NJ90523 commited on
Commit
cad8400
·
verified ·
1 Parent(s): 976b7c7

Upload folder using huggingface_hub

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 CHANGED
@@ -1,3 +1,209 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen3-4B
3
+ library_name: peft
4
+ pipeline_tag: text-generation
5
+ tags:
6
+ - base_model:adapter:Qwen/Qwen3-4B
7
+ - lora
8
+ - sft
9
+ - transformers
10
+ - trl
11
+ ---
12
+
13
+ # Model Card for Model ID
14
+
15
+ <!-- Provide a quick summary of what the model is/does. -->
16
+
17
+
18
+
19
+ ## Model Details
20
+
21
+ ### Model Description
22
+
23
+ <!-- Provide a longer summary of what this model is. -->
24
+
25
+
26
+
27
+ - **Developed by:** [More Information Needed]
28
+ - **Funded by [optional]:** [More Information Needed]
29
+ - **Shared by [optional]:** [More Information Needed]
30
+ - **Model type:** [More Information Needed]
31
+ - **Language(s) (NLP):** [More Information Needed]
32
+ - **License:** [More Information Needed]
33
+ - **Finetuned from model [optional]:** [More Information Needed]
34
+
35
+ ### Model Sources [optional]
36
+
37
+ <!-- Provide the basic links for the model. -->
38
+
39
+ - **Repository:** [More Information Needed]
40
+ - **Paper [optional]:** [More Information Needed]
41
+ - **Demo [optional]:** [More Information Needed]
42
+
43
+ ## Uses
44
+
45
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
46
+
47
+ ### Direct Use
48
+
49
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
50
+
51
+ [More Information Needed]
52
+
53
+ ### Downstream Use [optional]
54
+
55
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
56
+
57
+ [More Information Needed]
58
+
59
+ ### Out-of-Scope Use
60
+
61
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
62
+
63
+ [More Information Needed]
64
+
65
+ ## Bias, Risks, and Limitations
66
+
67
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
68
+
69
+ [More Information Needed]
70
+
71
+ ### Recommendations
72
+
73
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
74
+
75
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
76
+
77
+ ## How to Get Started with the Model
78
+
79
+ Use the code below to get started with the model.
80
+
81
+ [More Information Needed]
82
+
83
+ ## Training Details
84
+
85
+ ### Training Data
86
+
87
+ <!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
88
+
89
+ [More Information Needed]
90
+
91
+ ### Training Procedure
92
+
93
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
94
+
95
+ #### Preprocessing [optional]
96
+
97
+ [More Information Needed]
98
+
99
+
100
+ #### Training Hyperparameters
101
+
102
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
103
+
104
+ #### Speeds, Sizes, Times [optional]
105
+
106
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
107
+
108
+ [More Information Needed]
109
+
110
+ ## Evaluation
111
+
112
+ <!-- This section describes the evaluation protocols and provides the results. -->
113
+
114
+ ### Testing Data, Factors & Metrics
115
+
116
+ #### Testing Data
117
+
118
+ <!-- This should link to a Dataset Card if possible. -->
119
+
120
+ [More Information Needed]
121
+
122
+ #### Factors
123
+
124
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
125
+
126
+ [More Information Needed]
127
+
128
+ #### Metrics
129
+
130
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
131
+
132
+ [More Information Needed]
133
+
134
+ ### Results
135
+
136
+ [More Information Needed]
137
+
138
+ #### Summary
139
+
140
+
141
+
142
+ ## Model Examination [optional]
143
+
144
+ <!-- Relevant interpretability work for the model goes here -->
145
+
146
+ [More Information Needed]
147
+
148
+ ## Environmental Impact
149
+
150
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
151
+
152
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
153
+
154
+ - **Hardware Type:** [More Information Needed]
155
+ - **Hours used:** [More Information Needed]
156
+ - **Cloud Provider:** [More Information Needed]
157
+ - **Compute Region:** [More Information Needed]
158
+ - **Carbon Emitted:** [More Information Needed]
159
+
160
+ ## Technical Specifications [optional]
161
+
162
+ ### Model Architecture and Objective
163
+
164
+ [More Information Needed]
165
+
166
+ ### Compute Infrastructure
167
+
168
+ [More Information Needed]
169
+
170
+ #### Hardware
171
+
172
+ [More Information Needed]
173
+
174
+ #### Software
175
+
176
+ [More Information Needed]
177
+
178
+ ## Citation [optional]
179
+
180
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
181
+
182
+ **BibTeX:**
183
+
184
+ [More Information Needed]
185
+
186
+ **APA:**
187
+
188
+ [More Information Needed]
189
+
190
+ ## Glossary [optional]
191
+
192
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
193
+
194
+ [More Information Needed]
195
+
196
+ ## More Information [optional]
197
+
198
+ [More Information Needed]
199
+
200
+ ## Model Card Authors [optional]
201
+
202
+ [More Information Needed]
203
+
204
+ ## Model Card Contact
205
+
206
+ [More Information Needed]
207
+ ### Framework versions
208
+
209
+ - PEFT 0.18.0
adapter_config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "Qwen/Qwen3-4B",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": null,
25
+ "peft_type": "LORA",
26
+ "peft_version": "0.18.0",
27
+ "qalora_group_size": 16,
28
+ "r": 32,
29
+ "rank_pattern": {},
30
+ "revision": null,
31
+ "target_modules": [
32
+ "gate_proj",
33
+ "down_proj",
34
+ "o_proj",
35
+ "k_proj",
36
+ "v_proj",
37
+ "q_proj",
38
+ "up_proj"
39
+ ],
40
+ "target_parameters": null,
41
+ "task_type": "CAUSAL_LM",
42
+ "trainable_token_indices": null,
43
+ "use_dora": false,
44
+ "use_qalora": false,
45
+ "use_rslora": false
46
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81f39eea5aa0a63c99773c947d6eb69750f37f80ab1d8cdb92227e28c54bf1cf
3
+ size 264308896
added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
chat_template.jinja ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if message.content is string %}
27
+ {%- set content = message.content %}
28
+ {%- else %}
29
+ {%- set content = '' %}
30
+ {%- endif %}
31
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
32
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
33
+ {%- elif message.role == "assistant" %}
34
+ {%- set reasoning_content = '' %}
35
+ {%- if message.reasoning_content is string %}
36
+ {%- set reasoning_content = message.reasoning_content %}
37
+ {%- else %}
38
+ {%- if '</think>' in content %}
39
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
40
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
41
+ {%- endif %}
42
+ {%- endif %}
43
+ {%- if loop.index0 > ns.last_query_index %}
44
+ {%- if loop.last or (not loop.last and reasoning_content) %}
45
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
46
+ {%- else %}
47
+ {{- '<|im_start|>' + message.role + '\n' + content }}
48
+ {%- endif %}
49
+ {%- else %}
50
+ {{- '<|im_start|>' + message.role + '\n' + content }}
51
+ {%- endif %}
52
+ {%- if message.tool_calls %}
53
+ {%- for tool_call in message.tool_calls %}
54
+ {%- if (loop.first and content) or (not loop.first) %}
55
+ {{- '\n' }}
56
+ {%- endif %}
57
+ {%- if tool_call.function %}
58
+ {%- set tool_call = tool_call.function %}
59
+ {%- endif %}
60
+ {{- '<tool_call>\n{"name": "' }}
61
+ {{- tool_call.name }}
62
+ {{- '", "arguments": ' }}
63
+ {%- if tool_call.arguments is string %}
64
+ {{- tool_call.arguments }}
65
+ {%- else %}
66
+ {{- tool_call.arguments | tojson }}
67
+ {%- endif %}
68
+ {{- '}\n</tool_call>' }}
69
+ {%- endfor %}
70
+ {%- endif %}
71
+ {{- '<|im_end|>\n' }}
72
+ {%- elif message.role == "tool" %}
73
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
74
+ {{- '<|im_start|>user' }}
75
+ {%- endif %}
76
+ {{- '\n<tool_response>\n' }}
77
+ {{- content }}
78
+ {{- '\n</tool_response>' }}
79
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
80
+ {{- '<|im_end|>\n' }}
81
+ {%- endif %}
82
+ {%- endif %}
83
+ {%- endfor %}
84
+ {%- if add_generation_prompt %}
85
+ {{- '<|im_start|>assistant\n' }}
86
+ {%- if enable_thinking is defined and enable_thinking is false %}
87
+ {{- '<think>\n\n</think>\n\n' }}
88
+ {%- endif %}
89
+ {%- endif %}
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd559026c1ce4f03948c802024c22c3309d815a29e8808fcfdbc29ce0e9ff453
3
+ size 528908363
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f84d59a8df474e6bbfdf0347edf7a554e20dae5959ffbc14c766c46b36d4662a
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e80fc673aeb69b7ff7db3152c771456fc30542bbca30a6e0d62a8eb8e8dc1e01
3
+ size 1465
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ },
181
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|im_end|>",
232
+ "errors": "replace",
233
+ "extra_special_tokens": {},
234
+ "model_max_length": 131072,
235
+ "pad_token": "<|endoftext|>",
236
+ "split_special_tokens": false,
237
+ "tokenizer_class": "Qwen2Tokenizer",
238
+ "unk_token": null
239
+ }
trainer_state.json ADDED
@@ -0,0 +1,2226 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2170,
3
+ "best_metric": 0.7015999555587769,
4
+ "best_model_checkpoint": "./qwen-ocaml-sft/checkpoint-2170",
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2170,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "entropy": 0.6813220992684365,
14
+ "epoch": 0.009216589861751152,
15
+ "grad_norm": 3.2051875591278076,
16
+ "learning_rate": 2.7272727272727273e-05,
17
+ "loss": 3.3755,
18
+ "mean_token_accuracy": 0.6017551895231008,
19
