad_norm": 0.9182650879024069,
722
+ "learning_rate": 1.7335541586760928e-05,
723
+ "loss": 0.7678,
724
+ "mean_token_accuracy": 0.768033024075789,
725
+ "step": 425
726
+ },
727
+ {
728
+ "epoch": 0.3181650018497965,
729
+ "grad_norm": 0.9270374928264972,
730
+ "learning_rate": 1.7247065826951694e-05,
731
+ "loss": 0.7687,
732
+ "mean_token_accuracy": 0.7670009758238231,
733
+ "step": 430
734
+ },
735
+ {
736
+ "epoch": 0.3218645948945616,
737
+ "grad_norm": 0.9239947000531302,
738
+ "learning_rate": 1.715737879046483e-05,
739
+ "loss": 0.7807,
740
+ "mean_token_accuracy": 0.7646944309939067,
741
+ "step": 435
742
+ },
743
+ {
744
+ "epoch": 0.3255641879393267,
745
+ "grad_norm": 0.9234310473796695,
746
+ "learning_rate": 1.7066495467617552e-05,
747
+ "loss": 0.722,
748
+ "mean_token_accuracy": 0.7810521648421277,
749
+ "step": 440
750
+ },
751
+ {
752
+ "epoch": 0.32926378098409176,
753
+ "grad_norm": 0.9873873391290533,
754
+ "learning_rate": 1.6974431048674714e-05,
755
+ "loss": 0.7773,
756
+ "mean_token_accuracy": 0.7676746345471892,
757
+ "step": 445
758
+ },
759
+ {
760
+ "epoch": 0.33296337402885684,
761
+ "grad_norm": 0.8848520725498183,
762
+ "learning_rate": 1.6881200921309914e-05,
763
+ "loss": 0.7799,
764
+ "mean_token_accuracy": 0.7653884936966252,
765
+ "step": 450
766
+ },
767
+ {
768
+ "epoch": 0.3366629670736219,
769
+ "grad_norm": 0.9045509398039985,
770
+ "learning_rate": 1.6786820668033596e-05,
771
+ "loss": 0.7575,
772
+ "mean_token_accuracy": 0.7713310238062097,
773
+ "step": 455
774
+ },
775
+ {
776
+ "epoch": 0.340362560118387,
777
+ "grad_norm": 0.890193659220096,
778
+ "learning_rate": 1.6691306063588583e-05,
779
+ "loss": 0.7643,
780
+ "mean_token_accuracy": 0.7680450171610729,
781
+ "step": 460
782
+ },
783
+ {
784
+ "epoch": 0.34406215316315203,
785
+ "grad_norm": 0.9121601629876329,
786
+ "learning_rate": 1.6594673072313478e-05,
787
+ "loss": 0.7565,
788
+ "mean_token_accuracy": 0.7709845040717809,
789
+ "step": 465
790
+ },
791
+ {
792
+ "epoch": 0.3477617462079171,
793
+ "grad_norm": 0.8596218378746548,
794
+ "learning_rate": 1.6496937845474375e-05,
795
+ "loss": 0.7899,
796
+ "mean_token_accuracy": 0.762253394394126,
797
+ "step": 470
798
+ },
799
+ {
800
+ "epoch": 0.3514613392526822,
801
+ "grad_norm": 0.8881090780569305,
802
+ "learning_rate": 1.639811671856535e-05,
803
+ "loss": 0.7716,
804
+ "mean_token_accuracy": 0.766323517253549,
805
+ "step": 475
806
+ },
807
+ {
808
+ "epoch": 0.3551609322974473,
809
+ "grad_norm": 0.9129717947745375,
810
+ "learning_rate": 1.6298226208578127e-05,
811
+ "loss": 0.7773,
812
+ "mean_token_accuracy": 0.7662860024728196,
813
+ "step": 480
814
+ },
815
+ {
816
+ "epoch": 0.35886052534221236,
817
+ "grad_norm": 0.8773141616181648,
818
+ "learning_rate": 1.6197283011241423e-05,
819
+ "loss": 0.7823,
820
+ "mean_token_accuracy": 0.7632768240832929,
821
+ "step": 485
822
+ },
823
+ {
824
+ "epoch": 0.36256011838697744,
825
+ "grad_norm": 0.8872263111805669,
826
+ "learning_rate": 1.6095303998230432e-05,
827
+ "loss": 0.7683,
828
+ "mean_token_accuracy": 0.7674727430979753,
829
+ "step": 490
830
+ },
831
+ {
832
+ "epoch": 0.3662597114317425,
833
+ "grad_norm": 0.98466468959405,
834
+ "learning_rate": 1.599230621434687e-05,
835
+ "loss": 0.7591,
836
+ "mean_token_accuracy": 0.7693043297460778,
837
+ "step": 495
838
+ },
839
+ {
840
+ "epoch": 0.3699593044765076,
841
+ "grad_norm": 0.8908360555035981,
842
+ "learning_rate": 1.5888306874670112e-05,
843
+ "loss": 0.7362,
844
+ "mean_token_accuracy": 0.7770877269382522,
845
+ "step": 500
846
+ },
847
+ {
848
+ "epoch": 0.3699593044765076,
849
+ "eval_loss": 0.784197211265564,
850
+ "eval_mean_token_accuracy": 0.7620830191395729,
851
+ "eval_runtime": 12.9033,
852
+ "eval_samples_per_second": 9.997,
853
+ "eval_steps_per_second": 2.557,
854
+ "step": 500
855
+ },
856
+ {
857
+ "epoch": 0.3736588975212727,
858
+ "grad_norm": 0.9048447583159639,
859
+ "learning_rate": 1.5783323361679865e-05,
860
+ "loss": 0.7787,
861
+ "mean_token_accuracy": 0.7648656868298831,
862
+ "step": 505
863
+ },
864
+ {
865
+ "epoch": 0.37735849056603776,
866
+ "grad_norm": 0.9120327051425232,
867
+ "learning_rate": 1.567737322235084e-05,
868
+ "loss": 0.7593,
869
+ "mean_token_accuracy": 0.7714510339240377,
870
+ "step": 510
871
+ },
872
+ {
873
+ "epoch": 0.3810580836108028,
874
+ "grad_norm": 0.8195601280151146,
875
+ "learning_rate": 1.557047416521996e-05,
876
+ "loss": 0.7215,
877
+ "mean_token_accuracy": 0.7789905700934947,
878
+ "step": 515
879
+ },
880
+ {
881
+ "epoch": 0.38475767665556787,
882
+ "grad_norm": 0.9650761679887094,
883
+ "learning_rate": 1.546264405742654e-05,
884
+ "loss": 0.7782,
885
+ "mean_token_accuracy": 0.7636861487365787,
886
+ "step": 520
887
+ },
888
+ {
889
+ "epoch": 0.38845726970033295,
890
+ "grad_norm": 0.8237906709737827,
891
+ "learning_rate": 1.535390092172597e-05,
892
+ "loss": 0.7699,
893
+ "mean_token_accuracy": 0.7671755622376449,
894
+ "step": 525
895
+ },
896
+ {
897
+ "epoch": 0.39215686274509803,
898
+ "grad_norm": 0.8299457105893924,
899
+ "learning_rate": 1.5244262933477401e-05,
900
+ "loss": 0.7804,
901
+ "mean_token_accuracy": 0.7637937394911022,
902
+ "step": 530
903
+ },
904
+ {
905
+ "epoch": 0.3958564557898631,
906
+ "grad_norm": 1.0313966357425606,
907
+ "learning_rate": 1.5133748417605878e-05,
908
+ "loss": 0.7752,
909
+ "mean_token_accuracy": 0.7650544976847276,
910
+ "step": 535
911
+ },
912
+ {
913
+ "epoch": 0.3995560488346282,
914
+ "grad_norm": 0.8476613666163046,
915
+ "learning_rate": 1.5022375845539537e-05,
916
+ "loss": 0.7459,
917
+ "mean_token_accuracy": 0.7739755173371613,
918
+ "step": 540
919
+ },
920
+ {
921
+ "epoch": 0.4032556418793933,
922
+ "grad_norm": 0.8488122984336212,
923
+ "learning_rate": 1.4910163832122278e-05,
924
+ "loss": 0.7282,
925
+ "mean_token_accuracy": 0.7793907307750887,
926
