Laugh1ng commited on
Commit
94cc0a4
·
verified ·
1 Parent(s): 920ef29

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: other
4
+ # base_model: /inspire/hdd/global_user/25015/models/Qwen2.5-3B-Instruct
5
+ base_model: Qwen/Qwen2.5-3B-Instruct
6
+ tags:
7
+ - llama-factory
8
+ - lora
9
+ - generated_from_trainer
10
+ model-index:
11
+ - name: 6w1_en_e5_l1
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # 6w1_en_e5_l1
19
+
20
+ This model is a fine-tuned version of [/inspire/hdd/global_user/25015/models/Qwen2.5-3B-Instruct](https://huggingface.co//inspire/hdd/global_user/25015/models/Qwen2.5-3B-Instruct) on the 6w1_en dataset.
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - learning_rate: 1e-05
40
+ - train_batch_size: 2
41
+ - eval_batch_size: 8
42
+ - seed: 42
43
+ - distributed_type: multi-GPU
44
+ - num_devices: 4
45
+ - gradient_accumulation_steps: 8
46
+ - total_train_batch_size: 64
47
+ - total_eval_batch_size: 32
48
+ - optimizer: Use OptimizerNames.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: 0.1
51
+ - num_epochs: 5.0
52
+
53
+ ### Training results
54
+
55
+
56
+
57
+ ### Framework versions
58
+
59
+ - PEFT 0.12.0
60
+ - Transformers 4.49.0
61
+ - Pytorch 2.6.0+cu124
62
+ - Datasets 3.2.0
63
+ - Tokenizers 0.21.0
adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/inspire/hdd/global_user/25015/models/Qwen2.5-3B-Instruct",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 16,
14
+ "lora_dropout": 0.0,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 8,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "up_proj",
24
+ "gate_proj",
25
+ "v_proj",
26
+ "down_proj",
27
+ "q_proj",
28
+ "k_proj",
29
+ "o_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b0713c240d42d76d3a87d9ec09de177a2291d1f13fc4eb7913b26e04842f350
3
+ size 59933632
added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.9949212136997,
3
+ "total_flos": 0.0,
4
+ "train_loss": 0.03522291941339332,
5
+ "train_runtime": 16429.2929,
6
+ "train_samples_per_second": 18.696,
7
+ "train_steps_per_second": 0.292
8
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
tokenizer_config.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0]['role'] == 'system' %}\n {{- messages[0]['content'] }}\n {%- else %}\n {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n {%- endif %}\n {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n {%- if messages[0]['role'] == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n {%- else %}\n {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role }}\n {%- if message.content %}\n {{- '\\n' + message.content }}\n {%- endif %}\n {%- for tool_call in message.tool_calls %}\n {%- if tool_call.function is defined %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '\\n<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- message.content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
199
+ "clean_up_tokenization_spaces": false,
200
+ "eos_token": "<|im_end|>",
201
+ "errors": "replace",
202
+ "extra_special_tokens": {},
203
+ "model_max_length": 2048,
204
+ "pad_token": "<|endoftext|>",
205
+ "padding_side": "right",
206
+ "split_special_tokens": false,
207
+ "tokenizer_class": "Qwen2Tokenizer",
208
+ "unk_token": null
209
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.9949212136997,
3
+ "total_flos": 0.0,
4
+ "train_loss": 0.03522291941339332,
5
+ "train_runtime": 16429.2929,
6
+ "train_samples_per_second": 18.696,
7
+ "train_steps_per_second": 0.292
8
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,480 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 10, "total_steps": 4795, "loss": 1.0155, "lr": 2.0833333333333333e-07, "epoch": 0.010418023180101576, "percentage": 0.21, "elapsed_time": "0:00:37", "remaining_time": "5:00:07"}
2
+ {"current_steps": 20, "total_steps": 4795, "loss": 1.0247, "lr": 4.1666666666666667e-07, "epoch": 0.020836046360203152, "percentage": 0.42, "elapsed_time": "0:00:59", "remaining_time": "3:57:50"}
3
+ {"current_steps": 30, "total_steps": 4795, "loss": 0.9569, "lr": 6.25e-07, "epoch": 0.03125406954030473, "percentage": 0.63, "elapsed_time": "0:01:21", "remaining_time": "3:36:29"}
4
+ {"current_steps": 40, "total_steps": 4795, "loss": 0.9531, "lr": 8.333333333333333e-07, "epoch": 0.041672092720406305, "percentage": 0.83, "elapsed_time": "0:01:51", "remaining_time": "3:40:21"}
5
+ {"current_steps": 50, "total_steps": 4795, "loss": 0.913, "lr": 1.0416666666666667e-06, "epoch": 0.052090115900507876, "percentage": 1.04, "elapsed_time": "0:02:24", "remaining_time": "3:48:23"}
6
+ {"current_steps": 60, "total_steps": 4795, "loss": 0.9508, "lr": 1.25e-06, "epoch": 0.06250813908060945, "percentage": 1.25, "elapsed_time": "0:03:01", "remaining_time": "3:58:25"}
7
+ {"current_steps": 70, "total_steps": 4795, "loss": 0.9788, "lr": 1.4583333333333335e-06, "epoch": 0.07292616226071102, "percentage": 1.46, "elapsed_time": "0:03:38", "remaining_time": "4:06:13"}
8
+ {"current_steps": 80, "total_steps": 4795, "loss": 0.9664, "lr": 1.6666666666666667e-06, "epoch": 0.08334418544081261, "percentage": 1.67, "elapsed_time": "0:04:16", "remaining_time": "4:11:34"}
9
+ {"current_steps": 90, "total_steps": 4795, "loss": 0.9333, "lr": 1.8750000000000003e-06, "epoch": 0.09376220862091418, "percentage": 1.88, "elapsed_time": "0:04:54", "remaining_time": "4:16:26"}
10
+ {"current_steps": 100, "total_steps": 4795, "loss": 0.8748, "lr": 2.0833333333333334e-06, "epoch": 0.10418023180101575, "percentage": 2.09, "elapsed_time": "0:05:31", "remaining_time": "4:19:42"}
11
+ {"current_steps": 110, "total_steps": 4795, "loss": 0.8221, "lr": 2.2916666666666666e-06, "epoch": 0.11459825498111734, "percentage": 2.29, "elapsed_time": "0:06:09", "remaining_time": "4:22:19"}
12
+ {"current_steps": 120, "total_steps": 4795, "loss": 0.7869, "lr": 2.5e-06, "epoch": 0.1250162781612189, "percentage": 2.5, "elapsed_time": "0:06:46", "remaining_time": "4:24:01"}
13
+ {"current_steps": 130, "total_steps": 4795, "loss": 0.7208, "lr": 2.7083333333333334e-06, "epoch": 0.13543430134132048, "percentage": 2.71, "elapsed_time": "0:07:25", "remaining_time": "4:26:42"}
14
+ {"current_steps": 140, "total_steps": 4795, "loss": 0.7022, "lr": 2.916666666666667e-06, "epoch": 0.14585232452142205, "percentage": 2.92, "elapsed_time": "0:08:07", "remaining_time": "4:30:22"}
15
+ {"current_steps": 150, "total_steps": 4795, "loss": 0.6499, "lr": 3.125e-06, "epoch": 0.15627034770152365, "percentage": 3.13, "elapsed_time": "0:08:48", "remaining_time": "4:32:43"}
16
+ {"current_steps": 160, "total_steps": 4795, "loss": 0.5223, "lr": 3.3333333333333333e-06, "epoch": 0.16668837088162522, "percentage": 3.34, "elapsed_time": "0:09:25", "remaining_time": "4:33:07"}
17
+ {"current_steps": 170, "total_steps": 4795, "loss": 0.426, "lr": 3.5416666666666673e-06, "epoch": 0.1771063940617268, "percentage": 3.55, "elapsed_time": "0:10:02", "remaining_time": "4:33:16"}
18
+ {"current_steps": 180, "total_steps": 4795, "loss": 0.2739, "lr": 3.7500000000000005e-06, "epoch": 0.18752441724182836, "percentage": 3.75, "elapsed_time": "0:10:40", "remaining_time": "4:33:31"}
19
+ {"current_steps": 190, "total_steps": 4795, "loss": 0.2038, "lr": 3.958333333333333e-06, "epoch": 0.19794244042192993, "percentage": 3.96, "elapsed_time": "0:11:17", "remaining_time": "4:33:30"}
20