+ "num_tokens": 28960.0,
20
+ "step": 10
21
+ },
22
+ {
23
+ "entropy": 1.2283372402191162,
24
+ "epoch": 0.018433179723502304,
25
+ "grad_norm": 1.4653997421264648,
26
+ "learning_rate": 5.757575757575758e-05,
27
+ "loss": 2.155,
28
+ "mean_token_accuracy": 0.6499206535518169,
29
+ "num_tokens": 56242.0,
30
+ "step": 20
31
+ },
32
+ {
33
+ "entropy": 1.4246523216366769,
34
+ "epoch": 0.027649769585253458,
35
+ "grad_norm": 0.817514181137085,
36
+ "learning_rate": 8.787878787878789e-05,
37
+ "loss": 1.246,
38
+ "mean_token_accuracy": 0.7391218021512032,
39
+ "num_tokens": 81828.0,
40
+ "step": 30
41
+ },
42
+ {
43
+ "entropy": 0.8476247988641262,
44
+ "epoch": 0.03686635944700461,
45
+ "grad_norm": 0.5663232803344727,
46
+ "learning_rate": 0.0001181818181818182,
47
+ "loss": 0.8677,
48
+ "mean_token_accuracy": 0.8135000921785831,
49
+ "num_tokens": 108123.0,
50
+ "step": 40
51
+ },
52
+ {
53
+ "entropy": 0.8444311738014221,
54
+ "epoch": 0.04608294930875576,
55
+ "grad_norm": 0.34002283215522766,
56
+ "learning_rate": 0.00014848484848484849,
57
+ "loss": 0.8244,
58
+ "mean_token_accuracy": 0.809417700022459,
59
+ "num_tokens": 135208.0,
60
+ "step": 50
61
+ },
62
+ {
63
+ "entropy": 0.8123929310590029,
64
+ "epoch": 0.055299539170506916,
65
+ "grad_norm": 0.35241490602493286,
66
+ "learning_rate": 0.0001787878787878788,
67
+ "loss": 0.8167,
68
+ "mean_token_accuracy": 0.8164027214050293,
69
+ "num_tokens": 162440.0,
70
+ "step": 60
71
+ },
72
+ {
73
+ "entropy": 0.7658756393939257,
74
+ "epoch": 0.06451612903225806,
75
+ "grad_norm": 0.3681650459766388,
76
+ "learning_rate": 0.00019999899672483473,
77
+ "loss": 0.7657,
78
+ "mean_token_accuracy": 0.8210422351956368,
79
+ "num_tokens": 189807.0,
80
+ "step": 70
81
+ },
82
+ {
83
+ "entropy": 0.7832401964813471,
84
+ "epoch": 0.07373271889400922,
85
+ "grad_norm": 0.3235045373439789,
86
+ "learning_rate": 0.00019998116128192004,
87
+ "loss": 0.7576,
88
+ "mean_token_accuracy": 0.8224803328514099,
89
+ "num_tokens": 216034.0,
90
+ "step": 80
91
+ },
92
+ {
93
+ "entropy": 0.7970531325787306,
94
+ "epoch": 0.08294930875576037,
95
+ "grad_norm": 0.3650619685649872,
96
+ "learning_rate": 0.0001999410354122953,
97
+ "loss": 0.7725,
98
+ "mean_token_accuracy": 0.8188429720699787,
99
+ "num_tokens": 244444.0,
100
+ "step": 90
101
+ },
102
+ {
103
+ "entropy": 0.8288803774863481,
104
+ "epoch": 0.09216589861751152,
105
+ "grad_norm": 0.35073766112327576,
106
+ "learning_rate": 0.00019987862806187336,
107
+ "loss": 0.8067,
108
+ "mean_token_accuracy": 0.8131011754274369,
109
+ "num_tokens": 272396.0,
110
+ "step": 100
111
+ },
112
+ {
113
+ "entropy": 0.7561512392014265,
114
+ "epoch": 0.10138248847926268,
115
+ "grad_norm": 0.33590376377105713,
116
+ "learning_rate": 0.00019979395314414003,
117
+ "loss": 0.7478,
118
+ "mean_token_accuracy": 0.8274184472858905,
119
+ "num_tokens": 298632.0,
120
+ "step": 110
121
+ },
122
+ {
123
+ "entropy": 0.7232621975243092,
124
+ "epoch": 0.11059907834101383,
125
+ "grad_norm": 0.3256271779537201,
126
+ "learning_rate": 0.00019968702953705226,
127
+ "loss": 0.7314,
128
+ "mean_token_accuracy": 0.8329285934567452,
129
+ "num_tokens": 326029.0,
130
+ "step": 120
131
+ },
132
+ {
133
+ "entropy": 0.7907478779554367,
134
+ "epoch": 0.11981566820276497,
135
+ "grad_norm": 0.33607712388038635,
136
+ "learning_rate": 0.000199557881078829,
137
+ "loss": 0.7676,
138
+ "mean_token_accuracy": 0.8183201469480992,
139
+ "num_tokens": 351003.0,
140
+ "step": 130
141
+ },
142
+ {
143
+ "entropy": 0.7906187400221825,
144
+ "epoch": 0.12903225806451613,
145
+ "grad_norm": 0.2779998183250427,
146
+ "learning_rate": 0.00019940653656263705,
147
+ "loss": 0.7742,
148
+ "mean_token_accuracy": 0.8192058347165585,
149
+ "num_tokens": 379065.0,
150
+ "step": 140
151
+ },
152
+ {
153
+ "entropy": 0.7664360474795103,
154
+ "epoch": 0.1382488479262673,
155
+ "grad_norm": 0.3371148407459259,
156
+ "learning_rate": 0.00019923302973017128,
157
+ "loss": 0.749,
158
+ "mean_token_accuracy": 0.8246398828923702,
159
+ "num_tokens": 406059.0,
160
+ "step": 150
161
+ },
162
+ {
163
+ "entropy": 0.79318634159863,
164
+ "epoch": 0.14746543778801843,
165
+ "grad_norm": 0.35788455605506897,
166
+ "learning_rate": 0.00019903739926413244,
167
+ "loss": 0.7716,
168
+ "mean_token_accuracy": 0.8203366309404373,
169
+ "num_tokens": 432930.0,
170
+ "step": 160
171
+ },
172
+ {
173
+ "entropy": 0.7674953136593103,
174
+ "epoch": 0.15668202764976957,
175
+ "grad_norm": 0.3379119336605072,
176
+ "learning_rate": 0.00019881968877960262,
177
+ "loss": 0.7447,
178
+ "mean_token_accuracy": 0.8250967852771283,
179
+ "num_tokens": 459227.0,
180
+ "step": 170
181
+ },
182
+ {
183
+ "entropy": 0.7442232768982648,
184
+ "epoch": 0.16589861751152074,
185
+ "grad_norm": 0.2733379304409027,
186
+ "learning_rate": 0.00019857994681432174,
187
+ "loss": 0.7193,
188
+ "mean_token_accuracy": 0.8280802048742771,
189
+ "num_tokens": 488100.0,
190
+ "step": 180
191
+ },
192
+ {
193
+ "entropy": 0.8061734095215798,
194
+ "epoch": 0.17511520737327188,
195
+ "grad_norm": 0.39817723631858826,
196
+ "learning_rate": 0.0001983182268178661,
197
+ "loss": 0.7938,
198
+ "mean_token_accuracy": 0.8147257067263126,
199
+ "num_tokens": 514529.0,
200
+ "step": 190
201
+ },
202
+ {
203
+ "entropy": 0.730117141827941,
204
+ "epoch": 0.18433179723502305,
205
+ "grad_norm": 0.28748640418052673,
206
+ "learning_rate": 0.00019803458713973194,
207
+ "loss": 0.6923,
208
+ "mean_token_accuracy": 0.8354114755988121,
209
+ "num_tokens": 541967.0,
210
+ "step": 200
211
+ },
212
+ {
213
+ "entropy": 0.7647835846990347,
214
+ "epoch": 0.1935483870967742,
215
+ "grad_norm": 0.3064899146556854,
216
+ "learning_rate": 0.00019772909101632683,
217
+ "loss": 0.7855,
218
+ "mean_token_accuracy": 0.8232583694159985,
219
+ "num_tokens": 567869.0,
220
+ "step": 210
221
+ },
222
+ {
223
+ "entropy": 0.7511183556169272,
224
+ "epoch": 0.20276497695852536,
225
+ "grad_norm": 0.284687876701355,
226
+ "learning_rate": 0.00019740180655687114,
227
+ "loss": 0.7365,
228
+ "mean_token_accuracy": 0.8268198311328888,
229
+ "num_tokens": 594801.0,
230
+ "step": 220
231
+ },
232
+ {
233
+ "entropy": 0.7312648665159941,
234
+ "epoch": 0.2119815668202765,
235
+ "grad_norm": 0.2912767827510834,
236
+ "learning_rate": 0.00019705280672821342,
237
+ "loss": 0.7149,
238
+ "mean_token_accuracy": 0.8331556461751461,
239
+ "num_tokens": 623065.0,
240
+ "step": 230
241
+ },
242
+ {
243
+ "entropy": 0.7296199429780245,
244
+ "epoch": 0.22119815668202766,
245
+ "grad_norm": 0.2914985120296478,
246
+ "learning_rate": 0.0001966821693385628,
247
+ "loss": 0.7133,
248
+ "mean_token_accuracy": 0.8297220386564732,
249
+ "num_tokens": 651286.0,
250
+ "step": 240
251
+ },
252
+ {
253
+ "entropy": 0.7230386165902019,
254
+ "epoch": 0.2304147465437788,
255
+ "grad_norm": 0.2879176437854767,
256
+ "learning_rate": 0.0001962899770201418,
257
+ "loss": 0.7015,
258
+ "mean_token_accuracy": 0.8328824043273926,
259
+ "num_tokens": 679328.0,
260
+ "step": 250
261
+ },
262
+ {
263
+ "entropy": 0.7616100814193487,
264
+ "epoch": 0.23963133640552994,
265
+ "grad_norm": 0.2995160222053528,
266
+ "learning_rate": 0.00019587631721076394,
267
+ "loss": 0.7468,
268
+ "mean_token_accuracy": 0.8225546918809414,
269
+ "num_tokens": 706510.0,
270
+ "step": 260
271
+ },
272
+ {
273
+ "entropy": 0.7302906958386302,
274
+ "epoch": 0.2488479262672811,
275
+ "grad_norm": 0.32238849997520447,
276
+ "learning_rate": 0.0001954412821343396,
277
+ "loss": 0.695,
278
+ "mean_token_accuracy": 0.8321026556193829,
279
+ "num_tokens": 735903.0,
280
+ "step": 270
281
+ },
282
+ {
283
+ "entropy": 0.7462644644081593,
284
+ "epoch": 0.25806451612903225,
285
+ "grad_norm": 0.292940229177475,
286
+ "learning_rate": 0.00019498496878031526,
287
+ "loss": 0.738,
288
+ "mean_token_accuracy": 0.823956660181284,
289
+ "num_tokens": 763795.0,
290
+ "step": 280
291
+ },
292
+ {
293
+ "entropy": 0.7547617245465517,
294
+ "epoch": 0.2672811059907834,
295
+ "grad_norm": 0.31900840997695923,
296
+ "learning_rate": 0.0001945074788820499,
297
+ "loss": 0.7015,
298
+ "mean_token_accuracy": 0.8319075770676136,
299
+ "num_tokens": 793256.0,
300
+ "step": 290
301
+ },
302
+ {
303
+ "entropy": 0.7219028085470199,
304
+ "epoch": 0.2764976958525346,
305
+ "grad_norm": 0.21311832964420319,
306
+ "learning_rate": 0.00019400891889413409,
307
+ "loss": 0.6913,
308
+ "mean_token_accuracy": 0.8360162504017353,
309
+ "num_tokens": 822820.0,
310
+ "step": 300
311
+ },
312
+ {
313
+ "entropy": 0.7612819388508797,
314
+ "epoch": 0.2857142857142857,
315
+ "grad_norm": 0.24163030087947845,
316
+ "learning_rate": 0.00019348939996865604,
317
+ "loss": 0.7312,
318
+ "mean_token_accuracy": 0.8284988068044186,
319
+ "num_tokens": 849246.0,
320
+ "step": 310
321
+ },
322
+ {
323
+ "entropy": 0.7190706975758075,
324
+ "epoch": 0.29493087557603687,
325
+ "grad_norm": 0.2876209318637848,
326
+ "learning_rate": 0.0001929490379304211,
327
+ "loss": 0.6825,
328
+ "mean_token_accuracy": 0.8362526267766952,
329
+ "num_tokens": 876769.0,
330
+ "step": 320
331
+ },
332
+ {
333
+ "entropy": 0.7381139978766441,
334
+ "epoch": 0.30414746543778803,
335
+ "grad_norm": 0.3026144504547119,
336
+ "learning_rate": 0.0001923879532511287,
337
+ "loss": 0.737,
338
+ "mean_token_accuracy": 0.8290980480611324,
339
+ "num_tokens": 903354.0,
340
+ "step": 330
341
+ },
342
+ {
343
+ "entropy": 0.7504136189818382,
344
+ "epoch": 0.31336405529953915,
345