+ "step": 545
927
+ },
928
+ {
929
+ "epoch": 0.40695523492415836,
930
+ "grad_norm": 0.9246732865437337,
931
+ "learning_rate": 1.4797131132502464e-05,
932
+ "loss": 0.7387,
933
+ "mean_token_accuracy": 0.7771462284436527,
934
+ "step": 550
935
+ },
936
+ {
937
+ "epoch": 0.41065482796892344,
938
+ "grad_norm": 0.8693436586761074,
939
+ "learning_rate": 1.4683296638998192e-05,
940
+ "loss": 0.7449,
941
+ "mean_token_accuracy": 0.7733528571354785,
942
+ "step": 555
943
+ },
944
+ {
945
+ "epoch": 0.41435442101368847,
946
+ "grad_norm": 1.0132116370018853,
947
+ "learning_rate": 1.4568679377939619e-05,
948
+ "loss": 0.7338,
949
+ "mean_token_accuracy": 0.7772718301836594,
950
+ "step": 560
951
+ },
952
+ {
953
+ "epoch": 0.41805401405845355,
954
+ "grad_norm": 0.8274852436061462,
955
+ "learning_rate": 1.4453298506488896e-05,
956
+ "loss": 0.7468,
957
+ "mean_token_accuracy": 0.7725176751538921,
958
+ "step": 565
959
+ },
960
+ {
961
+ "epoch": 0.42175360710321863,
962
+ "grad_norm": 0.8143005464066339,
963
+ "learning_rate": 1.4337173309438236e-05,
964
+ "loss": 0.7423,
965
+ "mean_token_accuracy": 0.7735196406247817,
966
+ "step": 570
967
+ },
968
+ {
969
+ "epoch": 0.4254532001479837,
970
+ "grad_norm": 0.8624822737135253,
971
+ "learning_rate": 1.4220323195986649e-05,
972
+ "loss": 0.7448,
973
+ "mean_token_accuracy": 0.7735122705759122,
974
+ "step": 575
975
+ },
976
+ {
977
+ "epoch": 0.4291527931927488,
978
+ "grad_norm": 0.8889965581940739,
979
+ "learning_rate": 1.4102767696495885e-05,
980
+ "loss": 0.7674,
981
+ "mean_token_accuracy": 0.7686958478722382,
982
+ "step": 580
983
+ },
984
+ {
985
+ "epoch": 0.4328523862375139,
986
+ "grad_norm": 0.7782401935752693,
987
+ "learning_rate": 1.398452645922611e-05,
988
+ "loss": 0.765,
989
+ "mean_token_accuracy": 0.7691531672999341,
990
+ "step": 585
991
+ },
992
+ {
993
+ "epoch": 0.43655197928227896,
994
+ "grad_norm": 0.8590974972802418,
995
+ "learning_rate": 1.3865619247051916e-05,
996
+ "loss": 0.7507,
997
+ "mean_token_accuracy": 0.7729800254264771,
998
+ "step": 590
999
+ },
1000
+ {
1001
+ "epoch": 0.44025157232704404,
1002
+ "grad_norm": 0.7988473909688566,
1003
+ "learning_rate": 1.3746065934159123e-05,
1004
+ "loss": 0.7242,
1005
+ "mean_token_accuracy": 0.780331890399065,
1006
+ "step": 595
1007
+ },
1008
+ {
1009
+ "epoch": 0.4439511653718091,
1010
+ "grad_norm": 0.9282202265668916,
1011
+ "learning_rate": 1.3625886502723008e-05,
1012
+ "loss": 0.7587,
1013
+ "mean_token_accuracy": 0.7696747774936822,
1014
+ "step": 600
1015
+ },
1016
+ {
1017
+ "epoch": 0.4439511653718091,
1018
+ "eval_loss": 0.7762653827667236,
1019
+ "eval_mean_token_accuracy": 0.7637766286530999,
1020
+ "eval_runtime": 12.9012,
1021
+ "eval_samples_per_second": 9.999,
1022
+ "eval_steps_per_second": 2.558,
1023
+ "step": 600
1024
+ },
1025
+ {
1026
+ "epoch": 0.4476507584165742,
1027
+ "grad_norm": 0.8375068814255916,
1028
+ "learning_rate": 1.3505101039568494e-05,
1029
+ "loss": 0.7114,
1030
+ "mean_token_accuracy": 0.78174785615947,
1031
+ "step": 605
1032
+ },
1033
+ {
1034
+ "epoch": 0.4513503514613392,
1035
+ "grad_norm": 0.8406018488041918,
1036
+ "learning_rate": 1.3383729732812814e-05,
1037
+ "loss": 0.7309,
1038
+ "mean_token_accuracy": 0.7761469410168554,
1039
+ "step": 610
1040
+ },
1041
+ {
1042
+ "epoch": 0.4550499445061043,
1043
+ "grad_norm": 0.8214051664479577,
1044
+ "learning_rate": 1.3261792868491267e-05,
1045
+ "loss": 0.754,
1046
+ "mean_token_accuracy": 0.7719396256522534,
1047
+ "step": 615
1048
+ },
1049
+ {
1050
+ "epoch": 0.4587495375508694,
1051
+ "grad_norm": 0.8854429907993514,
1052
+ "learning_rate": 1.3139310827166613e-05,
1053
+ "loss": 0.7529,
1054
+ "mean_token_accuracy": 0.7723088441600371,
1055
+ "step": 620
1056
+ },
1057
+ {
1058
+ "epoch": 0.46244913059563447,
1059
+ "grad_norm": 0.8135757240558104,
1060
+ "learning_rate": 1.3016304080522657e-05,
1061
+ "loss": 0.7756,
1062
+ "mean_token_accuracy": 0.7642865483170395,
1063
+ "step": 625
1064
+ },
1065
+ {
1066
+ "epoch": 0.46614872364039955,
1067
+ "grad_norm": 0.8597128387115325,
1068
+ "learning_rate": 1.2892793187942588e-05,
1069
+ "loss": 0.7856,
1070
+ "mean_token_accuracy": 0.7615404441461953,
1071
+ "step": 630
1072
+ },
1073
+ {
1074
+ "epoch": 0.46984831668516464,
1075
+ "grad_norm": 0.8740917081939998,
1076
+ "learning_rate": 1.2768798793072708e-05,
1077
+ "loss": 0.7599,
1078
+ "mean_token_accuracy": 0.7694673674218602,
1079
+ "step": 635
1080
+ },
1081
+ {
1082
+ "epoch": 0.4735479097299297,
1083
+ "grad_norm": 0.8455035652800976,
1084
+ "learning_rate": 1.2644341620372025e-05,
1085
+ "loss": 0.7374,
1086
+ "mean_token_accuracy": 0.7753507196148239,
1087
+ "step": 640
1088
+ },
1089
+ {
1090
+ "epoch": 0.4772475027746948,
1091
+ "grad_norm": 0.8453775134661862,
1092
+ "learning_rate": 1.2519442471648364e-05,
1093
+ "loss": 0.7539,
1094
+ "mean_token_accuracy": 0.7711800403046617,
1095
+ "step": 645
1096
+ },
1097
+ {
1098
+ "epoch": 0.4809470958194599,
1099
+ "grad_norm": 0.8131753165282993,
1100
+ "learning_rate": 1.2394122222581557e-05,
1101
+ "loss": 0.7476,
1102
+ "mean_token_accuracy": 0.7740160766147529,
1103
+ "step": 650
1104
+ },
1105
+ {
1106
+ "epoch": 0.48464668886422496,
1107
+ "grad_norm": 0.8752309330635153,
1108
+ "learning_rate": 1.226840181923427e-05,
1109
+ "loss": 0.7234,
1110
+ "mean_token_accuracy": 0.7799259045764162,
1111
+ "step": 655
1112
+ },
1113
+ {
1114
+ "epoch": 0.48834628190899,
1115
+ "grad_norm": 0.9394868684222661,
1116
+ "learning_rate": 1.214230227455106e-05,
1117
+ "loss": 0.761,
1118
+ "mean_token_accuracy": 0.76914448788722,
1119
+ "step": 660
1120
+ },
1121
+ {
1122
+ "epoch": 0.49204587495375507,
1123
+ "grad_norm": 0.9190997939430626,
1124
+ "learning_rate": 1.201584466484629e-05,
1125
+ "loss": 0.7429,
1126
+ "mean_token_accuracy": 0.7749135561574454,
1127
+ "step": 665
1128