+ {"current_steps": 200, "total_steps": 4795, "loss": 0.1047, "lr": 4.166666666666667e-06, "epoch": 0.2083604636020315, "percentage": 4.17, "elapsed_time": "0:11:54", "remaining_time": "4:33:28"}
21
+ {"current_steps": 210, "total_steps": 4795, "loss": 0.0695, "lr": 4.3750000000000005e-06, "epoch": 0.2187784867821331, "percentage": 4.38, "elapsed_time": "0:12:31", "remaining_time": "4:33:27"}
22
+ {"current_steps": 220, "total_steps": 4795, "loss": 0.0448, "lr": 4.583333333333333e-06, "epoch": 0.22919650996223467, "percentage": 4.59, "elapsed_time": "0:13:08", "remaining_time": "4:33:14"}
23
+ {"current_steps": 230, "total_steps": 4795, "loss": 0.0482, "lr": 4.791666666666668e-06, "epoch": 0.23961453314233624, "percentage": 4.8, "elapsed_time": "0:13:46", "remaining_time": "4:33:15"}
24
+ {"current_steps": 240, "total_steps": 4795, "loss": 0.0464, "lr": 5e-06, "epoch": 0.2500325563224378, "percentage": 5.01, "elapsed_time": "0:14:23", "remaining_time": "4:33:05"}
25
+ {"current_steps": 250, "total_steps": 4795, "loss": 0.026, "lr": 5.208333333333334e-06, "epoch": 0.2604505795025394, "percentage": 5.21, "elapsed_time": "0:15:01", "remaining_time": "4:33:12"}
26
+ {"current_steps": 260, "total_steps": 4795, "loss": 0.0131, "lr": 5.416666666666667e-06, "epoch": 0.27086860268264096, "percentage": 5.42, "elapsed_time": "0:15:39", "remaining_time": "4:33:04"}
27
+ {"current_steps": 270, "total_steps": 4795, "loss": 0.0267, "lr": 5.625e-06, "epoch": 0.28128662586274256, "percentage": 5.63, "elapsed_time": "0:16:16", "remaining_time": "4:32:43"}
28
+ {"current_steps": 280, "total_steps": 4795, "loss": 0.0192, "lr": 5.833333333333334e-06, "epoch": 0.2917046490428441, "percentage": 5.84, "elapsed_time": "0:16:53", "remaining_time": "4:32:25"}
29
+ {"current_steps": 290, "total_steps": 4795, "loss": 0.0221, "lr": 6.041666666666667e-06, "epoch": 0.3021226722229457, "percentage": 6.05, "elapsed_time": "0:17:31", "remaining_time": "4:32:11"}
30
+ {"current_steps": 300, "total_steps": 4795, "loss": 0.043, "lr": 6.25e-06, "epoch": 0.3125406954030473, "percentage": 6.26, "elapsed_time": "0:18:08", "remaining_time": "4:31:49"}
31
+ {"current_steps": 310, "total_steps": 4795, "loss": 0.049, "lr": 6.458333333333334e-06, "epoch": 0.32295871858314884, "percentage": 6.47, "elapsed_time": "0:18:46", "remaining_time": "4:31:31"}
32
+ {"current_steps": 320, "total_steps": 4795, "loss": 0.0131, "lr": 6.666666666666667e-06, "epoch": 0.33337674176325044, "percentage": 6.67, "elapsed_time": "0:19:23", "remaining_time": "4:31:07"}
33
+ {"current_steps": 330, "total_steps": 4795, "loss": 0.0244, "lr": 6.875e-06, "epoch": 0.343794764943352, "percentage": 6.88, "elapsed_time": "0:20:00", "remaining_time": "4:30:40"}
34
+ {"current_steps": 340, "total_steps": 4795, "loss": 0.0149, "lr": 7.083333333333335e-06, "epoch": 0.3542127881234536, "percentage": 7.09, "elapsed_time": "0:20:37", "remaining_time": "4:30:15"}
35
+ {"current_steps": 350, "total_steps": 4795, "loss": 0.0086, "lr": 7.291666666666667e-06, "epoch": 0.3646308113035551, "percentage": 7.3, "elapsed_time": "0:21:14", "remaining_time": "4:29:50"}
36
+ {"current_steps": 360, "total_steps": 4795, "loss": 0.0158, "lr": 7.500000000000001e-06, "epoch": 0.3750488344836567, "percentage": 7.51, "elapsed_time": "0:21:52", "remaining_time": "4:29:24"}
37
+ {"current_steps": 370, "total_steps": 4795, "loss": 0.0331, "lr": 7.708333333333334e-06, "epoch": 0.3854668576637583, "percentage": 7.72, "elapsed_time": "0:22:29", "remaining_time": "4:29:04"}
38
+ {"current_steps": 380, "total_steps": 4795, "loss": 0.0109, "lr": 7.916666666666667e-06, "epoch": 0.39588488084385987, "percentage": 7.92, "elapsed_time": "0:23:07", "remaining_time": "4:28:42"}
39
+ {"current_steps": 390, "total_steps": 4795, "loss": 0.0116, "lr": 8.125000000000001e-06, "epoch": 0.40630290402396146, "percentage": 8.13, "elapsed_time": "0:23:44", "remaining_time": "4:28:14"}
40
+ {"current_steps": 400, "total_steps": 4795, "loss": 0.0086, "lr": 8.333333333333334e-06, "epoch": 0.416720927204063, "percentage": 8.34, "elapsed_time": "0:24:22", "remaining_time": "4:27:49"}
41
+ {"current_steps": 410, "total_steps": 4795, "loss": 0.0257, "lr": 8.541666666666666e-06, "epoch": 0.4271389503841646, "percentage": 8.55, "elapsed_time": "0:24:59", "remaining_time": "4:27:19"}
42
+ {"current_steps": 420, "total_steps": 4795, "loss": 0.0211, "lr": 8.750000000000001e-06, "epoch": 0.4375569735642662, "percentage": 8.76, "elapsed_time": "0:25:37", "remaining_time": "4:26:55"}
43
+ {"current_steps": 430, "total_steps": 4795, "loss": 0.0195, "lr": 8.958333333333334e-06, "epoch": 0.44797499674436775, "percentage": 8.97, "elapsed_time": "0:26:15", "remaining_time": "4:26:29"}
44
+ {"current_steps": 440, "total_steps": 4795, "loss": 0.0042, "lr": 9.166666666666666e-06, "epoch": 0.45839301992446935, "percentage": 9.18, "elapsed_time": "0:26:52", "remaining_time": "4:26:03"}
45
+ {"current_steps": 450, "total_steps": 4795, "loss": 0.0225, "lr": 9.375000000000001e-06, "epoch": 0.4688110431045709, "percentage": 9.38, "elapsed_time": "0:27:29", "remaining_time": "4:25:31"}
46
+ {"current_steps": 460, "total_steps": 4795, "loss": 0.0207, "lr": 9.583333333333335e-06, "epoch": 0.4792290662846725, "percentage": 9.59, "elapsed_time": "0:28:06", "remaining_time": "4:24:56"}
47
+ {"current_steps": 470, "total_steps": 4795, "loss": 0.0184, "lr": 9.791666666666666e-06, "epoch": 0.48964708946477403, "percentage": 9.8, "elapsed_time": "0:28:44", "remaining_time": "4:24:26"}
48
+ {"current_steps": 480, "total_steps": 4795, "loss": 0.0082, "lr": 1e-05, "epoch": 0.5000651126448756, "percentage": 10.01, "elapsed_time": "0:29:11", "remaining_time": "4:22:21"}
49
+ {"current_steps": 490, "total_steps": 4795, "loss": 0.0201, "lr": 9.999867481584167e-06, "epoch": 0.5104831358249772, "percentage": 10.22, "elapsed_time": "0:29:32", "remaining_time": "4:19:33"}
50
+ {"current_steps": 500, "total_steps": 4795, "loss": 0.0118, "lr": 9.99946993336112e-06, "epoch": 0.5209011590050788, "percentage": 10.43, "elapsed_time": "0:29:55", "remaining_time": "4:17:01"}
51
+ {"current_steps": 510, "total_steps": 4795, "loss": 0.0028, "lr": 9.998807376403843e-06, "epoch": 0.5313191821851804, "percentage": 10.64, "elapsed_time": "0:30:17", "remaining_time": "4:14:28"}
52
+ {"current_steps": 520, "total_steps": 4795, "loss": 0.0248, "lr": 9.997879845832736e-06, "epoch": 0.5417372053652819, "percentage": 10.84, "elapsed_time": "0:30:39", "remaining_time": "4:12:00"}
53
+ {"current_steps": 530, "total_steps": 4795, "loss": 0.0026, "lr": 9.996687390813751e-06, "epoch": 0.5521552285453835, "percentage": 11.05, "elapsed_time": "0:31:04", "remaining_time": "4:10:05"}
54
+ {"current_steps": 540, "total_steps": 4795, "loss": 0.0216, "lr": 9.995230074555788e-06, "epoch": 0.5625732517254851, "percentage": 11.26, "elapsed_time": "0:31:26", "remaining_time": "4:07:44"}
55
+ {"current_steps": 550, "total_steps": 4795, "loss": 0.0169, "lr": 9.993507974307346e-06, "epoch": 0.5729912749055867, "percentage": 11.47, "elapsed_time": "0:31:48", "remaining_time": "4:05:28"}
56
+ {"current_steps": 560, "total_steps": 4795, "loss": 0.0111, "lr": 9.991521181352419e-06, "epoch": 0.5834092980856882, "percentage": 11.68, "elapsed_time": "0:32:10", "remaining_time": "4:03:17"}
57