+ "grad_norm": 0.30887624621391296,
346
+ "learning_rate": 0.00019180627102251366,
347
+ "loss": 0.7179,
348
+ "mean_token_accuracy": 0.8358229048550129,
349
+ "num_tokens": 929002.0,
350
+ "step": 340
351
+ },
352
+ {
353
+ "entropy": 0.7561500292271376,
354
+ "epoch": 0.3225806451612903,
355
+ "grad_norm": 0.26517966389656067,
356
+ "learning_rate": 0.00019120412092845762,
357
+ "loss": 0.7543,
358
+ "mean_token_accuracy": 0.8260673426091671,
359
+ "num_tokens": 956160.0,
360
+ "step": 350
361
+ },
362
+ {
363
+ "entropy": 0.718599527142942,
364
+ "epoch": 0.3317972350230415,
365
+ "grad_norm": 0.3065567910671234,
366
+ "learning_rate": 0.0001905816372160765,
367
+ "loss": 0.6811,
368
+ "mean_token_accuracy": 0.8358904510736466,
369
+ "num_tokens": 983958.0,
370
+ "step": 360
371
+ },
372
+ {
373
+ "entropy": 0.7102599702775478,
374
+ "epoch": 0.34101382488479265,
375
+ "grad_norm": 0.3542201519012451,
376
+ "learning_rate": 0.0001899389586657904,
377
+ "loss": 0.706,
378
+ "mean_token_accuracy": 0.8331074707210064,
379
+ "num_tokens": 1011403.0,
380
+ "step": 370
381
+ },
382
+ {
383
+ "entropy": 0.723623962327838,
384
+ "epoch": 0.35023041474654376,
385
+ "grad_norm": 0.276357501745224,
386
+ "learning_rate": 0.00018927622856038313,
387
+ "loss": 0.6863,
388
+ "mean_token_accuracy": 0.8354352965950966,
389
+ "num_tokens": 1038520.0,
390
+ "step": 380
391
+ },
392
+ {
393
+ "entropy": 0.7096922701224685,
394
+ "epoch": 0.35944700460829493,
395
+ "grad_norm": 0.23507949709892273,
396
+ "learning_rate": 0.00018859359465305802,
397
+ "loss": 0.6646,
398
+ "mean_token_accuracy": 0.8400698743760586,
399
+ "num_tokens": 1068912.0,
400
+ "step": 390
401
+ },
402
+ {
403
+ "entropy": 0.7881791427731514,
404
+ "epoch": 0.3686635944700461,
405
+ "grad_norm": 0.30573952198028564,
406
+ "learning_rate": 0.0001878912091344966,
407
+ "loss": 0.7777,
408
+ "mean_token_accuracy": 0.8242236070334912,
409
+ "num_tokens": 1097236.0,
410
+ "step": 400
411
+ },
412
+ {
413
+ "entropy": 0.6701938826590776,
414
+ "epoch": 0.3778801843317972,
415
+ "grad_norm": 0.2812069058418274,
416
+ "learning_rate": 0.0001871692285989285,
417
+ "loss": 0.6481,
418
+ "mean_token_accuracy": 0.8438641004264354,
419
+ "num_tokens": 1124008.0,
420
+ "step": 410
421
+ },
422
+ {
423
+ "entropy": 0.781910614669323,
424
+ "epoch": 0.3870967741935484,
425
+ "grad_norm": 0.3356766998767853,
426
+ "learning_rate": 0.00018642781400921913,
427
+ "loss": 0.7539,
428
+ "mean_token_accuracy": 0.8259334497153759,
429
+ "num_tokens": 1151747.0,
430
+ "step": 420
431
+ },
432
+ {
433
+ "entropy": 0.7922669878229499,
434
+ "epoch": 0.39631336405529954,
435
+ "grad_norm": 0.2958647310733795,
436
+ "learning_rate": 0.0001856671306609839,
437
+ "loss": 0.7501,
438
+ "mean_token_accuracy": 0.8176686450839042,
439
+ "num_tokens": 1179637.0,
440
+ "step": 430
441
+ },
442
+ {
443
+ "entropy": 0.7031485013663769,
444
+ "epoch": 0.4055299539170507,
445
+ "grad_norm": 0.34288182854652405,
446
+ "learning_rate": 0.0001848873481457358,
447
+ "loss": 0.6797,
448
+ "mean_token_accuracy": 0.8355260133743286,
449
+ "num_tokens": 1206743.0,
450
+ "step": 440
451
+ },
452
+ {
453
+ "entropy": 0.7472133696079254,
454
+ "epoch": 0.4147465437788018,
455
+ "grad_norm": 0.3254289925098419,
456
+ "learning_rate": 0.00018408864031307603,
457
+ "loss": 0.714,
458
+ "mean_token_accuracy": 0.8313152596354485,
459
+ "num_tokens": 1235532.0,
460
+ "step": 450
461
+ },
462
+ {
463
+ "entropy": 0.7830955628305674,
464
+ "epoch": 0.423963133640553,
465
+ "grad_norm": 0.27417755126953125,
466
+ "learning_rate": 0.00018327118523193446,
467
+ "loss": 0.7618,
468
+ "mean_token_accuracy": 0.8231057062745094,
469
+ "num_tokens": 1260898.0,
470
+ "step": 460
471
+ },
472
+ {
473
+ "entropy": 0.7908357389271259,
474
+ "epoch": 0.43317972350230416,
475
+ "grad_norm": 0.31729546189308167,
476
+ "learning_rate": 0.00018243516515087022,
477
+ "loss": 0.7612,
478
+ "mean_token_accuracy": 0.8260951161384582,
479
+ "num_tokens": 1285377.0,
480
+ "step": 470
481
+ },
482
+ {
483
+ "entropy": 0.7501760717481375,
484
+ "epoch": 0.4423963133640553,
485
+ "grad_norm": 0.301329106092453,
486
+ "learning_rate": 0.00018158076645743973,
487
+ "loss": 0.7349,
488
+ "mean_token_accuracy": 0.8240762166678905,
489
+ "num_tokens": 1310688.0,
490
+ "step": 480
491
+ },
492
+ {
493
+ "entropy": 0.7087145041674375,
494
+ "epoch": 0.45161290322580644,
495
+ "grad_norm": 0.3549270033836365,
496
+ "learning_rate": 0.00018070817963664252,
497
+ "loss": 0.7009,
498
+ "mean_token_accuracy": 0.8339350454509258,
499
+ "num_tokens": 1336965.0,
500
+ "step": 490
501
+ },
502
+ {
503
+ "entropy": 0.7400309775024653,
504
+ "epoch": 0.4608294930875576,
505
+ "grad_norm": 0.362632691860199,
506
+ "learning_rate": 0.0001798175992284532,
507
+ "loss": 0.7154,
508
+ "mean_token_accuracy": 0.8314796030521393,
509
+ "num_tokens": 1363400.0,
510
+ "step": 500
511
+ },
512
+ {
513
+ "entropy": 0.6961963389068841,
514
+ "epoch": 0.4700460829493088,
515
+ "grad_norm": 0.31901466846466064,
516
+ "learning_rate": 0.00017890922378444949,
517
+ "loss": 0.669,
518
+ "mean_token_accuracy": 0.8373238749802112,
519
+ "num_tokens": 1389113.0,
520
+ "step": 510
521
+ },
522
+ {
523
+ "entropy": 0.7375663548707962,
524
+ "epoch": 0.4792626728110599,
525
+ "grad_norm": 0.27372968196868896,
526
+ "learning_rate": 0.00017798325582354602,
527
+ "loss": 0.7192,
528
+ "mean_token_accuracy": 0.8323125831782818,
529
+ "num_tokens": 1415275.0,
530
+ "step": 520
531
+ },
532
+ {
533
+ "entropy": 0.705496003292501,
534
+ "epoch": 0.48847926267281105,
535
+ "grad_norm": 0.3057447671890259,
536
+ "learning_rate": 0.00017703990178684326,
537
+ "loss": 0.699,
538
+ "mean_token_accuracy": 0.8369313515722752,
539
+ "num_tokens": 1443315.0,
540
+ "step": 530
541
+ },
542
+ {
543
+ "entropy": 0.7515531703829765,
544
+ "epoch": 0.4976958525345622,
545
+ "grad_norm": 0.32823631167411804,
546
+ "learning_rate": 0.0001760793719916025,
547
+ "loss": 0.725,
548
+ "mean_token_accuracy": 0.8294148877263069,
549
+ "num_tokens": 1470203.0,
550
+ "step": 540
551
+ },
552
+ {
553
+ "entropy": 0.7450070679187775,
554
+ "epoch": 0.5069124423963134,
555
+ "grad_norm": 0.2847367823123932,
556
+ "learning_rate": 0.00017510188058435618,
557
+ "loss": 0.7266,
558
+ "mean_token_accuracy": 0.8312779307365418,
559
+ "num_tokens": 1499344.0,
560
+ "step": 550
561
+ },
562
+ {
563
+ "entropy": 0.7544645313173532,
564
+ "epoch": 0.5161290322580645,
565
+ "grad_norm": 0.29744410514831543,
566
+ "learning_rate": 0.00017410764549316497,
567
+ "loss": 0.7335,
568
+ "mean_token_accuracy": 0.8273285463452339,
569
+ "num_tokens": 1524469.0,
570
+ "step": 560
571
+ },
572
+ {
573
+ "entropy": 0.7390379134565592,
574
+ "epoch": 0.5253456221198156,
575
+ "grad_norm": 0.28133663535118103,
576
+ "learning_rate": 0.00017309688837903125,
577
+ "loss": 0.7386,
578
+ "mean_token_accuracy": 0.8234638713300229,
579
+ "num_tokens": 1551793.0,
580
+ "step": 570
581
+ },
582
+ {
583
+ "entropy": 0.7615715146064759,
584
+ "epoch": 0.5345622119815668,
585
+ "grad_norm": 0.31992441415786743,
586
+ "learning_rate": 0.00017206983458648077,
587
+ "loss": 0.7443,
588
+ "mean_token_accuracy": 0.8309870153665543,
589
+ "num_tokens": 1575394.0,
590
+ "step": 580
591
+ },
592
+ {
593
+ "entropy": 0.7331630699336529,
594
+ "epoch": 0.543778801843318,
595
+ "grad_norm": 0.27732887864112854,
596
+ "learning_rate": 0.00017102671309332287,
597
+ "loss": 0.7342,
598
+ "mean_token_accuracy": 0.8291289664804935,
599
+ "num_tokens": 1601952.0,
600
+ "step": 590
601
+ },
602
+ {
603
+ "entropy": 0.7456940717995166,
604
+ "epoch": 0.5529953917050692,
605
+ "grad_norm": 0.2734481394290924,
606
+ "learning_rate": 0.0001699677564596009,
607
+ "loss": 0.718,
608
+ "mean_token_accuracy": 0.82945137321949,
609
+ "num_tokens": 1627428.0,
610
+ "step": 600
611
+ },
612
+ {
613
+ "entropy": 0.7992878194898367,
614
+ "epoch": 0.5622119815668203,
615
+ "grad_norm": 0.2915154695510864,
616
+ "learning_rate": 0.00016889320077574356,
617
+ "loss": 0.8216,
618
+ "mean_token_accuracy": 0.813803830742836,
619
+ "num_tokens": 1654199.0,
620
+ "step": 610
621
+ },
622
+ {
623
+ "entropy": 0.7602401621639728,
624
+ "epoch": 0.5714285714285714,
625
+ "grad_norm": 0.26141229271888733,
626
+ "learning_rate": 0.0001678032856099296,
627
+ "loss": 0.719,
628
+ "mean_token_accuracy": 0.8310174874961376,
629
+ "num_tokens": 1681115.0,
630
+ "step": 620
631
+ },
632
+ {
633
+ "entropy": 0.7364817965775728,
634
+ "epoch": 0.5806451612903226,
635
+ "grad_norm": 0.3432636559009552,
636
+ "learning_rate": 0.0001666982539546769,
637
+ "loss": 0.7112,
638
+ "mean_token_accuracy": 0.8290246821939945,
639
+ "num_tokens": 1708734.0,
640
+ "step": 630
641
+ },
642
+ {
643
+ "entropy": 0.7116030134260655,
644
+ "epoch": 0.5898617511520737,
645
+ "grad_norm": 0.3160955011844635,
646
+ "learning_rate": 0.00016557835217266833,
647
+ "loss": 0.712,
648
+ "mean_token_accuracy": 0.8317363314330578,
649
+ "num_tokens": 1735137.0,
650
+ "step": 640
651
+ },
652
+ {
653
+ "entropy": 0.6948481441475451,
654
+ "epoch": 0.5990783410138248,
655
+ "grad_norm": 0.3272562026977539,
656
+ "learning_rate": 0.0001644438299418259,
657
+ "loss": 0.6515,
658
+ "mean_token_accuracy": 0.8416998535394669,
659
+ "num_tokens": 1763024.0,
660
+ "step": 650
661
+ },
662
+ {
663
+ "entropy": 0.7368293164297939,
664
+ "epoch": 0.6082949308755761,
665
+ "grad_norm": 0.3517436385154724,
666
+ "learning_rate": 0.00016329494019964616,
