Hugging Face's logo

duxx
/
DeepSeek-R1-Distill-Qwen-1.5B-Turkish

Text Generation
Transformers
Safetensors
qwen2
llama-factory
full
Generated from Trainer
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
duxx commited on
Commit
674e25f
·
verified ·
1 Parent(s): 0f31ee6

Upload folder using huggingface_hub

Browse files
Files changed (16) hide show
  1. .gitattributes +1 -0
  2. README.md +73 -0
  3. all_results.json +12 -0
  4. config.json +30 -0
  5. eval_results.json +7 -0
  6. generation_config.json +9 -0
  7. model.safetensors +3 -0
  8. special_tokens_map.json +32 -0
  9. tokenizer.json +3 -0
  10. tokenizer_config.json +208 -0
  11. train_results.json +8 -0
  12. trainer_log.jsonl +263 -0
  13. trainer_state.json +1885 -0
  14. training_args.bin +3 -0
  15. training_eval_loss.png +0 -0
  16. training_loss.png +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: other
4
+ base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
5
+ tags:
6
+ - llama-factory
7
+ - full
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: reasoning-turkish-R1
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # reasoning-turkish-R1
18
+
19
+ This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) on the reasoning-turkish-R1 dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 1.1396
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 1e-05
41
+ - train_batch_size: 1
42
+ - eval_batch_size: 1
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 8
46
+ - total_train_batch_size: 8
47
+ - total_eval_batch_size: 8
48
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
49
+ - lr_scheduler_type: cosine
50
+ - lr_scheduler_warmup_ratio: 0.01
51
+ - num_epochs: 1.0
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss |
56
+ |:-------------:|:------:|:----:|:---------------:|
57
+ | 1.6161 | 0.1028 | 26 | 1.4920 |
58
+ | 1.3683 | 0.2055 | 52 | 1.3256 |
59
+ | 1.409 | 0.3083 | 78 | 1.2486 |
60
+ | 1.1914 | 0.4111 | 104 | 1.2049 |
61
+ | 1.1264 | 0.5138 | 130 | 1.1763 |
62
+ | 1.1721 | 0.6166 | 156 | 1.1584 |
63
+ | 1.2738 | 0.7194 | 182 | 1.1467 |
64
+ | 1.1729 | 0.8221 | 208 | 1.1415 |
65
+ | 1.4446 | 0.9249 | 234 | 1.1399 |
66
+
67
+
68
+ ### Framework versions
69
+
70
+ - Transformers 4.48.2
71
+ - Pytorch 2.6.0+cu124
72
+ - Datasets 3.2.0
73
+ - Tokenizers 0.21.0
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_loss": 1.139628291130066,
4
+ "eval_runtime": 0.6305,
5
+ "eval_samples_per_second": 33.308,
6
+ "eval_steps_per_second": 4.758,
7
+ "total_flos": 8.443588829301965e+16,
8
+ "train_loss": 1.3333067849219553,
9
+ "train_runtime": 1520.8803,
10
+ "train_samples_per_second": 1.329,
11
+ "train_steps_per_second": 0.166
12
+ }
config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B",
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "eos_token_id": 151643,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 1536,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 8960,
13
+ "max_position_embeddings": 131072,
14
+ "max_window_layers": 21,
15
+ "model_type": "qwen2",
16
+ "num_attention_heads": 12,
17
+ "num_hidden_layers": 28,
18
+ "num_key_value_heads": 2,
19
+ "rms_norm_eps": 1e-06,
20
+ "rope_scaling": null,
21
+ "rope_theta": 10000,
22
+ "sliding_window": null,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "bfloat16",
25
+ "transformers_version": "4.48.2",
26
+ "use_cache": false,
27
+ "use_mrope": false,
28
+ "use_sliding_window": false,
29
+ "vocab_size": 151936
30
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_loss": 1.139628291130066,
4
+ "eval_runtime": 0.6305,
5
+ "eval_samples_per_second": 33.308,
6
+ "eval_steps_per_second": 4.758
7
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 151646,
4
+ "do_sample": true,
5
+ "eos_token_id": 151643,
6
+ "temperature": 0.6,
7
+ "top_p": 0.95,
8
+ "transformers_version": "4.48.2"
9
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e13ed8a367f5366eb89d22121f8ce6f8a4d51b3568b6b577518f023317ceec9a
3
+ size 3554214752
special_tokens_map.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ {
4
+ "content": "<|im_end|>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ }
10
+ ],
11
+ "bos_token": {
12
+ "content": "<｜begin▁of▁sentence｜>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "eos_token": {
19
+ "content": "<｜end▁of▁sentence｜>",
20
+ "lstrip": false,
21
+ "normalized": false,
22
+ "rstrip": false,
23
+ "single_word": false
24
+ },
25
+ "pad_token": {
26
+ "content": "<｜end▁of▁sentence｜>",
27
+ "lstrip": false,
28
+ "normalized": false,
29
+ "rstrip": false,
30
+ "single_word": false
31
+ }
32
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02643f00207dfc5ed248992486bde04314c21dca556bf65ce520690962b8db63
3
+ size 11422965
tokenizer_config.json ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "151643": {
7
+ "content": "<｜end▁of▁sentence｜>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "151644": {
15
+ "content": "<｜User｜>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": false
21
+ },
22
+ "151645": {
23
+ "content": "<｜Assistant｜>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": false
29
+ },
30
+ "151646": {
31
+ "content": "<｜begin▁of▁sentence｜>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "151647": {
39
+ "content": "<|EOT|>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": false
45
+ },
46
+ "151648": {
47
+ "content": "<think>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": false
53
+ },
54
+ "151649": {
55
+ "content": "</think>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": false
61
+ },
62
+ "151650": {
63
+ "content": "<|quad_start|>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "151651": {
71
+ "content": "<|quad_end|>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "151652": {
79
+ "content": "<|vision_start|>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "151653": {
87
+ "content": "<|vision_end|>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "151654": {
95
+ "content": "<|vision_pad|>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "151655": {
103
+ "content": "<|image_pad|>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "151656": {
111
+ "content": "<|video_pad|>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "151657": {
119
+ "content": "<tool_call>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": false
125
+ },
126
+ "151658": {
127
+ "content": "</tool_call>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": false
133
+ },
134
+ "151659": {
135
+ "content": "<|fim_prefix|>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": false
141
+ },
142
+ "151660": {
143
+ "content": "<|fim_middle|>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": false
149
+ },
150
+ "151661": {
151
+ "content": "<|fim_suffix|>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": false
157
+ },
158
+ "151662": {
159
+ "content": "<|fim_pad|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": false
165
+ },
166
+ "151663": {
167
+ "content": "<|repo_name|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": false
173
+ },
174
+ "151664": {
175
+ "content": "<|file_sep|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": false
181
+ },
182
+ "151665": {
183
+ "content": "<|im_end|>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": false,
187
+ "single_word": false,
188
+ "special": true
189
+ }
190
+ },
191
+ "additional_special_tokens": [
192
+ "<|im_end|>"
193
+ ],
194
+ "bos_token": "<｜begin▁of▁sentence｜>",
195
+ "chat_template": "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='') %}{%- for message in messages %}{%- if message['role'] == 'system' %}{% set ns.system_prompt = message['content'] %}{%- endif %}{%- endfor %}{{bos_token}}{{ns.system_prompt}}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<｜User｜>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is none %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls']%}{%- if not ns.is_first %}{{'<｜Assistant｜><｜tool▁calls▁begin｜><｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<｜tool▁call▁end｜>'}}{%- set ns.is_first = true -%}{%- else %}{{'\\n' + '<｜tool▁call▁begin｜>' + tool['type'] + '<｜tool▁sep｜>' + tool['function']['name'] + '\\n' + '```json' + '\\n' + tool['function']['arguments'] + '\\n' + '```' + '<｜tool▁call▁end｜>'}}{{'<｜tool▁calls▁end｜><｜end▁of▁sentence｜>'}}{%- endif %}{%- endfor %}{%- endif %}{%- if message['role'] == 'assistant' and message['content'] is not none %}{%- if ns.is_tool %}{{'<｜tool▁outputs▁end｜>' + message['content'] + '<｜end▁of▁sentence｜>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<｜Assistant｜>' + content + '<｜end▁of▁sentence｜>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<｜tool▁outputs▁begin｜><｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- set ns.is_output_first = false %}{%- else %}{{'\\n<｜tool▁output▁begin｜>' + message['content'] + '<｜tool▁output▁end｜>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<｜tool▁outputs▁end｜>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<｜Assistant｜>'}}{% endif %}",
196
+ "clean_up_tokenization_spaces": false,
197
+ "eos_token": "<｜end▁of▁sentence｜>",
198
+ "extra_special_tokens": {},
199
+ "legacy": true,
200
+ "model_max_length": 4500,
201
+ "pad_token": "<｜end▁of▁sentence｜>",
202
+ "padding_side": "right",
203
+ "sp_model_kwargs": {},
204
+ "split_special_tokens": false,
205
+ "tokenizer_class": "LlamaTokenizer",
206
+ "unk_token": null,
207
+ "use_default_system_prompt": false
208
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 8.443588829301965e+16,
4
+ "train_loss": 1.3333067849219553,
5
+ "train_runtime": 1520.8803,
6
+ "train_samples_per_second": 1.329,
7
+ "train_steps_per_second": 0.166
8
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,263 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 1, "total_steps": 253, "loss": 2.3652, "lr": 3.3333333333333333e-06, "epoch": 0.003952569169960474, "percentage": 0.4, "elapsed_time": "0:00:06", "remaining_time": "0:25:40"}
2
+ {"current_steps": 2, "total_steps": 253, "loss": 2.1458, "lr": 6.666666666666667e-06, "epoch": 0.007905138339920948, "percentage": 0.79, "elapsed_time": "0:00:11", "remaining_time": "0:24:56"}
3
+ {"current_steps": 3, "total_steps": 253, "loss": 2.2628, "lr": 1e-05, "epoch": 0.011857707509881422, "percentage": 1.19, "elapsed_time": "0:00:17", "remaining_time": "0:24:23"}
4
+ {"current_steps": 4, "total_steps": 253, "loss": 2.0308, "lr": 9.999605221019082e-06, "epoch": 0.015810276679841896, "percentage": 1.58, "elapsed_time": "0:00:23", "remaining_time": "0:24:14"}
5
+ {"current_steps": 5, "total_steps": 253, "loss": 1.6917, "lr": 9.9984209464165e-06, "epoch": 0.019762845849802372, "percentage": 1.98, "elapsed_time": "0:00:29", "remaining_time": "0:24:16"}
6
+ {"current_steps": 6, "total_steps": 253, "loss": 1.8489, "lr": 9.996447363202947e-06, "epoch": 0.023715415019762844, "percentage": 2.37, "elapsed_time": "0:00:35", "remaining_time": "0:24:15"}
7
+ {"current_steps": 7, "total_steps": 253, "loss": 1.9894, "lr": 9.99368478303009e-06, "epoch": 0.02766798418972332, "percentage": 2.77, "elapsed_time": "0:00:41", "remaining_time": "0:24:15"}
8
+ {"current_steps": 8, "total_steps": 253, "loss": 1.8366, "lr": 9.990133642141359e-06, "epoch": 0.03162055335968379, "percentage": 3.16, "elapsed_time": "0:00:47", "remaining_time": "0:24:07"}
9
+ {"current_steps": 9, "total_steps": 253, "loss": 1.9013, "lr": 9.98579450130307e-06, "epoch": 0.03557312252964427, "percentage": 3.56, "elapsed_time": "0:00:53", "remaining_time": "0:23:57"}
10
+ {"current_steps": 10, "total_steps": 253, "loss": 2.034, "lr": 9.980668045715864e-06, "epoch": 0.039525691699604744, "percentage": 3.95, "elapsed_time": "0:00:58", "remaining_time": "0:23:51"}
11
+ {"current_steps": 11, "total_steps": 253, "loss": 1.8565, "lr": 9.974755084906503e-06, "epoch": 0.043478260869565216, "percentage": 4.35, "elapsed_time": "0:01:04", "remaining_time": "0:23:49"}
12
+ {"current_steps": 12, "total_steps": 253, "loss": 1.8154, "lr": 9.968056552600043e-06, "epoch": 0.04743083003952569, "percentage": 4.74, "elapsed_time": "0:01:11", "remaining_time": "0:23:46"}
13
+ {"current_steps": 13, "total_steps": 253, "loss": 1.7205, "lr": 9.960573506572391e-06, "epoch": 0.05138339920948617, "percentage": 5.14, "elapsed_time": "0:01:16", "remaining_time": "0:23:41"}
14
+ {"current_steps": 14, "total_steps": 253, "loss": 1.8481, "lr": 9.952307128483257e-06, "epoch": 0.05533596837944664, "percentage": 5.53, "elapsed_time": "0:01:22", "remaining_time": "0:23:35"}
15
+ {"current_steps": 15, "total_steps": 253, "loss": 1.5792, "lr": 9.94325872368957e-06, "epoch": 0.05928853754940711, "percentage": 5.93, "elapsed_time": "0:01:28", "remaining_time": "0:23:27"}
16
+ {"current_steps": 16, "total_steps": 253, "loss": 1.6527, "lr": 9.93342972103934e-06, "epoch": 0.06324110671936758, "percentage": 6.32, "elapsed_time": "0:01:34", "remaining_time": "0:23:23"}
17
+ {"current_steps": 17, "total_steps": 253, "loss": 1.3883, "lr": 9.922821672646028e-06, "epoch": 0.06719367588932806, "percentage": 6.72, "elapsed_time": "0:01:40", "remaining_time": "0:23:18"}
18
+ {"current_steps": 18, "total_steps": 253, "loss": 1.5297, "lr": 9.911436253643445e-06, "epoch": 0.07114624505928854, "percentage": 7.11, "elapsed_time": "0:01:46", "remaining_time": "0:23:13"}
19
+ {"current_steps": 19, "total_steps": 253, "loss": 1.5525, "lr": 9.899275261921236e-06, "epoch": 0.07509881422924901, "percentage": 7.51, "elapsed_time": "0:01:52", "remaining_time": "0:23:08"}
20
+ {"current_steps": 20, "total_steps": 253, "loss": 1.5752, "lr": 9.886340617840968e-06, "epoch": 0.07905138339920949, "percentage": 7.91, "elapsed_time": "0:01:58", "remaining_time": "0:23:02"}
21
+ {"current_steps": 21, "total_steps": 253, "loss": 1.5944, "lr": 9.872634363932887e-06, "epoch": 0.08300395256916997, "percentage": 8.3, "elapsed_time": "0:02:04", "remaining_time": "0:22:55"}