+ },
1129
+ {
1130
+ "epoch": 0.49574546799852015,
1131
+ "grad_norm": 0.991650882333127,
1132
+ "learning_rate": 1.1889050126281405e-05,
1133
+ "loss": 0.7624,
1134
+ "mean_token_accuracy": 0.7674541114303453,
1135
+ "step": 670
1136
+ },
1137
+ {
1138
+ "epoch": 0.49944506104328523,
1139
+ "grad_norm": 0.8727170236064898,
1140
+ "learning_rate": 1.1761939851332241e-05,
1141
+ "loss": 0.7686,
1142
+ "mean_token_accuracy": 0.7668983696182747,
1143
+ "step": 675
1144
+ },
1145
+ {
1146
+ "epoch": 0.5031446540880503,
1147
+ "grad_norm": 0.8423456719916217,
1148
+ "learning_rate": 1.1634535085246903e-05,
1149
+ "loss": 0.7502,
1150
+ "mean_token_accuracy": 0.7719282566060406,
1151
+ "step": 680
1152
+ },
1153
+ {
1154
+ "epoch": 0.5068442471328154,
1155
+ "grad_norm": 0.8041730146404278,
1156
+ "learning_rate": 1.1506857122494832e-05,
1157
+ "loss": 0.7557,
1158
+ "mean_token_accuracy": 0.7698715177090566,
1159
+ "step": 685
1160
+ },
1161
+ {
1162
+ "epoch": 0.5105438401775805,
1163
+ "grad_norm": 0.8008949159036595,
1164
+ "learning_rate": 1.1378927303207637e-05,
1165
+ "loss": 0.7555,
1166
+ "mean_token_accuracy": 0.7708746251231458,
1167
+ "step": 690
1168
+ },
1169
+ {
1170
+ "epoch": 0.5142434332223456,
1171
+ "grad_norm": 0.8535115805575514,
1172
+ "learning_rate": 1.12507670096123e-05,
1173
+ "loss": 0.7427,
1174
+ "mean_token_accuracy": 0.7726444687618006,
1175
+ "step": 695
1176
+ },
1177
+ {
1178
+ "epoch": 0.5179430262671106,
1179
+ "grad_norm": 0.9312344890724055,
1180
+ "learning_rate": 1.1122397662457352e-05,
1181
+ "loss": 0.7502,
1182
+ "mean_token_accuracy": 0.7726102122304674,
1183
+ "step": 700
1184
+ },
1185
+ {
1186
+ "epoch": 0.5179430262671106,
1187
+ "eval_loss": 0.7686853408813477,
1188
+ "eval_mean_token_accuracy": 0.7649829158123188,
1189
+ "eval_runtime": 12.9077,
1190
+ "eval_samples_per_second": 9.994,
1191
+ "eval_steps_per_second": 2.557,
1192
+ "step": 700
1193
+ },
1194
+ {
1195
+ "epoch": 0.5216426193118757,
1196
+ "grad_norm": 0.868812002235657,
1197
+ "learning_rate": 1.0993840717432582e-05,
1198
+ "loss": 0.739,
1199
+ "mean_token_accuracy": 0.7753342555426331,
1200
+ "step": 705
1201
+ },
1202
+ {
1203
+ "epoch": 0.5253422123566408,
1204
+ "grad_norm": 0.873440662985193,
1205
+ "learning_rate": 1.0865117661582958e-05,
1206
+ "loss": 0.7439,
1207
+ "mean_token_accuracy": 0.7739453320579491,
1208
+ "step": 710
1209
+ },
1210
+ {
1211
+ "epoch": 0.5290418054014059,
1212
+ "grad_norm": 0.862857533425593,
1213
+ "learning_rate": 1.0736250009717249e-05,
1214
+ "loss": 0.7569,
1215
+ "mean_token_accuracy": 0.770897334069314,
1216
+ "step": 715
1217
+ },
1218
+ {
1219
+ "epoch": 0.532741398446171,
1220
+ "grad_norm": 0.8639562204638955,
1221
+ "learning_rate": 1.0607259300812047e-05,
1222
+ "loss": 0.7074,
1223
+ "mean_token_accuracy": 0.7842015210192824,
1224
+ "step": 720
1225
+ },
1226
+ {
1227
+ "epoch": 0.536440991490936,
1228
+ "grad_norm": 0.899875919647952,
1229
+ "learning_rate": 1.0478167094411733e-05,
1230
+ "loss": 0.7363,
1231
+ "mean_token_accuracy": 0.7761369932840716,
1232
+ "step": 725
1233
+ },
1234
+ {
1235
+ "epoch": 0.540140584535701,
1236
+ "grad_norm": 0.8342218548841652,
1237
+ "learning_rate": 1.0348994967025012e-05,
1238
+ "loss": 0.7417,
1239
+ "mean_token_accuracy": 0.7746170812078568,
1240
+ "step": 730
1241
+ },
1242
+ {
1243
+ "epoch": 0.5438401775804661,
1244
+ "grad_norm": 0.8013624939670422,
1245
+ "learning_rate": 1.0219764508518595e-05,
1246
+ "loss": 0.7384,
1247
+ "mean_token_accuracy": 0.7767422075955539,
1248
+ "step": 735
1249
+ },
1250
+ {
1251
+ "epoch": 0.5475397706252312,
1252
+ "grad_norm": 0.8120468940513571,
1253
+ "learning_rate": 1.0090497318508687e-05,
1254
+ "loss": 0.7566,
1255
+ "mean_token_accuracy": 0.7691283622927295,
1256
+ "step": 740
1257
+ },
1258
+ {
1259
+ "epoch": 0.5512393636699963,
1260
+ "grad_norm": 0.8150917404467021,
1261
+ "learning_rate": 9.961215002750799e-06,
1262
+ "loss": 0.7624,
1263
+ "mean_token_accuracy": 0.7672601324412168,
1264
+ "step": 745
1265
+ },
1266
+ {
1267
+ "epoch": 0.5549389567147613,
1268
+ "grad_norm": 0.8358908477578574,
1269
+ "learning_rate": 9.831939169528565e-06,
1270
+ "loss": 0.7198,
1271
+ "mean_token_accuracy": 0.7803871901829526,
1272
+ "step": 750
1273
+ },
1274
+ {
1275
+ "epoch": 0.5586385497595264,
1276
+ "grad_norm": 0.8360497025012443,
1277
+ "learning_rate": 9.702691426042124e-06,
1278
+ "loss": 0.704,
1279
+ "mean_token_accuracy": 0.784279945314917,
1280
+ "step": 755
1281
+ },
1282
+ {
1283
+ "epoch": 0.5623381428042915,
1284
+ "grad_norm": 0.8484683109694328,
1285
+ "learning_rate": 9.573493374796694e-06,
1286
+ "loss": 0.742,
1287
+ "mean_token_accuracy": 0.773911456375648,
1288
+ "step": 760
1289
+ },
1290
+ {
1291
+ "epoch": 0.5660377358490566,
1292
+ "grad_norm": 0.801814751253701,
1293
+ "learning_rate": 9.444366609991916e-06,
1294
+ "loss": 0.7475,
1295
+ "mean_token_accuracy": 0.7704960987519774,
1296
+ "step": 765
1297
+ },
1298
+ {
1299
+ "epoch": 0.5697373288938217,
1300
+ "grad_norm": 0.7903451176947093,
1301
+ "learning_rate": 9.315332713912593e-06,
1302
+ "loss": 0.6802,
1303
+ "mean_token_accuracy": 0.7910297225776487,
1304
+ "step": 770
1305
+ },
1306
+ {
1307
+ "epoch": 0.5734369219385868,
1308
+ "grad_norm": 0.7640452473763807,
1309
+ "learning_rate": 9.18641325332142e-06,
1310
+ "loss": 0.7254,
1311
+ "mean_token_accuracy": 0.7779958468370307,
1312
+ "step": 775
1313
+ },
1314
+ {
1315
+ "epoch": 0.5771365149833518,
1316
+ "grad_norm": 0.7764311897247947,
1317
+ "learning_rate": 9.057629775854314e-06,
1318
+ "loss": 0.7317,
1319
+ "mean_token_accuracy": 0.7763394937851484,
1320
+ "step": 780
1321
+ },
1322
+ {
1323
+ "epoch": 0.5808361080281169,
1324
+ "grad_norm": 0.8245708624875354,
1325
+ "learning_rate": 8.929003806418934e-06,
1326
+ "loss": 0.7376,
1327
+ "mean_token_accuracy": 0.7750152545715012,