667
+ "loss": 0.7297,
668
+ "mean_token_accuracy": 0.8293934009969235,
669
+ "num_tokens": 1789240.0,
670
+ "step": 660
671
+ },
672
+ {
673
+ "entropy": 0.7496969614177942,
674
+ "epoch": 0.6175115207373272,
675
+ "grad_norm": 0.37232086062431335,
676
+ "learning_rate": 0.00016213193908680876,
677
+ "loss": 0.7259,
678
+ "mean_token_accuracy": 0.8272240906953812,
679
+ "num_tokens": 1817298.0,
680
+ "step": 670
681
+ },
682
+ {
683
+ "entropy": 0.7615457896143198,
684
+ "epoch": 0.6267281105990783,
685
+ "grad_norm": 0.3335306942462921,
686
+ "learning_rate": 0.00016095508589007062,
687
+ "loss": 0.7412,
688
+ "mean_token_accuracy": 0.8261737614870072,
689
+ "num_tokens": 1843687.0,
690
+ "step": 680
691
+ },
692
+ {
693
+ "entropy": 0.7249070111662149,
694
+ "epoch": 0.6359447004608295,
695
+ "grad_norm": 0.4050976037979126,
696
+ "learning_rate": 0.00015976464298445917,
697
+ "loss": 0.7164,
698
+ "mean_token_accuracy": 0.8301085084676743,
699
+ "num_tokens": 1873530.0,
700
+ "step": 690
701
+ },
702
+ {
703
+ "entropy": 0.7132208090275526,
704
+ "epoch": 0.6451612903225806,
705
+ "grad_norm": 0.25421249866485596,
706
+ "learning_rate": 0.0001585608757747767,
707
+ "loss": 0.6608,
708
+ "mean_token_accuracy": 0.8420857444405556,
709
+ "num_tokens": 1900690.0,
710
+ "step": 700
711
+ },
712
+ {
713
+ "entropy": 0.7460955917835236,
714
+ "epoch": 0.6543778801843319,
715
+ "grad_norm": 0.3034956753253937,
716
+ "learning_rate": 0.00015734405263642925,
717
+ "loss": 0.7343,
718
+ "mean_token_accuracy": 0.8272924281656742,
719
+ "num_tokens": 1927455.0,
720
+ "step": 710
721
+ },
722
+ {
723
+ "entropy": 0.7328020770102739,
724
+ "epoch": 0.663594470046083,
725
+ "grad_norm": 0.33788177371025085,
726
+ "learning_rate": 0.00015611444485559355,
727
+ "loss": 0.709,
728
+ "mean_token_accuracy": 0.83270103931427,
729
+ "num_tokens": 1953144.0,
730
+ "step": 720
731
+ },
732
+ {
733
+ "entropy": 0.7447477359324693,
734
+ "epoch": 0.6728110599078341,
735
+ "grad_norm": 0.3346586525440216,
736
+ "learning_rate": 0.00015487232656873466,
737
+ "loss": 0.7222,
738
+ "mean_token_accuracy": 0.8270196095108986,
739
+ "num_tokens": 1980596.0,
740
+ "step": 730
741
+ },
742
+ {
743
+ "entropy": 0.6749195521697402,
744
+ "epoch": 0.6820276497695853,
745
+ "grad_norm": 0.3452974259853363,
746
+ "learning_rate": 0.00015361797470148813,
747
+ "loss": 0.6646,
748
+ "mean_token_accuracy": 0.8394610390067101,
749
+ "num_tokens": 2008675.0,
750
+ "step": 740
751
+ },
752
+ {
753
+ "entropy": 0.7265764184296131,
754
+ "epoch": 0.6912442396313364,
755
+ "grad_norm": 0.2909655272960663,
756
+ "learning_rate": 0.00015235166890692066,
757
+ "loss": 0.7088,
758
+ "mean_token_accuracy": 0.8314410902559757,
759
+ "num_tokens": 2033771.0,
760
+ "step": 750
761
+ },
762
+ {
763
+ "entropy": 0.7315562825649977,
764
+ "epoch": 0.7004608294930875,
765
+ "grad_norm": 0.29648321866989136,
766
+ "learning_rate": 0.00015107369150318248,
767
+ "loss": 0.7098,
768
+ "mean_token_accuracy": 0.8290056876838208,
769
+ "num_tokens": 2062456.0,
770
+ "step": 760
771
+ },
772
+ {
773
+ "entropy": 0.7180851440876722,
774
+ "epoch": 0.7096774193548387,
775
+ "grad_norm": 0.3001059889793396,
776
+ "learning_rate": 0.00014978432741056546,
777
+ "loss": 0.7102,
778
+ "mean_token_accuracy": 0.8328296788036823,
779
+ "num_tokens": 2088056.0,
780
+ "step": 770
781
+ },
782
+ {
783
+ "entropy": 0.6930279418826103,
784
+ "epoch": 0.7188940092165899,
785
+ "grad_norm": 0.24208398163318634,
786
+ "learning_rate": 0.00014848386408798122,
787
+ "loss": 0.67,
788
+ "mean_token_accuracy": 0.8385166138410568,
789
+ "num_tokens": 2114543.0,
790
+ "step": 780
791
+ },
792
+ {
793
+ "entropy": 0.7212698049843311,
794
+ "epoch": 0.728110599078341,
795
+ "grad_norm": 0.29187795519828796,
796
+ "learning_rate": 0.0001471725914688734,
797
+ "loss": 0.7113,
798
+ "mean_token_accuracy": 0.830100017786026,
799
+ "num_tokens": 2141643.0,
800
+ "step": 790
801
+ },
802
+ {
803
+ "entropy": 0.7350654354318976,
804
+ "epoch": 0.7373271889400922,
805
+ "grad_norm": 0.25248026847839355,
806
+ "learning_rate": 0.0001458508018965778,
807
+ "loss": 0.7152,
808
+ "mean_token_accuracy": 0.8316237829625607,
809
+ "num_tokens": 2169695.0,
810
+ "step": 800
811
+ },
812
+ {
813
+ "entropy": 0.7286242093890906,
814
+ "epoch": 0.7465437788018433,
815
+ "grad_norm": 0.31015968322753906,
816
+ "learning_rate": 0.00014451879005914574,
817
+ "loss": 0.7034,
818
+ "mean_token_accuracy": 0.8364780396223068,
819
+ "num_tokens": 2196982.0,
820
+ "step": 810
821
+ },
822
+ {
823
+ "entropy": 0.7194942735135555,
824
+ "epoch": 0.7557603686635944,
825
+ "grad_norm": 0.32759010791778564,
826
+ "learning_rate": 0.00014317685292364443,
827
+ "loss": 0.7043,
828
+ "mean_token_accuracy": 0.8338726572692394,
829
+ "num_tokens": 2224474.0,
830
+ "step": 820
831
+ },
832
+ {
833
+ "entropy": 0.77793597728014,
834
+ "epoch": 0.7649769585253456,
835
+ "grad_norm": 0.2720436453819275,
836
+ "learning_rate": 0.00014182528966994926,
837
+ "loss": 0.7589,
838
+ "mean_token_accuracy": 0.8218803226947784,
839
+ "num_tokens": 2250739.0,
840
+ "step": 830
841
+ },
842
+ {
843
+ "entropy": 0.7076451467350126,
844
+ "epoch": 0.7741935483870968,
845
+ "grad_norm": 0.3766821324825287,
846
+ "learning_rate": 0.00014046440162404256,
847
+ "loss": 0.6806,
848
+ "mean_token_accuracy": 0.8397734023630619,
849
+ "num_tokens": 2277559.0,
850
+ "step": 840
851
+ },
852
+ {
853
+ "entropy": 0.6975365605205297,
854
+ "epoch": 0.783410138248848,
855
+ "grad_norm": 0.2798302471637726,
856
+ "learning_rate": 0.00013909449219083454,
857
+ "loss": 0.6818,
858
+ "mean_token_accuracy": 0.8393688075244427,
859
+ "num_tokens": 2305360.0,
860
+ "step": 850
861
+ },
862
+ {
863
+ "entropy": 0.6880889561027288,
864
+ "epoch": 0.7926267281105991,
865
+ "grad_norm": 0.40115004777908325,
866
+ "learning_rate": 0.0001377158667865198,
867
+ "loss": 0.67,
868
+ "mean_token_accuracy": 0.8419740565121174,
869
+ "num_tokens": 2332072.0,
870
+ "step": 860
871
+ },
872
+ {
873
+ "entropy": 0.7083487439900636,
874
+ "epoch": 0.8018433179723502,
875
+ "grad_norm": 0.33026403188705444,
876
+ "learning_rate": 0.0001363288327704863,
877
+ "loss": 0.7089,
878
+ "mean_token_accuracy": 0.8338705748319626,
879
+ "num_tokens": 2357608.0,
880
+ "step": 870
881
+ },
882
+ {
883
+ "entropy": 0.7598309967666864,
884
+ "epoch": 0.8110599078341014,
885
+ "grad_norm": 0.29955339431762695,
886
+ "learning_rate": 0.0001349336993767905,
887
+ "loss": 0.7444,
888
+ "mean_token_accuracy": 0.8298642814159394,
889
+ "num_tokens": 2383770.0,
890
+ "step": 880
891
+ },
892
+ {
893
+ "entropy": 0.6922020763158798,
894
+ "epoch": 0.8202764976958525,
895
+ "grad_norm": 0.28863799571990967,
896
+ "learning_rate": 0.00013353077764521504,
897
+ "loss": 0.6756,
898
+ "mean_token_accuracy": 0.8409150242805481,
899
+ "num_tokens": 2410579.0,
900
+ "step": 890
901
+ },
902
+ {
903
+ "entropy": 0.6869130529463291,
904
+ "epoch": 0.8294930875576036,
905
+ "grad_norm": 0.24954994022846222,
906
+ "learning_rate": 0.00013212038035192327,
907
+ "loss": 0.6487,
908
+ "mean_token_accuracy": 0.8432103365659713,
909
+ "num_tokens": 2436906.0,
910
+ "step": 900
911
+ },
912
+ {
913
+ "entropy": 0.7394245728850365,
914
+ "epoch": 0.8387096774193549,
915
+ "grad_norm": 0.3518418073654175,
916
+ "learning_rate": 0.00013070282193972717,
917
+ "loss": 0.7251,
918
+ "mean_token_accuracy": 0.8314510688185692,
919
+ "num_tokens": 2461086.0,
920
+ "step": 910
921
+ },
922
+ {
923
+ "entropy": 0.7510461054742337,
924
+ "epoch": 0.847926267281106,
925
+ "grad_norm": 0.2915824353694916,
926
+ "learning_rate": 0.00012927841844798346,
927
+ "loss": 0.7232,
928
+ "mean_token_accuracy": 0.8302491903305054,
929
+ "num_tokens": 2486522.0,
930
+ "step": 920
931
+ },
932
+ {
933
+ "entropy": 0.6820386406034231,
934
+ "epoch": 0.8571428571428571,
935
+ "grad_norm": 0.31238240003585815,
936
+ "learning_rate": 0.00012784748744213372,
937
+ "loss": 0.6716,
938
+ "mean_token_accuracy": 0.8376850992441177,
939
+ "num_tokens": 2515816.0,
940
+ "step": 930
941
+ },
942
+ {
943
+ "entropy": 0.721437955647707,
944
+ "epoch": 0.8663594470046083,
945
+ "grad_norm": 0.3243514597415924,
946
+ "learning_rate": 0.00012641034794290455,
947
+ "loss": 0.6695,
948
+ "mean_token_accuracy": 0.8371677421033382,
949
+ "num_tokens": 2543482.0,
950
+ "step": 940
951
+ },
952
+ {
953
+ "entropy": 0.6836452066898346,
954
+ "epoch": 0.8755760368663594,
955
+ "grad_norm": 0.27484971284866333,
956
+ "learning_rate": 0.00012496732035518292,
957
+ "loss": 0.6738,
958
+ "mean_token_accuracy": 0.8397900149226188,
959
+ "num_tokens": 2570037.0,
960
+ "step": 950
961
+ },
962
+ {
963
+ "entropy": 0.7361621217802167,
964
+ "epoch": 0.8847926267281107,
965
+ "grad_norm": 0.3037663400173187,
966
+ "learning_rate": 0.00012351872639658313,
967
+ "loss": 0.7052,
968
+ "mean_token_accuracy": 0.832957761734724,
969
+ "num_tokens": 2597157.0,
970
+ "step": 960
971
+ },
972
+ {
973
+ "entropy": 0.7077585969120264,
974
+ "epoch": 0.8940092165898618,
975
+ "grad_norm": 0.3264278173446655,
976
+ "learning_rate": 0.00012206488902572111,
977
+ "loss": 0.7059,
978
+ "mean_token_accuracy": 0.8371899336576462,
979
+ "num_tokens": 2625082.0,
980
+ "step": 970
981
+ },
982
+ {
983
+ "entropy": 0.7448218245059252,
984
+ "epoch": 0.9032258064516129,
985
+ "grad_norm": 0.2658604681491852,
986
+ "learning_rate": 0.000120606132370212,
987
+ "loss": 0.7165,