22
+ {"current_steps": 22, "total_steps": 253, "loss": 1.4019, "lr": 9.85815866457337e-06, "epoch": 0.08695652173913043, "percentage": 8.7, "elapsed_time": "0:02:10", "remaining_time": "0:22:49"}
23
+ {"current_steps": 23, "total_steps": 253, "loss": 1.706, "lr": 9.842915805643156e-06, "epoch": 0.09090909090909091, "percentage": 9.09, "elapsed_time": "0:02:16", "remaining_time": "0:22:44"}
24
+ {"current_steps": 24, "total_steps": 253, "loss": 1.6249, "lr": 9.82690819416637e-06, "epoch": 0.09486166007905138, "percentage": 9.49, "elapsed_time": "0:02:22", "remaining_time": "0:22:38"}
25
+ {"current_steps": 25, "total_steps": 253, "loss": 1.8374, "lr": 9.81013835793043e-06, "epoch": 0.09881422924901186, "percentage": 9.88, "elapsed_time": "0:02:28", "remaining_time": "0:22:33"}
26
+ {"current_steps": 26, "total_steps": 253, "loss": 1.6161, "lr": 9.79260894508688e-06, "epoch": 0.10276679841897234, "percentage": 10.28, "elapsed_time": "0:02:34", "remaining_time": "0:22:25"}
27
+ {"current_steps": 26, "total_steps": 253, "eval_loss": 1.4920315742492676, "epoch": 0.10276679841897234, "percentage": 10.28, "elapsed_time": "0:02:34", "remaining_time": "0:22:31"}
28
+ {"current_steps": 27, "total_steps": 253, "loss": 1.6593, "lr": 9.774322723733216e-06, "epoch": 0.1067193675889328, "percentage": 10.67, "elapsed_time": "0:02:40", "remaining_time": "0:22:25"}
29
+ {"current_steps": 28, "total_steps": 253, "loss": 1.6982, "lr": 9.755282581475769e-06, "epoch": 0.11067193675889328, "percentage": 11.07, "elapsed_time": "0:02:46", "remaining_time": "0:22:20"}
30
+ {"current_steps": 29, "total_steps": 253, "loss": 1.4423, "lr": 9.735491524973723e-06, "epoch": 0.11462450592885376, "percentage": 11.46, "elapsed_time": "0:02:52", "remaining_time": "0:22:14"}
31
+ {"current_steps": 30, "total_steps": 253, "loss": 1.3055, "lr": 9.714952679464324e-06, "epoch": 0.11857707509881422, "percentage": 11.86, "elapsed_time": "0:02:58", "remaining_time": "0:22:09"}
32
+ {"current_steps": 31, "total_steps": 253, "loss": 1.6302, "lr": 9.693669288269371e-06, "epoch": 0.1225296442687747, "percentage": 12.25, "elapsed_time": "0:03:04", "remaining_time": "0:22:03"}
33
+ {"current_steps": 32, "total_steps": 253, "loss": 1.7975, "lr": 9.671644712283061e-06, "epoch": 0.12648221343873517, "percentage": 12.65, "elapsed_time": "0:03:10", "remaining_time": "0:21:57"}
34
+ {"current_steps": 33, "total_steps": 253, "loss": 1.2709, "lr": 9.648882429441258e-06, "epoch": 0.13043478260869565, "percentage": 13.04, "elapsed_time": "0:03:16", "remaining_time": "0:21:52"}
35
+ {"current_steps": 34, "total_steps": 253, "loss": 1.4833, "lr": 9.62538603417229e-06, "epoch": 0.13438735177865613, "percentage": 13.44, "elapsed_time": "0:03:22", "remaining_time": "0:21:46"}
36
+ {"current_steps": 35, "total_steps": 253, "loss": 1.2978, "lr": 9.601159236829353e-06, "epoch": 0.1383399209486166, "percentage": 13.83, "elapsed_time": "0:03:28", "remaining_time": "0:21:39"}
37
+ {"current_steps": 36, "total_steps": 253, "loss": 1.5622, "lr": 9.576205863104588e-06, "epoch": 0.1422924901185771, "percentage": 14.23, "elapsed_time": "0:03:34", "remaining_time": "0:21:32"}
38
+ {"current_steps": 37, "total_steps": 253, "loss": 1.7766, "lr": 9.550529853424979e-06, "epoch": 0.14624505928853754, "percentage": 14.62, "elapsed_time": "0:03:40", "remaining_time": "0:21:26"}
39
+ {"current_steps": 38, "total_steps": 253, "loss": 1.5206, "lr": 9.524135262330098e-06, "epoch": 0.15019762845849802, "percentage": 15.02, "elapsed_time": "0:03:46", "remaining_time": "0:21:21"}
40
+ {"current_steps": 39, "total_steps": 253, "loss": 1.5341, "lr": 9.497026257831856e-06, "epoch": 0.1541501976284585, "percentage": 15.42, "elapsed_time": "0:03:52", "remaining_time": "0:21:15"}
41
+ {"current_steps": 40, "total_steps": 253, "loss": 1.4157, "lr": 9.46920712075632e-06, "epoch": 0.15810276679841898, "percentage": 15.81, "elapsed_time": "0:03:58", "remaining_time": "0:21:09"}
42
+ {"current_steps": 41, "total_steps": 253, "loss": 1.2846, "lr": 9.440682244067724e-06, "epoch": 0.16205533596837945, "percentage": 16.21, "elapsed_time": "0:04:04", "remaining_time": "0:21:03"}
43
+ {"current_steps": 42, "total_steps": 253, "loss": 1.4727, "lr": 9.411456132174768e-06, "epoch": 0.16600790513833993, "percentage": 16.6, "elapsed_time": "0:04:10", "remaining_time": "0:20:58"}
44
+ {"current_steps": 43, "total_steps": 253, "loss": 1.6619, "lr": 9.381533400219319e-06, "epoch": 0.16996047430830039, "percentage": 17.0, "elapsed_time": "0:04:16", "remaining_time": "0:20:52"}
45
+ {"current_steps": 44, "total_steps": 253, "loss": 1.4687, "lr": 9.35091877334763e-06, "epoch": 0.17391304347826086, "percentage": 17.39, "elapsed_time": "0:04:22", "remaining_time": "0:20:46"}
46
+ {"current_steps": 45, "total_steps": 253, "loss": 1.4871, "lr": 9.319617085964177e-06, "epoch": 0.17786561264822134, "percentage": 17.79, "elapsed_time": "0:04:28", "remaining_time": "0:20:40"}
47
+ {"current_steps": 46, "total_steps": 253, "loss": 1.2694, "lr": 9.287633280968263e-06, "epoch": 0.18181818181818182, "percentage": 18.18, "elapsed_time": "0:04:34", "remaining_time": "0:20:34"}
48
+ {"current_steps": 47, "total_steps": 253, "loss": 1.2148, "lr": 9.25497240897346e-06, "epoch": 0.1857707509881423, "percentage": 18.58, "elapsed_time": "0:04:40", "remaining_time": "0:20:28"}
49
+ {"current_steps": 48, "total_steps": 253, "loss": 1.5932, "lr": 9.221639627510076e-06, "epoch": 0.18972332015810275, "percentage": 18.97, "elapsed_time": "0:04:46", "remaining_time": "0:20:22"}
50
+ {"current_steps": 49, "total_steps": 253, "loss": 1.3074, "lr": 9.18764020021071e-06, "epoch": 0.19367588932806323, "percentage": 19.37, "elapsed_time": "0:04:52", "remaining_time": "0:20:16"}
51
+ {"current_steps": 50, "total_steps": 253, "loss": 1.5327, "lr": 9.152979495979064e-06, "epoch": 0.1976284584980237, "percentage": 19.76, "elapsed_time": "0:04:58", "remaining_time": "0:20:11"}
52
+ {"current_steps": 51, "total_steps": 253, "loss": 1.6561, "lr": 9.117662988142138e-06, "epoch": 0.2015810276679842, "percentage": 20.16, "elapsed_time": "0:05:04", "remaining_time": "0:20:04"}
53
+ {"current_steps": 52, "total_steps": 253, "loss": 1.3683, "lr": 9.08169625358592e-06, "epoch": 0.20553359683794467, "percentage": 20.55, "elapsed_time": "0:05:09", "remaining_time": "0:19:58"}
54
+ {"current_steps": 52, "total_steps": 253, "eval_loss": 1.325599193572998, "epoch": 0.20553359683794467, "percentage": 20.55, "elapsed_time": "0:05:10", "remaining_time": "0:20:00"}
55
+ {"current_steps": 53, "total_steps": 253, "loss": 1.5192, "lr": 9.045084971874738e-06, "epoch": 0.20948616600790515, "percentage": 20.95, "elapsed_time": "0:05:16", "remaining_time": "0:19:54"}
56
+ {"current_steps": 54, "total_steps": 253, "loss": 1.4107, "lr": 9.007834924354384e-06, "epoch": 0.2134387351778656, "percentage": 21.34, "elapsed_time": "0:05:22", "remaining_time": "0:19:48"}
57
+ {"current_steps": 55, "total_steps": 253, "loss": 1.392, "lr": 8.969951993239177e-06, "epoch": 0.21739130434782608, "percentage": 21.74, "elapsed_time": "0:05:28", "remaining_time": "0:19:42"}
58
+ {"current_steps": 56, "total_steps": 253, "loss": 1.2833, "lr": 8.931442160683094e-06, "epoch": 0.22134387351778656, "percentage": 22.13, "elapsed_time": "0:05:34", "remaining_time": "0:19:37"}
59
+ {"current_steps": 57, "total_steps": 253, "loss": 1.26, "lr": 8.892311507835118e-06, "epoch": 0.22529644268774704, "percentage": 22.53, "elapsed_time": "0:05:40", "remaining_time": "0:19:30"}
60
+ {"current_steps": 58, "total_steps": 253, "loss": 1.3223, "lr": 8.852566213878947e-06, "epoch": 0.22924901185770752, "percentage": 22.92, "elapsed_time": "0:05:46", "remaining_time": "0:19:24"}
61
+ {"current_steps": 59, "total_steps": 253, "loss": 1.3826, "lr": 8.81221255505724e-06, "epoch": 0.233201581027668, "percentage": 23.32, "elapsed_time": "0:05:52", "remaining_time": "0:19:18"}
62
+ {"current_steps": 60, "total_steps": 253, "loss": 1.4478, "lr": 8.77125690368052e-06, "epoch": 0.23715415019762845, "percentage": 23.72, "elapsed_time": "0:05:58", "remaining_time": "0:19:13"}
63
+ {"current_steps": 61, "total_steps": 253, "loss": 1.3503, "lr": 8.729705727120911e-06, "epoch": 0.24110671936758893, "percentage": 24.11, "elapsed_time": "0:06:04", "remaining_time": "0:19:07"}
64
+ {"current_steps": 62, "total_steps": 253, "loss": 1.041, "lr": 8.68756558679087e-06, "epoch": 0.2450592885375494, "percentage": 24.51, "elapsed_time": "0:06:10", "remaining_time": "0:19:01"}
65
+ {"current_steps": 63, "total_steps": 253, "loss": 1.323, "lr": 8.644843137107058e-06, "epoch": 0.2490118577075099, "percentage": 24.9, "elapsed_time": "0:06:16", "remaining_time": "0:18:55"}
66
+ {"current_steps": 64, "total_steps": 253, "loss": 1.3562, "lr": 8.601545124439535e-06, "epoch": 0.25296442687747034, "percentage": 25.3, "elapsed_time": "0:06:22", "remaining_time": "0:18:49"}
67
+ {"current_steps": 65, "total_steps": 253, "loss": 1.5575, "lr": 8.557678386046429e-06, "epoch": 0.25691699604743085, "percentage": 25.69, "elapsed_time": "0:06:28", "remaining_time": "0:18:43"}
68
+ {"current_steps": 66, "total_steps": 253, "loss": 1.2068, "lr": 8.513249848994248e-06, "epoch": 0.2608695652173913, "percentage": 26.09, "elapsed_time": "0:06:34", "remaining_time": "0:18:37"}
69
+ {"current_steps": 67, "total_steps": 253, "loss": 1.2409, "lr": 8.468266529064025e-06, "epoch": 0.2648221343873518, "percentage": 26.48, "elapsed_time": "0:06:40", "remaining_time": "0:18:31"}
70
+ {"current_steps": 68, "total_steps": 253, "loss": 1.4648, "lr": 8.422735529643445e-06, "epoch": 0.26877470355731226, "percentage": 26.88, "elapsed_time": "0:06:46", "remaining_time": "0:18:25"}
71
+ {"current_steps": 69, "total_steps": 253, "loss": 1.142, "lr": 8.376664040605122e-06, "epoch": 0.2727272727272727, "percentage": 27.27, "elapsed_time": "0:06:52", "remaining_time": "0:18:19"}
72
+ {"current_steps": 70, "total_steps": 253, "loss": 1.179, "lr": 8.33005933717126e-06, "epoch": 0.2766798418972332, "percentage": 27.67, "elapsed_time": "0:06:58", "remaining_time": "0:18:13"}
73
+ {"current_steps": 71, "total_steps": 253, "loss": 1.258, "lr": 8.282928778764783e-06, "epoch": 0.28063241106719367, "percentage": 28.06, "elapsed_time": "0:07:04", "remaining_time": "0:18:07"}
74
+ {"current_steps": 72, "total_steps": 253, "loss": 1.3277, "lr": 8.235279807847223e-06, "epoch": 0.2845849802371542, "percentage": 28.46, "elapsed_time": "0:07:10", "remaining_time": "0:18:01"}
75
+ {"current_steps": 73, "total_steps": 253, "loss": 1.1869, "lr": 8.18711994874345e-06, "epoch": 0.2885375494071146, "percentage": 28.85, "elapsed_time": "0:07:16", "remaining_time": "0:17:55"}
76
+ {"current_steps": 74, "total_steps": 253, "loss": 1.37, "lr": 8.138456806453503e-06, "epoch": 0.2924901185770751, "percentage": 29.25, "elapsed_time": "0:07:22", "remaining_time": "0:17:49"}
77
+ {"current_steps": 75, "total_steps": 253, "loss": 1.2171, "lr": 8.089298065451673e-06, "epoch": 0.2964426877470356, "percentage": 29.64, "elapsed_time": "0:07:28", "remaining_time": "0:17:43"}
78
+ {"current_steps": 76, "total_steps": 253, "loss": 1.3983, "lr": 8.039651488473028e-06, "epoch": 0.30039525691699603, "percentage": 30.04, "elapsed_time": "0:07:34", "remaining_time": "0:17:37"}
79
+ {"current_steps": 77, "total_steps": 253, "loss": 1.4224, "lr": 7.989524915287595e-06, "epoch": 0.30434782608695654, "percentage": 30.43, "elapsed_time": "0:07:40", "remaining_time": "0:17:31"}
80
+ {"current_steps": 78, "total_steps": 253, "loss": 1.409, "lr": 7.938926261462366e-06, "epoch": 0.308300395256917, "percentage": 30.83, "elapsed_time": "0:07:46", "remaining_time": "0:17:25"}
81
+ {"current_steps": 78, "total_steps": 253, "eval_loss": 1.2486090660095215, "epoch": 0.308300395256917, "percentage": 30.83, "elapsed_time": "0:07:46", "remaining_time": "0:17:27"}
82
+ {"current_steps": 79, "total_steps": 253, "loss": 1.3565, "lr": 7.887863517111337e-06, "epoch": 0.31225296442687744, "percentage": 31.23, "elapsed_time": "0:07:52", "remaining_time": "0:17:20"}
83
+ {"current_steps": 80, "total_steps": 253, "loss": 1.336, "lr": 7.836344745633785e-06, "epoch": 0.31620553359683795, "percentage": 31.62, "elapsed_time": "0:07:58", "remaining_time": "0:17:14"}
84
+ {"current_steps": 81, "total_steps": 253, "loss": 1.318, "lr": 7.78437808244094e-06, "epoch": 0.3201581027667984, "percentage": 32.02, "elapsed_time": "0:08:04", "remaining_time": "0:17:08"}
85
+ {"current_steps": 82, "total_steps": 253, "loss": 1.264, "lr": 7.731971733671347e-06, "epoch": 0.3241106719367589, "percentage": 32.41, "elapsed_time": "0:08:10", "remaining_time": "0:17:02"}
86
+ {"current_steps": 83, "total_steps": 253, "loss": 1.2758, "lr": 7.679133974894984e-06, "epoch": 0.32806324110671936, "percentage": 32.81, "elapsed_time": "0:08:16", "remaining_time": "0:16:57"}
87
+ {"current_steps": 84, "total_steps": 253, "loss": 1.1448, "lr": 7.6258731498064796e-06, "epoch": 0.33201581027667987, "percentage": 33.2, "elapsed_time": "0:08:22", "remaining_time": "0:16:51"}
88
+ {"current_steps": 85, "total_steps": 253, "loss": 1.2644, "lr": 7.572197668907533e-06, "epoch": 0.3359683794466403, "percentage": 33.6, "elapsed_time": "0:08:28", "remaining_time": "0:16:45"}
89
+ {"current_steps": 86, "total_steps": 253, "loss": 1.2656, "lr": 7.518116008178805e-06, "epoch": 0.33992094861660077, "percentage": 33.99, "elapsed_time": "0:08:34", "remaining_time": "0:16:39"}
90
+ {"current_steps": 87, "total_steps": 253, "loss": 1.2453, "lr": 7.463636707741458e-06, "epoch": 0.3438735177865613, "percentage": 34.39, "elapsed_time": "0:08:40", "remaining_time": "0:16:33"}
91
+ {"current_steps": 88, "total_steps": 253, "loss": 1.1815, "lr": 7.408768370508577e-06, "epoch": 0.34782608695652173, "percentage": 34.78, "elapsed_time": "0:08:46", "remaining_time": "0:16:27"}