+ {"current_steps": 570, "total_steps": 4795, "loss": 0.0115, "lr": 9.989269801005675e-06, "epoch": 0.5938273212657899, "percentage": 11.89, "elapsed_time": "0:32:32", "remaining_time": "4:01:08"}
58
+ {"current_steps": 580, "total_steps": 4795, "loss": 0.0072, "lr": 9.986753952606851e-06, "epoch": 0.6042453444458914, "percentage": 12.1, "elapsed_time": "0:33:03", "remaining_time": "4:00:12"}
59
+ {"current_steps": 590, "total_steps": 4795, "loss": 0.0056, "lr": 9.983973769514448e-06, "epoch": 0.6146633676259929, "percentage": 12.3, "elapsed_time": "0:33:34", "remaining_time": "3:59:17"}
60
+ {"current_steps": 600, "total_steps": 4795, "loss": 0.0153, "lr": 9.98092939909865e-06, "epoch": 0.6250813908060946, "percentage": 12.51, "elapsed_time": "0:34:06", "remaining_time": "3:58:30"}
61
+ {"current_steps": 610, "total_steps": 4795, "loss": 0.0186, "lr": 9.977621002733512e-06, "epoch": 0.6354994139861961, "percentage": 12.72, "elapsed_time": "0:34:39", "remaining_time": "3:57:46"}
62
+ {"current_steps": 620, "total_steps": 4795, "loss": 0.0042, "lr": 9.974048755788416e-06, "epoch": 0.6459174371662977, "percentage": 12.93, "elapsed_time": "0:35:12", "remaining_time": "3:57:02"}
63
+ {"current_steps": 630, "total_steps": 4795, "loss": 0.0136, "lr": 9.970212847618761e-06, "epoch": 0.6563354603463992, "percentage": 13.14, "elapsed_time": "0:35:44", "remaining_time": "3:56:18"}
64
+ {"current_steps": 640, "total_steps": 4795, "loss": 0.0068, "lr": 9.96611348155594e-06, "epoch": 0.6667534835265009, "percentage": 13.35, "elapsed_time": "0:36:17", "remaining_time": "3:55:36"}
65
+ {"current_steps": 650, "total_steps": 4795, "loss": 0.0075, "lr": 9.961750874896548e-06, "epoch": 0.6771715067066024, "percentage": 13.56, "elapsed_time": "0:36:49", "remaining_time": "3:54:51"}
66
+ {"current_steps": 660, "total_steps": 4795, "loss": 0.0154, "lr": 9.957125258890877e-06, "epoch": 0.687589529886704, "percentage": 13.76, "elapsed_time": "0:37:22", "remaining_time": "3:54:09"}
67
+ {"current_steps": 670, "total_steps": 4795, "loss": 0.008, "lr": 9.952236878730648e-06, "epoch": 0.6980075530668056, "percentage": 13.97, "elapsed_time": "0:37:54", "remaining_time": "3:53:25"}
68
+ {"current_steps": 680, "total_steps": 4795, "loss": 0.0293, "lr": 9.947085993536019e-06, "epoch": 0.7084255762469072, "percentage": 14.18, "elapsed_time": "0:38:27", "remaining_time": "3:52:44"}
69
+ {"current_steps": 690, "total_steps": 4795, "loss": 0.0225, "lr": 9.941672876341848e-06, "epoch": 0.7188435994270087, "percentage": 14.39, "elapsed_time": "0:39:00", "remaining_time": "3:52:02"}
70
+ {"current_steps": 700, "total_steps": 4795, "loss": 0.003, "lr": 9.935997814083221e-06, "epoch": 0.7292616226071102, "percentage": 14.6, "elapsed_time": "0:39:32", "remaining_time": "3:51:21"}
71
+ {"current_steps": 710, "total_steps": 4795, "loss": 0.0134, "lr": 9.930061107580245e-06, "epoch": 0.7396796457872119, "percentage": 14.81, "elapsed_time": "0:40:05", "remaining_time": "3:50:41"}
72
+ {"current_steps": 720, "total_steps": 4795, "loss": 0.0228, "lr": 9.923863071522092e-06, "epoch": 0.7500976689673134, "percentage": 15.02, "elapsed_time": "0:40:38", "remaining_time": "3:50:02"}
73
+ {"current_steps": 730, "total_steps": 4795, "loss": 0.0066, "lr": 9.917404034450333e-06, "epoch": 0.760515692147415, "percentage": 15.22, "elapsed_time": "0:41:10", "remaining_time": "3:49:17"}
74
+ {"current_steps": 740, "total_steps": 4795, "loss": 0.0034, "lr": 9.91068433874151e-06, "epoch": 0.7709337153275166, "percentage": 15.43, "elapsed_time": "0:41:43", "remaining_time": "3:48:38"}
75
+ {"current_steps": 750, "total_steps": 4795, "loss": 0.0172, "lr": 9.903704340589e-06, "epoch": 0.7813517385076182, "percentage": 15.64, "elapsed_time": "0:42:16", "remaining_time": "3:48:01"}
76
+ {"current_steps": 760, "total_steps": 4795, "loss": 0.0158, "lr": 9.896464409984115e-06, "epoch": 0.7917697616877197, "percentage": 15.85, "elapsed_time": "0:42:48", "remaining_time": "3:47:18"}
77
+ {"current_steps": 770, "total_steps": 4795, "loss": 0.0118, "lr": 9.888964930696514e-06, "epoch": 0.8021877848678214, "percentage": 16.06, "elapsed_time": "0:43:21", "remaining_time": "3:46:37"}
78
+ {"current_steps": 780, "total_steps": 4795, "loss": 0.0091, "lr": 9.881206300253838e-06, "epoch": 0.8126058080479229, "percentage": 16.27, "elapsed_time": "0:43:54", "remaining_time": "3:45:59"}
79
+ {"current_steps": 790, "total_steps": 4795, "loss": 0.0195, "lr": 9.87318892992066e-06, "epoch": 0.8230238312280245, "percentage": 16.48, "elapsed_time": "0:44:27", "remaining_time": "3:45:21"}
80
+ {"current_steps": 800, "total_steps": 4795, "loss": 0.0072, "lr": 9.864913244676661e-06, "epoch": 0.833441854408126, "percentage": 16.68, "elapsed_time": "0:45:00", "remaining_time": "3:44:46"}
81
+ {"current_steps": 810, "total_steps": 4795, "loss": 0.0102, "lr": 9.856379683194122e-06, "epoch": 0.8438598775882277, "percentage": 16.89, "elapsed_time": "0:45:33", "remaining_time": "3:44:06"}
82
+ {"current_steps": 820, "total_steps": 4795, "loss": 0.0037, "lr": 9.847588697814662e-06, "epoch": 0.8542779007683292, "percentage": 17.1, "elapsed_time": "0:46:05", "remaining_time": "3:43:26"}
83
+ {"current_steps": 830, "total_steps": 4795, "loss": 0.0057, "lr": 9.838540754525266e-06, "epoch": 0.8646959239484308, "percentage": 17.31, "elapsed_time": "0:46:37", "remaining_time": "3:42:45"}
84
+ {"current_steps": 840, "total_steps": 4795, "loss": 0.0094, "lr": 9.829236332933573e-06, "epoch": 0.8751139471285324, "percentage": 17.52, "elapsed_time": "0:47:11", "remaining_time": "3:42:09"}
85
+ {"current_steps": 850, "total_steps": 4795, "loss": 0.0051, "lr": 9.819675926242473e-06, "epoch": 0.885531970308634, "percentage": 17.73, "elapsed_time": "0:47:43", "remaining_time": "3:41:31"}
86
+ {"current_steps": 860, "total_steps": 4795, "loss": 0.0179, "lr": 9.809860041223945e-06, "epoch": 0.8959499934887355, "percentage": 17.94, "elapsed_time": "0:48:16", "remaining_time": "3:40:51"}
87
+ {"current_steps": 870, "total_steps": 4795, "loss": 0.0257, "lr": 9.799789198192197e-06, "epoch": 0.906368016668837, "percentage": 18.14, "elapsed_time": "0:48:48", "remaining_time": "3:40:13"}
88
+ {"current_steps": 880, "total_steps": 4795, "loss": 0.0044, "lr": 9.789463930976098e-06, "epoch": 0.9167860398489387, "percentage": 18.35, "elapsed_time": "0:49:21", "remaining_time": "3:39:34"}
89
+ {"current_steps": 890, "total_steps": 4795, "loss": 0.0096, "lr": 9.77888478689087e-06, "epoch": 0.9272040630290402, "percentage": 18.56, "elapsed_time": "0:49:54", "remaining_time": "3:38:58"}
90
+ {"current_steps": 900, "total_steps": 4795, "loss": 0.0132, "lr": 9.76805232670908e-06, "epoch": 0.9376220862091418, "percentage": 18.77, "elapsed_time": "0:50:27", "remaining_time": "3:38:20"}
91
+ {"current_steps": 910, "total_steps": 4795, "loss": 0.0021, "lr": 9.756967124630911e-06, "epoch": 0.9480401093892434, "percentage": 18.98, "elapsed_time": "0:51:00", "remaining_time": "3:37:44"}
92
+ {"current_steps": 920, "total_steps": 4795, "loss": 0.0027, "lr": 9.745629768253735e-06, "epoch": 0.958458132569345, "percentage": 19.19, "elapsed_time": "0:51:32", "remaining_time": "3:37:05"}
93
+ {"current_steps": 930, "total_steps": 4795, "loss": 0.0081, "lr": 9.734040858540947e-06, "epoch": 0.9688761557494465, "percentage": 19.4, "elapsed_time": "0:52:04", "remaining_time": "3:36:26"}
94