988
+ "mean_token_accuracy": 0.8308831945061683,
989
+ "num_tokens": 2653591.0,
990
+ "step": 980
991
+ },
992
+ {
993
+ "entropy": 0.7214519936591387,
994
+ "epoch": 0.9124423963133641,
995
+ "grad_norm": 0.273061066865921,
996
+ "learning_rate": 0.00011914278165440706,
997
+ "loss": 0.6913,
998
+ "mean_token_accuracy": 0.8331013038754463,
999
+ "num_tokens": 2681148.0,
1000
+ "step": 990
1001
+ },
1002
+ {
1003
+ "entropy": 0.7345532923936844,
1004
+ "epoch": 0.9216589861751152,
1005
+ "grad_norm": 0.3739384114742279,
1006
+ "learning_rate": 0.0001176751631268862,
1007
+ "loss": 0.7251,
1008
+ "mean_token_accuracy": 0.8301717408001423,
1009
+ "num_tokens": 2706874.0,
1010
+ "step": 1000
1011
+ },
1012
+ {
1013
+ "entropy": 0.6948579344898462,
1014
+ "epoch": 0.9308755760368663,
1015
+ "grad_norm": 0.2887601852416992,
1016
+ "learning_rate": 0.00011620360398772192,
1017
+ "loss": 0.69,
1018
+ "mean_token_accuracy": 0.834751196205616,
1019
+ "num_tokens": 2734030.0,
1020
+ "step": 1010
1021
+ },
1022
+ {
1023
+ "entropy": 0.7203499909490347,
1024
+ "epoch": 0.9400921658986175,
1025
+ "grad_norm": 0.31931382417678833,
1026
+ "learning_rate": 0.00011472843231553145,
1027
+ "loss": 0.7169,
1028
+ "mean_token_accuracy": 0.8334086969494819,
1029
+ "num_tokens": 2761616.0,
1030
+ "step": 1020
1031
+ },
1032
+ {
1033
+ "entropy": 0.6754552606493235,
1034
+ "epoch": 0.9493087557603687,
1035
+ "grad_norm": 0.32389092445373535,
1036
+ "learning_rate": 0.00011324997699433272,
1037
+ "loss": 0.6458,
1038
+ "mean_token_accuracy": 0.8474155105650425,
1039
+ "num_tokens": 2790760.0,
1040
+ "step": 1030
1041
+ },
1042
+ {
1043
+ "entropy": 0.7120772618800402,
1044
+ "epoch": 0.9585253456221198,
1045
+ "grad_norm": 0.24603354930877686,
1046
+ "learning_rate": 0.0001117685676402211,
1047
+ "loss": 0.7025,
1048
+ "mean_token_accuracy": 0.8360289856791496,
1049
+ "num_tokens": 2816382.0,
1050
+ "step": 1040
1051
+ },
1052
+ {
1053
+ "entropy": 0.7087072119116783,
1054
+ "epoch": 0.967741935483871,
1055
+ "grad_norm": 0.28773602843284607,
1056
+ "learning_rate": 0.00011028453452788241,
1057
+ "loss": 0.6794,
1058
+ "mean_token_accuracy": 0.8350553438067436,
1059
+ "num_tokens": 2843265.0,
1060
+ "step": 1050
1061
+ },
1062
+ {
1063
+ "entropy": 0.717174931243062,
1064
+ "epoch": 0.9769585253456221,
1065
+ "grad_norm": 0.37724316120147705,
1066
+ "learning_rate": 0.00010879820851695958,
1067
+ "loss": 0.7219,
1068
+ "mean_token_accuracy": 0.8334485366940498,
1069
+ "num_tokens": 2868077.0,
1070
+ "step": 1060
1071
+ },
1072
+ {
1073
+ "entropy": 0.7240816669538617,
1074
+ "epoch": 0.9861751152073732,
1075
+ "grad_norm": 0.3419313132762909,
1076
+ "learning_rate": 0.0001073099209782888,
1077
+ "loss": 0.6814,
1078
+ "mean_token_accuracy": 0.8350721500813961,
1079
+ "num_tokens": 2895044.0,
1080
+ "step": 1070
1081
+ },
1082
+ {
1083
+ "entropy": 0.7110250508412719,
1084
+ "epoch": 0.9953917050691244,
1085
+ "grad_norm": 0.2590339779853821,
1086
+ "learning_rate": 0.00010582000372002152,
1087
+ "loss": 0.6907,
1088
+ "mean_token_accuracy": 0.8350939892232419,
1089
+ "num_tokens": 2921165.0,
1090
+ "step": 1080
1091
+ },
1092
+ {
1093
+ "epoch": 1.0,
1094
+ "eval_entropy": 0.7250895947870296,
1095
+ "eval_loss": 0.7114262580871582,
1096
+ "eval_mean_token_accuracy": 0.8371727046329305,
1097
+ "eval_num_tokens": 2934806.0,
1098
+ "eval_runtime": 197.3082,
1099
+ "eval_samples_per_second": 10.998,
1100
+ "eval_steps_per_second": 10.998,
1101
+ "step": 1085
1102
+ },
1103
+ {
1104
+ "entropy": 0.7308962121605873,
1105
+ "epoch": 1.0046082949308757,
1106
+ "grad_norm": 0.2802174389362335,
1107
+ "learning_rate": 0.0001043287889136491,
1108
+ "loss": 0.668,
1109
+ "mean_token_accuracy": 0.8353404991328717,
1110
+ "num_tokens": 2947161.0,
1111
+ "step": 1090
1112
+ },
1113
+ {
1114
+ "entropy": 0.6342746868729592,
1115
+ "epoch": 1.0138248847926268,
1116
+ "grad_norm": 0.37129464745521545,
1117
+ "learning_rate": 0.00010283660901994632,
1118
+ "loss": 0.6233,
1119
+ "mean_token_accuracy": 0.8470003835856914,
1120
+ "num_tokens": 2974641.0,
1121
+ "step": 1100
1122
+ },
1123
+ {
1124
+ "entropy": 0.6460907947272062,
1125
+ "epoch": 1.023041474654378,
1126
+ "grad_norm": 0.3424113690853119,
1127
+ "learning_rate": 0.00010134379671485073,
1128
+ "loss": 0.6369,
1129
+ "mean_token_accuracy": 0.8495963260531425,
1130
+ "num_tokens": 3001379.0,
1131
+ "step": 1110
1132
+ },
1133
+ {
1134
+ "entropy": 0.6835248069837689,
1135
+ "epoch": 1.032258064516129,
1136
+ "grad_norm": 0.29605832695961,
1137
+ "learning_rate": 9.985068481529354e-05,
1138
+ "loss": 0.6327,
1139
+ "mean_token_accuracy": 0.8451249115169048,
1140
+ "num_tokens": 3027806.0,
1141
+ "step": 1120
1142
+ },
1143
+ {
1144
+ "entropy": 0.6169640816748142,
1145
+ "epoch": 1.0414746543778801,
1146
+ "grad_norm": 0.37180233001708984,
1147
+ "learning_rate": 9.83576062049994e-05,
1148
+ "loss": 0.5968,
1149
+ "mean_token_accuracy": 0.8531802453100681,
1150
+ "num_tokens": 3052892.0,
1151
+ "step": 1130
1152
+ },
1153
+ {
1154
+ "entropy": 0.6297633443027735,
1155
+ "epoch": 1.0506912442396312,
1156
+ "grad_norm": 0.3378913402557373,
1157
+ "learning_rate": 9.686489376027136e-05,
1158
+ "loss": 0.6206,
1159
+ "mean_token_accuracy": 0.851526065915823,
1160
+ "num_tokens": 3079924.0,
1161
+ "step": 1140
1162
+ },
1163
+ {
1164
+ "entropy": 0.6136038340628147,
1165
+ "epoch": 1.0599078341013826,
1166
+ "grad_norm": 0.3650684356689453,
1167
+ "learning_rate": 9.537288027577711e-05,
1168
+ "loss": 0.6093,
1169
+ "mean_token_accuracy": 0.8550195962190628,
1170
+ "num_tokens": 3105837.0,
1171
+ "step": 1150
1172
+ },
1173
+ {
1174
+ "entropy": 0.6491836536675691,
1175
+ "epoch": 1.0691244239631337,
1176
+ "grad_norm": 0.3491412401199341,
1177
+ "learning_rate": 9.388189839035379e-05,
1178
+ "loss": 0.6144,
1179
+ "mean_token_accuracy": 0.8473505489528179,
1180
+ "num_tokens": 3132962.0,
1181
+ "step": 1160
1182
+ },
1183
+ {
1184
+ "entropy": 0.6371682230383158,
1185
+ "epoch": 1.0783410138248848,
1186
+ "grad_norm": 0.38217079639434814,
1187
+ "learning_rate": 9.239228051284743e-05,
1188
+ "loss": 0.6358,
1189
+ "mean_token_accuracy": 0.8470985740423203,
1190
+ "num_tokens": 3159515.0,
1191
+ "step": 1170
1192
+ },
1193
+ {
1194
+ "entropy": 0.6246544945985079,
1195
+ "epoch": 1.087557603686636,
1196
+ "grad_norm": 0.3524271249771118,
1197
+ "learning_rate": 9.090435874800354e-05,
1198
+ "loss": 0.5815,
1199
+ "mean_token_accuracy": 0.8581700764596463,
1200
+ "num_tokens": 3186296.0,
1201
+ "step": 1180
1202
+ },
1203
+ {
1204
+ "entropy": 0.6239660510793328,
1205
+ "epoch": 1.096774193548387,
1206
+ "grad_norm": 0.38734525442123413,
1207
+ "learning_rate": 8.941846482242566e-05,
1208
+ "loss": 0.6149,
1209
+ "mean_token_accuracy": 0.8517151668667793,
1210
+ "num_tokens": 3213934.0,
1211
+ "step": 1190
1212
+ },
1213
+ {
1214
+ "entropy": 0.6256657449528575,
1215
+ "epoch": 1.1059907834101383,
1216
+ "grad_norm": 0.3149538040161133,
1217
+ "learning_rate": 8.793493001061815e-05,
1218
+ "loss": 0.5964,
1219
+ "mean_token_accuracy": 0.8548560328781605,
1220
+ "num_tokens": 3242142.0,
1221
+ "step": 1200
1222
+ },
1223
+ {
1224
+ "entropy": 0.5848878964781761,
1225
+ "epoch": 1.1152073732718895,
1226
+ "grad_norm": 0.30550169944763184,
1227
+ "learning_rate": 8.645408506112966e-05,
1228
+ "loss": 0.5425,
1229
+ "mean_token_accuracy": 0.8643457405269146,
1230
+ "num_tokens": 3270400.0,
1231
+ "step": 1210
1232
+ },
1233
+ {
1234
+ "entropy": 0.6510914264246821,
1235
+ "epoch": 1.1244239631336406,
1236
+ "grad_norm": 0.3443332016468048,
1237
+ "learning_rate": 8.497626012281427e-05,
1238
+ "loss": 0.6214,
1239
+ "mean_token_accuracy": 0.845578433573246,
1240
+ "num_tokens": 3298889.0,
1241
+ "step": 1220
1242
+ },
1243
+ {
1244
+ "entropy": 0.6255331162363291,
1245
+ "epoch": 1.1336405529953917,
1246
+ "grad_norm": 0.35833534598350525,
1247
+ "learning_rate": 8.350178467122565e-05,
1248
+ "loss": 0.5779,
1249
+ "mean_token_accuracy": 0.8540341056883335,
1250
+ "num_tokens": 3326093.0,
1251
+ "step": 1230
1252
+ },
1253
+ {
1254
+ "entropy": 0.6205343475565315,
1255
+ "epoch": 1.1428571428571428,
1256
+ "grad_norm": 0.3398672640323639,
1257
+ "learning_rate": 8.203098743516202e-05,
1258
+ "loss": 0.6122,
1259
+ "mean_token_accuracy": 0.8538564696907998,
1260
+ "num_tokens": 3353277.0,
1261
+ "step": 1240
1262
+ },
1263
+ {
1264
+ "entropy": 0.591589437611401,
1265
+ "epoch": 1.1520737327188941,
1266
+ "grad_norm": 0.29090237617492676,
1267
+ "learning_rate": 8.056419632337713e-05,
1268
+ "loss": 0.5647,
1269
+ "mean_token_accuracy": 0.8593504391610622,
1270
+ "num_tokens": 3381753.0,
1271
+ "step": 1250
1272
+ },
1273
+ {
1274
+ "entropy": 0.5821992674842477,
1275
+ "epoch": 1.1612903225806452,
1276
+ "grad_norm": 0.29793089628219604,
1277
+ "learning_rate": 7.910173835147428e-05,
1278
+ "loss": 0.5356,
1279
+ "mean_token_accuracy": 0.8655187331140042,
1280
+ "num_tokens": 3410051.0,
1281
+ "step": 1260
1282
+ },
1283
+ {
1284
+ "entropy": 0.5886814070865511,
1285
+ "epoch": 1.1705069124423964,
1286
+ "grad_norm": 0.3498776853084564,
1287
+ "learning_rate": 7.764393956899944e-05,
1288
+ "loss": 0.5544,
1289
+ "mean_token_accuracy": 0.8588730558753014,
1290
+ "num_tokens": 3438209.0,
1291
+ "step": 1270
1292
+ },
1293
+ {
1294
+ "entropy": 0.5947490762919188,
1295
+ "epoch": 1.1797235023041475,
1296
+ "grad_norm": 0.35999783873558044,
1297
+ "learning_rate": 7.619112498674969e-05,