92
+ {"current_steps": 89, "total_steps": 253, "loss": 1.2232, "lr": 7.353519660826665e-06, "epoch": 0.35177865612648224, "percentage": 35.18, "elapsed_time": "0:08:52", "remaining_time": "0:16:21"}
93
+ {"current_steps": 90, "total_steps": 253, "loss": 1.1386, "lr": 7.297899303107441e-06, "epoch": 0.3557312252964427, "percentage": 35.57, "elapsed_time": "0:08:58", "remaining_time": "0:16:14"}
94
+ {"current_steps": 91, "total_steps": 253, "loss": 1.2962, "lr": 7.241916080450163e-06, "epoch": 0.35968379446640314, "percentage": 35.97, "elapsed_time": "0:09:04", "remaining_time": "0:16:08"}
95
+ {"current_steps": 92, "total_steps": 253, "loss": 1.3109, "lr": 7.185578833254665e-06, "epoch": 0.36363636363636365, "percentage": 36.36, "elapsed_time": "0:09:10", "remaining_time": "0:16:02"}
96
+ {"current_steps": 93, "total_steps": 253, "loss": 1.4728, "lr": 7.128896457825364e-06, "epoch": 0.3675889328063241, "percentage": 36.76, "elapsed_time": "0:09:16", "remaining_time": "0:15:56"}
97
+ {"current_steps": 94, "total_steps": 253, "loss": 1.2597, "lr": 7.071877904966422e-06, "epoch": 0.3715415019762846, "percentage": 37.15, "elapsed_time": "0:09:22", "remaining_time": "0:15:50"}
98
+ {"current_steps": 95, "total_steps": 253, "loss": 1.2041, "lr": 7.014532178568314e-06, "epoch": 0.37549407114624506, "percentage": 37.55, "elapsed_time": "0:09:28", "remaining_time": "0:15:44"}
99
+ {"current_steps": 96, "total_steps": 253, "loss": 1.1369, "lr": 6.9568683341860135e-06, "epoch": 0.3794466403162055, "percentage": 37.94, "elapsed_time": "0:09:34", "remaining_time": "0:15:39"}
100
+ {"current_steps": 97, "total_steps": 253, "loss": 1.2103, "lr": 6.898895477609007e-06, "epoch": 0.383399209486166, "percentage": 38.34, "elapsed_time": "0:09:40", "remaining_time": "0:15:33"}
101
+ {"current_steps": 98, "total_steps": 253, "loss": 1.2609, "lr": 6.840622763423391e-06, "epoch": 0.38735177865612647, "percentage": 38.74, "elapsed_time": "0:09:46", "remaining_time": "0:15:27"}
102
+ {"current_steps": 99, "total_steps": 253, "loss": 1.1807, "lr": 6.782059393566254e-06, "epoch": 0.391304347826087, "percentage": 39.13, "elapsed_time": "0:09:52", "remaining_time": "0:15:20"}
103
+ {"current_steps": 100, "total_steps": 253, "loss": 1.3748, "lr": 6.723214615872585e-06, "epoch": 0.3952569169960474, "percentage": 39.53, "elapsed_time": "0:09:58", "remaining_time": "0:15:15"}
104
+ {"current_steps": 101, "total_steps": 253, "loss": 1.271, "lr": 6.664097722614934e-06, "epoch": 0.39920948616600793, "percentage": 39.92, "elapsed_time": "0:10:03", "remaining_time": "0:15:08"}
105
+ {"current_steps": 102, "total_steps": 253, "loss": 1.3016, "lr": 6.604718049036047e-06, "epoch": 0.4031620553359684, "percentage": 40.32, "elapsed_time": "0:10:10", "remaining_time": "0:15:03"}
106
+ {"current_steps": 103, "total_steps": 253, "loss": 1.3072, "lr": 6.545084971874738e-06, "epoch": 0.40711462450592883, "percentage": 40.71, "elapsed_time": "0:10:16", "remaining_time": "0:14:57"}
107
+ {"current_steps": 104, "total_steps": 253, "loss": 1.1914, "lr": 6.485207907885175e-06, "epoch": 0.41106719367588934, "percentage": 41.11, "elapsed_time": "0:10:21", "remaining_time": "0:14:51"}
108
+ {"current_steps": 104, "total_steps": 253, "eval_loss": 1.2048839330673218, "epoch": 0.41106719367588934, "percentage": 41.11, "elapsed_time": "0:10:22", "remaining_time": "0:14:52"}
109
+ {"current_steps": 105, "total_steps": 253, "loss": 1.3336, "lr": 6.425096312349881e-06, "epoch": 0.4150197628458498, "percentage": 41.5, "elapsed_time": "0:10:28", "remaining_time": "0:14:45"}
110
+ {"current_steps": 106, "total_steps": 253, "loss": 1.275, "lr": 6.364759677586627e-06, "epoch": 0.4189723320158103, "percentage": 41.9, "elapsed_time": "0:10:34", "remaining_time": "0:14:40"}
111
+ {"current_steps": 107, "total_steps": 253, "loss": 1.237, "lr": 6.304207531449486e-06, "epoch": 0.42292490118577075, "percentage": 42.29, "elapsed_time": "0:10:40", "remaining_time": "0:14:33"}
112
+ {"current_steps": 108, "total_steps": 253, "loss": 1.1993, "lr": 6.243449435824276e-06, "epoch": 0.4268774703557312, "percentage": 42.69, "elapsed_time": "0:10:46", "remaining_time": "0:14:27"}
113
+ {"current_steps": 109, "total_steps": 253, "loss": 1.3898, "lr": 6.182494985118625e-06, "epoch": 0.4308300395256917, "percentage": 43.08, "elapsed_time": "0:10:52", "remaining_time": "0:14:21"}
114
+ {"current_steps": 110, "total_steps": 253, "loss": 1.1805, "lr": 6.121353804746907e-06, "epoch": 0.43478260869565216, "percentage": 43.48, "elapsed_time": "0:10:58", "remaining_time": "0:14:15"}
115
+ {"current_steps": 111, "total_steps": 253, "loss": 0.8793, "lr": 6.060035549610275e-06, "epoch": 0.43873517786561267, "percentage": 43.87, "elapsed_time": "0:11:04", "remaining_time": "0:14:09"}
116
+ {"current_steps": 112, "total_steps": 253, "loss": 1.3551, "lr": 5.9985499025720354e-06, "epoch": 0.4426877470355731, "percentage": 44.27, "elapsed_time": "0:11:10", "remaining_time": "0:14:03"}
117
+ {"current_steps": 113, "total_steps": 253, "loss": 1.0515, "lr": 5.936906572928625e-06, "epoch": 0.44664031620553357, "percentage": 44.66, "elapsed_time": "0:11:16", "remaining_time": "0:13:57"}
118
+ {"current_steps": 114, "total_steps": 253, "loss": 1.0884, "lr": 5.8751152948763815e-06, "epoch": 0.4505928853754941, "percentage": 45.06, "elapsed_time": "0:11:22", "remaining_time": "0:13:51"}
119
+ {"current_steps": 115, "total_steps": 253, "loss": 1.4432, "lr": 5.813185825974419e-06, "epoch": 0.45454545454545453, "percentage": 45.45, "elapsed_time": "0:11:27", "remaining_time": "0:13:45"}
120
+ {"current_steps": 116, "total_steps": 253, "loss": 1.0239, "lr": 5.751127945603786e-06, "epoch": 0.45849802371541504, "percentage": 45.85, "elapsed_time": "0:11:33", "remaining_time": "0:13:39"}
121
+ {"current_steps": 117, "total_steps": 253, "loss": 1.5178, "lr": 5.68895145342319e-06, "epoch": 0.4624505928853755, "percentage": 46.25, "elapsed_time": "0:11:39", "remaining_time": "0:13:33"}
122
+ {"current_steps": 118, "total_steps": 253, "loss": 1.3872, "lr": 5.626666167821522e-06, "epoch": 0.466403162055336, "percentage": 46.64, "elapsed_time": "0:11:45", "remaining_time": "0:13:27"}
123
+ {"current_steps": 119, "total_steps": 253, "loss": 1.3254, "lr": 5.5642819243674085e-06, "epoch": 0.47035573122529645, "percentage": 47.04, "elapsed_time": "0:11:51", "remaining_time": "0:13:21"}
124
+ {"current_steps": 120, "total_steps": 253, "loss": 1.1636, "lr": 5.5018085742560745e-06, "epoch": 0.4743083003952569, "percentage": 47.43, "elapsed_time": "0:11:57", "remaining_time": "0:13:15"}
125
+ {"current_steps": 121, "total_steps": 253, "loss": 1.0787, "lr": 5.439255982753717e-06, "epoch": 0.4782608695652174, "percentage": 47.83, "elapsed_time": "0:12:03", "remaining_time": "0:13:09"}
126
+ {"current_steps": 122, "total_steps": 253, "loss": 1.3438, "lr": 5.376634027639664e-06, "epoch": 0.48221343873517786, "percentage": 48.22, "elapsed_time": "0:12:09", "remaining_time": "0:13:03"}
127
+ {"current_steps": 123, "total_steps": 253, "loss": 1.2148, "lr": 5.3139525976465675e-06, "epoch": 0.48616600790513836, "percentage": 48.62, "elapsed_time": "0:12:15", "remaining_time": "0:12:57"}
128
+ {"current_steps": 124, "total_steps": 253, "loss": 1.1715, "lr": 5.251221590898848e-06, "epoch": 0.4901185770750988, "percentage": 49.01, "elapsed_time": "0:12:21", "remaining_time": "0:12:51"}
129
+ {"current_steps": 125, "total_steps": 253, "loss": 1.1467, "lr": 5.188450913349674e-06, "epoch": 0.49407114624505927, "percentage": 49.41, "elapsed_time": "0:12:27", "remaining_time": "0:12:45"}
130
+ {"current_steps": 126, "total_steps": 253, "loss": 1.2558, "lr": 5.1256504772166885e-06, "epoch": 0.4980237154150198, "percentage": 49.8, "elapsed_time": "0:12:33", "remaining_time": "0:12:39"}
131
+ {"current_steps": 127, "total_steps": 253, "loss": 1.0128, "lr": 5.062830199416764e-06, "epoch": 0.5019762845849802, "percentage": 50.2, "elapsed_time": "0:12:39", "remaining_time": "0:12:33"}
132
+ {"current_steps": 128, "total_steps": 253, "loss": 1.2622, "lr": 5e-06, "epoch": 0.5059288537549407, "percentage": 50.59, "elapsed_time": "0:12:45", "remaining_time": "0:12:27"}
133
+ {"current_steps": 129, "total_steps": 253, "loss": 1.3989, "lr": 4.937169800583237e-06, "epoch": 0.5098814229249012, "percentage": 50.99, "elapsed_time": "0:12:51", "remaining_time": "0:12:21"}
134
+ {"current_steps": 130, "total_steps": 253, "loss": 1.1264, "lr": 4.874349522783313e-06, "epoch": 0.5138339920948617, "percentage": 51.38, "elapsed_time": "0:12:57", "remaining_time": "0:12:15"}
135
+ {"current_steps": 130, "total_steps": 253, "eval_loss": 1.1763263940811157, "epoch": 0.5138339920948617, "percentage": 51.38, "elapsed_time": "0:12:57", "remaining_time": "0:12:15"}
136
+ {"current_steps": 131, "total_steps": 253, "loss": 1.2868, "lr": 4.811549086650327e-06, "epoch": 0.5177865612648221, "percentage": 51.78, "elapsed_time": "0:13:03", "remaining_time": "0:12:09"}
137
+ {"current_steps": 132, "total_steps": 253, "loss": 1.221, "lr": 4.748778409101153e-06, "epoch": 0.5217391304347826, "percentage": 52.17, "elapsed_time": "0:13:09", "remaining_time": "0:12:03"}
138
+ {"current_steps": 133, "total_steps": 253, "loss": 1.2466, "lr": 4.686047402353433e-06, "epoch": 0.525691699604743, "percentage": 52.57, "elapsed_time": "0:13:15", "remaining_time": "0:11:57"}
139
+ {"current_steps": 134, "total_steps": 253, "loss": 1.3095, "lr": 4.6233659723603374e-06, "epoch": 0.5296442687747036, "percentage": 52.96, "elapsed_time": "0:13:21", "remaining_time": "0:11:51"}
140
+ {"current_steps": 135, "total_steps": 253, "loss": 1.4281, "lr": 4.560744017246284e-06, "epoch": 0.5335968379446641, "percentage": 53.36, "elapsed_time": "0:13:27", "remaining_time": "0:11:45"}
141
+ {"current_steps": 136, "total_steps": 253, "loss": 1.1636, "lr": 4.4981914257439254e-06, "epoch": 0.5375494071146245, "percentage": 53.75, "elapsed_time": "0:13:33", "remaining_time": "0:11:39"}
142
+ {"current_steps": 137, "total_steps": 253, "loss": 1.2409, "lr": 4.4357180756325915e-06, "epoch": 0.541501976284585, "percentage": 54.15, "elapsed_time": "0:13:39", "remaining_time": "0:11:33"}
143
+ {"current_steps": 138, "total_steps": 253, "loss": 1.1857, "lr": 4.373333832178478e-06, "epoch": 0.5454545454545454, "percentage": 54.55, "elapsed_time": "0:13:45", "remaining_time": "0:11:27"}
144
+ {"current_steps": 139, "total_steps": 253, "loss": 1.1947, "lr": 4.31104854657681e-06, "epoch": 0.549407114624506, "percentage": 54.94, "elapsed_time": "0:13:51", "remaining_time": "0:11:21"}
145
+ {"current_steps": 140, "total_steps": 253, "loss": 1.1367, "lr": 4.248872054396215e-06, "epoch": 0.5533596837944664, "percentage": 55.34, "elapsed_time": "0:13:57", "remaining_time": "0:11:15"}
146
+ {"current_steps": 141, "total_steps": 253, "loss": 1.1754, "lr": 4.186814174025582e-06, "epoch": 0.5573122529644269, "percentage": 55.73, "elapsed_time": "0:14:03", "remaining_time": "0:11:09"}
147
+ {"current_steps": 142, "total_steps": 253, "loss": 1.1789, "lr": 4.124884705123619e-06, "epoch": 0.5612648221343873, "percentage": 56.13, "elapsed_time": "0:14:08", "remaining_time": "0:11:03"}
148
+ {"current_steps": 143, "total_steps": 253, "loss": 1.2189, "lr": 4.063093427071376e-06, "epoch": 0.5652173913043478, "percentage": 56.52, "elapsed_time": "0:14:14", "remaining_time": "0:10:57"}
149
+ {"current_steps": 144, "total_steps": 253, "loss": 1.3021, "lr": 4.001450097427965e-06, "epoch": 0.5691699604743083, "percentage": 56.92, "elapsed_time": "0:14:20", "remaining_time": "0:10:51"}
150
+ {"current_steps": 145, "total_steps": 253, "loss": 1.2536, "lr": 3.939964450389728e-06, "epoch": 0.5731225296442688, "percentage": 57.31, "elapsed_time": "0:14:26", "remaining_time": "0:10:45"}
151
+ {"current_steps": 146, "total_steps": 253, "loss": 1.2669, "lr": 3.8786461952530955e-06, "epoch": 0.5770750988142292, "percentage": 57.71, "elapsed_time": "0:14:32", "remaining_time": "0:10:39"}
152
+ {"current_steps": 147, "total_steps": 253, "loss": 1.2246, "lr": 3.817505014881378e-06, "epoch": 0.5810276679841897, "percentage": 58.1, "elapsed_time": "0:14:38", "remaining_time": "0:10:33"}
153
+ {"current_steps": 148, "total_steps": 253, "loss": 1.2705, "lr": 3.756550564175727e-06, "epoch": 0.5849802371541502, "percentage": 58.5, "elapsed_time": "0:14:44", "remaining_time": "0:10:27"}
154
+ {"current_steps": 149, "total_steps": 253, "loss": 1.3191, "lr": 3.695792468550517e-06, "epoch": 0.5889328063241107, "percentage": 58.89, "elapsed_time": "0:14:50", "remaining_time": "0:10:21"}
155
+ {"current_steps": 150, "total_steps": 253, "loss": 1.3388, "lr": 3.635240322413375e-06, "epoch": 0.5928853754940712, "percentage": 59.29, "elapsed_time": "0:14:56", "remaining_time": "0:10:15"}
156
+ {"current_steps": 151, "total_steps": 253, "loss": 1.2169, "lr": 3.5749036876501196e-06, "epoch": 0.5968379446640316, "percentage": 59.68, "elapsed_time": "0:15:02", "remaining_time": "0:10:09"}
157
+ {"current_steps": 152, "total_steps": 253, "loss": 1.4159, "lr": 3.5147920921148267e-06, "epoch": 0.6007905138339921, "percentage": 60.08, "elapsed_time": "0:15:08", "remaining_time": "0:10:03"}
158
+ {"current_steps": 153, "total_steps": 253, "loss": 1.3918, "lr": 3.4549150281252635e-06, "epoch": 0.6047430830039525, "percentage": 60.47, "elapsed_time": "0:15:14", "remaining_time": "0:09:57"}
159
+ {"current_steps": 154, "total_steps": 253, "loss": 1.0842, "lr": 3.3952819509639534e-06, "epoch": 0.6086956521739131, "percentage": 60.87, "elapsed_time": "0:15:20", "remaining_time": "0:09:51"}
160
+ {"current_steps": 155, "total_steps": 253, "loss": 1.3536, "lr": 3.3359022773850673e-06, "epoch": 0.6126482213438735, "percentage": 61.26, "elapsed_time": "0:15:26", "remaining_time": "0:09:45"}
161