+ {"current_steps": 940, "total_steps": 4795, "loss": 0.0013, "lr": 9.722201009790135e-06, "epoch": 0.9792941789295481, "percentage": 19.6, "elapsed_time": "0:52:37", "remaining_time": "3:35:48"}
95
+ {"current_steps": 950, "total_steps": 4795, "loss": 0.0154, "lr": 9.710110849600498e-06, "epoch": 0.9897122021096497, "percentage": 19.81, "elapsed_time": "0:53:09", "remaining_time": "3:35:10"}
96
+ {"current_steps": 960, "total_steps": 4795, "loss": 0.0182, "lr": 9.697771018839586e-06, "epoch": 1.0, "percentage": 20.02, "elapsed_time": "0:53:42", "remaining_time": "3:34:31"}
97
+ {"current_steps": 970, "total_steps": 4795, "loss": 0.0078, "lr": 9.685182171609328e-06, "epoch": 1.0104180231801017, "percentage": 20.23, "elapsed_time": "0:54:15", "remaining_time": "3:33:56"}
98
+ {"current_steps": 980, "total_steps": 4795, "loss": 0.0026, "lr": 9.672344975211361e-06, "epoch": 1.020836046360203, "percentage": 20.44, "elapsed_time": "0:54:48", "remaining_time": "3:33:19"}
99
+ {"current_steps": 990, "total_steps": 4795, "loss": 0.0013, "lr": 9.659260110111659e-06, "epoch": 1.0312540695403047, "percentage": 20.65, "elapsed_time": "0:55:20", "remaining_time": "3:32:43"}
100
+ {"current_steps": 1000, "total_steps": 4795, "loss": 0.0006, "lr": 9.645928269904459e-06, "epoch": 1.0416720927204064, "percentage": 20.86, "elapsed_time": "0:55:54", "remaining_time": "3:32:08"}
101
+ {"current_steps": 1010, "total_steps": 4795, "loss": 0.0078, "lr": 9.632350161275496e-06, "epoch": 1.0520901159005078, "percentage": 21.06, "elapsed_time": "0:56:26", "remaining_time": "3:31:31"}
102
+ {"current_steps": 1020, "total_steps": 4795, "loss": 0.0033, "lr": 9.618526503964552e-06, "epoch": 1.0625081390806095, "percentage": 21.27, "elapsed_time": "0:56:59", "remaining_time": "3:30:55"}
103
+ {"current_steps": 1030, "total_steps": 4795, "loss": 0.0041, "lr": 9.604458030727292e-06, "epoch": 1.072926162260711, "percentage": 21.48, "elapsed_time": "0:57:31", "remaining_time": "3:30:17"}
104
+ {"current_steps": 1040, "total_steps": 4795, "loss": 0.0015, "lr": 9.590145487296431e-06, "epoch": 1.0833441854408126, "percentage": 21.69, "elapsed_time": "0:58:04", "remaining_time": "3:29:41"}
105
+ {"current_steps": 1050, "total_steps": 4795, "loss": 0.0003, "lr": 9.575589632342201e-06, "epoch": 1.0937622086209142, "percentage": 21.9, "elapsed_time": "0:58:36", "remaining_time": "3:29:03"}
106
+ {"current_steps": 1060, "total_steps": 4795, "loss": 0.0157, "lr": 9.560791237432141e-06, "epoch": 1.1041802318010157, "percentage": 22.11, "elapsed_time": "0:59:09", "remaining_time": "3:28:27"}
107
+ {"current_steps": 1070, "total_steps": 4795, "loss": 0.0074, "lr": 9.545751086990187e-06, "epoch": 1.1145982549811173, "percentage": 22.31, "elapsed_time": "0:59:42", "remaining_time": "3:27:50"}
108
+ {"current_steps": 1080, "total_steps": 4795, "loss": 0.0001, "lr": 9.530469978255105e-06, "epoch": 1.125016278161219, "percentage": 22.52, "elapsed_time": "1:00:14", "remaining_time": "3:27:13"}
109
+ {"current_steps": 1090, "total_steps": 4795, "loss": 0.0074, "lr": 9.514948721238227e-06, "epoch": 1.1354343013413204, "percentage": 22.73, "elapsed_time": "1:00:46", "remaining_time": "3:26:35"}
110
+ {"current_steps": 1100, "total_steps": 4795, "loss": 0.0108, "lr": 9.499188138680504e-06, "epoch": 1.145852324521422, "percentage": 22.94, "elapsed_time": "1:01:19", "remaining_time": "3:25:59"}
111
+ {"current_steps": 1110, "total_steps": 4795, "loss": 0.002, "lr": 9.48318906600891e-06, "epoch": 1.1562703477015237, "percentage": 23.15, "elapsed_time": "1:01:52", "remaining_time": "3:25:24"}
112
+ {"current_steps": 1120, "total_steps": 4795, "loss": 0.0033, "lr": 9.466952351292158e-06, "epoch": 1.1666883708816251, "percentage": 23.36, "elapsed_time": "1:02:25", "remaining_time": "3:24:48"}
113
+ {"current_steps": 1130, "total_steps": 4795, "loss": 0.0002, "lr": 9.450478855195724e-06, "epoch": 1.1771063940617268, "percentage": 23.57, "elapsed_time": "1:02:57", "remaining_time": "3:24:12"}
114
+ {"current_steps": 1140, "total_steps": 4795, "loss": 0.0033, "lr": 9.433769450936254e-06, "epoch": 1.1875244172418284, "percentage": 23.77, "elapsed_time": "1:03:30", "remaining_time": "3:23:36"}
115
+ {"current_steps": 1150, "total_steps": 4795, "loss": 0.0005, "lr": 9.416825024235262e-06, "epoch": 1.1979424404219299, "percentage": 23.98, "elapsed_time": "1:04:03", "remaining_time": "3:23:00"}
116
+ {"current_steps": 1160, "total_steps": 4795, "loss": 0.0062, "lr": 9.399646473272181e-06, "epoch": 1.2083604636020315, "percentage": 24.19, "elapsed_time": "1:04:35", "remaining_time": "3:22:25"}
117
+ {"current_steps": 1170, "total_steps": 4795, "loss": 0.0018, "lr": 9.382234708636753e-06, "epoch": 1.2187784867821332, "percentage": 24.4, "elapsed_time": "1:05:08", "remaining_time": "3:21:48"}
118
+ {"current_steps": 1180, "total_steps": 4795, "loss": 0.0023, "lr": 9.364590653280767e-06, "epoch": 1.2291965099622346, "percentage": 24.61, "elapsed_time": "1:05:40", "remaining_time": "3:21:13"}
119
+ {"current_steps": 1190, "total_steps": 4795, "loss": 0.0065, "lr": 9.346715242469128e-06, "epoch": 1.2396145331423363, "percentage": 24.82, "elapsed_time": "1:06:13", "remaining_time": "3:20:37"}
120
+ {"current_steps": 1200, "total_steps": 4795, "loss": 0.0004, "lr": 9.328609423730285e-06, "epoch": 1.250032556322438, "percentage": 25.03, "elapsed_time": "1:06:45", "remaining_time": "3:19:59"}
121
+ {"current_steps": 1210, "total_steps": 4795, "loss": 0.0122, "lr": 9.310274156806006e-06, "epoch": 1.2604505795025394, "percentage": 25.23, "elapsed_time": "1:07:17", "remaining_time": "3:19:21"}
122
+ {"current_steps": 1220, "total_steps": 4795, "loss": 0.0105, "lr": 9.291710413600498e-06, "epoch": 1.270868602682641, "percentage": 25.44, "elapsed_time": "1:07:49", "remaining_time": "3:18:45"}
123
+ {"current_steps": 1230, "total_steps": 4795, "loss": 0.0005, "lr": 9.272919178128902e-06, "epoch": 1.2812866258627427, "percentage": 25.65, "elapsed_time": "1:08:22", "remaining_time": "3:18:10"}
124
+ {"current_steps": 1240, "total_steps": 4795, "loss": 0.0032, "lr": 9.253901446465116e-06, "epoch": 1.291704649042844, "percentage": 25.86, "elapsed_time": "1:08:55", "remaining_time": "3:17:37"}
125
+ {"current_steps": 1250, "total_steps": 4795, "loss": 0.0003, "lr": 9.234658226689015e-06, "epoch": 1.3021226722229458, "percentage": 26.07, "elapsed_time": "1:09:29", "remaining_time": "3:17:03"}
126
+ {"current_steps": 1260, "total_steps": 4795, "loss": 0.0046, "lr": 9.215190538832995e-06, "epoch": 1.3125406954030474, "percentage": 26.28, "elapsed_time": "1:10:01", "remaining_time": "3:16:27"}
127
+ {"current_steps": 1270, "total_steps": 4795, "loss": 0.0001, "lr": 9.195499414827917e-06, "epoch": 1.3229587185831488, "percentage": 26.49, "elapsed_time": "1:10:33", "remaining_time": "3:15:51"}
128
+ {"current_steps": 1280, "total_steps": 4795, "loss": 0.0267, "lr": 9.175585898448408e-06, "epoch": 1.3333767417632505, "percentage": 26.69, "elapsed_time": "1:11:06", "remaining_time": "3:15:15"}
129
+ {"current_steps": 1290, "total_steps": 4795, "loss": 0.0157, "lr": 9.155451045257523e-06, "epoch": 1.343794764943352, "percentage": 26.9, "elapsed_time": "1:11:38", "remaining_time": "3:14:40"}
130
+ {"current_steps": 1300, "total_steps": 4795, "loss": 0.002, "lr": 9.135095922550801e-06, "epoch": 1.3542127881234536, "percentage": 27.11, "elapsed_time": "1:12:12", "remaining_time": "3:14:06"}