1298
+ "loss": 0.5511,
1299
+ "mean_token_accuracy": 0.8604489989578724,
1300
+ "num_tokens": 3467191.0,
1301
+ "step": 1280
1302
+ },
1303
+ {
1304
+ "entropy": 0.6096777388826012,
1305
+ "epoch": 1.1889400921658986,
1306
+ "grad_norm": 0.4569447338581085,
1307
+ "learning_rate": 7.474361850431315e-05,
1308
+ "loss": 0.5803,
1309
+ "mean_token_accuracy": 0.8564766734838486,
1310
+ "num_tokens": 3495301.0,
1311
+ "step": 1290
1312
+ },
1313
+ {
1314
+ "entropy": 0.6138049149885774,
1315
+ "epoch": 1.1981566820276497,
1316
+ "grad_norm": 0.40922898054122925,
1317
+ "learning_rate": 7.330174283785711e-05,
1318
+ "loss": 0.5919,
1319
+ "mean_token_accuracy": 0.8560373924672604,
1320
+ "num_tokens": 3522210.0,
1321
+ "step": 1300
1322
+ },
1323
+ {
1324
+ "entropy": 0.5668233297765255,
1325
+ "epoch": 1.2073732718894008,
1326
+ "grad_norm": 0.3519320487976074,
1327
+ "learning_rate": 7.186581944817925e-05,
1328
+ "loss": 0.5589,
1329
+ "mean_token_accuracy": 0.863438468426466,
1330
+ "num_tokens": 3550910.0,
1331
+ "step": 1310
1332
+ },
1333
+ {
1334
+ "entropy": 0.6422557730227709,
1335
+ "epoch": 1.2165898617511521,
1336
+ "grad_norm": 0.40460196137428284,
1337
+ "learning_rate": 7.043616846903953e-05,
1338
+ "loss": 0.6295,
1339
+ "mean_token_accuracy": 0.844844151288271,
1340
+ "num_tokens": 3577947.0,
1341
+ "step": 1320
1342
+ },
1343
+ {
1344
+ "entropy": 0.6118434071540833,
1345
+ "epoch": 1.2258064516129032,
1346
+ "grad_norm": 0.3850909471511841,
1347
+ "learning_rate": 6.901310863578732e-05,
1348
+ "loss": 0.5792,
1349
+ "mean_token_accuracy": 0.8563656531274318,
1350
+ "num_tokens": 3605135.0,
1351
+ "step": 1330
1352
+ },
1353
+ {
1354
+ "entropy": 0.631221866607666,
1355
+ "epoch": 1.2350230414746544,
1356
+ "grad_norm": 0.30839139223098755,
1357
+ "learning_rate": 6.759695721430055e-05,
1358
+ "loss": 0.6155,
1359
+ "mean_token_accuracy": 0.8490869931876659,
1360
+ "num_tokens": 3633169.0,
1361
+ "step": 1340
1362
+ },
1363
+ {
1364
+ "entropy": 0.623043118789792,
1365
+ "epoch": 1.2442396313364055,
1366
+ "grad_norm": 0.37266048789024353,
1367
+ "learning_rate": 6.618802993025266e-05,
1368
+ "loss": 0.6443,
1369
+ "mean_token_accuracy": 0.8491488955914974,
1370
+ "num_tokens": 3660421.0,
1371
+ "step": 1350
1372
+ },
1373
+ {
1374
+ "entropy": 0.5951266692951321,
1375
+ "epoch": 1.2534562211981566,
1376
+ "grad_norm": 0.32506951689720154,
1377
+ "learning_rate": 6.478664089872249e-05,
1378
+ "loss": 0.5569,
1379
+ "mean_token_accuracy": 0.8649635158479214,
1380
+ "num_tokens": 3687798.0,
1381
+ "step": 1360
1382
+ },
1383
+ {
1384
+ "entropy": 0.6104681365191936,
1385
+ "epoch": 1.262672811059908,
1386
+ "grad_norm": 0.3316543698310852,
1387
+ "learning_rate": 6.339310255416378e-05,
1388
+ "loss": 0.5858,
1389
+ "mean_token_accuracy": 0.856657163053751,
1390
+ "num_tokens": 3715941.0,
1391
+ "step": 1370
1392
+ },
1393
+ {
1394
+ "entropy": 0.5855855537578464,
1395
+ "epoch": 1.271889400921659,
1396
+ "grad_norm": 0.4697554409503937,
1397
+ "learning_rate": 6.200772558074873e-05,
1398
+ "loss": 0.5614,
1399
+ "mean_token_accuracy": 0.8604506768286229,
1400
+ "num_tokens": 3741133.0,
1401
+ "step": 1380
1402
+ },
1403
+ {
1404
+ "entropy": 0.6417946748435497,
1405
+ "epoch": 1.2811059907834101,
1406
+ "grad_norm": 0.3982338011264801,
1407
+ "learning_rate": 6.06308188431024e-05,
1408
+ "loss": 0.623,
1409
+ "mean_token_accuracy": 0.8485914342105388,
1410
+ "num_tokens": 3766133.0,
1411
+ "step": 1390
1412
+ },
1413
+ {
1414
+ "entropy": 0.6297718342393637,
1415
+ "epoch": 1.2903225806451613,
1416
+ "grad_norm": 0.4881386458873749,
1417
+ "learning_rate": 5.926268931744243e-05,
1418
+ "loss": 0.6192,
1419
+ "mean_token_accuracy": 0.8497353821992875,
1420
+ "num_tokens": 3791970.0,
1421
+ "step": 1400
1422
+ },
1423
+ {
1424
+ "entropy": 0.5673586906865239,
1425
+ "epoch": 1.2995391705069124,
1426
+ "grad_norm": 0.4258589446544647,
1427
+ "learning_rate": 5.79036420231398e-05,
1428
+ "loss": 0.5419,
1429
+ "mean_token_accuracy": 0.8655278801918029,
1430
+ "num_tokens": 3818832.0,
1431
+ "step": 1410
1432
+ },
1433
+ {
1434
+ "entropy": 0.6246172869578004,
1435
+ "epoch": 1.3087557603686637,
1436
+ "grad_norm": 0.3208440840244293,
1437
+ "learning_rate": 5.655397995471579e-05,
1438
+ "loss": 0.5975,
1439
+ "mean_token_accuracy": 0.8529197834432125,
1440
+ "num_tokens": 3845334.0,
1441
+ "step": 1420
1442
+ },
1443
+ {
1444
+ "entropy": 0.6241569153964519,
1445
+ "epoch": 1.3179723502304148,
1446
+ "grad_norm": 0.35756826400756836,
1447
+ "learning_rate": 5.5214004014290755e-05,
1448
+ "loss": 0.6038,
1449
+ "mean_token_accuracy": 0.8531735584139823,
1450
+ "num_tokens": 3871208.0,
1451
+ "step": 1430
1452
+ },
1453
+ {
1454
+ "entropy": 0.5929729694500565,
1455
+ "epoch": 1.327188940092166,
1456
+ "grad_norm": 0.34781602025032043,
1457
+ "learning_rate": 5.3884012944498895e-05,
1458
+ "loss": 0.5709,
1459
+ "mean_token_accuracy": 0.8586051426827908,
1460
+ "num_tokens": 3899041.0,
1461
+ "step": 1440
1462
+ },
1463
+ {
1464
+ "entropy": 0.5982153214514255,
1465
+ "epoch": 1.336405529953917,
1466
+ "grad_norm": 0.42399492859840393,
1467
+ "learning_rate": 5.256430326188484e-05,
1468
+ "loss": 0.5724,
1469
+ "mean_token_accuracy": 0.8588254593312741,
1470
+ "num_tokens": 3924718.0,
1471
+ "step": 1450
1472
+ },
1473
+ {
1474
+ "entropy": 0.6157030487433076,
1475
+ "epoch": 1.3456221198156681,
1476
+ "grad_norm": 0.4058624804019928,
1477
+ "learning_rate": 5.125516919079625e-05,
1478
+ "loss": 0.597,
1479
+ "mean_token_accuracy": 0.8565631859004498,
1480
+ "num_tokens": 3950704.0,
1481
+ "step": 1460
1482
+ },
1483
+ {
1484
+ "entropy": 0.6145446136593818,
1485
+ "epoch": 1.3548387096774195,
1486
+ "grad_norm": 0.41059333086013794,
1487
+ "learning_rate": 4.995690259778762e-05,
1488
+ "loss": 0.5909,
1489
+ "mean_token_accuracy": 0.8548187188804149,
1490
+ "num_tokens": 3977937.0,
1491
+ "step": 1470
1492
+ },
1493
+ {
1494
+ "entropy": 0.6084423962980509,
1495
+ "epoch": 1.3640552995391704,
1496
+ "grad_norm": 0.3891129791736603,
1497
+ "learning_rate": 4.866979292654995e-05,
1498
+ "loss": 0.5937,
1499
+ "mean_token_accuracy": 0.8557461231946946,
1500
+ "num_tokens": 4006038.0,
1501
+ "step": 1480
1502
+ },
1503
+ {
1504
+ "entropy": 0.6311043184250593,
1505
+ "epoch": 1.3732718894009217,
1506
+ "grad_norm": 0.3631453514099121,
1507
+ "learning_rate": 4.7394127133379996e-05,
1508
+ "loss": 0.6024,
1509
+ "mean_token_accuracy": 0.851103562861681,
1510
+ "num_tokens": 4033093.0,
1511
+ "step": 1490
1512
+ },
1513
+ {
1514
+ "entropy": 0.57787942904979,
1515
+ "epoch": 1.3824884792626728,
1516
+ "grad_norm": 0.32106077671051025,
1517
+ "learning_rate": 4.6130189623204766e-05,
1518
+ "loss": 0.5412,
1519
+ "mean_token_accuracy": 0.8636514358222485,
1520
+ "num_tokens": 4060651.0,
1521
+ "step": 1500
1522
+ },
1523
+ {
1524
+ "entropy": 0.6655275866389274,
1525
+ "epoch": 1.391705069124424,
1526
+ "grad_norm": 0.41340014338493347,
1527
+ "learning_rate": 4.4878262186174226e-05,
1528
+ "loss": 0.6377,
1529
+ "mean_token_accuracy": 0.8466010347008706,
1530
+ "num_tokens": 4087640.0,
1531
+ "step": 1510
1532
+ },
1533
+ {
1534
+ "entropy": 0.6205959409475327,
1535
+ "epoch": 1.400921658986175,
1536
+ "grad_norm": 0.3026493191719055,
1537
+ "learning_rate": 4.3638623934837394e-05,
1538
+ "loss": 0.5781,
1539
+ "mean_token_accuracy": 0.856752559542656,
1540
+ "num_tokens": 4114791.0,
1541
+ "step": 1520
1542
+ },
1543
+ {
1544
+ "entropy": 0.5819996166974306,
1545
+ "epoch": 1.4101382488479262,
1546
+ "grad_norm": 0.37101536989212036,
1547
+ "learning_rate": 4.2411551241915305e-05,
1548
+ "loss": 0.5485,
1549
+ "mean_token_accuracy": 0.8607306607067585,
1550
+ "num_tokens": 4141094.0,
1551
+ "step": 1530
1552
+ },
1553
+ {
1554
+ "entropy": 0.6253220835700631,
1555
+ "epoch": 1.4193548387096775,
1556
+ "grad_norm": 0.3750545084476471,
1557
+ "learning_rate": 4.119731767868453e-05,
1558
+ "loss": 0.6086,
1559
+ "mean_token_accuracy": 0.8513987928628921,
1560
+ "num_tokens": 4166035.0,
1561
+ "step": 1540
1562
+ },
1563
+ {
1564
+ "entropy": 0.6176507495343685,
1565
+ "epoch": 1.4285714285714286,
1566
+ "grad_norm": 0.5628491044044495,
1567
+ "learning_rate": 3.9996193953985625e-05,
1568
+ "loss": 0.6086,
1569
+ "mean_token_accuracy": 0.8544401422142982,
1570
+ "num_tokens": 4191198.0,
1571
+ "step": 1550
1572
+ },
1573
+ {
1574
+ "entropy": 0.6267381351441145,
1575
+ "epoch": 1.4377880184331797,
1576
+ "grad_norm": 0.3852846622467041,
1577
+ "learning_rate": 3.8808447853869656e-05,
1578
+ "loss": 0.6011,
1579
+ "mean_token_accuracy": 0.8528468564152718,
1580
+ "num_tokens": 4218796.0,
1581
+ "step": 1560
1582
+ },
1583
+ {
1584
+ "entropy": 0.5719573536887765,
1585
+ "epoch": 1.4470046082949308,
1586
+ "grad_norm": 0.4029249846935272,
1587
+ "learning_rate": 3.76343441818961e-05,
1588
+ "loss": 0.539,
1589
+ "mean_token_accuracy": 0.8656464323401452,
1590
+ "num_tokens": 4246865.0,
1591
+ "step": 1570
1592
+ },
1593
+ {
1594
+ "entropy": 0.5876866010949016,
1595
+ "epoch": 1.456221198156682,
1596
+ "grad_norm": 0.44668325781822205,
1597
+ "learning_rate": 3.64741447000961e-05,
1598
+ "loss": 0.5472,
1599
+ "mean_token_accuracy": 0.8622240588068962,
1600
+ "num_tokens": 4272706.0,
1601
+ "step": 1580
1602
+ },
1603
+ {
1604
+ "entropy": 0.6029468834400177,
1605
+ "epoch": 1.4654377880184333,
1606
+ "grad_norm": 0.35470443964004517,
1607
+ "learning_rate": 3.532810807061351e-05,
1608
+ "loss": 0.5987,