+ {"current_steps": 156, "total_steps": 253, "loss": 1.1721, "lr": 3.2767853841274154e-06, "epoch": 0.616600790513834, "percentage": 61.66, "elapsed_time": "0:15:32", "remaining_time": "0:09:39"}
162
+ {"current_steps": 156, "total_steps": 253, "eval_loss": 1.1584229469299316, "epoch": 0.616600790513834, "percentage": 61.66, "elapsed_time": "0:15:33", "remaining_time": "0:09:40"}
163
+ {"current_steps": 157, "total_steps": 253, "loss": 1.0664, "lr": 3.217940606433747e-06, "epoch": 0.6205533596837944, "percentage": 62.06, "elapsed_time": "0:15:39", "remaining_time": "0:09:34"}
164
+ {"current_steps": 158, "total_steps": 253, "loss": 1.2761, "lr": 3.1593772365766107e-06, "epoch": 0.6245059288537549, "percentage": 62.45, "elapsed_time": "0:15:45", "remaining_time": "0:09:28"}
165
+ {"current_steps": 159, "total_steps": 253, "loss": 1.2633, "lr": 3.1011045223909954e-06, "epoch": 0.6284584980237155, "percentage": 62.85, "elapsed_time": "0:15:51", "remaining_time": "0:09:22"}
166
+ {"current_steps": 160, "total_steps": 253, "loss": 1.2919, "lr": 3.043131665813988e-06, "epoch": 0.6324110671936759, "percentage": 63.24, "elapsed_time": "0:15:56", "remaining_time": "0:09:16"}
167
+ {"current_steps": 161, "total_steps": 253, "loss": 1.1463, "lr": 2.9854678214316875e-06, "epoch": 0.6363636363636364, "percentage": 63.64, "elapsed_time": "0:16:02", "remaining_time": "0:09:10"}
168
+ {"current_steps": 162, "total_steps": 253, "loss": 1.0966, "lr": 2.92812209503358e-06, "epoch": 0.6403162055335968, "percentage": 64.03, "elapsed_time": "0:16:08", "remaining_time": "0:09:04"}
169
+ {"current_steps": 163, "total_steps": 253, "loss": 1.2748, "lr": 2.871103542174637e-06, "epoch": 0.6442687747035574, "percentage": 64.43, "elapsed_time": "0:16:14", "remaining_time": "0:08:58"}
170
+ {"current_steps": 164, "total_steps": 253, "loss": 1.3632, "lr": 2.814421166745337e-06, "epoch": 0.6482213438735178, "percentage": 64.82, "elapsed_time": "0:16:20", "remaining_time": "0:08:52"}
171
+ {"current_steps": 165, "total_steps": 253, "loss": 1.1984, "lr": 2.7580839195498397e-06, "epoch": 0.6521739130434783, "percentage": 65.22, "elapsed_time": "0:16:26", "remaining_time": "0:08:46"}
172
+ {"current_steps": 166, "total_steps": 253, "loss": 1.5325, "lr": 2.7021006968925613e-06, "epoch": 0.6561264822134387, "percentage": 65.61, "elapsed_time": "0:16:32", "remaining_time": "0:08:40"}
173
+ {"current_steps": 167, "total_steps": 253, "loss": 1.4192, "lr": 2.646480339173337e-06, "epoch": 0.6600790513833992, "percentage": 66.01, "elapsed_time": "0:16:38", "remaining_time": "0:08:34"}
174
+ {"current_steps": 168, "total_steps": 253, "loss": 1.2821, "lr": 2.5912316294914232e-06, "epoch": 0.6640316205533597, "percentage": 66.4, "elapsed_time": "0:16:44", "remaining_time": "0:08:28"}
175
+ {"current_steps": 169, "total_steps": 253, "loss": 1.3371, "lr": 2.536363292258543e-06, "epoch": 0.6679841897233202, "percentage": 66.8, "elapsed_time": "0:16:50", "remaining_time": "0:08:22"}
176
+ {"current_steps": 170, "total_steps": 253, "loss": 1.2789, "lr": 2.4818839918211963e-06, "epoch": 0.6719367588932806, "percentage": 67.19, "elapsed_time": "0:16:56", "remaining_time": "0:08:16"}
177
+ {"current_steps": 171, "total_steps": 253, "loss": 1.2545, "lr": 2.4278023310924676e-06, "epoch": 0.6758893280632411, "percentage": 67.59, "elapsed_time": "0:17:02", "remaining_time": "0:08:10"}
178
+ {"current_steps": 172, "total_steps": 253, "loss": 0.9872, "lr": 2.3741268501935212e-06, "epoch": 0.6798418972332015, "percentage": 67.98, "elapsed_time": "0:17:08", "remaining_time": "0:08:04"}
179
+ {"current_steps": 173, "total_steps": 253, "loss": 1.0876, "lr": 2.320866025105016e-06, "epoch": 0.6837944664031621, "percentage": 68.38, "elapsed_time": "0:17:13", "remaining_time": "0:07:58"}
180
+ {"current_steps": 174, "total_steps": 253, "loss": 1.2257, "lr": 2.268028266328655e-06, "epoch": 0.6877470355731226, "percentage": 68.77, "elapsed_time": "0:17:19", "remaining_time": "0:07:52"}
181
+ {"current_steps": 175, "total_steps": 253, "loss": 1.1499, "lr": 2.2156219175590623e-06, "epoch": 0.691699604743083, "percentage": 69.17, "elapsed_time": "0:17:25", "remaining_time": "0:07:46"}
182
+ {"current_steps": 176, "total_steps": 253, "loss": 1.3324, "lr": 2.1636552543662187e-06, "epoch": 0.6956521739130435, "percentage": 69.57, "elapsed_time": "0:17:31", "remaining_time": "0:07:40"}
183
+ {"current_steps": 177, "total_steps": 253, "loss": 1.3478, "lr": 2.112136482888663e-06, "epoch": 0.6996047430830039, "percentage": 69.96, "elapsed_time": "0:17:37", "remaining_time": "0:07:34"}
184
+ {"current_steps": 178, "total_steps": 253, "loss": 1.2586, "lr": 2.061073738537635e-06, "epoch": 0.7035573122529645, "percentage": 70.36, "elapsed_time": "0:17:43", "remaining_time": "0:07:28"}
185
+ {"current_steps": 179, "total_steps": 253, "loss": 1.2455, "lr": 2.0104750847124075e-06, "epoch": 0.7075098814229249, "percentage": 70.75, "elapsed_time": "0:17:49", "remaining_time": "0:07:22"}
186
+ {"current_steps": 180, "total_steps": 253, "loss": 1.2892, "lr": 1.9603485115269743e-06, "epoch": 0.7114624505928854, "percentage": 71.15, "elapsed_time": "0:17:55", "remaining_time": "0:07:16"}
187
+ {"current_steps": 181, "total_steps": 253, "loss": 1.1216, "lr": 1.910701934548329e-06, "epoch": 0.7154150197628458, "percentage": 71.54, "elapsed_time": "0:18:01", "remaining_time": "0:07:10"}
188
+ {"current_steps": 182, "total_steps": 253, "loss": 1.2738, "lr": 1.8615431935464984e-06, "epoch": 0.7193675889328063, "percentage": 71.94, "elapsed_time": "0:18:07", "remaining_time": "0:07:04"}
189
+ {"current_steps": 182, "total_steps": 253, "eval_loss": 1.1467477083206177, "epoch": 0.7193675889328063, "percentage": 71.94, "elapsed_time": "0:18:07", "remaining_time": "0:07:04"}
190
+ {"current_steps": 183, "total_steps": 253, "loss": 1.3237, "lr": 1.8128800512565514e-06, "epoch": 0.7233201581027668, "percentage": 72.33, "elapsed_time": "0:18:13", "remaining_time": "0:06:58"}
191
+ {"current_steps": 184, "total_steps": 253, "loss": 1.1454, "lr": 1.7647201921527802e-06, "epoch": 0.7272727272727273, "percentage": 72.73, "elapsed_time": "0:18:19", "remaining_time": "0:06:52"}
192
+ {"current_steps": 185, "total_steps": 253, "loss": 1.5128, "lr": 1.7170712212352187e-06, "epoch": 0.7312252964426877, "percentage": 73.12, "elapsed_time": "0:18:25", "remaining_time": "0:06:46"}
193
+ {"current_steps": 186, "total_steps": 253, "loss": 0.9779, "lr": 1.6699406628287423e-06, "epoch": 0.7351778656126482, "percentage": 73.52, "elapsed_time": "0:18:31", "remaining_time": "0:06:40"}
194
+ {"current_steps": 187, "total_steps": 253, "loss": 0.9658, "lr": 1.6233359593948777e-06, "epoch": 0.7391304347826086, "percentage": 73.91, "elapsed_time": "0:18:37", "remaining_time": "0:06:34"}
195
+ {"current_steps": 188, "total_steps": 253, "loss": 1.2752, "lr": 1.5772644703565564e-06, "epoch": 0.7430830039525692, "percentage": 74.31, "elapsed_time": "0:18:43", "remaining_time": "0:06:28"}
196
+ {"current_steps": 189, "total_steps": 253, "loss": 1.1755, "lr": 1.531733470935976e-06, "epoch": 0.7470355731225297, "percentage": 74.7, "elapsed_time": "0:18:49", "remaining_time": "0:06:22"}
197
+ {"current_steps": 190, "total_steps": 253, "loss": 1.1529, "lr": 1.4867501510057548e-06, "epoch": 0.7509881422924901, "percentage": 75.1, "elapsed_time": "0:18:54", "remaining_time": "0:06:16"}
198
+ {"current_steps": 191, "total_steps": 253, "loss": 1.2361, "lr": 1.4423216139535735e-06, "epoch": 0.7549407114624506, "percentage": 75.49, "elapsed_time": "0:19:00", "remaining_time": "0:06:10"}
199
+ {"current_steps": 192, "total_steps": 253, "loss": 1.2842, "lr": 1.3984548755604655e-06, "epoch": 0.758893280632411, "percentage": 75.89, "elapsed_time": "0:19:06", "remaining_time": "0:06:04"}
200
+ {"current_steps": 193, "total_steps": 253, "loss": 1.256, "lr": 1.3551568628929434e-06, "epoch": 0.7628458498023716, "percentage": 76.28, "elapsed_time": "0:19:12", "remaining_time": "0:05:58"}
201
+ {"current_steps": 194, "total_steps": 253, "loss": 1.1356, "lr": 1.312434413209131e-06, "epoch": 0.766798418972332, "percentage": 76.68, "elapsed_time": "0:19:18", "remaining_time": "0:05:52"}
202
+ {"current_steps": 195, "total_steps": 253, "loss": 1.151, "lr": 1.2702942728790897e-06, "epoch": 0.7707509881422925, "percentage": 77.08, "elapsed_time": "0:19:24", "remaining_time": "0:05:46"}
203
+ {"current_steps": 196, "total_steps": 253, "loss": 1.3267, "lr": 1.2287430963194807e-06, "epoch": 0.7747035573122529, "percentage": 77.47, "elapsed_time": "0:19:30", "remaining_time": "0:05:40"}
204
+ {"current_steps": 197, "total_steps": 253, "loss": 1.1943, "lr": 1.18778744494276e-06, "epoch": 0.7786561264822134, "percentage": 77.87, "elapsed_time": "0:19:36", "remaining_time": "0:05:34"}
205
+ {"current_steps": 198, "total_steps": 253, "loss": 1.4392, "lr": 1.1474337861210543e-06, "epoch": 0.782608695652174, "percentage": 78.26, "elapsed_time": "0:19:42", "remaining_time": "0:05:28"}
206
+ {"current_steps": 199, "total_steps": 253, "loss": 1.2678, "lr": 1.1076884921648834e-06, "epoch": 0.7865612648221344, "percentage": 78.66, "elapsed_time": "0:19:47", "remaining_time": "0:05:22"}
207
+ {"current_steps": 200, "total_steps": 253, "loss": 1.2738, "lr": 1.0685578393169054e-06, "epoch": 0.7905138339920948, "percentage": 79.05, "elapsed_time": "0:19:53", "remaining_time": "0:05:16"}
208
+ {"current_steps": 201, "total_steps": 253, "loss": 1.4097, "lr": 1.0300480067608232e-06, "epoch": 0.7944664031620553, "percentage": 79.45, "elapsed_time": "0:19:59", "remaining_time": "0:05:10"}
209
+ {"current_steps": 202, "total_steps": 253, "loss": 1.1463, "lr": 9.921650756456164e-07, "epoch": 0.7984189723320159, "percentage": 79.84, "elapsed_time": "0:20:05", "remaining_time": "0:05:04"}
210
+ {"current_steps": 203, "total_steps": 253, "loss": 1.3288, "lr": 9.549150281252633e-07, "epoch": 0.8023715415019763, "percentage": 80.24, "elapsed_time": "0:20:11", "remaining_time": "0:04:58"}
211
+ {"current_steps": 204, "total_steps": 253, "loss": 1.1852, "lr": 9.183037464140804e-07, "epoch": 0.8063241106719368, "percentage": 80.63, "elapsed_time": "0:20:17", "remaining_time": "0:04:52"}
212
+ {"current_steps": 205, "total_steps": 253, "loss": 1.2328, "lr": 8.823370118578628e-07, "epoch": 0.8102766798418972, "percentage": 81.03, "elapsed_time": "0:20:23", "remaining_time": "0:04:46"}
213
+ {"current_steps": 206, "total_steps": 253, "loss": 1.2409, "lr": 8.470205040209362e-07, "epoch": 0.8142292490118577, "percentage": 81.42, "elapsed_time": "0:20:29", "remaining_time": "0:04:40"}
214
+ {"current_steps": 207, "total_steps": 253, "loss": 1.4195, "lr": 8.123597997892918e-07, "epoch": 0.8181818181818182, "percentage": 81.82, "elapsed_time": "0:20:35", "remaining_time": "0:04:34"}
215
+ {"current_steps": 208, "total_steps": 253, "loss": 1.1729, "lr": 7.783603724899258e-07, "epoch": 0.8221343873517787, "percentage": 82.21, "elapsed_time": "0:20:41", "remaining_time": "0:04:28"}
216
+ {"current_steps": 208, "total_steps": 253, "eval_loss": 1.1414591073989868, "epoch": 0.8221343873517787, "percentage": 82.21, "elapsed_time": "0:20:41", "remaining_time": "0:04:28"}
217
+ {"current_steps": 209, "total_steps": 253, "loss": 1.1787, "lr": 7.450275910265415e-07, "epoch": 0.8260869565217391, "percentage": 82.61, "elapsed_time": "0:20:47", "remaining_time": "0:04:22"}
218
+ {"current_steps": 210, "total_steps": 253, "loss": 1.3486, "lr": 7.123667190317396e-07, "epoch": 0.8300395256916996, "percentage": 83.0, "elapsed_time": "0:20:53", "remaining_time": "0:04:16"}
219
+ {"current_steps": 211, "total_steps": 253, "loss": 1.2067, "lr": 6.803829140358237e-07, "epoch": 0.83399209486166, "percentage": 83.4, "elapsed_time": "0:20:59", "remaining_time": "0:04:10"}
220
+ {"current_steps": 212, "total_steps": 253, "loss": 1.0929, "lr": 6.490812266523716e-07, "epoch": 0.8379446640316206, "percentage": 83.79, "elapsed_time": "0:21:05", "remaining_time": "0:04:04"}
221
+ {"current_steps": 213, "total_steps": 253, "loss": 1.0544, "lr": 6.184665997806832e-07, "epoch": 0.841897233201581, "percentage": 84.19, "elapsed_time": "0:21:11", "remaining_time": "0:03:58"}
222
+ {"current_steps": 214, "total_steps": 253, "loss": 1.2409, "lr": 5.885438678252342e-07, "epoch": 0.8458498023715415, "percentage": 84.58, "elapsed_time": "0:21:17", "remaining_time": "0:03:52"}
223
+ {"current_steps": 215, "total_steps": 253, "loss": 1.3816, "lr": 5.593177559322776e-07, "epoch": 0.849802371541502, "percentage": 84.98, "elapsed_time": "0:21:23", "remaining_time": "0:03:46"}
224
+ {"current_steps": 216, "total_steps": 253, "loss": 1.0972, "lr": 5.307928792436812e-07, "epoch": 0.8537549407114624, "percentage": 85.38, "elapsed_time": "0:21:29", "remaining_time": "0:03:40"}
225
+ {"current_steps": 217, "total_steps": 253, "loss": 1.3786, "lr": 5.029737421681446e-07, "epoch": 0.857707509881423, "percentage": 85.77, "elapsed_time": "0:21:35", "remaining_time": "0:03:34"}
226
+ {"current_steps": 218, "total_steps": 253, "loss": 1.3592, "lr": 4.758647376699033e-07, "epoch": 0.8616600790513834, "percentage": 86.17, "elapsed_time": "0:21:40", "remaining_time": "0:03:28"}
227
+ {"current_steps": 219, "total_steps": 253, "loss": 1.3265, "lr": 4.494701465750217e-07, "epoch": 0.8656126482213439, "percentage": 86.56, "elapsed_time": "0:21:46", "remaining_time": "0:03:22"}
228
+ {"current_steps": 220, "total_steps": 253, "loss": 1.1705, "lr": 4.237941368954124e-07, "epoch": 0.8695652173913043, "percentage": 86.96, "elapsed_time": "0:21:52", "remaining_time": "0:03:16"}
229
+ {"current_steps": 221, "total_steps": 253, "loss": 1.2725, "lr": 3.9884076317064813e-07, "epoch": 0.8735177865612648, "percentage": 87.35, "elapsed_time": "0:21:58", "remaining_time": "0:03:10"}
230