131
+ {"current_steps": 1310, "total_steps": 4795, "loss": 0.008, "lr": 9.114521609299691e-06, "epoch": 1.364630811303555, "percentage": 27.32, "elapsed_time": "1:12:45", "remaining_time": "3:13:32"}
132
+ {"current_steps": 1320, "total_steps": 4795, "loss": 0.0023, "lr": 9.09372919609435e-06, "epoch": 1.3750488344836567, "percentage": 27.53, "elapsed_time": "1:13:18", "remaining_time": "3:12:58"}
133
+ {"current_steps": 1330, "total_steps": 4795, "loss": 0.013, "lr": 9.072719785085842e-06, "epoch": 1.3854668576637583, "percentage": 27.74, "elapsed_time": "1:13:50", "remaining_time": "3:12:23"}
134
+ {"current_steps": 1340, "total_steps": 4795, "loss": 0.0098, "lr": 9.051494489927714e-06, "epoch": 1.3958848808438598, "percentage": 27.95, "elapsed_time": "1:14:22", "remaining_time": "3:11:46"}
135
+ {"current_steps": 1350, "total_steps": 4795, "loss": 0.0152, "lr": 9.030054435716961e-06, "epoch": 1.4063029040239614, "percentage": 28.15, "elapsed_time": "1:14:54", "remaining_time": "3:11:09"}
136
+ {"current_steps": 1360, "total_steps": 4795, "loss": 0.0014, "lr": 9.008400758934392e-06, "epoch": 1.416720927204063, "percentage": 28.36, "elapsed_time": "1:15:27", "remaining_time": "3:10:35"}
137
+ {"current_steps": 1370, "total_steps": 4795, "loss": 0.0089, "lr": 8.986534607384383e-06, "epoch": 1.4271389503841645, "percentage": 28.57, "elapsed_time": "1:16:00", "remaining_time": "3:10:02"}
138
+ {"current_steps": 1380, "total_steps": 4795, "loss": 0.0008, "lr": 8.96445714013404e-06, "epoch": 1.4375569735642661, "percentage": 28.78, "elapsed_time": "1:16:33", "remaining_time": "3:09:27"}
139
+ {"current_steps": 1390, "total_steps": 4795, "loss": 0.0177, "lr": 8.942169527451756e-06, "epoch": 1.4479749967443678, "percentage": 28.99, "elapsed_time": "1:17:06", "remaining_time": "3:08:53"}
140
+ {"current_steps": 1400, "total_steps": 4795, "loss": 0.0177, "lr": 8.919672950745185e-06, "epoch": 1.4583930199244692, "percentage": 29.2, "elapsed_time": "1:17:39", "remaining_time": "3:08:19"}
141
+ {"current_steps": 1410, "total_steps": 4795, "loss": 0.0087, "lr": 8.896968602498605e-06, "epoch": 1.468811043104571, "percentage": 29.41, "elapsed_time": "1:18:12", "remaining_time": "3:07:44"}
142
+ {"current_steps": 1420, "total_steps": 4795, "loss": 0.0023, "lr": 8.874057686209727e-06, "epoch": 1.4792290662846725, "percentage": 29.61, "elapsed_time": "1:18:44", "remaining_time": "3:07:08"}
143
+ {"current_steps": 1430, "total_steps": 4795, "loss": 0.01, "lr": 8.850941416325876e-06, "epoch": 1.489647089464774, "percentage": 29.82, "elapsed_time": "1:19:16", "remaining_time": "3:06:33"}
144
+ {"current_steps": 1440, "total_steps": 4795, "loss": 0.0129, "lr": 8.827621018179644e-06, "epoch": 1.5000651126448756, "percentage": 30.03, "elapsed_time": "1:19:49", "remaining_time": "3:05:59"}
145
+ {"current_steps": 1450, "total_steps": 4795, "loss": 0.0049, "lr": 8.804097727923916e-06, "epoch": 1.5104831358249773, "percentage": 30.24, "elapsed_time": "1:20:22", "remaining_time": "3:05:24"}
146
+ {"current_steps": 1460, "total_steps": 4795, "loss": 0.0104, "lr": 8.780372792466356e-06, "epoch": 1.5209011590050787, "percentage": 30.45, "elapsed_time": "1:20:54", "remaining_time": "3:04:48"}
147
+ {"current_steps": 1470, "total_steps": 4795, "loss": 0.0032, "lr": 8.75644746940331e-06, "epoch": 1.5313191821851804, "percentage": 30.66, "elapsed_time": "1:21:27", "remaining_time": "3:04:15"}
148
+ {"current_steps": 1480, "total_steps": 4795, "loss": 0.008, "lr": 8.732323026953141e-06, "epoch": 1.541737205365282, "percentage": 30.87, "elapsed_time": "1:22:00", "remaining_time": "3:03:40"}
149
+ {"current_steps": 1490, "total_steps": 4795, "loss": 0.0052, "lr": 8.708000743889009e-06, "epoch": 1.5521552285453835, "percentage": 31.07, "elapsed_time": "1:22:33", "remaining_time": "3:03:07"}
150
+ {"current_steps": 1500, "total_steps": 4795, "loss": 0.0151, "lr": 8.68348190947108e-06, "epoch": 1.5625732517254851, "percentage": 31.28, "elapsed_time": "1:23:05", "remaining_time": "3:02:32"}
151
+ {"current_steps": 1510, "total_steps": 4795, "loss": 0.0012, "lr": 8.658767823378198e-06, "epoch": 1.5729912749055868, "percentage": 31.49, "elapsed_time": "1:23:38", "remaining_time": "3:01:57"}
152
+ {"current_steps": 1520, "total_steps": 4795, "loss": 0.0039, "lr": 8.63385979563897e-06, "epoch": 1.5834092980856882, "percentage": 31.7, "elapsed_time": "1:24:11", "remaining_time": "3:01:22"}
153
+ {"current_steps": 1530, "total_steps": 4795, "loss": 0.0013, "lr": 8.608759146562352e-06, "epoch": 1.5938273212657899, "percentage": 31.91, "elapsed_time": "1:24:43", "remaining_time": "3:00:48"}
154
+ {"current_steps": 1540, "total_steps": 4795, "loss": 0.0007, "lr": 8.583467206667643e-06, "epoch": 1.6042453444458915, "percentage": 32.12, "elapsed_time": "1:25:15", "remaining_time": "3:00:13"}
155
+ {"current_steps": 1550, "total_steps": 4795, "loss": 0.0005, "lr": 8.557985316613967e-06, "epoch": 1.614663367625993, "percentage": 32.33, "elapsed_time": "1:25:48", "remaining_time": "2:59:39"}
156
+ {"current_steps": 1560, "total_steps": 4795, "loss": 0.0003, "lr": 8.532314827129207e-06, "epoch": 1.6250813908060946, "percentage": 32.53, "elapsed_time": "1:26:20", "remaining_time": "2:59:03"}
157
+ {"current_steps": 1570, "total_steps": 4795, "loss": 0.0042, "lr": 8.5064570989384e-06, "epoch": 1.6354994139861962, "percentage": 32.74, "elapsed_time": "1:26:53", "remaining_time": "2:58:28"}
158
+ {"current_steps": 1580, "total_steps": 4795, "loss": 0.0042, "lr": 8.480413502691618e-06, "epoch": 1.6459174371662977, "percentage": 32.95, "elapsed_time": "1:27:25", "remaining_time": "2:57:54"}
159
+ {"current_steps": 1590, "total_steps": 4795, "loss": 0.0009, "lr": 8.454185418891305e-06, "epoch": 1.6563354603463991, "percentage": 33.16, "elapsed_time": "1:27:57", "remaining_time": "2:57:18"}
160
+ {"current_steps": 1600, "total_steps": 4795, "loss": 0.0071, "lr": 8.427774237819113e-06, "epoch": 1.666753483526501, "percentage": 33.37, "elapsed_time": "1:28:30", "remaining_time": "2:56:45"}
161
+ {"current_steps": 1610, "total_steps": 4795, "loss": 0.0078, "lr": 8.401181359462187e-06, "epoch": 1.6771715067066024, "percentage": 33.58, "elapsed_time": "1:29:03", "remaining_time": "2:56:11"}
162
+ {"current_steps": 1620, "total_steps": 4795, "loss": 0.0003, "lr": 8.374408193438977e-06, "epoch": 1.6875895298867039, "percentage": 33.79, "elapsed_time": "1:29:35", "remaining_time": "2:55:36"}
163
+ {"current_steps": 1630, "total_steps": 4795, "loss": 0.0054, "lr": 8.347456158924496e-06, "epoch": 1.6980075530668057, "percentage": 33.99, "elapsed_time": "1:30:08", "remaining_time": "2:55:01"}
164
+ {"current_steps": 1640, "total_steps": 4795, "loss": 0.0016, "lr": 8.320326684575116e-06, "epoch": 1.7084255762469072, "percentage": 34.2, "elapsed_time": "1:30:41", "remaining_time": "2:54:27"}
165
+ {"current_steps": 1650, "total_steps": 4795, "loss": 0.0008, "lr": 8.29302120845282e-06, "epoch": 1.7188435994270086, "percentage": 34.41, "elapsed_time": "1:31:13", "remaining_time": "2:53:52"}
166
+ {"current_steps": 1660, "total_steps": 4795, "loss": 0.0011, "lr": 8.265541177948986e-06, "epoch": 1.7292616226071102, "percentage": 34.62, "elapsed_time": "1:31:46", "remaining_time": "2:53:19"}
167