1328
+ "step": 785
1329
+ },
1330
+ {
1331
+ "epoch": 0.584535701072882,
1332
+ "grad_norm": 0.8103882551202022,
1333
+ "learning_rate": 8.800556843597002e-06,
1334
+ "loss": 0.7259,
1335
+ "mean_token_accuracy": 0.7782374392513656,
1336
+ "step": 790
1337
+ },
1338
+ {
1339
+ "epoch": 0.5882352941176471,
1340
+ "grad_norm": 0.7928991222424273,
1341
+ "learning_rate": 8.672310356051023e-06,
1342
+ "loss": 0.7431,
1343
+ "mean_token_accuracy": 0.7736308218387102,
1344
+ "step": 795
1345
+ },
1346
+ {
1347
+ "epoch": 0.5919348871624122,
1348
+ "grad_norm": 0.8289989830065249,
1349
+ "learning_rate": 8.544285778936004e-06,
1350
+ "loss": 0.7352,
1351
+ "mean_token_accuracy": 0.7764898230184529,
1352
+ "step": 800
1353
+ },
1354
+ {
1355
+ "epoch": 0.5919348871624122,
1356
+ "eval_loss": 0.7624932527542114,
1357
+ "eval_mean_token_accuracy": 0.7669661829206922,
1358
+ "eval_runtime": 12.9119,
1359
+ "eval_samples_per_second": 9.991,
1360
+ "eval_steps_per_second": 2.556,
1361
+ "step": 800
1362
+ },
1363
+ {
1364
+ "epoch": 0.5956344802071772,
1365
+ "grad_norm": 0.8482327309493489,
1366
+ "learning_rate": 8.416504510316774e-06,
1367
+ "loss": 0.7194,
1368
+ "mean_token_accuracy": 0.7805804982214577,
1369
+ "step": 805
1370
+ },
1371
+ {
1372
+ "epoch": 0.5993340732519423,
1373
+ "grad_norm": 0.8542094892921783,
1374
+ "learning_rate": 8.28898790759152e-06,
1375
+ "loss": 0.7332,
1376
+ "mean_token_accuracy": 0.7760156726467715,
1377
+ "step": 810
1378
+ },
1379
+ {
1380
+ "epoch": 0.6030336662967074,
1381
+ "grad_norm": 0.8315942810202602,
1382
+ "learning_rate": 8.161757283922084e-06,
1383
+ "loss": 0.7186,
1384
+ "mean_token_accuracy": 0.7800488930789287,
1385
+ "step": 815
1386
+ },
1387
+ {
1388
+ "epoch": 0.6067332593414725,
1389
+ "grad_norm": 0.8561047694162326,
1390
+ "learning_rate": 8.034833904671698e-06,
1391
+ "loss": 0.7286,
1392
+ "mean_token_accuracy": 0.7782032355305302,
1393
+ "step": 820
1394
+ },
1395
+ {
1396
+ "epoch": 0.6104328523862376,
1397
+ "grad_norm": 0.8322803740232512,
1398
+ "learning_rate": 7.908238983850666e-06,
1399
+ "loss": 0.7475,
1400
+ "mean_token_accuracy": 0.7716142551709843,
1401
+ "step": 825
1402
+ },
1403
+ {
1404
+ "epoch": 0.6141324454310025,
1405
+ "grad_norm": 0.8093783949232168,
1406
+ "learning_rate": 7.781993680570656e-06,
1407
+ "loss": 0.7419,
1408
+ "mean_token_accuracy": 0.7753669766613711,
1409
+ "step": 830
1410
+ },
1411
+ {
1412
+ "epoch": 0.6178320384757676,
1413
+ "grad_norm": 0.8251647916013486,
1414
+ "learning_rate": 7.656119095508155e-06,
1415
+ "loss": 0.7163,
1416
+ "mean_token_accuracy": 0.7810057582729616,
1417
+ "step": 835
1418
+ },
1419
+ {
1420
+ "epoch": 0.6215316315205327,
1421
+ "grad_norm": 0.8936545458969576,
1422
+ "learning_rate": 7.530636267377706e-06,
1423
+ "loss": 0.7212,
1424
+ "mean_token_accuracy": 0.7782000869575987,
1425
+ "step": 840
1426
+ },
1427
+ {
1428
+ "epoch": 0.6252312245652978,
1429
+ "grad_norm": 0.8327340922107708,
1430
+ "learning_rate": 7.405566169415481e-06,
1431
+ "loss": 0.7417,
1432
+ "mean_token_accuracy": 0.7713689918387485,
1433
+ "step": 845
1434
+ },
1435
+ {
1436
+ "epoch": 0.6289308176100629,
1437
+ "grad_norm": 0.9240689034622575,
1438
+ "learning_rate": 7.280929705873818e-06,
1439
+ "loss": 0.7671,
1440
+ "mean_token_accuracy": 0.7674969695134786,
1441
+ "step": 850
1442
+ },
1443
+ {
1444
+ "epoch": 0.632630410654828,
1445
+ "grad_norm": 0.8604187816835673,
1446
+ "learning_rate": 7.15674770852727e-06,
1447
+ "loss": 0.7714,
1448
+ "mean_token_accuracy": 0.765074378427229,
1449
+ "step": 855
1450
+ },
1451
+ {
1452
+ "epoch": 0.636330003699593,
1453
+ "grad_norm": 0.8575462475833365,
1454
+ "learning_rate": 7.033040933190776e-06,
1455
+ "loss": 0.7485,
1456
+ "mean_token_accuracy": 0.7725655493857382,
1457
+ "step": 860
1458
+ },
1459
+ {
1460
+ "epoch": 0.6400295967443581,
1461
+ "grad_norm": 0.7868065852245846,
1462
+ "learning_rate": 6.909830056250527e-06,
1463
+ "loss": 0.703,
1464
+ "mean_token_accuracy": 0.7851972669426284,
1465
+ "step": 865
1466
+ },
1467
+ {
1468
+ "epoch": 0.6437291897891232,
1469
+ "grad_norm": 0.817598798759793,
1470
+ "learning_rate": 6.787135671208126e-06,
1471
+ "loss": 0.7576,
1472
+ "mean_token_accuracy": 0.7701196194143121,
1473
+ "step": 870
1474
+ },
1475
+ {
1476
+ "epoch": 0.6474287828338883,
1477
+ "grad_norm": 0.7951083662310198,
1478
+ "learning_rate": 6.6649782852385554e-06,
1479
+ "loss": 0.7459,
1480
+ "mean_token_accuracy": 0.772123421166796,
1481
+ "step": 875
1482
+ },
1483
+ {
1484
+ "epoch": 0.6511283758786534,
1485
+ "grad_norm": 0.7271974698252879,
1486
+ "learning_rate": 6.543378315762634e-06,
1487
+ "loss": 0.7247,
1488
+ "mean_token_accuracy": 0.7791351612791593,
1489
+ "step": 880
1490
+ },
1491
+ {
1492
+ "epoch": 0.6548279689234184,
1493
+ "grad_norm": 0.8824398500626421,
1494
+ "learning_rate": 6.42235608703441e-06,
1495
+ "loss": 0.7606,
1496
+ "mean_token_accuracy": 0.7677733208877957,
1497
+ "step": 885
1498
+ },
1499
+ {
1500
+ "epoch": 0.6585275619681835,
1501
+ "grad_norm": 0.8552464856465032,
1502
+ "learning_rate": 6.301931826744189e-06,
1503
+ "loss": 0.7419,
1504
+ "mean_token_accuracy": 0.7743753446438143,
1505
+ "step": 890
1506
+ },
1507
+ {
1508
+ "epoch": 0.6622271550129486,
1509
+ "grad_norm": 0.8306173604508144,
1510
+ "learning_rate": 6.18212566263765e-06,
1511
+ "loss": 0.7378,
1512
+ "mean_token_accuracy": 0.7754989605528264,
1513
+ "step": 895
1514
+ },
1515
+ {
1516
+ "epoch": 0.6659267480577137,
1517
+ "grad_norm": 0.8046739019546305,
1518
+ "learning_rate": 6.0629576191517035e-06,
1519
+ "loss": 0.7254,
1520
+ "mean_token_accuracy": 0.7763354265372273,
1521
+ "step": 900
1522
+ },
1523
+ {
1524
+ "epoch": 0.6659267480577137,
1525
+ "eval_loss": 0.7582561373710632,
1526
+ "eval_mean_token_accuracy": 0.7673597290333886,
1527
+ "eval_runtime": 12.9256,
1528