+ {"current_steps": 222, "total_steps": 253, "loss": 1.2205, "lr": 3.7461396582771035e-07, "epoch": 0.8774703557312253, "percentage": 87.75, "elapsed_time": "0:22:04", "remaining_time": "0:03:04"}
231
+ {"current_steps": 223, "total_steps": 253, "loss": 1.2246, "lr": 3.511175705587433e-07, "epoch": 0.8814229249011858, "percentage": 88.14, "elapsed_time": "0:22:10", "remaining_time": "0:02:58"}
232
+ {"current_steps": 224, "total_steps": 253, "loss": 1.0818, "lr": 3.283552877169399e-07, "epoch": 0.8853754940711462, "percentage": 88.54, "elapsed_time": "0:22:16", "remaining_time": "0:02:53"}
233
+ {"current_steps": 225, "total_steps": 253, "loss": 1.328, "lr": 3.0633071173062966e-07, "epoch": 0.8893280632411067, "percentage": 88.93, "elapsed_time": "0:22:22", "remaining_time": "0:02:47"}
234
+ {"current_steps": 226, "total_steps": 253, "loss": 1.0848, "lr": 2.850473205356774e-07, "epoch": 0.8932806324110671, "percentage": 89.33, "elapsed_time": "0:22:28", "remaining_time": "0:02:41"}
235
+ {"current_steps": 227, "total_steps": 253, "loss": 1.3192, "lr": 2.6450847502627883e-07, "epoch": 0.8972332015810277, "percentage": 89.72, "elapsed_time": "0:22:34", "remaining_time": "0:02:35"}
236
+ {"current_steps": 228, "total_steps": 253, "loss": 1.2458, "lr": 2.447174185242324e-07, "epoch": 0.9011857707509882, "percentage": 90.12, "elapsed_time": "0:22:40", "remaining_time": "0:02:29"}
237
+ {"current_steps": 229, "total_steps": 253, "loss": 1.2568, "lr": 2.2567727626678527e-07, "epoch": 0.9051383399209486, "percentage": 90.51, "elapsed_time": "0:22:45", "remaining_time": "0:02:23"}
238
+ {"current_steps": 230, "total_steps": 253, "loss": 1.4233, "lr": 2.0739105491312028e-07, "epoch": 0.9090909090909091, "percentage": 90.91, "elapsed_time": "0:22:51", "remaining_time": "0:02:17"}
239
+ {"current_steps": 231, "total_steps": 253, "loss": 1.3435, "lr": 1.8986164206957037e-07, "epoch": 0.9130434782608695, "percentage": 91.3, "elapsed_time": "0:22:57", "remaining_time": "0:02:11"}
240
+ {"current_steps": 232, "total_steps": 253, "loss": 1.1919, "lr": 1.7309180583363062e-07, "epoch": 0.9169960474308301, "percentage": 91.7, "elapsed_time": "0:23:03", "remaining_time": "0:02:05"}
241
+ {"current_steps": 233, "total_steps": 253, "loss": 1.0833, "lr": 1.5708419435684463e-07, "epoch": 0.9209486166007905, "percentage": 92.09, "elapsed_time": "0:23:09", "remaining_time": "0:01:59"}
242
+ {"current_steps": 234, "total_steps": 253, "loss": 1.4446, "lr": 1.4184133542663014e-07, "epoch": 0.924901185770751, "percentage": 92.49, "elapsed_time": "0:23:15", "remaining_time": "0:01:53"}
243
+ {"current_steps": 234, "total_steps": 253, "eval_loss": 1.1398719549179077, "epoch": 0.924901185770751, "percentage": 92.49, "elapsed_time": "0:23:16", "remaining_time": "0:01:53"}
244
+ {"current_steps": 235, "total_steps": 253, "loss": 1.3058, "lr": 1.2736563606711384e-07, "epoch": 0.9288537549407114, "percentage": 92.89, "elapsed_time": "0:23:22", "remaining_time": "0:01:47"}
245
+ {"current_steps": 236, "total_steps": 253, "loss": 1.1822, "lr": 1.136593821590326e-07, "epoch": 0.932806324110672, "percentage": 93.28, "elapsed_time": "0:23:28", "remaining_time": "0:01:41"}
246
+ {"current_steps": 237, "total_steps": 253, "loss": 1.2477, "lr": 1.007247380787657e-07, "epoch": 0.9367588932806324, "percentage": 93.68, "elapsed_time": "0:23:33", "remaining_time": "0:01:35"}
247
+ {"current_steps": 238, "total_steps": 253, "loss": 0.9579, "lr": 8.856374635655696e-08, "epoch": 0.9407114624505929, "percentage": 94.07, "elapsed_time": "0:23:39", "remaining_time": "0:01:29"}
248
+ {"current_steps": 239, "total_steps": 253, "loss": 1.2895, "lr": 7.717832735397335e-08, "epoch": 0.9446640316205533, "percentage": 94.47, "elapsed_time": "0:23:45", "remaining_time": "0:01:23"}
249
+ {"current_steps": 240, "total_steps": 253, "loss": 1.2901, "lr": 6.657027896065982e-08, "epoch": 0.9486166007905138, "percentage": 94.86, "elapsed_time": "0:23:51", "remaining_time": "0:01:17"}
250
+ {"current_steps": 241, "total_steps": 253, "loss": 1.1185, "lr": 5.674127631043025e-08, "epoch": 0.9525691699604744, "percentage": 95.26, "elapsed_time": "0:23:57", "remaining_time": "0:01:11"}
251
+ {"current_steps": 242, "total_steps": 253, "loss": 1.1315, "lr": 4.769287151674407e-08, "epoch": 0.9565217391304348, "percentage": 95.65, "elapsed_time": "0:24:03", "remaining_time": "0:01:05"}
252
+ {"current_steps": 243, "total_steps": 253, "loss": 1.3511, "lr": 3.9426493427611177e-08, "epoch": 0.9604743083003953, "percentage": 96.05, "elapsed_time": "0:24:09", "remaining_time": "0:00:59"}
253
+ {"current_steps": 244, "total_steps": 253, "loss": 1.2501, "lr": 3.194344739995803e-08, "epoch": 0.9644268774703557, "percentage": 96.44, "elapsed_time": "0:24:15", "remaining_time": "0:00:53"}
254
+ {"current_steps": 245, "total_steps": 253, "loss": 1.1408, "lr": 2.5244915093499134e-08, "epoch": 0.9683794466403162, "percentage": 96.84, "elapsed_time": "0:24:21", "remaining_time": "0:00:47"}
255
+ {"current_steps": 246, "total_steps": 253, "loss": 1.2169, "lr": 1.9331954284137476e-08, "epoch": 0.9723320158102767, "percentage": 97.23, "elapsed_time": "0:24:27", "remaining_time": "0:00:41"}
256
+ {"current_steps": 247, "total_steps": 253, "loss": 1.4156, "lr": 1.4205498696930332e-08, "epoch": 0.9762845849802372, "percentage": 97.63, "elapsed_time": "0:24:33", "remaining_time": "0:00:35"}
257
+ {"current_steps": 248, "total_steps": 253, "loss": 1.1922, "lr": 9.866357858642206e-09, "epoch": 0.9802371541501976, "percentage": 98.02, "elapsed_time": "0:24:39", "remaining_time": "0:00:29"}
258
+ {"current_steps": 249, "total_steps": 253, "loss": 1.4098, "lr": 6.315216969912663e-09, "epoch": 0.9841897233201581, "percentage": 98.42, "elapsed_time": "0:24:45", "remaining_time": "0:00:23"}
259
+ {"current_steps": 250, "total_steps": 253, "loss": 1.3601, "lr": 3.5526367970539765e-09, "epoch": 0.9881422924901185, "percentage": 98.81, "elapsed_time": "0:24:51", "remaining_time": "0:00:17"}
260
+ {"current_steps": 251, "total_steps": 253, "loss": 1.2039, "lr": 1.5790535835003006e-09, "epoch": 0.9920948616600791, "percentage": 99.21, "elapsed_time": "0:24:57", "remaining_time": "0:00:11"}
261
+ {"current_steps": 252, "total_steps": 253, "loss": 1.1514, "lr": 3.9477898091944135e-10, "epoch": 0.9960474308300395, "percentage": 99.6, "elapsed_time": "0:25:02", "remaining_time": "0:00:05"}
262
+ {"current_steps": 253, "total_steps": 253, "loss": 1.2887, "lr": 0.0, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "0:25:08", "remaining_time": "0:00:00"}
263
+ {"current_steps": 253, "total_steps": 253, "epoch": 1.0, "percentage": 100.0, "elapsed_time": "0:25:20", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,1885 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 26,
6
+ "global_step": 253,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.003952569169960474,
13
+ "grad_norm": 3.0407819747924805,
14
+ "learning_rate": 3.3333333333333333e-06,
15
+ "loss": 2.3652,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.007905138339920948,
20
+ "grad_norm": 3.093733310699463,
21
+ "learning_rate": 6.666666666666667e-06,
22
+ "loss": 2.1458,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.011857707509881422,
27
+ "grad_norm": 3.0263988971710205,
28
+ "learning_rate": 1e-05,
29
+ "loss": 2.2628,
30
+ "step": 3
31
+ },
32
+ {
33
+ "epoch": 0.015810276679841896,
34
+ "grad_norm": 2.6713979244232178,
35
+ "learning_rate": 9.999605221019082e-06,
36
+ "loss": 2.0308,
37
+ "step": 4
38
+ },
39
+ {
40
+ "epoch": 0.019762845849802372,
41
+ "grad_norm": 2.4664857387542725,
42
+ "learning_rate": 9.9984209464165e-06,
43
+ "loss": 1.6917,
44
+ "step": 5
45
+ },
46
+ {
47
+ "epoch": 0.023715415019762844,
48
+ "grad_norm": 2.100984811782837,
49
+ "learning_rate": 9.996447363202947e-06,
50
+ "loss": 1.8489,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 0.02766798418972332,
55
+ "grad_norm": 1.7837852239608765,
56
+ "learning_rate": 9.99368478303009e-06,
57
+ "loss": 1.9894,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 0.03162055335968379,
62
+ "grad_norm": 1.6338787078857422,
63
+ "learning_rate": 9.990133642141359e-06,
64
+ "loss": 1.8366,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 0.03557312252964427,
69
+ "grad_norm": 1.7102547883987427,
70
+ "learning_rate": 9.98579450130307e-06,
71
+ "loss": 1.9013,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 0.039525691699604744,
76
+ "grad_norm": 1.8172574043273926,
77
+ "learning_rate": 9.980668045715864e-06,
78
+ "loss": 2.034,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 0.043478260869565216,
83
+ "grad_norm": 1.7970800399780273,
84
+ "learning_rate": 9.974755084906503e-06,
85
+ "loss": 1.8565,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 0.04743083003952569,
90
+ "grad_norm": 1.9662507772445679,
91
+ "learning_rate": 9.968056552600043e-06,
92
+ "loss": 1.8154,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 0.05138339920948617,
97
+ "grad_norm": 1.6226023435592651,
98
+ "learning_rate": 9.960573506572391e-06,
99
+ "loss": 1.7205,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 0.05533596837944664,
104
+ "grad_norm": 1.4996356964111328,
105
+ "learning_rate": 9.952307128483257e-06,
106
+ "loss": 1.8481,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 0.05928853754940711,
111
+ "grad_norm": 1.2719931602478027,
112
+ "learning_rate": 9.94325872368957e-06,
113
+ "loss": 1.5792,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 0.06324110671936758,
118
+ "grad_norm": 1.2413394451141357,
119
+ "learning_rate": 9.93342972103934e-06,
120
+ "loss": 1.6527,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 0.06719367588932806,
125
+ "grad_norm": 1.1712374687194824,
126
+ "learning_rate": 9.922821672646028e-06,
127
+ "loss": 1.3883,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 0.07114624505928854,
132
+ "grad_norm": 1.1695728302001953,
133
+ "learning_rate": 9.911436253643445e-06,
134
+ "loss": 1.5297,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 0.07509881422924901,
139
+ "grad_norm": 1.1794648170471191,
140
+ "learning_rate": 9.899275261921236e-06,
141
+ "loss": 1.5525,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 0.07905138339920949,
146
+ "grad_norm": 1.131502628326416,
147
+ "learning_rate": 9.886340617840968e-06,
148
+ "loss": 1.5752,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 0.08300395256916997,
153
+ "grad_norm": 1.1554540395736694,
154
+ "learning_rate": 9.872634363932887e-06,
155
+ "loss": 1.5944,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 0.08695652173913043,
160
+ "grad_norm": 1.1099374294281006,
161
+ "learning_rate": 9.85815866457337e-06,
162
+ "loss": 1.4019,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 0.09090909090909091,
167
+ "grad_norm": 1.2356349229812622,
168
+ "learning_rate": 9.842915805643156e-06,
169
+ "loss": 1.706,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 0.09486166007905138,
174
+ "grad_norm": 1.1689324378967285,
175
+ "learning_rate": 9.82690819416637e-06,
176
+ "loss": 1.6249,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 0.09881422924901186,
181
+ "grad_norm": 1.2375783920288086,
182
+ "learning_rate": 9.81013835793043e-06,
183
+ "loss": 1.8374,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 0.10276679841897234,
188
+ "grad_norm": 1.1797913312911987,
189
+ "learning_rate": 9.79260894508688e-06,
190
+ "loss": 1.6161,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 0.10276679841897234,
195
+ "eval_loss": 1.4920315742492676,
196
+ "eval_runtime": 0.6276,
197
+ "eval_samples_per_second": 33.462,
198
+ "eval_steps_per_second": 4.78,
199
+ "step": 26
200
+ },
201
+ {
202
+ "epoch": 0.1067193675889328,
203
+ "grad_norm": 1.1279163360595703,
204
+ "learning_rate": 9.774322723733216e-06,
205
+ "loss": 1.6593,
206
+ "step": 27
207
+ },
208
+ {
209
+ "epoch": 0.11067193675889328,
210
+ "grad_norm": 1.1455029249191284,
211
+ "learning_rate": 9.755282581475769e-06,
212
+ "loss": 1.6982,
213
+ "step": 28
214
+ },
215
+ {
216
+ "epoch": 0.11462450592885376,
217
+ "grad_norm": 1.0429457426071167,
218
+ "learning_rate": 9.735491524973723e-06,
219
+ "loss": 1.4423,
220
+ "step": 29
221
+ },
222
+ {
223
+ "epoch": 0.11857707509881422,
224
+ "grad_norm": 0.9471132755279541,
225
+ "learning_rate": 9.714952679464324e-06,
226
+ "loss": 1.3055,
227
+ "step": 30
228
+ },
229
+ {
230
+ "epoch": 0.1225296442687747,
231
+ "grad_norm": 1.0985954999923706,
232
+ "learning_rate": 9.693669288269371e-06,
233
+ "loss": 1.6302,
234
+ "step": 31
235
+ },
236
+ {
237
+ "epoch": 0.12648221343873517,
238
+ "grad_norm": 1.11372709274292,
239
+ "learning_rate": 9.671644712283061e-06,
240
+ "loss": 1.7975,
241
+ "step": 32
242
+ },
243
+ {
244
+ "epoch": 0.13043478260869565,
245
+ "grad_norm": 0.9115369915962219,
246
+ "learning_rate": 9.648882429441258e-06,
247
+ "loss": 1.2709,
248
+ "step": 33
249
+ },
250
+ {
251
+ "epoch": 0.13438735177865613,
252
+ "grad_norm": 1.0038697719573975,
253
+ "learning_rate": 9.62538603417229e-06,
254
+ "loss": 1.4833,
255
+ "step": 34
256
+ },
257
+ {
258
+ "epoch": 0.1383399209486166,
259
+ "grad_norm": 1.0109611749649048,
260
+ "learning_rate": 9.601159236829353e-06,
261
+ "loss": 1.2978,
262
+ "step": 35
263
+ },
264
+ {
265
+ "epoch": 0.1422924901185771,
266
+ "grad_norm": 0.966667115688324,
267
+ "learning_rate": 9.576205863104588e-06,
268
+ "loss": 1.5622,
269
+ "step": 36
270
+ },
271
+ {
272
+ "epoch": 0.14624505928853754,
273
+ "grad_norm": 1.0922729969024658,
274
+ "learning_rate": 9.550529853424979e-06,
275
+ "loss": 1.7766,
276
+ "step": 37
277
+ },
278
+ {
279
+ "epoch": 0.15019762845849802,
280
+ "grad_norm": 1.010776162147522,
281
+ "learning_rate": 9.524135262330098e-06,
282
+ "loss": 1.5206,
283
+ "step": 38
284
+ },
285
+ {
286
+ "epoch": 0.1541501976284585,
287
+ "grad_norm": 0.9982693791389465,
288
+ "learning_rate": 9.497026257831856e-06,