1609
+ "mean_token_accuracy": 0.8555419981479645,
1610
+ "num_tokens": 4299873.0,
1611
+ "step": 1590
1612
+ },
1613
+ {
1614
+ "entropy": 0.5952557157725096,
1615
+ "epoch": 1.4746543778801844,
1616
+ "grad_norm": 0.34689727425575256,
1617
+ "learning_rate": 3.4196489798037025e-05,
1618
+ "loss": 0.5656,
1619
+ "mean_token_accuracy": 0.8606510788202286,
1620
+ "num_tokens": 4326365.0,
1621
+ "step": 1600
1622
+ },
1623
+ {
1624
+ "entropy": 0.5785195421427488,
1625
+ "epoch": 1.4838709677419355,
1626
+ "grad_norm": 0.40470728278160095,
1627
+ "learning_rate": 3.30795421724366e-05,
1628
+ "loss": 0.5674,
1629
+ "mean_token_accuracy": 0.8623944170773029,
1630
+ "num_tokens": 4352052.0,
1631
+ "step": 1610
1632
+ },
1633
+ {
1634
+ "entropy": 0.6490232923999428,
1635
+ "epoch": 1.4930875576036866,
1636
+ "grad_norm": 0.386251300573349,
1637
+ "learning_rate": 3.1977514213116087e-05,
1638
+ "loss": 0.6283,
1639
+ "mean_token_accuracy": 0.8493647269904614,
1640
+ "num_tokens": 4378378.0,
1641
+ "step": 1620
1642
+ },
1643
+ {
1644
+ "entropy": 0.6235011581331491,
1645
+ "epoch": 1.5023041474654377,
1646
+ "grad_norm": 0.3535836637020111,
1647
+ "learning_rate": 3.0890651613095564e-05,
1648
+ "loss": 0.5623,
1649
+ "mean_token_accuracy": 0.8560213819146156,
1650
+ "num_tokens": 4405029.0,
1651
+ "step": 1630
1652
+ },
1653
+ {
1654
+ "entropy": 0.6310895785689354,
1655
+ "epoch": 1.511520737327189,
1656
+ "grad_norm": 0.3408556580543518,
1657
+ "learning_rate": 2.981919668433477e-05,
1658
+ "loss": 0.5782,
1659
+ "mean_token_accuracy": 0.8549186430871487,
1660
+ "num_tokens": 4432095.0,
1661
+ "step": 1640
1662
+ },
1663
+ {
1664
+ "entropy": 0.5660297932103276,
1665
+ "epoch": 1.52073732718894,
1666
+ "grad_norm": 0.40132173895835876,
1667
+ "learning_rate": 2.8763388303710636e-05,
1668
+ "loss": 0.5407,
1669
+ "mean_token_accuracy": 0.8663985416293144,
1670
+ "num_tokens": 4458205.0,
1671
+ "step": 1650
1672
+ },
1673
+ {
1674
+ "entropy": 0.5760546216741205,
1675
+ "epoch": 1.5299539170506913,
1676
+ "grad_norm": 0.41320013999938965,
1677
+ "learning_rate": 2.772346185976069e-05,
1678
+ "loss": 0.5745,
1679
+ "mean_token_accuracy": 0.8599842116236687,
1680
+ "num_tokens": 4485047.0,
1681
+ "step": 1660
1682
+ },
1683
+ {
1684
+ "entropy": 0.5921478979289532,
1685
+ "epoch": 1.5391705069124424,
1686
+ "grad_norm": 0.3433243930339813,
1687
+ "learning_rate": 2.669964920020387e-05,
1688
+ "loss": 0.5583,
1689
+ "mean_token_accuracy": 0.861686623096466,
1690
+ "num_tokens": 4514056.0,
1691
+ "step": 1670
1692
+ },
1693
+ {
1694
+ "entropy": 0.5948457010090351,
1695
+ "epoch": 1.5483870967741935,
1696
+ "grad_norm": 0.3089885711669922,
1697
+ "learning_rate": 2.5692178580251014e-05,
1698
+ "loss": 0.5748,
1699
+ "mean_token_accuracy": 0.8604165323078632,
1700
+ "num_tokens": 4540939.0,
1701
+ "step": 1680
1702
+ },
1703
+ {
1704
+ "entropy": 0.627241513505578,
1705
+ "epoch": 1.5576036866359448,
1706
+ "grad_norm": 0.40662682056427,
1707
+ "learning_rate": 2.470127461171635e-05,
1708
+ "loss": 0.6055,
1709
+ "mean_token_accuracy": 0.8513796024024487,
1710
+ "num_tokens": 4568381.0,
1711
+ "step": 1690
1712
+ },
1713
+ {
1714
+ "entropy": 0.5909772358834744,
1715
+ "epoch": 1.5668202764976957,
1716
+ "grad_norm": 0.38541415333747864,
1717
+ "learning_rate": 2.3727158212940868e-05,
1718
+ "loss": 0.5525,
1719
+ "mean_token_accuracy": 0.8601111464202404,
1720
+ "num_tokens": 4593965.0,
1721
+ "step": 1700
1722
+ },
1723
+ {
1724
+ "entropy": 0.5652065623551608,
1725
+ "epoch": 1.576036866359447,
1726
+ "grad_norm": 0.4092041850090027,
1727
+ "learning_rate": 2.277004655953968e-05,
1728
+ "loss": 0.5236,
1729
+ "mean_token_accuracy": 0.8660901002585888,
1730
+ "num_tokens": 4621313.0,
1731
+ "step": 1710
1732
+ },
1733
+ {
1734
+ "entropy": 0.5846194272860885,
1735
+ "epoch": 1.5852534562211982,
1736
+ "grad_norm": 1.227542519569397,
1737
+ "learning_rate": 2.1830153035983235e-05,
1738
+ "loss": 0.5679,
1739
+ "mean_token_accuracy": 0.8617964021861553,
1740
+ "num_tokens": 4647684.0,
1741
+ "step": 1720
1742
+ },
1743
+ {
1744
+ "entropy": 0.5993610793724656,
1745
+ "epoch": 1.5944700460829493,
1746
+ "grad_norm": 0.3254019320011139,
1747
+ "learning_rate": 2.090768718802435e-05,
1748
+ "loss": 0.5809,
1749
+ "mean_token_accuracy": 0.8566695757210254,
1750
+ "num_tokens": 4676108.0,
1751
+ "step": 1730
1752
+ },
1753
+ {
1754
+ "entropy": 0.6142864376306534,
1755
+ "epoch": 1.6036866359447006,
1756
+ "grad_norm": 0.43382972478866577,
1757
+ "learning_rate": 2.0002854675980555e-05,
1758
+ "loss": 0.5895,
1759
+ "mean_token_accuracy": 0.8540547095239163,
1760
+ "num_tokens": 4703326.0,
1761
+ "step": 1740
1762
+ },
1763
+ {
1764
+ "entropy": 0.5973653512075543,
1765
+ "epoch": 1.6129032258064515,
1766
+ "grad_norm": 0.35976386070251465,
1767
+ "learning_rate": 1.9115857228882915e-05,
1768
+ "loss": 0.5765,
1769
+ "mean_token_accuracy": 0.8610024131834507,
1770
+ "num_tokens": 4730931.0,
1771
+ "step": 1750
1772
+ },
1773
+ {
1774
+ "entropy": 0.5743678430095315,
1775
+ "epoch": 1.6221198156682028,
1776
+ "grad_norm": 0.3067852854728699,
1777
+ "learning_rate": 1.824689259950144e-05,
1778
+ "loss": 0.5249,
1779
+ "mean_token_accuracy": 0.8677471876144409,
1780
+ "num_tokens": 4758011.0,
1781
+ "step": 1760
1782
+ },
1783
+ {
1784
+ "entropy": 0.5814758617430925,
1785
+ "epoch": 1.631336405529954,
1786
+ "grad_norm": 0.31184518337249756,
1787
+ "learning_rate": 1.7396154520256658e-05,
1788
+ "loss": 0.5571,
1789
+ "mean_token_accuracy": 0.8642784260213375,
1790
+ "num_tokens": 4784883.0,
1791
+ "step": 1770
1792
+ },
1793
+ {
1794
+ "entropy": 0.571275531500578,
1795
+ "epoch": 1.640552995391705,
1796
+ "grad_norm": 0.3559126853942871,
1797
+ "learning_rate": 1.6563832660027778e-05,
1798
+ "loss": 0.5458,
1799
+ "mean_token_accuracy": 0.8658527344465256,
1800
+ "num_tokens": 4811732.0,
1801
+ "step": 1780
1802
+ },
1803
+ {
1804
+ "entropy": 0.6137392023578286,
1805
+ "epoch": 1.6497695852534562,
1806
+ "grad_norm": 0.39478200674057007,
1807
+ "learning_rate": 1.5750112581866782e-05,
1808
+ "loss": 0.6095,
1809
+ "mean_token_accuracy": 0.8507423147559166,
1810
+ "num_tokens": 4836675.0,
1811
+ "step": 1790
1812
+ },
1813
+ {
1814
+ "entropy": 0.582203283533454,
1815
+ "epoch": 1.6589861751152073,
1816
+ "grad_norm": 0.34633445739746094,
1817
+ "learning_rate": 1.4955175701627721e-05,
1818
+ "loss": 0.5655,
1819
+ "mean_token_accuracy": 0.8587138280272484,
1820
+ "num_tokens": 4864037.0,
1821
+ "step": 1800
1822
+ },
1823
+ {
1824
+ "entropy": 0.6082916561514139,
1825
+ "epoch": 1.6682027649769586,
1826
+ "grad_norm": 0.4591638147830963,
1827
+ "learning_rate": 1.417919924752078e-05,
1828
+ "loss": 0.5801,
1829
+ "mean_token_accuracy": 0.8564680516719818,
1830
+ "num_tokens": 4888440.0,
1831
+ "step": 1810
1832
+ },
1833
+ {
1834
+ "entropy": 0.6297502366825938,
1835
+ "epoch": 1.6774193548387095,
1836
+ "grad_norm": 0.3770333230495453,
1837
+ "learning_rate": 1.3422356220599985e-05,
1838
+ "loss": 0.5886,
1839
+ "mean_token_accuracy": 0.8504323914647103,
1840
+ "num_tokens": 4915729.0,
1841
+ "step": 1820
1842
+ },
1843
+ {
1844
+ "entropy": 0.5782902050763369,
1845
+ "epoch": 1.6866359447004609,
1846
+ "grad_norm": 0.37898045778274536,
1847
+ "learning_rate": 1.2684815356193069e-05,
1848
+ "loss": 0.5488,
1849
+ "mean_token_accuracy": 0.8620311908423901,
1850
+ "num_tokens": 4943824.0,
1851
+ "step": 1830
1852
+ },
1853
+ {
1854
+ "entropy": 0.5877890868112445,
1855
+ "epoch": 1.695852534562212,
1856
+ "grad_norm": 0.3805868625640869,
1857
+ "learning_rate": 1.1966741086282795e-05,
1858
+ "loss": 0.5516,
1859
+ "mean_token_accuracy": 0.8623282983899117,
1860
+ "num_tokens": 4971942.0,
1861
+ "step": 1840
1862
+ },
1863
+ {
1864
+ "entropy": 0.6066540889441967,
1865
+ "epoch": 1.705069124423963,
1866
+ "grad_norm": 0.41600632667541504,
1867
+ "learning_rate": 1.1268293502847294e-05,
1868
+ "loss": 0.5697,
1869
+ "mean_token_accuracy": 0.8601134240627288,
1870
+ "num_tokens": 4998123.0,
1871
+ "step": 1850
1872
+ },
1873
+ {
1874
+ "entropy": 0.5986152492463589,
1875
+ "epoch": 1.7142857142857144,
1876
+ "grad_norm": 0.3279772996902466,
1877
+ "learning_rate": 1.0589628322168166e-05,
1878
+ "loss": 0.555,
1879
+ "mean_token_accuracy": 0.8551080651581288,
1880
+ "num_tokens": 5027515.0,
1881
+ "step": 1860
1882
+ },
1883
+ {
1884
+ "entropy": 0.5761340187862516,
1885
+ "epoch": 1.7235023041474653,
1886
+ "grad_norm": 0.36064383387565613,
1887
+ "learning_rate": 9.930896850114235e-06,
1888
+ "loss": 0.5528,
1889
+ "mean_token_accuracy": 0.8606743112206459,
1890
+ "num_tokens": 5056367.0,
1891
+ "step": 1870
1892
+ },
1893
+ {
1894
+ "entropy": 0.582815052382648,
1895
+ "epoch": 1.7327188940092166,
1896
+ "grad_norm": 0.39327871799468994,
1897
+ "learning_rate": 9.292245948408373e-06,
1898
+ "loss": 0.5498,
1899
+ "mean_token_accuracy": 0.8631025187671184,
1900
+ "num_tokens": 5084309.0,
1901
+ "step": 1880
1902
+ },
1903
+ {
1904
+ "entropy": 0.5940475927665829,
1905
+ "epoch": 1.7419354838709677,
1906
+ "grad_norm": 0.42712950706481934,
1907
+ "learning_rate": 8.673818001885237e-06,
1908
+ "loss": 0.5559,
1909
+ "mean_token_accuracy": 0.8595978386700154,
1910
+ "num_tokens": 5111002.0,
1911
+ "step": 1890
1912
+ },
1913
+ {
1914
+ "entropy": 0.6012541469186544,
1915
+ "epoch": 1.7511520737327189,
1916
+ "grad_norm": 0.38366585969924927,
1917
+ "learning_rate": 8.075750886747091e-06,
1918
+ "loss": 0.5865,