+ "eval_samples_per_second": 9.98,
1529
+ "eval_steps_per_second": 2.553,
1530
+ "step": 900
1531
+ },
1532
+ {
1533
+ "epoch": 0.6696263411024788,
1534
+ "grad_norm": 0.7871630029994926,
1535
+ "learning_rate": 5.944447614067588e-06,
1536
+ "loss": 0.7365,
1537
+ "mean_token_accuracy": 0.7753696170648604,
1538
+ "step": 905
1539
+ },
1540
+ {
1541
+ "epoch": 0.6733259341472438,
1542
+ "grad_norm": 0.8274033791731158,
1543
+ "learning_rate": 5.8266154551818225e-06,
1544
+ "loss": 0.7148,
1545
+ "mean_token_accuracy": 0.7819856387093902,
1546
+ "step": 910
1547
+ },
1548
+ {
1549
+ "epoch": 0.6770255271920089,
1550
+ "grad_norm": 0.8585256390323572,
1551
+ "learning_rate": 5.709480836995509e-06,
1552
+ "loss": 0.7099,
1553
+ "mean_token_accuracy": 0.7823146568890846,
1554
+ "step": 915
1555
+ },
1556
+ {
1557
+ "epoch": 0.680725120236774,
1558
+ "grad_norm": 0.8245071508434019,
1559
+ "learning_rate": 5.593063337422595e-06,
1560
+ "loss": 0.7205,
1561
+ "mean_token_accuracy": 0.7803457311979414,
1562
+ "step": 920
1563
+ },
1564
+ {
1565
+ "epoch": 0.684424713281539,
1566
+ "grad_norm": 0.8137995619905571,
1567
+ "learning_rate": 5.477382414517625e-06,
1568
+ "loss": 0.7443,
1569
+ "mean_token_accuracy": 0.7742054658480322,
1570
+ "step": 925
1571
+ },
1572
+ {
1573
+ "epoch": 0.6881243063263041,
1574
+ "grad_norm": 0.8435356879873597,
1575
+ "learning_rate": 5.362457403223495e-06,
1576
+ "loss": 0.7321,
1577
+ "mean_token_accuracy": 0.7777223904380888,
1578
+ "step": 930
1579
+ },
1580
+ {
1581
+ "epoch": 0.6918238993710691,
1582
+ "grad_norm": 0.8275951168813332,
1583
+ "learning_rate": 5.248307512139818e-06,
1584
+ "loss": 0.7327,
1585
+ "mean_token_accuracy": 0.7760629303675801,
1586
+ "step": 935
1587
+ },
1588
+ {
1589
+ "epoch": 0.6955234924158342,
1590
+ "grad_norm": 0.7986090408289105,
1591
+ "learning_rate": 5.134951820312402e-06,
1592
+ "loss": 0.6797,
1593
+ "mean_token_accuracy": 0.792751892181402,
1594
+ "step": 940
1595
+ },
1596
+ {
1597
+ "epoch": 0.6992230854605993,
1598
+ "grad_norm": 0.8344358039184492,
1599
+ "learning_rate": 5.022409274044346e-06,
1600
+ "loss": 0.7484,
1601
+ "mean_token_accuracy": 0.7706997735792959,
1602
+ "step": 945
1603
+ },
1604
+ {
1605
+ "epoch": 0.7029226785053644,
1606
+ "grad_norm": 0.7590018472193605,
1607
+ "learning_rate": 4.910698683729371e-06,
1608
+ "loss": 0.7246,
1609
+ "mean_token_accuracy": 0.7794280766358616,
1610
+ "step": 950
1611
+ },
1612
+ {
1613
+ "epoch": 0.7066222715501295,
1614
+ "grad_norm": 0.8418460117578578,
1615
+ "learning_rate": 4.799838720707847e-06,
1616
+ "loss": 0.6786,
1617
+ "mean_token_accuracy": 0.7908993228344331,
1618
+ "step": 955
1619
+ },
1620
+ {
1621
+ "epoch": 0.7103218645948945,
1622
+ "grad_norm": 0.8709719228420416,
1623
+ "learning_rate": 4.6898479141460415e-06,
1624
+ "loss": 0.7067,
1625
+ "mean_token_accuracy": 0.7843742462359516,
1626
+ "step": 960
1627
+ },
1628
+ {
1629
+ "epoch": 0.7140214576396596,
1630
+ "grad_norm": 0.8535611787813129,
1631
+ "learning_rate": 4.580744647939163e-06,
1632
+ "loss": 0.7435,
1633
+ "mean_token_accuracy": 0.7730417779644787,
1634
+ "step": 965
1635
+ },
1636
+ {
1637
+ "epoch": 0.7177210506844247,
1638
+ "grad_norm": 0.8885996504303695,
1639
+ "learning_rate": 4.472547157638674e-06,
1640
+ "loss": 0.694,
1641
+ "mean_token_accuracy": 0.7880904268803789,
1642
+ "step": 970
1643
+ },
1644
+ {
1645
+ "epoch": 0.7214206437291898,
1646
+ "grad_norm": 0.9587623922743913,
1647
+ "learning_rate": 4.365273527404384e-06,
1648
+ "loss": 0.7401,
1649
+ "mean_token_accuracy": 0.7739410892854838,
1650
+ "step": 975
1651
+ },
1652
+ {
1653
+ "epoch": 0.7251202367739549,
1654
+ "grad_norm": 0.8107566238251871,
1655
+ "learning_rate": 4.258941686981864e-06,
1656
+ "loss": 0.7242,
1657
+ "mean_token_accuracy": 0.779501991954344,
1658
+ "step": 980
1659
+ },
1660
+ {
1661
+ "epoch": 0.72881982981872,
1662
+ "grad_norm": 0.8788408769548844,
1663
+ "learning_rate": 4.15356940870567e-06,
1664
+ "loss": 0.7382,
1665
+ "mean_token_accuracy": 0.7752977269534177,
1666
+ "step": 985
1667
+ },
1668
+ {
1669
+ "epoch": 0.732519422863485,
1670
+ "grad_norm": 0.8161841348783122,
1671
+ "learning_rate": 4.049174304528857e-06,
1672
+ "loss": 0.7122,
1673
+ "mean_token_accuracy": 0.7834159122342004,
1674
+ "step": 990
1675
+ },
1676
+ {
1677
+ "epoch": 0.7362190159082501,
1678
+ "grad_norm": 0.8045647610652492,
1679
+ "learning_rate": 3.945773823079315e-06,
1680
+ "loss": 0.7085,
1681
+ "mean_token_accuracy": 0.7838271814299385,
1682
+ "step": 995
1683
+ },
1684
+ {
1685
+ "epoch": 0.7399186089530152,
1686
+ "grad_norm": 0.7888951101393026,
1687
+ "learning_rate": 3.8433852467434175e-06,
1688
+ "loss": 0.7436,
1689
+ "mean_token_accuracy": 0.7715282648856208,
1690
+ "step": 1000
1691
+ },
1692
+ {
1693
+ "epoch": 0.7399186089530152,
1694
+ "eval_loss": 0.7540702819824219,
1695
+ "eval_mean_token_accuracy": 0.7684317578583987,
1696
+ "eval_runtime": 12.9145,
1697
+ "eval_samples_per_second": 9.989,
1698
+ "eval_steps_per_second": 2.555,
1699
+ "step": 1000
1700
+ },
1701
+ {
1702
+ "epoch": 0.7436182019977803,
1703
+ "grad_norm": 0.8476150102495247,
1704
+ "learning_rate": 3.742025688777413e-06,
1705
+ "loss": 0.755,
1706
+ "mean_token_accuracy": 0.769747059368865,
1707
+ "step": 1005
1708
+ },
1709
+ {
1710
+ "epoch": 0.7473177950425454,
1711
+ "grad_norm": 0.8447327457715914,
1712
+ "learning_rate": 3.641712090447125e-06,
1713
+ "loss": 0.7084,
1714
+ "mean_token_accuracy": 0.7835149860526573,
1715
+ "step": 1010
1716
+ },
1717
+ {
1718
+ "epoch": 0.7510173880873104,
1719
+ "grad_norm": 0.7966093391523419,
1720
+ "learning_rate": 3.542461218196379e-06,
1721
+ "loss": 0.7215,
1722
+ "mean_token_accuracy": 0.7790490334715525,
1723
+ "step": 1015
1724
+ },
1725
+ {
1726
+ "epoch": 0.7547169811320755,
1727
+ "grad_norm": 0.8909915448158038,
1728