289
+ "loss": 1.5341,
290
+ "step": 39
291
+ },
292
+ {
293
+ "epoch": 0.15810276679841898,
294
+ "grad_norm": 0.9735735058784485,
295
+ "learning_rate": 9.46920712075632e-06,
296
+ "loss": 1.4157,
297
+ "step": 40
298
+ },
299
+ {
300
+ "epoch": 0.16205533596837945,
301
+ "grad_norm": 0.9151104092597961,
302
+ "learning_rate": 9.440682244067724e-06,
303
+ "loss": 1.2846,
304
+ "step": 41
305
+ },
306
+ {
307
+ "epoch": 0.16600790513833993,
308
+ "grad_norm": 0.9795275330543518,
309
+ "learning_rate": 9.411456132174768e-06,
310
+ "loss": 1.4727,
311
+ "step": 42
312
+ },
313
+ {
314
+ "epoch": 0.16996047430830039,
315
+ "grad_norm": 1.1112526655197144,
316
+ "learning_rate": 9.381533400219319e-06,
317
+ "loss": 1.6619,
318
+ "step": 43
319
+ },
320
+ {
321
+ "epoch": 0.17391304347826086,
322
+ "grad_norm": 0.9755843877792358,
323
+ "learning_rate": 9.35091877334763e-06,
324
+ "loss": 1.4687,
325
+ "step": 44
326
+ },
327
+ {
328
+ "epoch": 0.17786561264822134,
329
+ "grad_norm": 0.9589361548423767,
330
+ "learning_rate": 9.319617085964177e-06,
331
+ "loss": 1.4871,
332
+ "step": 45
333
+ },
334
+ {
335
+ "epoch": 0.18181818181818182,
336
+ "grad_norm": 1.105637788772583,
337
+ "learning_rate": 9.287633280968263e-06,
338
+ "loss": 1.2694,
339
+ "step": 46
340
+ },
341
+ {
342
+ "epoch": 0.1857707509881423,
343
+ "grad_norm": 0.9662937521934509,
344
+ "learning_rate": 9.25497240897346e-06,
345
+ "loss": 1.2148,
346
+ "step": 47
347
+ },
348
+ {
349
+ "epoch": 0.18972332015810275,
350
+ "grad_norm": 1.0539385080337524,
351
+ "learning_rate": 9.221639627510076e-06,
352
+ "loss": 1.5932,
353
+ "step": 48
354
+ },
355
+ {
356
+ "epoch": 0.19367588932806323,
357
+ "grad_norm": 0.9249448180198669,
358
+ "learning_rate": 9.18764020021071e-06,
359
+ "loss": 1.3074,
360
+ "step": 49
361
+ },
362
+ {
363
+ "epoch": 0.1976284584980237,
364
+ "grad_norm": 1.0217876434326172,
365
+ "learning_rate": 9.152979495979064e-06,
366
+ "loss": 1.5327,
367
+ "step": 50
368
+ },
369
+ {
370
+ "epoch": 0.2015810276679842,
371
+ "grad_norm": 1.070178747177124,
372
+ "learning_rate": 9.117662988142138e-06,
373
+ "loss": 1.6561,
374
+ "step": 51
375
+ },
376
+ {
377
+ "epoch": 0.20553359683794467,
378
+ "grad_norm": 0.9971439242362976,
379
+ "learning_rate": 9.08169625358592e-06,
380
+ "loss": 1.3683,
381
+ "step": 52
382
+ },
383
+ {
384
+ "epoch": 0.20553359683794467,
385
+ "eval_loss": 1.325599193572998,
386
+ "eval_runtime": 0.6316,
387
+ "eval_samples_per_second": 33.25,
388
+ "eval_steps_per_second": 4.75,
389
+ "step": 52
390
+ },
391
+ {
392
+ "epoch": 0.20948616600790515,
393
+ "grad_norm": 0.9724494814872742,
394
+ "learning_rate": 9.045084971874738e-06,
395
+ "loss": 1.5192,
396
+ "step": 53
397
+ },
398
+ {
399
+ "epoch": 0.2134387351778656,
400
+ "grad_norm": 0.9346773624420166,
401
+ "learning_rate": 9.007834924354384e-06,
402
+ "loss": 1.4107,
403
+ "step": 54
404
+ },
405
+ {
406
+ "epoch": 0.21739130434782608,
407
+ "grad_norm": 0.9584831595420837,
408
+ "learning_rate": 8.969951993239177e-06,
409
+ "loss": 1.392,
410
+ "step": 55
411
+ },
412
+ {
413
+ "epoch": 0.22134387351778656,
414
+ "grad_norm": 0.8788594603538513,
415
+ "learning_rate": 8.931442160683094e-06,
416
+ "loss": 1.2833,
417
+ "step": 56
418
+ },
419
+ {
420
+ "epoch": 0.22529644268774704,
421
+ "grad_norm": 0.9413896203041077,
422
+ "learning_rate": 8.892311507835118e-06,
423
+ "loss": 1.26,
424
+ "step": 57
425
+ },
426
+ {
427
+ "epoch": 0.22924901185770752,
428
+ "grad_norm": 0.944340169429779,
429
+ "learning_rate": 8.852566213878947e-06,
430
+ "loss": 1.3223,
431
+ "step": 58
432
+ },
433
+ {
434
+ "epoch": 0.233201581027668,
435
+ "grad_norm": 0.8935737609863281,
436
+ "learning_rate": 8.81221255505724e-06,
437
+ "loss": 1.3826,
438
+ "step": 59
439
+ },
440
+ {
441
+ "epoch": 0.23715415019762845,
442
+ "grad_norm": 0.9506601095199585,
443
+ "learning_rate": 8.77125690368052e-06,
444
+ "loss": 1.4478,
445
+ "step": 60
446
+ },
447
+ {
448
+ "epoch": 0.24110671936758893,
449
+ "grad_norm": 0.9523410201072693,
450
+ "learning_rate": 8.729705727120911e-06,
451
+ "loss": 1.3503,
452
+ "step": 61
453
+ },
454
+ {
455
+ "epoch": 0.2450592885375494,
456
+ "grad_norm": 0.8316011428833008,
457
+ "learning_rate": 8.68756558679087e-06,
458
+ "loss": 1.041,
459
+ "step": 62
460
+ },
461
+ {
462
+ "epoch": 0.2490118577075099,
463
+ "grad_norm": 0.9092923998832703,
464
+ "learning_rate": 8.644843137107058e-06,
465
+ "loss": 1.323,
466
+ "step": 63
467
+ },
468
+ {
469
+ "epoch": 0.25296442687747034,
470
+ "grad_norm": 0.9320958852767944,
471
+ "learning_rate": 8.601545124439535e-06,
472
+ "loss": 1.3562,
473
+ "step": 64
474
+ },
475
+ {
476
+ "epoch": 0.25691699604743085,
477
+ "grad_norm": 0.9427369832992554,
478
+ "learning_rate": 8.557678386046429e-06,
479
+ "loss": 1.5575,
480
+ "step": 65
481
+ },
482
+ {
483
+ "epoch": 0.2608695652173913,
484
+ "grad_norm": 0.9649813175201416,
485
+ "learning_rate": 8.513249848994248e-06,
486
+ "loss": 1.2068,
487
+ "step": 66
488
+ },
489
+ {
490
+ "epoch": 0.2648221343873518,
491
+ "grad_norm": 0.9272534251213074,
492
+ "learning_rate": 8.468266529064025e-06,
493
+ "loss": 1.2409,
494
+ "step": 67
495
+ },
496
+ {
497
+ "epoch": 0.26877470355731226,
498
+ "grad_norm": 0.952318549156189,
499
+ "learning_rate": 8.422735529643445e-06,
500
+ "loss": 1.4648,
501
+ "step": 68
502
+ },
503
+ {
504
+ "epoch": 0.2727272727272727,
505
+ "grad_norm": 0.7945717573165894,
506
+ "learning_rate": 8.376664040605122e-06,
507
+ "loss": 1.142,
508
+ "step": 69
509
+ },
510
+ {
511
+ "epoch": 0.2766798418972332,
512
+ "grad_norm": 0.8417496085166931,
513
+ "learning_rate": 8.33005933717126e-06,
514
+ "loss": 1.179,
515
+ "step": 70
516
+ },
517
+ {
518
+ "epoch": 0.28063241106719367,
519
+ "grad_norm": 0.8613891005516052,
520
+ "learning_rate": 8.282928778764783e-06,
521
+ "loss": 1.258,
522
+ "step": 71
523
+ },
524
+ {
525
+ "epoch": 0.2845849802371542,
526
+ "grad_norm": 0.8771772980690002,
527
+ "learning_rate": 8.235279807847223e-06,
528
+ "loss": 1.3277,
529
+ "step": 72
530
+ },
531
+ {
532
+ "epoch": 0.2885375494071146,
533
+ "grad_norm": 0.9778826832771301,
534
+ "learning_rate": 8.18711994874345e-06,
535
+ "loss": 1.1869,
536
+ "step": 73
537
+ },
538
+ {
539
+ "epoch": 0.2924901185770751,
540
+ "grad_norm": 0.9443055391311646,
541
+ "learning_rate": 8.138456806453503e-06,
542
+ "loss": 1.37,
543
+ "step": 74
544
+ },
545
+ {
546
+ "epoch": 0.2964426877470356,
547
+ "grad_norm": 0.8966163992881775,
548
+ "learning_rate": 8.089298065451673e-06,
549
+ "loss": 1.2171,
550
+ "step": 75
551
+ },
552
+ {
553
+ "epoch": 0.30039525691699603,
554
+ "grad_norm": 0.9195836782455444,
555
+ "learning_rate": 8.039651488473028e-06,
556
+ "loss": 1.3983,
557
+ "step": 76
558
+ },
559
+ {
560
+ "epoch": 0.30434782608695654,
561
+ "grad_norm": 0.952008843421936,
562
+ "learning_rate": 7.989524915287595e-06,
563
+ "loss": 1.4224,
564
+ "step": 77
565
+ },
566
+ {
567
+ "epoch": 0.308300395256917,
568
+ "grad_norm": 1.0114339590072632,
569
+ "learning_rate": 7.938926261462366e-06,
570
+ "loss": 1.409,
571
+ "step": 78
572
+ },
573
+ {
574
+ "epoch": 0.308300395256917,
575
+ "eval_loss": 1.2486090660095215,
576
+ "eval_runtime": 0.6312,
577
+ "eval_samples_per_second": 33.269,
578
+ "eval_steps_per_second": 4.753,
579
+ "step": 78
580
+ },
581
+ {
582
+ "epoch": 0.31225296442687744,
583
+ "grad_norm": 0.8975977301597595,
584
+ "learning_rate": 7.887863517111337e-06,
585
+ "loss": 1.3565,
586
+ "step": 79
587
+ },
588
+ {
589
+ "epoch": 0.31620553359683795,
590
+ "grad_norm": 0.9322842955589294,
591
+ "learning_rate": 7.836344745633785e-06,
592
+ "loss": 1.336,
593
+ "step": 80
594
+ },
595
+ {
596
+ "epoch": 0.3201581027667984,
597
+ "grad_norm": 0.934016227722168,
598
+ "learning_rate": 7.78437808244094e-06,
599
+ "loss": 1.318,
600
+ "step": 81
601
+ },
602
+ {
603
+ "epoch": 0.3241106719367589,
604
+ "grad_norm": 0.8984368443489075,
605
+ "learning_rate": 7.731971733671347e-06,
606
+ "loss": 1.264,
607
+ "step": 82
608
+ },
609
+ {
610
+ "epoch": 0.32806324110671936,
611
+ "grad_norm": 0.9175249338150024,
612
+ "learning_rate": 7.679133974894984e-06,
613
+ "loss": 1.2758,
614
+ "step": 83
615
+ },
616
+ {
617
+ "epoch": 0.33201581027667987,
618
+ "grad_norm": 1.0880227088928223,
619
+ "learning_rate": 7.6258731498064796e-06,
620
+ "loss": 1.1448,
621
+ "step": 84
622
+ },
623
+ {
624
+ "epoch": 0.3359683794466403,
625
+ "grad_norm": 0.8632190823554993,
626
+ "learning_rate": 7.572197668907533e-06,
627
+ "loss": 1.2644,
628
+ "step": 85
629
+ },
630
+ {
631
+ "epoch": 0.33992094861660077,
632
+ "grad_norm": 0.8895804286003113,
633
+ "learning_rate": 7.518116008178805e-06,
634
+ "loss": 1.2656,
635
+ "step": 86
636
+ },
637
+ {
638
+ "epoch": 0.3438735177865613,
639
+ "grad_norm": 0.9582334756851196,
640
+ "learning_rate": 7.463636707741458e-06,
641
+ "loss": 1.2453,
642
+ "step": 87
643
+ },
644
+ {
645
+ "epoch": 0.34782608695652173,
646
+ "grad_norm": 0.8999531865119934,
647
+ "learning_rate": 7.408768370508577e-06,
648
+ "loss": 1.1815,
649
+ "step": 88
650
+ },
651
+ {
652
+ "epoch": 0.35177865612648224,
653
+ "grad_norm": 0.8813087940216064,
654
+ "learning_rate": 7.353519660826665e-06,
655
+ "loss": 1.2232,
656
+ "step": 89
657
+ },
658
+ {
659
+ "epoch": 0.3557312252964427,
660
+ "grad_norm": 0.8202010989189148,
661
+ "learning_rate": 7.297899303107441e-06,
662
+ "loss": 1.1386,
663
+ "step": 90
664
+ },
665
+ {
666
+ "epoch": 0.35968379446640314,
667
+ "grad_norm": 0.9317258596420288,
668
+ "learning_rate": 7.241916080450163e-06,
669
+ "loss": 1.2962,
670
+ "step": 91
671
+ },
672
+ {
673
+ "epoch": 0.36363636363636365,
674
+ "grad_norm": 0.8814587593078613,
675
+ "learning_rate": 7.185578833254665e-06,
676
+ "loss": 1.3109,
677
+ "step": 92
678
+ },
679
+ {
680
+ "epoch": 0.3675889328063241,
681
+ "grad_norm": 0.9203696846961975,
682
+ "learning_rate": 7.128896457825364e-06,
683
+ "loss": 1.4728,
684
+ "step": 93
685
+ },
686
+ {
687
+ "epoch": 0.3715415019762846,
688
+ "grad_norm": 0.842402994632721,
689
+ "learning_rate": 7.071877904966422e-06,
690
+ "loss": 1.2597,
691
+ "step": 94
692
+ },
693
+ {
694
+ "epoch": 0.37549407114624506,
695
+ "grad_norm": 0.8974599838256836,
696
+ "learning_rate": 7.014532178568314e-06,
697
+ "loss": 1.2041,
698
+ "step": 95
699
+ },
700
+ {
701
+ "epoch": 0.3794466403162055,
702
+ "grad_norm": 0.8709033727645874,
703
+ "learning_rate": 6.9568683341860135e-06,
704
+ "loss": 1.1369,
705
+ "step": 96
706
+ },
707
+ {
708
+ "epoch": 0.383399209486166,
709
+ "grad_norm": 0.8162269592285156,
710
+ "learning_rate": 6.898895477609007e-06,
711
+ "loss": 1.2103,
712
+ "step": 97
713
+ },
714
+ {
715
+ "epoch": 0.38735177865612647,
716
+ "grad_norm": 0.8966283798217773,
717
+ "learning_rate": 6.840622763423391e-06,
718
+ "loss": 1.2609,
719
+ "step": 98
720
+ },
721
+ {
722
+ "epoch": 0.391304347826087,
723
+ "grad_norm": 0.8919047117233276,
724
+ "learning_rate": 6.782059393566254e-06,
725
+ "loss": 1.1807,
726
+ "step": 99
727
+ },
728
+ {
729
+ "epoch": 0.3952569169960474,
730
+ "grad_norm": 0.9296970367431641,
731
+ "learning_rate": 6.723214615872585e-06,
732
+ "loss": 1.3748,
733
+ "step": 100
734
+ },
735
+ {
736
+ "epoch": 0.39920948616600793,
737
+ "grad_norm": 0.8833282589912415,
738
+ "learning_rate": 6.664097722614934e-06,
739
+ "loss": 1.271,
740
+ "step": 101
741
+ },
742
+ {
743
+ "epoch": 0.4031620553359684,
744
+ "grad_norm": 0.9111071228981018,
745
+ "learning_rate": 6.604718049036047e-06,
746
+ "loss": 1.3016,
747
+ "step": 102
748
+ },
749
+ {
750
+ "epoch": 0.40711462450592883,
751
+ "grad_norm": 0.9141144156455994,
752
+ "learning_rate": 6.545084971874738e-06,
753
+ "loss": 1.3072,
754
+ "step": 103
755
+ },
756
+ {
757
+ "epoch": 0.41106719367588934,
758
+ "grad_norm": 0.8589788675308228,
759
+ "learning_rate": 6.485207907885175e-06,
760
+ "loss": 1.1914,
761
+ "step": 104
762
+ },
763
+ {
764
+ "epoch": 0.41106719367588934,
765
+ "eval_loss": 1.2048839330673218,
766
+ "eval_runtime": 0.6355,
767
+ "eval_samples_per_second": 33.045,
768
+ "eval_steps_per_second": 4.721,
769
+ "step": 104
770
+ },
771
+ {
772
+ "epoch": 0.4150197628458498,
773
+ "grad_norm": 0.9179502725601196,
774
+ "learning_rate": 6.425096312349881e-06,
775
+ "loss": 1.3336,
776
+ "step": 105
777
+ },
778
+ {
779
+ "epoch": 0.4189723320158103,
780
+ "grad_norm": 0.8490063548088074,
781
+ "learning_rate": 6.364759677586627e-06,
782
+ "loss": 1.275,
783
+ "step": 106
784
+ },
785
+ {
786
+ "epoch": 0.42292490118577075,
787
+ "grad_norm": 0.8626274466514587,
788
+ "learning_rate": 6.304207531449486e-06,
789
+ "loss": 1.237,
790
+ "step": 107
791
+ },
792
+ {
793
+ "epoch": 0.4268774703557312,
794
+ "grad_norm": 0.855128824710846,
795
+ "learning_rate": 6.243449435824276e-06,
796
+ "loss": 1.1993,
797
+ "step": 108
798
+ },
799
+ {
800
+ "epoch": 0.4308300395256917,
801
+ "grad_norm": 0.8959077000617981,
802
+ "learning_rate": 6.182494985118625e-06,
803
+ "loss": 1.3898,
804
+ "step": 109
805
+ },
806
+ {
807
+ "epoch": 0.43478260869565216,
808
+ "grad_norm": 0.9411128163337708,
809
+ "learning_rate": 6.121353804746907e-06,
810
+ "loss": 1.1805,
811
+ "step": 110
812
+ },
813
+ {
814
+ "epoch": 0.43873517786561267,
815
+ "grad_norm": 0.8071564435958862,
816
+ "learning_rate": 6.060035549610275e-06,
817
+ "loss": 0.8793,
818
+ "step": 111
819
+ },
820
+ {
821
+ "epoch": 0.4426877470355731,
822
+ "grad_norm": 0.873630702495575,
823
+ "learning_rate": 5.9985499025720354e-06,
824
+ "loss": 1.3551,
825
+ "step": 112
826
+ },
827
+ {
828
+ "epoch": 0.44664031620553357,
829
+ "grad_norm": 0.7491222620010376,
830
+ "learning_rate": 5.936906572928625e-06,
831
+ "loss": 1.0515,
832
+ "step": 113
833
+ },
834
+ {
835
+ "epoch": 0.4505928853754941,
836
+ "grad_norm": 0.8194934129714966,