1919
+ "mean_token_accuracy": 0.8598495259881019,
1920
+ "num_tokens": 5137046.0,
1921
+ "step": 1900
1922
+ },
1923
+ {
1924
+ "entropy": 0.6162947248667479,
1925
+ "epoch": 1.7603686635944702,
1926
+ "grad_norm": 0.3927824795246124,
1927
+ "learning_rate": 7.498177939824858e-06,
1928
+ "loss": 0.5924,
1929
+ "mean_token_accuracy": 0.8556206963956356,
1930
+ "num_tokens": 5165425.0,
1931
+ "step": 1910
1932
+ },
1933
+ {
1934
+ "entropy": 0.5893848940730095,
1935
+ "epoch": 1.769585253456221,
1936
+ "grad_norm": 0.38203760981559753,
1937
+ "learning_rate": 6.941227928851179e-06,
1938
+ "loss": 0.5489,
1939
+ "mean_token_accuracy": 0.8609202355146408,
1940
+ "num_tokens": 5192293.0,
1941
+ "step": 1920
1942
+ },
1943
+ {
1944
+ "entropy": 0.5728636143729091,
1945
+ "epoch": 1.7788018433179724,
1946
+ "grad_norm": 0.3983209729194641,
1947
+ "learning_rate": 6.405025023752076e-06,
1948
+ "loss": 0.5426,
1949
+ "mean_token_accuracy": 0.8639165438711643,
1950
+ "num_tokens": 5223029.0,
1951
+ "step": 1930
1952
+ },
1953
+ {
1954
+ "entropy": 0.5681716226041317,
1955
+ "epoch": 1.7880184331797235,
1956
+ "grad_norm": 0.5423939824104309,
1957
+ "learning_rate": 5.889688768963742e-06,
1958
+ "loss": 0.5383,
1959
+ "mean_token_accuracy": 0.8677833467721939,
1960
+ "num_tokens": 5249251.0,
1961
+ "step": 1940
1962
+ },
1963
+ {
1964
+ "entropy": 0.5826122839003801,
1965
+ "epoch": 1.7972350230414746,
1966
+ "grad_norm": 0.443512499332428,
1967
+ "learning_rate": 5.395334056780643e-06,
1968
+ "loss": 0.5565,
1969
+ "mean_token_accuracy": 0.8625184670090675,
1970
+ "num_tokens": 5274986.0,
1971
+ "step": 1950
1972
+ },
1973
+ {
1974
+ "entropy": 0.6113853309303522,
1975
+ "epoch": 1.8064516129032258,
1976
+ "grad_norm": 0.3869023621082306,
1977
+ "learning_rate": 4.922071101740544e-06,
1978
+ "loss": 0.5807,
1979
+ "mean_token_accuracy": 0.8560661487281322,
1980
+ "num_tokens": 5300217.0,
1981
+ "step": 1960
1982
+ },
1983
+ {
1984
+ "entropy": 0.5974413625895977,
1985
+ "epoch": 1.8156682027649769,
1986
+ "grad_norm": 0.4994750916957855,
1987
+ "learning_rate": 4.470005416052769e-06,
1988
+ "loss": 0.5911,
1989
+ "mean_token_accuracy": 0.8553378529846668,
1990
+ "num_tokens": 5325240.0,
1991
+ "step": 1970
1992
+ },
1993
+ {
1994
+ "entropy": 0.5828492052853107,
1995
+ "epoch": 1.8248847926267282,
1996
+ "grad_norm": 0.37490594387054443,
1997
+ "learning_rate": 4.0392377860745125e-06,
1998
+ "loss": 0.5504,
1999
+ "mean_token_accuracy": 0.8645182229578495,
2000
+ "num_tokens": 5353087.0,
2001
+ "step": 1980
2002
+ },
2003
+ {
2004
+ "entropy": 0.614453304745257,
2005
+ "epoch": 1.8341013824884793,
2006
+ "grad_norm": 0.4105876088142395,
2007
+ "learning_rate": 3.629864249840953e-06,
2008
+ "loss": 0.6052,
2009
+ "mean_token_accuracy": 0.8532455086708068,
2010
+ "num_tokens": 5382519.0,
2011
+ "step": 1990
2012
+ },
2013
+ {
2014
+ "entropy": 0.5777437072247267,
2015
+ "epoch": 1.8433179723502304,
2016
+ "grad_norm": 0.4026196002960205,
2017
+ "learning_rate": 3.241976075653941e-06,
2018
+ "loss": 0.5425,
2019
+ "mean_token_accuracy": 0.86264653429389,
2020
+ "num_tokens": 5409627.0,
2021
+ "step": 2000
2022
+ },
2023
+ {
2024
+ "entropy": 0.6032186593860388,
2025
+ "epoch": 1.8525345622119815,
2026
+ "grad_norm": 0.4425666332244873,
2027
+ "learning_rate": 2.875659741734038e-06,
2028
+ "loss": 0.581,
2029
+ "mean_token_accuracy": 0.8562591679394245,
2030
+ "num_tokens": 5437899.0,
2031
+ "step": 2010
2032
+ },
2033
+ {
2034
+ "entropy": 0.5591373711824417,
2035
+ "epoch": 1.8617511520737327,
2036
+ "grad_norm": 0.39283913373947144,
2037
+ "learning_rate": 2.5309969169405532e-06,
2038
+ "loss": 0.5424,
2039
+ "mean_token_accuracy": 0.8625755675137043,
2040
+ "num_tokens": 5463949.0,
2041
+ "step": 2020
2042
+ },
2043
+ {
2044
+ "entropy": 0.5898947510868311,
2045
+ "epoch": 1.870967741935484,
2046
+ "grad_norm": 0.34159189462661743,
2047
+ "learning_rate": 2.2080644425637843e-06,
2048
+ "loss": 0.5512,
2049
+ "mean_token_accuracy": 0.8619050934910775,
2050
+ "num_tokens": 5490406.0,
2051
+ "step": 2030
2052
+ },
2053
+ {
2054
+ "entropy": 0.5845193408429623,
2055
+ "epoch": 1.8801843317972349,
2056
+ "grad_norm": 0.343017578125,
2057
+ "learning_rate": 1.9069343151934426e-06,
2058
+ "loss": 0.5537,
2059
+ "mean_token_accuracy": 0.8643649816513062,
2060
+ "num_tokens": 5518077.0,
2061
+ "step": 2040
2062
+ },
2063
+ {
2064
+ "entropy": 0.6067619472742081,
2065
+ "epoch": 1.8894009216589862,
2066
+ "grad_norm": 0.3805777132511139,
2067
+ "learning_rate": 1.627673670667451e-06,
2068
+ "loss": 0.5962,
2069
+ "mean_token_accuracy": 0.8592830084264278,
2070
+ "num_tokens": 5546303.0,
2071
+ "step": 2050
2072
+ },
2073
+ {
2074
+ "entropy": 0.5860567558556795,
2075
+ "epoch": 1.8986175115207373,
2076
+ "grad_norm": 0.38409003615379333,
2077
+ "learning_rate": 1.3703447691040816e-06,
2078
+ "loss": 0.5597,
2079
+ "mean_token_accuracy": 0.8619082234799862,
2080
+ "num_tokens": 5574574.0,
2081
+ "step": 2060
2082
+ },
2083
+ {
2084
+ "entropy": 0.5938012059777975,
2085
+ "epoch": 1.9078341013824884,
2086
+ "grad_norm": 0.4770345985889435,
2087
+ "learning_rate": 1.1350049810214258e-06,
2088
+ "loss": 0.5798,
2089
+ "mean_token_accuracy": 0.8559273667633533,
2090
+ "num_tokens": 5601062.0,
2091
+ "step": 2070
2092
+ },
2093
+ {
2094
+ "entropy": 0.6473757315427064,
2095
+ "epoch": 1.9170506912442398,
2096
+ "grad_norm": 0.4445556402206421,
2097
+ "learning_rate": 9.217067745467822e-07,
2098
+ "loss": 0.6203,
2099
+ "mean_token_accuracy": 0.8487160302698612,
2100
+ "num_tokens": 5627227.0,
2101
+ "step": 2080
2102
+ },
2103
+ {
2104
+ "entropy": 0.5940365632995963,
2105
+ "epoch": 1.9262672811059907,
2106
+ "grad_norm": 0.31166237592697144,
2107
+ "learning_rate": 7.304977037191129e-07,
2108
+ "loss": 0.5474,
2109
+ "mean_token_accuracy": 0.8592085830867291,
2110
+ "num_tokens": 5655152.0,
2111
+ "step": 2090
2112
+ },
2113
+ {
2114
+ "entropy": 0.6300271216779947,
2115
+ "epoch": 1.935483870967742,
2116
+ "grad_norm": 0.4227037727832794,
2117
+ "learning_rate": 5.614203978870358e-07,
2118
+ "loss": 0.618,
2119
+ "mean_token_accuracy": 0.84925981387496,
2120
+ "num_tokens": 5681343.0,
2121
+ "step": 2100
2122
+ },
2123
+ {
2124
+ "entropy": 0.5858591459691525,
2125
+ "epoch": 1.944700460829493,
2126
+ "grad_norm": 0.4059496223926544,
2127
+ "learning_rate": 4.1451255220478214e-07,
2128
+ "loss": 0.5661,
2129
+ "mean_token_accuracy": 0.8605487152934075,
2130
+ "num_tokens": 5708751.0,
2131
+ "step": 2110
2132
+ },
2133
+ {
2134
+ "entropy": 0.6246747657656669,
2135
+ "epoch": 1.9539170506912442,
2136
+ "grad_norm": 0.3966886103153229,
2137
+ "learning_rate": 2.898069192281749e-07,
2138
+ "loss": 0.5753,
2139
+ "mean_token_accuracy": 0.8555384337902069,
2140
+ "num_tokens": 5736242.0,
2141
+ "step": 2120
2142
+ },
2143
+ {
2144
+ "entropy": 0.5809266928583383,
2145
+ "epoch": 1.9631336405529956,
2146
+ "grad_norm": 0.4525570571422577,
2147
+ "learning_rate": 1.8733130161260327e-07,
2148
+ "loss": 0.5542,
2149
+ "mean_token_accuracy": 0.8635524161159992,
2150
+ "num_tokens": 5762920.0,
2151
+ "step": 2130
2152
+ },
2153
+ {
2154
+ "entropy": 0.5966292692348361,
2155
+ "epoch": 1.9723502304147464,
2156
+ "grad_norm": 0.4521016776561737,
2157
+ "learning_rate": 1.071085459145027e-07,
2158
+ "loss": 0.5698,
2159
+ "mean_token_accuracy": 0.8587224587798119,
2160
+ "num_tokens": 5788940.0,
2161
+ "step": 2140
2162
+ },
2163
+ {
2164
+ "entropy": 0.6092048080638051,
2165
+ "epoch": 1.9815668202764978,
2166
+ "grad_norm": 0.39693912863731384,
2167
+ "learning_rate": 4.9156537497818676e-08,
2168
+ "loss": 0.5942,
2169
+ "mean_token_accuracy": 0.8605172969400883,
2170
+ "num_tokens": 5815750.0,
2171
+ "step": 2150
2172
+ },
2173
+ {
2174
+ "entropy": 0.6100379537791014,
2175
+ "epoch": 1.9907834101382489,
2176
+ "grad_norm": 0.3836905062198639,
2177
+ "learning_rate": 1.348819654651834e-08,
2178
+ "loss": 0.5987,
2179
+ "mean_token_accuracy": 0.8537491798400879,
2180
+ "num_tokens": 5842446.0,
2181
+ "step": 2160
2182
+ },
2183
+ {
2184
+ "entropy": 0.6183315929025411,
2185
+ "epoch": 2.0,
2186
+ "grad_norm": 0.34127843379974365,
2187
+ "learning_rate": 1.1147518405030254e-10,
2188
+ "loss": 0.59,
2189
+ "mean_token_accuracy": 0.8559867747128009,
2190
+ "num_tokens": 5869612.0,
2191
+ "step": 2170
2192
+ },
2193
+ {
2194
+ "epoch": 2.0,
2195
+ "eval_entropy": 0.6346796658662607,
2196
+ "eval_loss": 0.7015999555587769,
2197
+ "eval_mean_token_accuracy": 0.8404892720934433,
2198
+ "eval_num_tokens": 5869612.0,
2199
+ "eval_runtime": 198.7416,
2200
+ "eval_samples_per_second": 10.919,
2201
+ "eval_steps_per_second": 10.919,
2202
+ "step": 2170
2203
+ }
2204
+ ],
2205
+ "logging_steps": 10,
2206
+ "max_steps": 2170,
2207
+ "num_input_tokens_seen": 0,
2208
+ "num_train_epochs": 2,
2209
+ "save_steps": 500,
2210
+ "stateful_callbacks": {
2211
+ "TrainerControl": {
2212
+ "args": {
2213
+ "should_epoch_stop": false,
2214
+ "should_evaluate": false,
2215
+ "should_log": false,
2216
+ "should_save": true,
2217
+ "should_training_stop": true
2218
+ },
2219
+ "attributes": {}
2220
+ }
2221
+ },
2222
+ "total_flos": 1.6397328361459507e+17,
2223
+ "train_batch_size": 2,
2224
+ "trial_name": null,
2225
+ "trial_params": null
2226
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:280dcf0271fa8b23a5440ba23ea98249f53c46f228cbc5314b1147c080890807
3
+ size 6225
vocab.json ADDED
The diff for this file is too large to render. See raw diff