+ {"current_steps": 1670, "total_steps": 4795, "loss": 0.0087, "lr": 8.237888049707656e-06, "epoch": 1.739679645787212, "percentage": 34.83, "elapsed_time": "1:32:18", "remaining_time": "2:52:44"}
168
+ {"current_steps": 1680, "total_steps": 4795, "loss": 0.0025, "lr": 8.210063289548328e-06, "epoch": 1.7500976689673133, "percentage": 35.04, "elapsed_time": "1:32:51", "remaining_time": "2:52:09"}
169
+ {"current_steps": 1690, "total_steps": 4795, "loss": 0.0044, "lr": 8.182068372388259e-06, "epoch": 1.760515692147415, "percentage": 35.25, "elapsed_time": "1:33:23", "remaining_time": "2:51:35"}
170
+ {"current_steps": 1700, "total_steps": 4795, "loss": 0.0011, "lr": 8.153904782164275e-06, "epoch": 1.7709337153275166, "percentage": 35.45, "elapsed_time": "1:33:56", "remaining_time": "2:51:00"}
171
+ {"current_steps": 1710, "total_steps": 4795, "loss": 0.0028, "lr": 8.125574011754125e-06, "epoch": 1.781351738507618, "percentage": 35.66, "elapsed_time": "1:34:28", "remaining_time": "2:50:26"}
172
+ {"current_steps": 1720, "total_steps": 4795, "loss": 0.0003, "lr": 8.097077562897332e-06, "epoch": 1.7917697616877197, "percentage": 35.87, "elapsed_time": "1:35:01", "remaining_time": "2:49:53"}
173
+ {"current_steps": 1730, "total_steps": 4795, "loss": 0.0082, "lr": 8.0684169461156e-06, "epoch": 1.8021877848678214, "percentage": 36.08, "elapsed_time": "1:35:33", "remaining_time": "2:49:18"}
174
+ {"current_steps": 1740, "total_steps": 4795, "loss": 0.0069, "lr": 8.03959368063274e-06, "epoch": 1.8126058080479228, "percentage": 36.29, "elapsed_time": "1:36:06", "remaining_time": "2:48:44"}
175
+ {"current_steps": 1750, "total_steps": 4795, "loss": 0.0107, "lr": 8.01060929429415e-06, "epoch": 1.8230238312280245, "percentage": 36.5, "elapsed_time": "1:36:39", "remaining_time": "2:48:10"}
176
+ {"current_steps": 1760, "total_steps": 4795, "loss": 0.0082, "lr": 7.981465323485808e-06, "epoch": 1.8334418544081261, "percentage": 36.7, "elapsed_time": "1:37:11", "remaining_time": "2:47:36"}
177
+ {"current_steps": 1770, "total_steps": 4795, "loss": 0.0009, "lr": 7.952163313052856e-06, "epoch": 1.8438598775882276, "percentage": 36.91, "elapsed_time": "1:37:44", "remaining_time": "2:47:02"}
178
+ {"current_steps": 1780, "total_steps": 4795, "loss": 0.0082, "lr": 7.922704816217693e-06, "epoch": 1.8542779007683292, "percentage": 37.12, "elapsed_time": "1:38:16", "remaining_time": "2:46:28"}
179
+ {"current_steps": 1790, "total_steps": 4795, "loss": 0.0239, "lr": 7.893091394497651e-06, "epoch": 1.8646959239484309, "percentage": 37.33, "elapsed_time": "1:38:49", "remaining_time": "2:45:54"}
180
+ {"current_steps": 1800, "total_steps": 4795, "loss": 0.0019, "lr": 7.863324617622227e-06, "epoch": 1.8751139471285323, "percentage": 37.54, "elapsed_time": "1:39:22", "remaining_time": "2:45:20"}
181
+ {"current_steps": 1810, "total_steps": 4795, "loss": 0.0004, "lr": 7.833406063449866e-06, "epoch": 1.885531970308634, "percentage": 37.75, "elapsed_time": "1:39:55", "remaining_time": "2:44:47"}
182
+ {"current_steps": 1820, "total_steps": 4795, "loss": 0.0025, "lr": 7.803337317884328e-06, "epoch": 1.8959499934887356, "percentage": 37.96, "elapsed_time": "1:40:27", "remaining_time": "2:44:12"}
183
+ {"current_steps": 1830, "total_steps": 4795, "loss": 0.0008, "lr": 7.773119974790626e-06, "epoch": 1.906368016668837, "percentage": 38.16, "elapsed_time": "1:41:00", "remaining_time": "2:43:39"}
184
+ {"current_steps": 1840, "total_steps": 4795, "loss": 0.0, "lr": 7.742755635910532e-06, "epoch": 1.9167860398489387, "percentage": 38.37, "elapsed_time": "1:41:33", "remaining_time": "2:43:06"}
185
+ {"current_steps": 1850, "total_steps": 4795, "loss": 0.0004, "lr": 7.712245910777684e-06, "epoch": 1.9272040630290403, "percentage": 38.58, "elapsed_time": "1:42:06", "remaining_time": "2:42:33"}
186
+ {"current_steps": 1860, "total_steps": 4795, "loss": 0.0108, "lr": 7.68159241663226e-06, "epoch": 1.9376220862091418, "percentage": 38.79, "elapsed_time": "1:42:39", "remaining_time": "2:41:58"}
187
+ {"current_steps": 1870, "total_steps": 4795, "loss": 0.0006, "lr": 7.650796778335248e-06, "epoch": 1.9480401093892434, "percentage": 39.0, "elapsed_time": "1:43:11", "remaining_time": "2:41:24"}
188
+ {"current_steps": 1880, "total_steps": 4795, "loss": 0.0178, "lr": 7.619860628282333e-06, "epoch": 1.958458132569345, "percentage": 39.21, "elapsed_time": "1:43:45", "remaining_time": "2:40:52"}
189
+ {"current_steps": 1890, "total_steps": 4795, "loss": 0.0031, "lr": 7.5887856063173525e-06, "epoch": 1.9688761557494465, "percentage": 39.42, "elapsed_time": "1:44:16", "remaining_time": "2:40:17"}
190
+ {"current_steps": 1900, "total_steps": 4795, "loss": 0.0006, "lr": 7.5575733596453805e-06, "epoch": 1.979294178929548, "percentage": 39.62, "elapsed_time": "1:44:48", "remaining_time": "2:39:42"}
191
+ {"current_steps": 1910, "total_steps": 4795, "loss": 0.0086, "lr": 7.526225542745408e-06, "epoch": 1.9897122021096498, "percentage": 39.83, "elapsed_time": "1:45:21", "remaining_time": "2:39:09"}
192
+ {"current_steps": 1920, "total_steps": 4795, "loss": 0.0021, "lr": 7.494743817282651e-06, "epoch": 2.0, "percentage": 40.04, "elapsed_time": "1:45:53", "remaining_time": "2:38:33"}
193
+ {"current_steps": 1930, "total_steps": 4795, "loss": 0.001, "lr": 7.463129852020465e-06, "epoch": 2.0104180231801014, "percentage": 40.25, "elapsed_time": "1:46:26", "remaining_time": "2:38:00"}
194
+ {"current_steps": 1940, "total_steps": 4795, "loss": 0.0002, "lr": 7.431385322731885e-06, "epoch": 2.0208360463602033, "percentage": 40.46, "elapsed_time": "1:46:59", "remaining_time": "2:37:27"}
195
+ {"current_steps": 1950, "total_steps": 4795, "loss": 0.0003, "lr": 7.399511912110807e-06, "epoch": 2.0312540695403047, "percentage": 40.67, "elapsed_time": "1:47:32", "remaining_time": "2:36:53"}
196
+ {"current_steps": 1960, "total_steps": 4795, "loss": 0.0009, "lr": 7.367511309682782e-06, "epoch": 2.041672092720406, "percentage": 40.88, "elapsed_time": "1:48:04", "remaining_time": "2:36:19"}
197
+ {"current_steps": 1970, "total_steps": 4795, "loss": 0.0064, "lr": 7.335385211715467e-06, "epoch": 2.052090115900508, "percentage": 41.08, "elapsed_time": "1:48:36", "remaining_time": "2:35:45"}
198
+ {"current_steps": 1980, "total_steps": 4795, "loss": 0.0002, "lr": 7.303135321128705e-06, "epoch": 2.0625081390806095, "percentage": 41.29, "elapsed_time": "1:49:09", "remaining_time": "2:35:11"}
199
+ {"current_steps": 1990, "total_steps": 4795, "loss": 0.0017, "lr": 7.270763347404262e-06, "epoch": 2.072926162260711, "percentage": 41.5, "elapsed_time": "1:49:42", "remaining_time": "2:34:37"}
200
+ {"current_steps": 2000, "total_steps": 4795, "loss": 0.0015, "lr": 7.238271006495206e-06, "epoch": 2.083344185440813, "percentage": 41.71, "elapsed_time": "1:50:15", "remaining_time": "2:34:04"}
201
+ {"current_steps": 2010, "total_steps": 4795, "loss": 0.0002, "lr": 7.205660020734955e-06, "epoch": 2.0937622086209142, "percentage": 41.92, "elapsed_time": "1:50:47", "remaining_time": "2:33:30"}
202
+ {"current_steps": 2020, "total_steps": 4795, "loss": 0.0001, "lr": 7.172932118745978e-06, "epoch": 2.1041802318010157, "percentage": 42.13, "elapsed_time": "1:51:20", "remaining_time": "2:32:57"}
203
+ {"current_steps": 2030, "total_steps": 4795, "loss": 0.0018, "lr": 7.140089035348166e-06, "epoch": 2.1145982549811175, "percentage": 42.34, "elapsed_time": "1:51:52", "remaining_time": "2:32:23"}