837
+ "learning_rate": 5.8751152948763815e-06,
838
+ "loss": 1.0884,
839
+ "step": 114
840
+ },
841
+ {
842
+ "epoch": 0.45454545454545453,
843
+ "grad_norm": 0.9355419874191284,
844
+ "learning_rate": 5.813185825974419e-06,
845
+ "loss": 1.4432,
846
+ "step": 115
847
+ },
848
+ {
849
+ "epoch": 0.45849802371541504,
850
+ "grad_norm": 0.760794460773468,
851
+ "learning_rate": 5.751127945603786e-06,
852
+ "loss": 1.0239,
853
+ "step": 116
854
+ },
855
+ {
856
+ "epoch": 0.4624505928853755,
857
+ "grad_norm": 0.9574670791625977,
858
+ "learning_rate": 5.68895145342319e-06,
859
+ "loss": 1.5178,
860
+ "step": 117
861
+ },
862
+ {
863
+ "epoch": 0.466403162055336,
864
+ "grad_norm": 0.9131107330322266,
865
+ "learning_rate": 5.626666167821522e-06,
866
+ "loss": 1.3872,
867
+ "step": 118
868
+ },
869
+ {
870
+ "epoch": 0.47035573122529645,
871
+ "grad_norm": 0.9126591682434082,
872
+ "learning_rate": 5.5642819243674085e-06,
873
+ "loss": 1.3254,
874
+ "step": 119
875
+ },
876
+ {
877
+ "epoch": 0.4743083003952569,
878
+ "grad_norm": 0.8607676029205322,
879
+ "learning_rate": 5.5018085742560745e-06,
880
+ "loss": 1.1636,
881
+ "step": 120
882
+ },
883
+ {
884
+ "epoch": 0.4782608695652174,
885
+ "grad_norm": 0.8300676345825195,
886
+ "learning_rate": 5.439255982753717e-06,
887
+ "loss": 1.0787,
888
+ "step": 121
889
+ },
890
+ {
891
+ "epoch": 0.48221343873517786,
892
+ "grad_norm": 0.8748376965522766,
893
+ "learning_rate": 5.376634027639664e-06,
894
+ "loss": 1.3438,
895
+ "step": 122
896
+ },
897
+ {
898
+ "epoch": 0.48616600790513836,
899
+ "grad_norm": 0.8148254156112671,
900
+ "learning_rate": 5.3139525976465675e-06,
901
+ "loss": 1.2148,
902
+ "step": 123
903
+ },
904
+ {
905
+ "epoch": 0.4901185770750988,
906
+ "grad_norm": 0.7850709557533264,
907
+ "learning_rate": 5.251221590898848e-06,
908
+ "loss": 1.1715,
909
+ "step": 124
910
+ },
911
+ {
912
+ "epoch": 0.49407114624505927,
913
+ "grad_norm": 0.8794577717781067,
914
+ "learning_rate": 5.188450913349674e-06,
915
+ "loss": 1.1467,
916
+ "step": 125
917
+ },
918
+ {
919
+ "epoch": 0.4980237154150198,
920
+ "grad_norm": 0.8375518321990967,
921
+ "learning_rate": 5.1256504772166885e-06,
922
+ "loss": 1.2558,
923
+ "step": 126
924
+ },
925
+ {
926
+ "epoch": 0.5019762845849802,
927
+ "grad_norm": 0.7941477298736572,
928
+ "learning_rate": 5.062830199416764e-06,
929
+ "loss": 1.0128,
930
+ "step": 127
931
+ },
932
+ {
933
+ "epoch": 0.5059288537549407,
934
+ "grad_norm": 0.9250499606132507,
935
+ "learning_rate": 5e-06,
936
+ "loss": 1.2622,
937
+ "step": 128
938
+ },
939
+ {
940
+ "epoch": 0.5098814229249012,
941
+ "grad_norm": 0.9209011793136597,
942
+ "learning_rate": 4.937169800583237e-06,
943
+ "loss": 1.3989,
944
+ "step": 129
945
+ },
946
+ {
947
+ "epoch": 0.5138339920948617,
948
+ "grad_norm": 0.8219020962715149,
949
+ "learning_rate": 4.874349522783313e-06,
950
+ "loss": 1.1264,
951
+ "step": 130
952
+ },
953
+ {
954
+ "epoch": 0.5138339920948617,
955
+ "eval_loss": 1.1763263940811157,
956
+ "eval_runtime": 0.6328,
957
+ "eval_samples_per_second": 33.184,
958
+ "eval_steps_per_second": 4.741,
959
+ "step": 130
960
+ },
961
+ {
962
+ "epoch": 0.5177865612648221,
963
+ "grad_norm": 0.9172279238700867,
964
+ "learning_rate": 4.811549086650327e-06,
965
+ "loss": 1.2868,
966
+ "step": 131
967
+ },
968
+ {
969
+ "epoch": 0.5217391304347826,
970
+ "grad_norm": 0.8297377228736877,
971
+ "learning_rate": 4.748778409101153e-06,
972
+ "loss": 1.221,
973
+ "step": 132
974
+ },
975
+ {
976
+ "epoch": 0.525691699604743,
977
+ "grad_norm": 0.8089572787284851,
978
+ "learning_rate": 4.686047402353433e-06,
979
+ "loss": 1.2466,
980
+ "step": 133
981
+ },
982
+ {
983
+ "epoch": 0.5296442687747036,
984
+ "grad_norm": 0.8571174144744873,
985
+ "learning_rate": 4.6233659723603374e-06,
986
+ "loss": 1.3095,
987
+ "step": 134
988
+ },
989
+ {
990
+ "epoch": 0.5335968379446641,
991
+ "grad_norm": 0.9645123481750488,
992
+ "learning_rate": 4.560744017246284e-06,
993
+ "loss": 1.4281,
994
+ "step": 135
995
+ },
996
+ {
997
+ "epoch": 0.5375494071146245,
998
+ "grad_norm": 0.8215659856796265,
999
+ "learning_rate": 4.4981914257439254e-06,
1000
+ "loss": 1.1636,
1001
+ "step": 136
1002
+ },
1003
+ {
1004
+ "epoch": 0.541501976284585,
1005
+ "grad_norm": 0.8466573357582092,
1006
+ "learning_rate": 4.4357180756325915e-06,
1007
+ "loss": 1.2409,
1008
+ "step": 137
1009
+ },
1010
+ {
1011
+ "epoch": 0.5454545454545454,
1012
+ "grad_norm": 0.8330338597297668,
1013
+ "learning_rate": 4.373333832178478e-06,
1014
+ "loss": 1.1857,
1015
+ "step": 138
1016
+ },
1017
+ {
1018
+ "epoch": 0.549407114624506,
1019
+ "grad_norm": 0.8727869987487793,
1020
+ "learning_rate": 4.31104854657681e-06,
1021
+ "loss": 1.1947,
1022
+ "step": 139
1023
+ },
1024
+ {
1025
+ "epoch": 0.5533596837944664,
1026
+ "grad_norm": 0.8822157979011536,
1027
+ "learning_rate": 4.248872054396215e-06,
1028
+ "loss": 1.1367,
1029
+ "step": 140
1030
+ },
1031
+ {
1032
+ "epoch": 0.5573122529644269,
1033
+ "grad_norm": 0.876277506351471,
1034
+ "learning_rate": 4.186814174025582e-06,
1035
+ "loss": 1.1754,
1036
+ "step": 141
1037
+ },
1038
+ {
1039
+ "epoch": 0.5612648221343873,
1040
+ "grad_norm": 0.8091803789138794,
1041
+ "learning_rate": 4.124884705123619e-06,
1042
+ "loss": 1.1789,
1043
+ "step": 142
1044
+ },
1045
+ {
1046
+ "epoch": 0.5652173913043478,
1047
+ "grad_norm": 0.9175235629081726,
1048
+ "learning_rate": 4.063093427071376e-06,
1049
+ "loss": 1.2189,
1050
+ "step": 143
1051
+ },
1052
+ {
1053
+ "epoch": 0.5691699604743083,
1054
+ "grad_norm": 0.8601986765861511,
1055
+ "learning_rate": 4.001450097427965e-06,
1056
+ "loss": 1.3021,
1057
+ "step": 144
1058
+ },
1059
+ {
1060
+ "epoch": 0.5731225296442688,
1061
+ "grad_norm": 0.821972668170929,
1062
+ "learning_rate": 3.939964450389728e-06,
1063
+ "loss": 1.2536,
1064
+ "step": 145
1065
+ },
1066
+ {
1067
+ "epoch": 0.5770750988142292,
1068
+ "grad_norm": 0.8965577483177185,
1069
+ "learning_rate": 3.8786461952530955e-06,
1070
+ "loss": 1.2669,
1071
+ "step": 146
1072
+ },
1073
+ {
1074
+ "epoch": 0.5810276679841897,
1075
+ "grad_norm": 0.8901940584182739,
1076
+ "learning_rate": 3.817505014881378e-06,
1077
+ "loss": 1.2246,
1078
+ "step": 147
1079
+ },
1080
+ {
1081
+ "epoch": 0.5849802371541502,
1082
+ "grad_norm": 0.8637762069702148,
1083
+ "learning_rate": 3.756550564175727e-06,
1084
+ "loss": 1.2705,
1085
+ "step": 148
1086
+ },
1087
+ {
1088
+ "epoch": 0.5889328063241107,
1089
+ "grad_norm": 0.8907751441001892,
1090
+ "learning_rate": 3.695792468550517e-06,
1091
+ "loss": 1.3191,
1092
+ "step": 149
1093
+ },
1094
+ {
1095
+ "epoch": 0.5928853754940712,
1096
+ "grad_norm": 0.9000714421272278,
1097
+ "learning_rate": 3.635240322413375e-06,
1098
+ "loss": 1.3388,
1099
+ "step": 150
1100
+ },
1101
+ {
1102
+ "epoch": 0.5968379446640316,
1103
+ "grad_norm": 0.8116331696510315,
1104
+ "learning_rate": 3.5749036876501196e-06,
1105
+ "loss": 1.2169,
1106
+ "step": 151
1107
+ },
1108
+ {
1109
+ "epoch": 0.6007905138339921,
1110
+ "grad_norm": 0.9376166462898254,
1111
+ "learning_rate": 3.5147920921148267e-06,
1112
+ "loss": 1.4159,
1113
+ "step": 152
1114
+ },
1115
+ {
1116
+ "epoch": 0.6047430830039525,
1117
+ "grad_norm": 0.8815961480140686,
1118
+ "learning_rate": 3.4549150281252635e-06,
1119
+ "loss": 1.3918,
1120
+ "step": 153
1121
+ },
1122
+ {
1123
+ "epoch": 0.6086956521739131,
1124
+ "grad_norm": 0.8155955672264099,
1125
+ "learning_rate": 3.3952819509639534e-06,
1126
+ "loss": 1.0842,
1127
+ "step": 154
1128
+ },
1129
+ {
1130
+ "epoch": 0.6126482213438735,
1131
+ "grad_norm": 0.8554919958114624,
1132
+ "learning_rate": 3.3359022773850673e-06,
1133
+ "loss": 1.3536,
1134
+ "step": 155
1135
+ },
1136
+ {
1137
+ "epoch": 0.616600790513834,
1138
+ "grad_norm": 0.8661870956420898,
1139
+ "learning_rate": 3.2767853841274154e-06,
1140
+ "loss": 1.1721,
1141
+ "step": 156
1142
+ },
1143
+ {
1144
+ "epoch": 0.616600790513834,
1145
+ "eval_loss": 1.1584229469299316,
1146
+ "eval_runtime": 0.6329,
1147
+ "eval_samples_per_second": 33.182,
1148
+ "eval_steps_per_second": 4.74,
1149
+ "step": 156
1150
+ },
1151
+ {
1152
+ "epoch": 0.6205533596837944,
1153
+ "grad_norm": 0.8114671111106873,
1154
+ "learning_rate": 3.217940606433747e-06,
1155
+ "loss": 1.0664,
1156
+ "step": 157
1157
+ },
1158
+ {
1159
+ "epoch": 0.6245059288537549,
1160
+ "grad_norm": 0.863220751285553,
1161
+ "learning_rate": 3.1593772365766107e-06,
1162
+ "loss": 1.2761,
1163
+ "step": 158
1164
+ },
1165
+ {
1166
+ "epoch": 0.6284584980237155,
1167
+ "grad_norm": 0.8586646318435669,
1168
+ "learning_rate": 3.1011045223909954e-06,
1169
+ "loss": 1.2633,
1170
+ "step": 159
1171
+ },
1172
+ {
1173
+ "epoch": 0.6324110671936759,
1174
+ "grad_norm": 0.8601418137550354,
1175
+ "learning_rate": 3.043131665813988e-06,
1176
+ "loss": 1.2919,
1177
+ "step": 160
1178
+ },
1179
+ {
1180
+ "epoch": 0.6363636363636364,
1181
+ "grad_norm": 0.8035183548927307,
1182
+ "learning_rate": 2.9854678214316875e-06,
1183
+ "loss": 1.1463,
1184
+ "step": 161
1185
+ },
1186
+ {
1187
+ "epoch": 0.6403162055335968,
1188
+ "grad_norm": 0.8140465021133423,
1189
+ "learning_rate": 2.92812209503358e-06,
1190
+ "loss": 1.0966,
1191
+ "step": 162
1192
+ },
1193
+ {
1194
+ "epoch": 0.6442687747035574,
1195
+ "grad_norm": 0.8538705706596375,
1196
+ "learning_rate": 2.871103542174637e-06,
1197
+ "loss": 1.2748,
1198
+ "step": 163
1199
+ },
1200
+ {
1201
+ "epoch": 0.6482213438735178,
1202
+ "grad_norm": 0.8856031894683838,
1203
+ "learning_rate": 2.814421166745337e-06,
1204
+ "loss": 1.3632,
1205
+ "step": 164
1206
+ },
1207
+ {
1208
+ "epoch": 0.6521739130434783,
1209
+ "grad_norm": 0.8147869110107422,
1210
+ "learning_rate": 2.7580839195498397e-06,
1211
+ "loss": 1.1984,
1212
+ "step": 165
1213
+ },
1214
+ {
1215
+ "epoch": 0.6561264822134387,
1216
+ "grad_norm": 0.9556436538696289,
1217
+ "learning_rate": 2.7021006968925613e-06,
1218
+ "loss": 1.5325,
1219
+ "step": 166
1220
+ },
1221
+ {
1222
+ "epoch": 0.6600790513833992,
1223
+ "grad_norm": 0.8936392664909363,
1224
+ "learning_rate": 2.646480339173337e-06,
1225
+ "loss": 1.4192,
1226
+ "step": 167
1227
+ },
1228
+ {
1229
+ "epoch": 0.6640316205533597,
1230
+ "grad_norm": 0.8458386063575745,
1231
+ "learning_rate": 2.5912316294914232e-06,
1232
+ "loss": 1.2821,
1233
+ "step": 168
1234
+ },
1235
+ {
1236
+ "epoch": 0.6679841897233202,
1237
+ "grad_norm": 0.9311079382896423,
1238
+ "learning_rate": 2.536363292258543e-06,
1239
+ "loss": 1.3371,
1240
+ "step": 169
1241
+ },
1242
+ {
1243
+ "epoch": 0.6719367588932806,
1244
+ "grad_norm": 0.8807729482650757,
1245
+ "learning_rate": 2.4818839918211963e-06,
1246
+ "loss": 1.2789,
1247
+ "step": 170
1248
+ },
1249
+ {
1250
+ "epoch": 0.6758893280632411,
1251
+ "grad_norm": 0.8326642513275146,
1252
+ "learning_rate": 2.4278023310924676e-06,
1253
+ "loss": 1.2545,
1254
+ "step": 171
1255
+ },
1256
+ {
1257
+ "epoch": 0.6798418972332015,
1258
+ "grad_norm": 0.7561216950416565,
1259
+ "learning_rate": 2.3741268501935212e-06,
1260
+ "loss": 0.9872,
1261
+ "step": 172
1262
+ },
1263
+ {
1264
+ "epoch": 0.6837944664031621,
1265
+ "grad_norm": 0.9049252867698669,
1266
+ "learning_rate": 2.320866025105016e-06,
1267
+ "loss": 1.0876,
1268
+ "step": 173
1269
+ },
1270
+ {
1271
+ "epoch": 0.6877470355731226,
1272
+ "grad_norm": 0.8050791621208191,
1273
+ "learning_rate": 2.268028266328655e-06,
1274
+ "loss": 1.2257,
1275
+ "step": 174
1276
+ },
1277
+ {
1278
+ "epoch": 0.691699604743083,
1279
+ "grad_norm": 0.7893306016921997,
1280
+ "learning_rate": 2.2156219175590623e-06,
1281
+ "loss": 1.1499,
1282
+ "step": 175
1283
+ },
1284
+ {
1285
+ "epoch": 0.6956521739130435,
1286
+ "grad_norm": 0.8402442336082458,
1287
+ "learning_rate": 2.1636552543662187e-06,
1288
+ "loss": 1.3324,
1289
+ "step": 176
1290
+ },
1291
+ {
1292
+ "epoch": 0.6996047430830039,
1293
+ "grad_norm": 0.847458004951477,
1294
+ "learning_rate": 2.112136482888663e-06,
1295
+ "loss": 1.3478,
1296
+ "step": 177
1297
+ },
1298
+ {
1299
+ "epoch": 0.7035573122529645,
1300
+ "grad_norm": 0.8522107601165771,
1301
+ "learning_rate": 2.061073738537635e-06,
1302
+ "loss": 1.2586,
1303
+ "step": 178
1304
+ },
1305
+ {
1306
+ "epoch": 0.7075098814229249,
1307
+ "grad_norm": 0.8222721219062805,
1308
+ "learning_rate": 2.0104750847124075e-06,
1309
+ "loss": 1.2455,
1310
+ "step": 179
1311
+ },
1312
+ {
1313
+ "epoch": 0.7114624505928854,
1314
+ "grad_norm": 0.8545589447021484,
1315
+ "learning_rate": 1.9603485115269743e-06,
1316
+ "loss": 1.2892,
1317
+ "step": 180
1318
+ },
1319
+ {
1320
+ "epoch": 0.7154150197628458,
1321
+ "grad_norm": 0.7990689873695374,
1322
+ "learning_rate": 1.910701934548329e-06,
1323
+ "loss": 1.1216,
1324
+ "step": 181
1325
+ },
1326
+ {
1327
+ "epoch": 0.7193675889328063,
1328
+ "grad_norm": 0.8381747007369995,
1329
+ "learning_rate": 1.8615431935464984e-06,
1330
+ "loss": 1.2738,
1331
+ "step": 182
1332
+ },
1333
+ {
1334
+ "epoch": 0.7193675889328063,
1335
+ "eval_loss": 1.1467477083206177,
1336
+ "eval_runtime": 0.6308,
1337
+ "eval_samples_per_second": 33.29,
1338
+ "eval_steps_per_second": 4.756,
1339
+ "step": 182
1340
+ },
1341
+ {
1342
+ "epoch": 0.7233201581027668,
1343
+ "grad_norm": 0.8528211712837219,
1344
+ "learning_rate": 1.8128800512565514e-06,
1345
+ "loss": 1.3237,
1346
+ "step": 183
1347
+ },
1348
+ {
1349
+ "epoch": 0.7272727272727273,
1350
+ "grad_norm": 0.8386929035186768,
1351
+ "learning_rate": 1.7647201921527802e-06,
1352
+ "loss": 1.1454,
1353
+ "step": 184
1354
+ },
1355
+ {
1356
+ "epoch": 0.7312252964426877,
1357
+ "grad_norm": 0.9228261709213257,
1358
+ "learning_rate": 1.7170712212352187e-06,
1359
+ "loss": 1.5128,
1360
+ "step": 185
1361
+ },
1362
+ {
1363
+ "epoch": 0.7351778656126482,
1364
+ "grad_norm": 0.763982892036438,
1365
+ "learning_rate": 1.6699406628287423e-06,