+ "learning_rate": 3.444289660844665e-06,
1729
+ "loss": 0.7361,
1730
+ "mean_token_accuracy": 0.7758344142261316,
1731
+ "step": 1020
1732
+ },
1733
+ {
1734
+ "epoch": 0.7584165741768405,
1735
+ "grad_norm": 0.828582805131554,
1736
+ "learning_rate": 3.347213826814456e-06,
1737
+ "loss": 0.7317,
1738
+ "mean_token_accuracy": 0.7749168212587448,
1739
+ "step": 1025
1740
+ },
1741
+ {
1742
+ "epoch": 0.7621161672216056,
1743
+ "grad_norm": 0.8172943841676832,
1744
+ "learning_rate": 3.2512499413887255e-06,
1745
+ "loss": 0.7206,
1746
+ "mean_token_accuracy": 0.7795061516510636,
1747
+ "step": 1030
1748
+ },
1749
+ {
1750
+ "epoch": 0.7658157602663707,
1751
+ "grad_norm": 0.7948969347509628,
1752
+ "learning_rate": 3.1564140439990256e-06,
1753
+ "loss": 0.7534,
1754
+ "mean_token_accuracy": 0.7695401774020444,
1755
+ "step": 1035
1756
+ },
1757
+ {
1758
+ "epoch": 0.7695153533111357,
1759
+ "grad_norm": 0.7639527363991839,
1760
+ "learning_rate": 3.0627219855446667e-06,
1761
+ "loss": 0.6907,
1762
+ "mean_token_accuracy": 0.7891346245991027,
1763
+ "step": 1040
1764
+ },
1765
+ {
1766
+ "epoch": 0.7732149463559008,
1767
+ "grad_norm": 0.796687026157432,
1768
+ "learning_rate": 2.970189425743383e-06,
1769
+ "loss": 0.7365,
1770
+ "mean_token_accuracy": 0.7739980038896167,
1771
+ "step": 1045
1772
+ },
1773
+ {
1774
+ "epoch": 0.7769145394006659,
1775
+ "grad_norm": 0.7960037813992309,
1776
+ "learning_rate": 2.8788318305139808e-06,
1777
+ "loss": 0.7513,
1778
+ "mean_token_accuracy": 0.7689661221516821,
1779
+ "step": 1050
1780
+ },
1781
+ {
1782
+ "epoch": 0.780614132445431,
1783
+ "grad_norm": 0.7733589592006201,
1784
+ "learning_rate": 2.7886644693913333e-06,
1785
+ "loss": 0.7443,
1786
+ "mean_token_accuracy": 0.7728263504337974,
1787
+ "step": 1055
1788
+ },
1789
+ {
1790
+ "epoch": 0.7843137254901961,
1791
+ "grad_norm": 0.8565688482800564,
1792
+ "learning_rate": 2.6997024129742544e-06,
1793
+ "loss": 0.7328,
1794
+ "mean_token_accuracy": 0.7762133218500764,
1795
+ "step": 1060
1796
+ },
1797
+ {
1798
+ "epoch": 0.7880133185349611,
1799
+ "grad_norm": 0.8923008390952016,
1800
+ "learning_rate": 2.611960530406572e-06,
1801
+ "loss": 0.728,
1802
+ "mean_token_accuracy": 0.7796289219427248,
1803
+ "step": 1065
1804
+ },
1805
+ {
1806
+ "epoch": 0.7917129115797262,
1807
+ "grad_norm": 0.7873211808203069,
1808
+ "learning_rate": 2.5254534868919077e-06,
1809
+ "loss": 0.726,
1810
+ "mean_token_accuracy": 0.7775058698315347,
1811
+ "step": 1070
1812
+ },
1813
+ {
1814
+ "epoch": 0.7954125046244913,
1815
+ "grad_norm": 0.7719831428886607,
1816
+ "learning_rate": 2.4401957412425213e-06,
1817
+ "loss": 0.7309,
1818
+ "mean_token_accuracy": 0.7749563569631559,
1819
+ "step": 1075
1820
+ },
1821
+ {
1822
+ "epoch": 0.7991120976692564,
1823
+ "grad_norm": 0.835137021064406,
1824
+ "learning_rate": 2.3562015434626784e-06,
1825
+ "loss": 0.7145,
1826
+ "mean_token_accuracy": 0.7820543807444283,
1827
+ "step": 1080
1828
+ },
1829
+ {
1830
+ "epoch": 0.8028116907140215,
1831
+ "grad_norm": 0.7671221716084607,
1832
+ "learning_rate": 2.273484932366874e-06,
1833
+ "loss": 0.7021,
1834
+ "mean_token_accuracy": 0.7854601544650027,
1835
+ "step": 1085
1836
+ },
1837
+ {
1838
+ "epoch": 0.8065112837587866,
1839
+ "grad_norm": 0.7788468903413968,
1840
+ "learning_rate": 2.192059733233408e-06,
1841
+ "loss": 0.7244,
1842
+ "mean_token_accuracy": 0.7782192310784752,
1843
+ "step": 1090
1844
+ },
1845
+ {
1846
+ "epoch": 0.8102108768035516,
1847
+ "grad_norm": 0.8041148765506401,
1848
+ "learning_rate": 2.111939555493603e-06,
1849
+ "loss": 0.7225,
1850
+ "mean_token_accuracy": 0.7786757617228395,
1851
+ "step": 1095
1852
+ },
1853
+ {
1854
+ "epoch": 0.8139104698483167,
1855
+ "grad_norm": 0.9148318869053712,
1856
+ "learning_rate": 2.0331377904571303e-06,
1857
+ "loss": 0.745,
1858
+ "mean_token_accuracy": 0.7731225924993291,
1859
+ "step": 1100
1860
+ },
1861
+ {
1862
+ "epoch": 0.8139104698483167,
1863
+ "eval_loss": 0.7510696649551392,
1864
+ "eval_mean_token_accuracy": 0.7692631147293293,
1865
+ "eval_runtime": 12.9268,
1866
+ "eval_samples_per_second": 9.979,
1867
+ "eval_steps_per_second": 2.553,
1868
+ "step": 1100
1869
+ },
1870
+ {
1871
+ "epoch": 0.8176100628930818,
1872
+ "grad_norm": 0.8176644589533959,
1873
+ "learning_rate": 1.9556676090737803e-06,
1874
+ "loss": 0.7544,
1875
+ "mean_token_accuracy": 0.7705814013113146,
1876
+ "step": 1105
1877
+ },
1878
+ {
1879
+ "epoch": 0.8213096559378469,
1880
+ "grad_norm": 0.751484746704062,
1881
+ "learning_rate": 1.879541959732072e-06,
1882
+ "loss": 0.7133,
1883
+ "mean_token_accuracy": 0.7810593845260246,
1884
+ "step": 1110
1885
+ },
1886
+ {
1887
+ "epoch": 0.825009248982612,
1888
+ "grad_norm": 0.8448196083310406,
1889
+ "learning_rate": 1.8047735660950427e-06,
1890
+ "loss": 0.7088,
1891
+ "mean_token_accuracy": 0.7839263073608779,
1892
+ "step": 1115
1893
+ },
1894
+ {
1895
+ "epoch": 0.8287088420273769,
1896
+ "grad_norm": 0.7666551359898016,
1897
+ "learning_rate": 1.7313749249736266e-06,
1898
+ "loss": 0.7225,
1899
+ "mean_token_accuracy": 0.7800409694989299,
1900
+ "step": 1120
1901
+ },
1902
+ {
1903
+ "epoch": 0.832408435072142,
1904
+ "grad_norm": 0.7750630686515836,
1905
+ "learning_rate": 1.6593583042379192e-06,
1906
+ "loss": 0.7302,
1907
+ "mean_token_accuracy": 0.7764794004807236,
1908
+ "step": 1125
1909
+ },
1910
+ {
1911
+ "epoch": 0.8361080281169071,
1912
+ "grad_norm": 0.7448783923471559,
1913
+ "learning_rate": 1.5887357407667314e-06,
1914
+ "loss": 0.7303,
1915
+ "mean_token_accuracy": 0.7770985998568949,
1916
+ "step": 1130
1917
+ },
1918
+ {
1919
+ "epoch": 0.8398076211616722,
1920
+ "grad_norm": 0.8617250383965689,
1921
+ "learning_rate": 1.5195190384357405e-06,
1922
+ "loss": 0.7261,
1923
+ "mean_token_accuracy": 0.7790343243952108,
1924
+ "step": 1135
1925
+ },
1926
+ {
1927
+ "epoch": 0.8435072142064373,