204
+ {"current_steps": 2040, "total_steps": 4795, "loss": 0.0005, "lr": 7.107132511466872e-06, "epoch": 2.125016278161219, "percentage": 42.54, "elapsed_time": "1:52:25", "remaining_time": "2:31:49"}
205
+ {"current_steps": 2050, "total_steps": 4795, "loss": 0.0026, "lr": 7.074064294040629e-06, "epoch": 2.1354343013413204, "percentage": 42.75, "elapsed_time": "1:52:58", "remaining_time": "2:31:16"}
206
+ {"current_steps": 2060, "total_steps": 4795, "loss": 0.0004, "lr": 7.040886135928554e-06, "epoch": 2.145852324521422, "percentage": 42.96, "elapsed_time": "1:53:30", "remaining_time": "2:30:41"}
207
+ {"current_steps": 2070, "total_steps": 4795, "loss": 0.0, "lr": 7.007599795817426e-06, "epoch": 2.1562703477015237, "percentage": 43.17, "elapsed_time": "1:54:02", "remaining_time": "2:30:07"}
208
+ {"current_steps": 2080, "total_steps": 4795, "loss": 0.0006, "lr": 6.9742070381284724e-06, "epoch": 2.166688370881625, "percentage": 43.38, "elapsed_time": "1:54:34", "remaining_time": "2:29:33"}
209
+ {"current_steps": 2090, "total_steps": 4795, "loss": 0.0001, "lr": 6.94070963292383e-06, "epoch": 2.1771063940617266, "percentage": 43.59, "elapsed_time": "1:55:07", "remaining_time": "2:29:00"}
210
+ {"current_steps": 2100, "total_steps": 4795, "loss": 0.0014, "lr": 6.907109355812729e-06, "epoch": 2.1875244172418284, "percentage": 43.8, "elapsed_time": "1:55:40", "remaining_time": "2:28:26"}
211
+ {"current_steps": 2110, "total_steps": 4795, "loss": 0.0005, "lr": 6.873407987857367e-06, "epoch": 2.19794244042193, "percentage": 44.0, "elapsed_time": "1:56:11", "remaining_time": "2:27:51"}
212
+ {"current_steps": 2120, "total_steps": 4795, "loss": 0.0, "lr": 6.8396073154785e-06, "epoch": 2.2083604636020313, "percentage": 44.21, "elapsed_time": "1:56:45", "remaining_time": "2:27:18"}
213
+ {"current_steps": 2130, "total_steps": 4795, "loss": 0.0017, "lr": 6.805709130360754e-06, "epoch": 2.218778486782133, "percentage": 44.42, "elapsed_time": "1:57:17", "remaining_time": "2:26:45"}
214
+ {"current_steps": 2140, "total_steps": 4795, "loss": 0.0002, "lr": 6.771715229357643e-06, "epoch": 2.2291965099622346, "percentage": 44.63, "elapsed_time": "1:57:50", "remaining_time": "2:26:11"}
215
+ {"current_steps": 2150, "total_steps": 4795, "loss": 0.0018, "lr": 6.73762741439633e-06, "epoch": 2.239614533142336, "percentage": 44.84, "elapsed_time": "1:58:23", "remaining_time": "2:25:38"}
216
+ {"current_steps": 2160, "total_steps": 4795, "loss": 0.0034, "lr": 6.7034474923821135e-06, "epoch": 2.250032556322438, "percentage": 45.05, "elapsed_time": "1:58:56", "remaining_time": "2:25:06"}
217
+ {"current_steps": 2170, "total_steps": 4795, "loss": 0.0004, "lr": 6.669177275102637e-06, "epoch": 2.2604505795025394, "percentage": 45.26, "elapsed_time": "1:59:29", "remaining_time": "2:24:33"}
218
+ {"current_steps": 2180, "total_steps": 4795, "loss": 0.0007, "lr": 6.634818579131865e-06, "epoch": 2.270868602682641, "percentage": 45.46, "elapsed_time": "2:00:02", "remaining_time": "2:23:59"}
219
+ {"current_steps": 2190, "total_steps": 4795, "loss": 0.0009, "lr": 6.6003732257337805e-06, "epoch": 2.2812866258627427, "percentage": 45.67, "elapsed_time": "2:00:35", "remaining_time": "2:23:26"}
220
+ {"current_steps": 2200, "total_steps": 4795, "loss": 0.0003, "lr": 6.565843040765849e-06, "epoch": 2.291704649042844, "percentage": 45.88, "elapsed_time": "2:01:07", "remaining_time": "2:22:52"}
221
+ {"current_steps": 2210, "total_steps": 4795, "loss": 0.0002, "lr": 6.531229854582235e-06, "epoch": 2.3021226722229455, "percentage": 46.09, "elapsed_time": "2:01:40", "remaining_time": "2:22:19"}
222
+ {"current_steps": 2220, "total_steps": 4795, "loss": 0.0, "lr": 6.496535501936781e-06, "epoch": 2.3125406954030474, "percentage": 46.3, "elapsed_time": "2:02:13", "remaining_time": "2:21:45"}
223
+ {"current_steps": 2230, "total_steps": 4795, "loss": 0.0008, "lr": 6.461761821885742e-06, "epoch": 2.322958718583149, "percentage": 46.51, "elapsed_time": "2:02:45", "remaining_time": "2:21:11"}
224
+ {"current_steps": 2240, "total_steps": 4795, "loss": 0.0002, "lr": 6.42691065769032e-06, "epoch": 2.3333767417632503, "percentage": 46.72, "elapsed_time": "2:03:18", "remaining_time": "2:20:38"}
225
+ {"current_steps": 2250, "total_steps": 4795, "loss": 0.0138, "lr": 6.391983856718941e-06, "epoch": 2.343794764943352, "percentage": 46.92, "elapsed_time": "2:03:50", "remaining_time": "2:20:05"}
226
+ {"current_steps": 2260, "total_steps": 4795, "loss": 0.0047, "lr": 6.356983270349339e-06, "epoch": 2.3542127881234536, "percentage": 47.13, "elapsed_time": "2:04:24", "remaining_time": "2:19:32"}
227
+ {"current_steps": 2270, "total_steps": 4795, "loss": 0.0003, "lr": 6.321910753870416e-06, "epoch": 2.364630811303555, "percentage": 47.34, "elapsed_time": "2:04:56", "remaining_time": "2:18:58"}
228
+ {"current_steps": 2280, "total_steps": 4795, "loss": 0.0001, "lr": 6.286768166383905e-06, "epoch": 2.375048834483657, "percentage": 47.55, "elapsed_time": "2:05:29", "remaining_time": "2:18:25"}
229
+ {"current_steps": 2290, "total_steps": 4795, "loss": 0.0002, "lr": 6.251557370705809e-06, "epoch": 2.3854668576637583, "percentage": 47.76, "elapsed_time": "2:06:01", "remaining_time": "2:17:51"}
230
+ {"current_steps": 2300, "total_steps": 4795, "loss": 0.0135, "lr": 6.216280233267681e-06, "epoch": 2.3958848808438598, "percentage": 47.97, "elapsed_time": "2:06:33", "remaining_time": "2:17:17"}
231
+ {"current_steps": 2310, "total_steps": 4795, "loss": 0.0067, "lr": 6.180938624017663e-06, "epoch": 2.4063029040239616, "percentage": 48.18, "elapsed_time": "2:07:06", "remaining_time": "2:16:44"}
232
+ {"current_steps": 2320, "total_steps": 4795, "loss": 0.0002, "lr": 6.145534416321384e-06, "epoch": 2.416720927204063, "percentage": 48.38, "elapsed_time": "2:07:38", "remaining_time": "2:16:10"}
233
+ {"current_steps": 2330, "total_steps": 4795, "loss": 0.0083, "lr": 6.110069486862652e-06, "epoch": 2.4271389503841645, "percentage": 48.59, "elapsed_time": "2:08:11", "remaining_time": "2:15:36"}
234
+ {"current_steps": 2340, "total_steps": 4795, "loss": 0.0173, "lr": 6.074545715543975e-06, "epoch": 2.4375569735642664, "percentage": 48.8, "elapsed_time": "2:08:44", "remaining_time": "2:15:04"}
235
+ {"current_steps": 2350, "total_steps": 4795, "loss": 0.0001, "lr": 6.0389649853869116e-06, "epoch": 2.447974996744368, "percentage": 49.01, "elapsed_time": "2:09:16", "remaining_time": "2:14:30"}
236
+ {"current_steps": 2360, "total_steps": 4795, "loss": 0.0003, "lr": 6.00332918243226e-06, "epoch": 2.4583930199244692, "percentage": 49.22, "elapsed_time": "2:09:49", "remaining_time": "2:13:57"}
237
+ {"current_steps": 2370, "total_steps": 4795, "loss": 0.0016, "lr": 5.967640195640083e-06, "epoch": 2.468811043104571, "percentage": 49.43, "elapsed_time": "2:10:22", "remaining_time": "2:13:23"}
238
+ {"current_steps": 2380, "total_steps": 4795, "loss": 0.001, "lr": 5.931899916789576e-06, "epoch": 2.4792290662846725, "percentage": 49.64, "elapsed_time": "2:10:55", "remaining_time": "2:12:51"}
239
+ {"current_steps": 2390, "total_steps": 4795, "loss": 0.001, "lr": 5.896110240378794e-06, "epoch": 2.489647089464774, "percentage": 49.84, "elapsed_time": "2:11:28", "remaining_time": "2:12:18"}
240