1366
+ "loss": 0.9779,
1367
+ "step": 186
1368
+ },
1369
+ {
1370
+ "epoch": 0.7391304347826086,
1371
+ "grad_norm": 0.9443770051002502,
1372
+ "learning_rate": 1.6233359593948777e-06,
1373
+ "loss": 0.9658,
1374
+ "step": 187
1375
+ },
1376
+ {
1377
+ "epoch": 0.7430830039525692,
1378
+ "grad_norm": 0.8337656855583191,
1379
+ "learning_rate": 1.5772644703565564e-06,
1380
+ "loss": 1.2752,
1381
+ "step": 188
1382
+ },
1383
+ {
1384
+ "epoch": 0.7470355731225297,
1385
+ "grad_norm": 0.8052669763565063,
1386
+ "learning_rate": 1.531733470935976e-06,
1387
+ "loss": 1.1755,
1388
+ "step": 189
1389
+ },
1390
+ {
1391
+ "epoch": 0.7509881422924901,
1392
+ "grad_norm": 0.8237268328666687,
1393
+ "learning_rate": 1.4867501510057548e-06,
1394
+ "loss": 1.1529,
1395
+ "step": 190
1396
+ },
1397
+ {
1398
+ "epoch": 0.7549407114624506,
1399
+ "grad_norm": 0.8448726534843445,
1400
+ "learning_rate": 1.4423216139535735e-06,
1401
+ "loss": 1.2361,
1402
+ "step": 191
1403
+ },
1404
+ {
1405
+ "epoch": 0.758893280632411,
1406
+ "grad_norm": 0.8243027329444885,
1407
+ "learning_rate": 1.3984548755604655e-06,
1408
+ "loss": 1.2842,
1409
+ "step": 192
1410
+ },
1411
+ {
1412
+ "epoch": 0.7628458498023716,
1413
+ "grad_norm": 0.8658849000930786,
1414
+ "learning_rate": 1.3551568628929434e-06,
1415
+ "loss": 1.256,
1416
+ "step": 193
1417
+ },
1418
+ {
1419
+ "epoch": 0.766798418972332,
1420
+ "grad_norm": 0.8200253248214722,
1421
+ "learning_rate": 1.312434413209131e-06,
1422
+ "loss": 1.1356,
1423
+ "step": 194
1424
+ },
1425
+ {
1426
+ "epoch": 0.7707509881422925,
1427
+ "grad_norm": 0.8177831768989563,
1428
+ "learning_rate": 1.2702942728790897e-06,
1429
+ "loss": 1.151,
1430
+ "step": 195
1431
+ },
1432
+ {
1433
+ "epoch": 0.7747035573122529,
1434
+ "grad_norm": 0.8590410351753235,
1435
+ "learning_rate": 1.2287430963194807e-06,
1436
+ "loss": 1.3267,
1437
+ "step": 196
1438
+ },
1439
+ {
1440
+ "epoch": 0.7786561264822134,
1441
+ "grad_norm": 0.8247159719467163,
1442
+ "learning_rate": 1.18778744494276e-06,
1443
+ "loss": 1.1943,
1444
+ "step": 197
1445
+ },
1446
+ {
1447
+ "epoch": 0.782608695652174,
1448
+ "grad_norm": 0.9330562949180603,
1449
+ "learning_rate": 1.1474337861210543e-06,
1450
+ "loss": 1.4392,
1451
+ "step": 198
1452
+ },
1453
+ {
1454
+ "epoch": 0.7865612648221344,
1455
+ "grad_norm": 0.8507537841796875,
1456
+ "learning_rate": 1.1076884921648834e-06,
1457
+ "loss": 1.2678,
1458
+ "step": 199
1459
+ },
1460
+ {
1461
+ "epoch": 0.7905138339920948,
1462
+ "grad_norm": 0.8290367126464844,
1463
+ "learning_rate": 1.0685578393169054e-06,
1464
+ "loss": 1.2738,
1465
+ "step": 200
1466
+ },
1467
+ {
1468
+ "epoch": 0.7944664031620553,
1469
+ "grad_norm": 0.8783449530601501,
1470
+ "learning_rate": 1.0300480067608232e-06,
1471
+ "loss": 1.4097,
1472
+ "step": 201
1473
+ },
1474
+ {
1475
+ "epoch": 0.7984189723320159,
1476
+ "grad_norm": 0.8614851236343384,
1477
+ "learning_rate": 9.921650756456164e-07,
1478
+ "loss": 1.1463,
1479
+ "step": 202
1480
+ },
1481
+ {
1482
+ "epoch": 0.8023715415019763,
1483
+ "grad_norm": 0.8946048617362976,
1484
+ "learning_rate": 9.549150281252633e-07,
1485
+ "loss": 1.3288,
1486
+ "step": 203
1487
+ },
1488
+ {
1489
+ "epoch": 0.8063241106719368,
1490
+ "grad_norm": 0.8253918290138245,
1491
+ "learning_rate": 9.183037464140804e-07,
1492
+ "loss": 1.1852,
1493
+ "step": 204
1494
+ },
1495
+ {
1496
+ "epoch": 0.8102766798418972,
1497
+ "grad_norm": 0.8394426107406616,
1498
+ "learning_rate": 8.823370118578628e-07,
1499
+ "loss": 1.2328,
1500
+ "step": 205
1501
+ },
1502
+ {
1503
+ "epoch": 0.8142292490118577,
1504
+ "grad_norm": 0.7891396880149841,
1505
+ "learning_rate": 8.470205040209362e-07,
1506
+ "loss": 1.2409,
1507
+ "step": 206
1508
+ },
1509
+ {
1510
+ "epoch": 0.8181818181818182,
1511
+ "grad_norm": 0.8969730138778687,
1512
+ "learning_rate": 8.123597997892918e-07,
1513
+ "loss": 1.4195,
1514
+ "step": 207
1515
+ },
1516
+ {
1517
+ "epoch": 0.8221343873517787,
1518
+ "grad_norm": 0.7859252095222473,
1519
+ "learning_rate": 7.783603724899258e-07,
1520
+ "loss": 1.1729,
1521
+ "step": 208
1522
+ },
1523
+ {
1524
+ "epoch": 0.8221343873517787,
1525
+ "eval_loss": 1.1414591073989868,
1526
+ "eval_runtime": 0.63,
1527
+ "eval_samples_per_second": 33.334,
1528
+ "eval_steps_per_second": 4.762,
1529
+ "step": 208
1530
+ },
1531
+ {
1532
+ "epoch": 0.8260869565217391,
1533
+ "grad_norm": 0.8004709482192993,
1534
+ "learning_rate": 7.450275910265415e-07,
1535
+ "loss": 1.1787,
1536
+ "step": 209
1537
+ },
1538
+ {
1539
+ "epoch": 0.8300395256916996,
1540
+ "grad_norm": 0.8863723874092102,
1541
+ "learning_rate": 7.123667190317396e-07,
1542
+ "loss": 1.3486,
1543
+ "step": 210
1544
+ },
1545
+ {
1546
+ "epoch": 0.83399209486166,
1547
+ "grad_norm": 0.7677893042564392,
1548
+ "learning_rate": 6.803829140358237e-07,
1549
+ "loss": 1.2067,
1550
+ "step": 211
1551
+ },
1552
+ {
1553
+ "epoch": 0.8379446640316206,
1554
+ "grad_norm": 0.7935658097267151,
1555
+ "learning_rate": 6.490812266523716e-07,
1556
+ "loss": 1.0929,
1557
+ "step": 212
1558
+ },
1559
+ {
1560
+ "epoch": 0.841897233201581,
1561
+ "grad_norm": 0.7727168798446655,
1562
+ "learning_rate": 6.184665997806832e-07,
1563
+ "loss": 1.0544,
1564
+ "step": 213
1565
+ },
1566
+ {
1567
+ "epoch": 0.8458498023715415,
1568
+ "grad_norm": 0.8302650451660156,
1569
+ "learning_rate": 5.885438678252342e-07,
1570
+ "loss": 1.2409,
1571
+ "step": 214
1572
+ },
1573
+ {
1574
+ "epoch": 0.849802371541502,
1575
+ "grad_norm": 0.8888508081436157,
1576
+ "learning_rate": 5.593177559322776e-07,
1577
+ "loss": 1.3816,
1578
+ "step": 215
1579
+ },
1580
+ {
1581
+ "epoch": 0.8537549407114624,
1582
+ "grad_norm": 0.76105135679245,
1583
+ "learning_rate": 5.307928792436812e-07,
1584
+ "loss": 1.0972,
1585
+ "step": 216
1586
+ },
1587
+ {
1588
+ "epoch": 0.857707509881423,
1589
+ "grad_norm": 0.8869984149932861,
1590
+ "learning_rate": 5.029737421681446e-07,
1591
+ "loss": 1.3786,
1592
+ "step": 217
1593
+ },
1594
+ {
1595
+ "epoch": 0.8616600790513834,
1596
+ "grad_norm": 0.8762255907058716,
1597
+ "learning_rate": 4.758647376699033e-07,
1598
+ "loss": 1.3592,
1599
+ "step": 218
1600
+ },
1601
+ {
1602
+ "epoch": 0.8656126482213439,
1603
+ "grad_norm": 0.8182483911514282,
1604
+ "learning_rate": 4.494701465750217e-07,
1605
+ "loss": 1.3265,
1606
+ "step": 219
1607
+ },
1608
+ {
1609
+ "epoch": 0.8695652173913043,
1610
+ "grad_norm": 0.8035054206848145,
1611
+ "learning_rate": 4.237941368954124e-07,
1612
+ "loss": 1.1705,
1613
+ "step": 220
1614
+ },
1615
+ {
1616
+ "epoch": 0.8735177865612648,
1617
+ "grad_norm": 0.8334324359893799,
1618
+ "learning_rate": 3.9884076317064813e-07,
1619
+ "loss": 1.2725,
1620
+ "step": 221
1621
+ },
1622
+ {
1623
+ "epoch": 0.8774703557312253,
1624
+ "grad_norm": 0.8486202955245972,
1625
+ "learning_rate": 3.7461396582771035e-07,
1626
+ "loss": 1.2205,
1627
+ "step": 222
1628
+ },
1629
+ {
1630
+ "epoch": 0.8814229249011858,
1631
+ "grad_norm": 0.8365004658699036,
1632
+ "learning_rate": 3.511175705587433e-07,
1633
+ "loss": 1.2246,
1634
+ "step": 223
1635
+ },
1636
+ {
1637
+ "epoch": 0.8853754940711462,
1638
+ "grad_norm": 0.7247976660728455,
1639
+ "learning_rate": 3.283552877169399e-07,
1640
+ "loss": 1.0818,
1641
+ "step": 224
1642
+ },
1643
+ {
1644
+ "epoch": 0.8893280632411067,
1645
+ "grad_norm": 0.9177906513214111,
1646
+ "learning_rate": 3.0633071173062966e-07,
1647
+ "loss": 1.328,
1648
+ "step": 225
1649
+ },
1650
+ {
1651
+ "epoch": 0.8932806324110671,
1652
+ "grad_norm": 0.8105781078338623,
1653
+ "learning_rate": 2.850473205356774e-07,
1654
+ "loss": 1.0848,
1655
+ "step": 226
1656
+ },
1657
+ {
1658
+ "epoch": 0.8972332015810277,
1659
+ "grad_norm": 0.8612114191055298,
1660
+ "learning_rate": 2.6450847502627883e-07,
1661
+ "loss": 1.3192,
1662
+ "step": 227
1663
+ },
1664
+ {
1665
+ "epoch": 0.9011857707509882,
1666
+ "grad_norm": 0.8288993239402771,
1667
+ "learning_rate": 2.447174185242324e-07,
1668
+ "loss": 1.2458,
1669
+ "step": 228
1670
+ },
1671
+ {
1672
+ "epoch": 0.9051383399209486,
1673
+ "grad_norm": 0.8506494760513306,
1674
+ "learning_rate": 2.2567727626678527e-07,
1675
+ "loss": 1.2568,
1676
+ "step": 229
1677
+ },
1678
+ {
1679
+ "epoch": 0.9090909090909091,
1680
+ "grad_norm": 0.9084119200706482,
1681
+ "learning_rate": 2.0739105491312028e-07,
1682
+ "loss": 1.4233,
1683
+ "step": 230
1684
+ },
1685
+ {
1686
+ "epoch": 0.9130434782608695,
1687
+ "grad_norm": 0.9303674101829529,
1688
+ "learning_rate": 1.8986164206957037e-07,
1689
+ "loss": 1.3435,
1690
+ "step": 231
1691
+ },
1692
+ {
1693
+ "epoch": 0.9169960474308301,
1694
+ "grad_norm": 0.8007397055625916,
1695
+ "learning_rate": 1.7309180583363062e-07,
1696
+ "loss": 1.1919,
1697
+ "step": 232
1698
+ },
1699
+ {
1700
+ "epoch": 0.9209486166007905,
1701
+ "grad_norm": 0.7730187177658081,
1702
+ "learning_rate": 1.5708419435684463e-07,
1703
+ "loss": 1.0833,
1704
+ "step": 233
1705
+ },
1706
+ {
1707
+ "epoch": 0.924901185770751,
1708
+ "grad_norm": 0.9434295296669006,
1709
+ "learning_rate": 1.4184133542663014e-07,
1710
+ "loss": 1.4446,
1711
+ "step": 234
1712
+ },
1713
+ {
1714
+ "epoch": 0.924901185770751,
1715
+ "eval_loss": 1.1398719549179077,
1716
+ "eval_runtime": 0.6371,
1717
+ "eval_samples_per_second": 32.96,
1718
+ "eval_steps_per_second": 4.709,
1719
+ "step": 234
1720
+ },
1721
+ {
1722
+ "epoch": 0.9288537549407114,
1723
+ "grad_norm": 0.8378840684890747,
1724
+ "learning_rate": 1.2736563606711384e-07,
1725
+ "loss": 1.3058,
1726
+ "step": 235
1727
+ },
1728
+ {
1729
+ "epoch": 0.932806324110672,
1730
+ "grad_norm": 0.7894282341003418,
1731
+ "learning_rate": 1.136593821590326e-07,
1732
+ "loss": 1.1822,
1733
+ "step": 236
1734
+ },
1735
+ {
1736
+ "epoch": 0.9367588932806324,
1737
+ "grad_norm": 0.8614757657051086,
1738
+ "learning_rate": 1.007247380787657e-07,
1739
+ "loss": 1.2477,
1740
+ "step": 237
1741
+ },
1742
+ {
1743
+ "epoch": 0.9407114624505929,
1744
+ "grad_norm": 0.7437376976013184,
1745
+ "learning_rate": 8.856374635655696e-08,
1746
+ "loss": 0.9579,
1747
+ "step": 238
1748
+ },
1749
+ {
1750
+ "epoch": 0.9446640316205533,
1751
+ "grad_norm": 0.8177477717399597,
1752
+ "learning_rate": 7.717832735397335e-08,
1753
+ "loss": 1.2895,
1754
+ "step": 239
1755
+ },
1756
+ {
1757
+ "epoch": 0.9486166007905138,
1758
+ "grad_norm": 0.8371565937995911,
1759
+ "learning_rate": 6.657027896065982e-08,
1760
+ "loss": 1.2901,
1761
+ "step": 240
1762
+ },
1763
+ {
1764
+ "epoch": 0.9525691699604744,
1765
+ "grad_norm": 0.8062224388122559,
1766
+ "learning_rate": 5.674127631043025e-08,
1767
+ "loss": 1.1185,
1768
+ "step": 241
1769
+ },
1770
+ {
1771
+ "epoch": 0.9565217391304348,
1772
+ "grad_norm": 0.782124936580658,
1773
+ "learning_rate": 4.769287151674407e-08,
1774
+ "loss": 1.1315,
1775
+ "step": 242
1776
+ },
1777
+ {
1778
+ "epoch": 0.9604743083003953,
1779
+ "grad_norm": 0.8169525265693665,
1780
+ "learning_rate": 3.9426493427611177e-08,
1781
+ "loss": 1.3511,
1782
+ "step": 243
1783
+ },
1784
+ {
1785
+ "epoch": 0.9644268774703557,
1786
+ "grad_norm": 0.791191577911377,
1787
+ "learning_rate": 3.194344739995803e-08,
1788
+ "loss": 1.2501,
1789
+ "step": 244
1790
+ },
1791
+ {
1792
+ "epoch": 0.9683794466403162,
1793
+ "grad_norm": 0.7813810706138611,
1794
+ "learning_rate": 2.5244915093499134e-08,
1795
+ "loss": 1.1408,
1796
+ "step": 245
1797
+ },
1798
+ {
1799
+ "epoch": 0.9723320158102767,
1800
+ "grad_norm": 0.8077353239059448,
1801
+ "learning_rate": 1.9331954284137476e-08,
1802
+ "loss": 1.2169,
1803
+ "step": 246
1804
+ },
1805
+ {
1806
+ "epoch": 0.9762845849802372,
1807
+ "grad_norm": 0.8798792362213135,
1808
+ "learning_rate": 1.4205498696930332e-08,
1809
+ "loss": 1.4156,
1810
+ "step": 247
1811
+ },
1812
+ {
1813
+ "epoch": 0.9802371541501976,
1814
+ "grad_norm": 0.7928077578544617,
1815
+ "learning_rate": 9.866357858642206e-09,
1816
+ "loss": 1.1922,
1817
+ "step": 248
1818
+ },
1819
+ {
1820
+ "epoch": 0.9841897233201581,
1821
+ "grad_norm": 0.8758398294448853,
1822
+ "learning_rate": 6.315216969912663e-09,
1823
+ "loss": 1.4098,
1824
+ "step": 249
1825
+ },
1826
+ {
1827
+ "epoch": 0.9881422924901185,
1828
+ "grad_norm": 0.8929165005683899,
1829
+ "learning_rate": 3.5526367970539765e-09,
1830
+ "loss": 1.3601,
1831
+ "step": 250
1832
+ },
1833
+ {
1834
+ "epoch": 0.9920948616600791,
1835
+ "grad_norm": 0.9001562595367432,
1836
+ "learning_rate": 1.5790535835003006e-09,
1837
+ "loss": 1.2039,
1838
+ "step": 251
1839
+ },
1840
+ {
1841
+ "epoch": 0.9960474308300395,
1842
+ "grad_norm": 0.8405194878578186,
1843
+ "learning_rate": 3.9477898091944135e-10,
1844
+ "loss": 1.1514,
1845
+ "step": 252
1846
+ },
1847
+ {
1848
+ "epoch": 1.0,
1849
+ "grad_norm": 0.8833491206169128,
1850
+ "learning_rate": 0.0,
1851
+ "loss": 1.2887,
1852
+ "step": 253
1853
+ },
1854
+ {
1855
+ "epoch": 1.0,
1856
+ "step": 253,
1857
+ "total_flos": 8.443588829301965e+16,
1858
+ "train_loss": 1.3333067849219553,
1859
+ "train_runtime": 1520.8803,
1860
+ "train_samples_per_second": 1.329,
1861
+ "train_steps_per_second": 0.166
1862
+ }
1863
+ ],
1864
+ "logging_steps": 1,
1865
+ "max_steps": 253,
1866
+ "num_input_tokens_seen": 0,
1867
+ "num_train_epochs": 1,
1868
+ "save_steps": 253,
1869
+ "stateful_callbacks": {
1870
+ "TrainerControl": {
1871
+ "args": {
1872
+ "should_epoch_stop": false,
1873
+ "should_evaluate": false,
1874
+ "should_log": false,
1875
+ "should_save": true,
1876
+ "should_training_stop": true
1877
+ },
1878
+ "attributes": {}
1879
+ }
1880
+ },
1881
+ "total_flos": 8.443588829301965e+16,
1882
+ "train_batch_size": 1,
1883
+ "trial_name": null,
1884
+ "trial_params": null
1885
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6efe1ceae43d0c9c5ff8ab3995e3488753b6c4a723f869bbc2032f0d0fa2a91
3
+ size 7352
training_eval_loss.png ADDED
training_loss.png ADDED