1928
+ "grad_norm": 0.7315929117396289,
1929
+ "learning_rate": 1.4517197661445893e-06,
1930
+ "loss": 0.7103,
1931
+ "mean_token_accuracy": 0.7820174506567448,
1932
+ "step": 1140
1933
+ },
1934
+ {
1935
+ "epoch": 0.8472068072512023,
1936
+ "grad_norm": 0.7464816300853588,
1937
+ "learning_rate": 1.3853492558832472e-06,
1938
+ "loss": 0.7248,
1939
+ "mean_token_accuracy": 0.7778668411101556,
1940
+ "step": 1145
1941
+ },
1942
+ {
1943
+ "epoch": 0.8509064002959674,
1944
+ "grad_norm": 0.8027284787026119,
1945
+ "learning_rate": 1.3204186008379926e-06,
1946
+ "loss": 0.7142,
1947
+ "mean_token_accuracy": 0.7828477066310621,
1948
+ "step": 1150
1949
+ },
1950
+ {
1951
+ "epoch": 0.8546059933407325,
1952
+ "grad_norm": 0.7867579697995757,
1953
+ "learning_rate": 1.2569386535372807e-06,
1954
+ "loss": 0.7411,
1955
+ "mean_token_accuracy": 0.7738658084540277,
1956
+ "step": 1155
1957
+ },
1958
+ {
1959
+ "epoch": 0.8583055863854976,
1960
+ "grad_norm": 0.8069270526220779,
1961
+ "learning_rate": 1.1949200240378577e-06,
1962
+ "loss": 0.7066,
1963
+ "mean_token_accuracy": 0.7833467087206057,
1964
+ "step": 1160
1965
+ },
1966
+ {
1967
+ "epoch": 0.8620051794302627,
1968
+ "grad_norm": 0.8115083552759281,
1969
+ "learning_rate": 1.1343730781513896e-06,
1970
+ "loss": 0.7117,
1971
+ "mean_token_accuracy": 0.782034573182628,
1972
+ "step": 1165
1973
+ },
1974
+ {
1975
+ "epoch": 0.8657047724750278,
1976
+ "grad_norm": 0.7379062498914578,
1977
+ "learning_rate": 1.0753079357119134e-06,
1978
+ "loss": 0.7334,
1979
+ "mean_token_accuracy": 0.7753454314583148,
1980
+ "step": 1170
1981
+ },
1982
+ {
1983
+ "epoch": 0.8694043655197928,
1984
+ "grad_norm": 0.7894649463121637,
1985
+ "learning_rate": 1.017734468884417e-06,
1986
+ "loss": 0.6873,
1987
+ "mean_token_accuracy": 0.7897845425207853,
1988
+ "step": 1175
1989
+ },
1990
+ {
1991
+ "epoch": 0.8731039585645579,
1992
+ "grad_norm": 0.8023692788963809,
1993
+ "learning_rate": 9.616623005147952e-07,
1994
+ "loss": 0.7416,
1995
+ "mean_token_accuracy": 0.7721576809793975,
1996
+ "step": 1180
1997
+ },
1998
+ {
1999
+ "epoch": 0.876803551609323,
2000
+ "grad_norm": 0.7552635215864901,
2001
+ "learning_rate": 9.071008025214767e-07,
2002
+ "loss": 0.6686,
2003
+ "mean_token_accuracy": 0.7955690867836535,
2004
+ "step": 1185
2005
+ },
2006
+ {
2007
+ "epoch": 0.8805031446540881,
2008
+ "grad_norm": 0.7427939874056083,
2009
+ "learning_rate": 8.540590943290128e-07,
2010
+ "loss": 0.7366,
2011
+ "mean_token_accuracy": 0.7745106495210125,
2012
+ "step": 1190
2013
+ },
2014
+ {
2015
+ "epoch": 0.8842027376988532,
2016
+ "grad_norm": 0.7579131248407016,
2017
+ "learning_rate": 8.025460413438457e-07,
2018
+ "loss": 0.7146,
2019
+ "mean_token_accuracy": 0.7814567025093309,
2020
+ "step": 1195
2021
+ },
2022
+ {
2023
+ "epoch": 0.8879023307436182,
2024
+ "grad_norm": 0.8173415852977716,
2025
+ "learning_rate": 7.525702534725443e-07,
2026
+ "loss": 0.7231,
2027
+ "mean_token_accuracy": 0.7785201805258471,
2028
+ "step": 1200
2029
+ },
2030
+ {
2031
+ "epoch": 0.8879023307436182,
2032
+ "eval_loss": 0.749515950679779,
2033
+ "eval_mean_token_accuracy": 0.7696483845818836,
2034
+ "eval_runtime": 12.9328,
2035
+ "eval_samples_per_second": 9.975,
2036
+ "eval_steps_per_second": 2.552,
2037
+ "step": 1200
2038
+ },
2039
+ {
2040
+ "epoch": 0.8916019237883833,
2041
+ "grad_norm": 0.7711527562330054,
2042
+ "learning_rate": 7.041400836827439e-07,
2043
+ "loss": 0.7143,
2044
+ "mean_token_accuracy": 0.7821882194464815,
2045
+ "step": 1205
2046
+ },
2047
+ {
2048
+ "epoch": 0.8953015168331484,
2049
+ "grad_norm": 0.7623772941452016,
2050
+ "learning_rate": 6.572636266070265e-07,
2051
+ "loss": 0.7387,
2052
+ "mean_token_accuracy": 0.7745962709444631,
2053
+ "step": 1210
2054
+ },
2055
+ {
2056
+ "epoch": 0.8990011098779135,
2057
+ "grad_norm": 0.7652994503797836,
2058
+ "learning_rate": 6.119487171899807e-07,
2059
+ "loss": 0.6961,
2060
+ "mean_token_accuracy": 0.7874255931848155,
2061
+ "step": 1215
2062
+ },
2063
+ {
2064
+ "epoch": 0.9027007029226785,
2065
+ "grad_norm": 0.8269678511841839,
2066
+ "learning_rate": 5.682029293786673e-07,
2067
+ "loss": 0.7255,
2068
+ "mean_token_accuracy": 0.7770277914333097,
2069
+ "step": 1220
2070
+ },
2071
+ {
2072
+ "epoch": 0.9064002959674435,
2073
+ "grad_norm": 0.7302387175249154,
2074
+ "learning_rate": 5.26033574856708e-07,
2075
+ "loss": 0.7036,
2076
+ "mean_token_accuracy": 0.7842749808544124,
2077
+ "step": 1225
2078
+ },
2079
+ {
2080
+ "epoch": 0.9100998890122086,
2081
+ "grad_norm": 0.7426310574145085,
2082
+ "learning_rate": 4.854477018222103e-07,
2083
+ "loss": 0.7085,
2084
+ "mean_token_accuracy": 0.7827587579577575,
2085
+ "step": 1230
2086
+ },
2087
+ {
2088
+ "epoch": 0.9137994820569737,
2089
+ "grad_norm": 0.7805083509005499,
2090
+ "learning_rate": 4.464520938097294e-07,
2091
+ "loss": 0.6552,
2092
+ "mean_token_accuracy": 0.7984315941552589,
2093
+ "step": 1235
2094
+ },
2095
+ {
2096
+ "epoch": 0.9174990751017388,
2097
+ "grad_norm": 0.7838715084886818,
2098
+ "learning_rate": 4.0905326855646186e-07,
2099
+ "loss": 0.6978,
2100
+ "mean_token_accuracy": 0.7866137937789542,
2101
+ "step": 1240
2102
+ },
2103
+ {
2104
+ "epoch": 0.9211986681465039,
2105
+ "grad_norm": 0.764142848066667,
2106
+ "learning_rate": 3.732574769128738e-07,
2107
+ "loss": 0.7425,
2108
+ "mean_token_accuracy": 0.7735163959357182,
2109
+ "step": 1245
2110
+ },
2111
+ {
2112
+ "epoch": 0.9248982611912689,
2113
+ "grad_norm": 0.8510657590112316,
2114
+ "learning_rate": 3.390707017979311e-07,
2115
+ "loss": 0.714,
2116
+ "mean_token_accuracy": 0.7817031945118419,
2117
+ "step": 1250
2118
+ },
2119
+ {
2120
+ "epoch": 0.928597854236034,
2121
+ "grad_norm": 0.8104897922544609,
2122
+ "learning_rate": 3.06498657199108e-07,
2123
+ "loss": 0.6972,
2124
+ "mean_token_accuracy": 0.7874145607415896,
2125
+ "step": 1255
2126
+ },
2127
+ {
2128
+ "epoch": 0.9322974472807991,
2129
+ "grad_norm": 0.8439497662820569,
2130
+ "learning_rate": 2.7554678721735675e-07,
2131
+ "loss": 0.7267,
2132
+ "mean_token_accuracy": 0.7782783773976328,
2133
+ "step": 1260
2134
+ },
2135
+ {
2136
+ "epoch": 0.9359970403255642,
2137
+ "grad_norm": 0.7506475456675276,
2138
+ "learning_rate": 2.4622026515717654e-07,
2139
+ "loss": 0.7116,
2140
+ "mean_token_accuracy": 0.7825011111093741,
2141
+ "step": 1265
2142
+ },
2143
+ {
2144
+ "epoch": 0.9396966333703293,
2145
+ "grad_norm": 0.796131840794368,
2146
+ "learning_rate": 2.1852399266194312e-07,
2147
+ "loss": 0.7043,
2148
+ "mean_token_accuracy": 0.7831285774300467,
2149
+ "step": 1270
2150
+ },
2151
+ {
2152
+ "epoch": 0.9433962264150944,
2153
+ "grad_norm": 0.8874806771851743,
2154
+ "learning_rate": 1.9246259889464935e-07,
2155
+ "loss": 0.7372,
2156
+ "mean_token_accuracy": 0.7754287878050505,
2157
+ "step": 1275
2158
+ },
2159
+ {
2160
+ "epoch": 0.9470958194598594,
2161
+ "grad_norm": 0.7616413651005322,
2162
+ "learning_rate": 1.6804043976418438e-07,
2163
+ "loss": 0.7115,
2164
+ "mean_token_accuracy": 0.782469409255144,
2165
+ "step": 1280
2166
+ },
2167
+ {
2168
+ "epoch": 0.9507954125046245,
2169
+ "grad_norm": 0.7354908780262112,
2170
+ "learning_rate": 1.4526159719728595e-07,
2171
+ "loss": 0.6923,
2172
+ "mean_token_accuracy": 0.7877402800804598,
2173
+ "step": 1285
2174
+ },
2175
+ {
2176
+ "epoch": 0.9544950055493896,
2177
+ "grad_norm": 0.8147962775571531,
2178
+ "learning_rate": 1.24129878456285e-07,
2179
+ "loss": 0.7266,
2180
+ "mean_token_accuracy": 0.7786802824447104,
2181
+ "step": 1290
2182
+ },
2183
+ {
2184
+ "epoch": 0.9581945985941547,
2185
+ "grad_norm": 0.7892855143587276,
2186
+ "learning_rate": 1.0464881550276362e-07,
2187
+ "loss": 0.7106,
2188
+ "mean_token_accuracy": 0.7812839457332936,
2189
+ "step": 1295
2190
+ },
2191
+ {
2192
+ "epoch": 0.9618941916389198,
2193
+ "grad_norm": 0.7807996872371684,
2194
+ "learning_rate": 8.682166440721729e-08,
2195
+ "loss": 0.6851,
2196
+ "mean_token_accuracy": 0.7892695119285057,
2197
+ "step": 1300
2198
+ },
2199
+ {
2200
+ "epoch": 0.9618941916389198,
2201
+ "eval_loss": 0.7488865852355957,
2202
+ "eval_mean_token_accuracy": 0.7697332311412013,
2203
+ "eval_runtime": 12.9253,
2204
+ "eval_samples_per_second": 9.98,
2205
+ "eval_steps_per_second": 2.553,
2206
+ "step": 1300
2207
+ },
2208
+ {
2209
+ "epoch": 0.9655937846836848,
2210
+ "grad_norm": 0.7500272096924462,
2211
+ "learning_rate": 7.065140480483235e-08,
2212
+ "loss": 0.722,
2213
+ "mean_token_accuracy": 0.7809713685380442,
2214
+ "step": 1305
2215
+ },
2216
+ {
2217
+ "epoch": 0.9692933777284499,
2218
+ "grad_norm": 0.7740539516194233,
2219
+ "learning_rate": 5.6140739397474445e-08,
2220
+ "loss": 0.741,
2221
+ "mean_token_accuracy": 0.7735731347336795,
2222
+ "step": 1310
2223
+ },
2224
+ {
2225
+ "epoch": 0.9729929707732149,
2226
+ "grad_norm": 0.7621995001113339,
2227
+ "learning_rate": 4.329209350195651e-08,
2228
+ "loss": 0.7277,
2229
+ "mean_token_accuracy": 0.7768363708086847,
2230
+ "step": 1315
2231
+ },
2232
+ {
2233
+ "epoch": 0.97669256381798,
2234
+ "grad_norm": 0.7190084013715407,
2235
+ "learning_rate": 3.210761464466639e-08,
2236
+ "loss": 0.6751,
2237
+ "mean_token_accuracy": 0.7917913948419976,
2238
+ "step": 1320
2239
+ },
2240
+ {
2241
+ "epoch": 0.9803921568627451,
2242
+ "grad_norm": 0.832622944037137,
2243
+ "learning_rate": 2.2589172202635014e-08,
2244
+ "loss": 0.7183,
2245
+ "mean_token_accuracy": 0.7814518443094862,
2246
+ "step": 1325
2247
+ },
2248
+ {
2249
+ "epoch": 0.9840917499075101,
2250
+ "grad_norm": 0.8517126035217566,
2251
+ "learning_rate": 1.4738357091084177e-08,
2252
+ "loss": 0.7223,
2253
+ "mean_token_accuracy": 0.7779738218649835,
2254
+ "step": 1330
2255
+ },
2256
+ {
2257
+ "epoch": 0.9877913429522752,
2258
+ "grad_norm": 0.7735765973323167,
2259
+ "learning_rate": 8.556481497521418e-09,
2260
+ "loss": 0.7064,
2261
+ "mean_token_accuracy": 0.7825683170955192,
2262
+ "step": 1335
2263
+ },
2264
+ {
2265
+ "epoch": 0.9914909359970403,
2266
+ "grad_norm": 0.7713072693524988,
2267
+ "learning_rate": 4.044578662419918e-09,
2268
+ "loss": 0.7179,
2269
+ "mean_token_accuracy": 0.7801742579673625,
2270
+ "step": 1340
2271
+ },
2272
+ {
2273
+ "epoch": 0.9951905290418054,
2274
+ "grad_norm": 0.7685841540071269,
2275
+ "learning_rate": 1.203402706525525e-09,
2276
+ "loss": 0.7493,
2277
+ "mean_token_accuracy": 0.7706208056883479,
2278
+ "step": 1345
2279
+ },
2280
+ {
2281
+ "epoch": 0.9988901220865705,
2282
+ "grad_norm": 0.8159997368959132,
2283
+ "learning_rate": 3.342850480869686e-11,
2284
+ "loss": 0.7328,
2285
+ "mean_token_accuracy": 0.7759899768837795,
2286
+ "step": 1350
2287
+ },
2288
+ {
2289
+ "epoch": 0.9996300406955235,
2290
+ "mean_token_accuracy": 0.7755799181439557,
2291
+ "step": 1351,
2292
+ "total_flos": 76966677970944.0,
2293
+ "train_loss": 0.759784622734163,
2294
+ "train_runtime": 8483.4218,
2295
+ "train_samples_per_second": 2.549,
2296
+ "train_steps_per_second": 0.159
2297
+ }
2298
+ ],
2299
+ "logging_steps": 5,
2300
+ "max_steps": 1351,
2301
+ "num_input_tokens_seen": 0,
2302
+ "num_train_epochs": 1,
2303
+ "save_steps": 500,
2304
+ "stateful_callbacks": {
2305
+ "TrainerControl": {
2306
+ "args": {
2307
+ "should_epoch_stop": false,
2308
+ "should_evaluate": false,
2309
+ "should_log": false,
2310
+ "should_save": false,
2311
+ "should_training_stop": false
2312
+ },
2313
+ "attributes": {}
2314
+ }
2315
+ },
2316
+ "total_flos": 76966677970944.0,
2317
+ "train_batch_size": 2,
2318
+ "trial_name": null,
2319
+ "trial_params": null
2320
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b629b9ebd2b4da1315199a8d55becb6466748be95be89f27edb77f11c1fe438b
3
+ size 7416
vocab.json ADDED
The diff for this file is too large to render. See raw diff