+ {"current_steps": 2400, "total_steps": 4795, "loss": 0.0019, "lr": 5.860273063524227e-06, "epoch": 2.500065112644876, "percentage": 50.05, "elapsed_time": "2:12:01", "remaining_time": "2:11:44"}
241
+ {"current_steps": 2410, "total_steps": 4795, "loss": 0.0, "lr": 5.824390285860233e-06, "epoch": 2.5104831358249773, "percentage": 50.26, "elapsed_time": "2:12:34", "remaining_time": "2:11:11"}
242
+ {"current_steps": 2420, "total_steps": 4795, "loss": 0.0, "lr": 5.788463809438356e-06, "epoch": 2.5209011590050787, "percentage": 50.47, "elapsed_time": "2:13:07", "remaining_time": "2:10:38"}
243
+ {"current_steps": 2430, "total_steps": 4795, "loss": 0.0042, "lr": 5.752495538626493e-06, "epoch": 2.5313191821851806, "percentage": 50.68, "elapsed_time": "2:13:39", "remaining_time": "2:10:04"}
244
+ {"current_steps": 2440, "total_steps": 4795, "loss": 0.009, "lr": 5.7164873800079475e-06, "epoch": 2.541737205365282, "percentage": 50.89, "elapsed_time": "2:14:11", "remaining_time": "2:09:31"}
245
+ {"current_steps": 2450, "total_steps": 4795, "loss": 0.0067, "lr": 5.680441242280378e-06, "epoch": 2.5521552285453835, "percentage": 51.09, "elapsed_time": "2:14:44", "remaining_time": "2:08:58"}
246
+ {"current_steps": 2460, "total_steps": 4795, "loss": 0.0017, "lr": 5.6443590361546095e-06, "epoch": 2.5625732517254853, "percentage": 51.3, "elapsed_time": "2:15:17", "remaining_time": "2:08:24"}
247
+ {"current_steps": 2470, "total_steps": 4795, "loss": 0.0, "lr": 5.608242674253362e-06, "epoch": 2.5729912749055868, "percentage": 51.51, "elapsed_time": "2:15:49", "remaining_time": "2:07:51"}
248
+ {"current_steps": 2480, "total_steps": 4795, "loss": 0.0101, "lr": 5.57209407100986e-06, "epoch": 2.583409298085688, "percentage": 51.72, "elapsed_time": "2:16:22", "remaining_time": "2:07:18"}
249
+ {"current_steps": 2490, "total_steps": 4795, "loss": 0.0003, "lr": 5.535915142566361e-06, "epoch": 2.59382732126579, "percentage": 51.93, "elapsed_time": "2:16:55", "remaining_time": "2:06:45"}
250
+ {"current_steps": 2500, "total_steps": 4795, "loss": 0.0077, "lr": 5.499707806672575e-06, "epoch": 2.6042453444458915, "percentage": 52.14, "elapsed_time": "2:17:27", "remaining_time": "2:06:11"}
251
+ {"current_steps": 2510, "total_steps": 4795, "loss": 0.0019, "lr": 5.463473982584023e-06, "epoch": 2.614663367625993, "percentage": 52.35, "elapsed_time": "2:18:00", "remaining_time": "2:05:38"}
252
+ {"current_steps": 2520, "total_steps": 4795, "loss": 0.0161, "lr": 5.4272155909602875e-06, "epoch": 2.625081390806095, "percentage": 52.55, "elapsed_time": "2:18:33", "remaining_time": "2:05:04"}
253
+ {"current_steps": 2530, "total_steps": 4795, "loss": 0.0153, "lr": 5.3909345537632205e-06, "epoch": 2.6354994139861962, "percentage": 52.76, "elapsed_time": "2:19:06", "remaining_time": "2:04:31"}
254
+ {"current_steps": 2540, "total_steps": 4795, "loss": 0.0008, "lr": 5.354632794155049e-06, "epoch": 2.6459174371662977, "percentage": 52.97, "elapsed_time": "2:19:38", "remaining_time": "2:03:58"}
255
+ {"current_steps": 2550, "total_steps": 4795, "loss": 0.0025, "lr": 5.318312236396445e-06, "epoch": 2.656335460346399, "percentage": 53.18, "elapsed_time": "2:20:11", "remaining_time": "2:03:25"}
256
+ {"current_steps": 2560, "total_steps": 4795, "loss": 0.0009, "lr": 5.281974805744516e-06, "epoch": 2.666753483526501, "percentage": 53.39, "elapsed_time": "2:20:43", "remaining_time": "2:02:51"}
257
+ {"current_steps": 2570, "total_steps": 4795, "loss": 0.0019, "lr": 5.245622428350764e-06, "epoch": 2.6771715067066024, "percentage": 53.6, "elapsed_time": "2:21:16", "remaining_time": "2:02:18"}
258
+ {"current_steps": 2580, "total_steps": 4795, "loss": 0.0042, "lr": 5.209257031158972e-06, "epoch": 2.687589529886704, "percentage": 53.81, "elapsed_time": "2:21:48", "remaining_time": "2:01:45"}
259
+ {"current_steps": 2590, "total_steps": 4795, "loss": 0.002, "lr": 5.1728805418030725e-06, "epoch": 2.6980075530668057, "percentage": 54.01, "elapsed_time": "2:22:21", "remaining_time": "2:01:12"}
260
+ {"current_steps": 2600, "total_steps": 4795, "loss": 0.011, "lr": 5.13649488850496e-06, "epoch": 2.708425576246907, "percentage": 54.22, "elapsed_time": "2:22:54", "remaining_time": "2:00:38"}
261
+ {"current_steps": 2610, "total_steps": 4795, "loss": 0.0004, "lr": 5.100101999972291e-06, "epoch": 2.7188435994270086, "percentage": 54.43, "elapsed_time": "2:23:26", "remaining_time": "2:00:05"}
262
+ {"current_steps": 2620, "total_steps": 4795, "loss": 0.0001, "lr": 5.063703805296239e-06, "epoch": 2.72926162260711, "percentage": 54.64, "elapsed_time": "2:23:58", "remaining_time": "1:59:31"}
263
+ {"current_steps": 2630, "total_steps": 4795, "loss": 0.0028, "lr": 5.027302233849243e-06, "epoch": 2.739679645787212, "percentage": 54.85, "elapsed_time": "2:24:31", "remaining_time": "1:58:58"}
264
+ {"current_steps": 2640, "total_steps": 4795, "loss": 0.0032, "lr": 4.990899215182735e-06, "epoch": 2.7500976689673133, "percentage": 55.06, "elapsed_time": "2:25:03", "remaining_time": "1:58:24"}
265
+ {"current_steps": 2650, "total_steps": 4795, "loss": 0.0004, "lr": 4.954496678924861e-06, "epoch": 2.7605156921474148, "percentage": 55.27, "elapsed_time": "2:25:36", "remaining_time": "1:57:51"}
266
+ {"current_steps": 2660, "total_steps": 4795, "loss": 0.0001, "lr": 4.9180965546781985e-06, "epoch": 2.7709337153275166, "percentage": 55.47, "elapsed_time": "2:26:08", "remaining_time": "1:57:18"}
267
+ {"current_steps": 2670, "total_steps": 4795, "loss": 0.0014, "lr": 4.8817007719174635e-06, "epoch": 2.781351738507618, "percentage": 55.68, "elapsed_time": "2:26:41", "remaining_time": "1:56:45"}
268
+ {"current_steps": 2680, "total_steps": 4795, "loss": 0.0017, "lr": 4.845311259887251e-06, "epoch": 2.7917697616877195, "percentage": 55.89, "elapsed_time": "2:27:14", "remaining_time": "1:56:11"}
269
+ {"current_steps": 2690, "total_steps": 4795, "loss": 0.0001, "lr": 4.808929947499751e-06, "epoch": 2.8021877848678214, "percentage": 56.1, "elapsed_time": "2:27:46", "remaining_time": "1:55:38"}
270
+ {"current_steps": 2700, "total_steps": 4795, "loss": 0.0, "lr": 4.772558763232521e-06, "epoch": 2.812605808047923, "percentage": 56.31, "elapsed_time": "2:28:19", "remaining_time": "1:55:05"}
271
+ {"current_steps": 2710, "total_steps": 4795, "loss": 0.0004, "lr": 4.736199635026247e-06, "epoch": 2.8230238312280242, "percentage": 56.52, "elapsed_time": "2:28:51", "remaining_time": "1:54:31"}
272
+ {"current_steps": 2720, "total_steps": 4795, "loss": 0.0001, "lr": 4.699854490182558e-06, "epoch": 2.833441854408126, "percentage": 56.73, "elapsed_time": "2:29:23", "remaining_time": "1:53:57"}
273
+ {"current_steps": 2730, "total_steps": 4795, "loss": 0.0026, "lr": 4.663525255261861e-06, "epoch": 2.8438598775882276, "percentage": 56.93, "elapsed_time": "2:29:55", "remaining_time": "1:53:24"}
274
+ {"current_steps": 2740, "total_steps": 4795, "loss": 0.0002, "lr": 4.627213855981221e-06, "epoch": 2.854277900768329, "percentage": 57.14, "elapsed_time": "2:30:29", "remaining_time": "1:52:52"}
275
+ {"current_steps": 2750, "total_steps": 4795, "loss": 0.0005, "lr": 4.590922217112279e-06, "epoch": 2.864695923948431, "percentage": 57.35, "elapsed_time": "2:31:02", "remaining_time": "1:52:18"}
276
+ {"current_steps": 2760, "total_steps": 4795, "loss": 0.0031, "lr": 4.554652262379236e-06, "epoch": 2.8751139471285323, "percentage": 57.56, "elapsed_time": "2:31:34", "remaining_time": "1:51:45"}
277
+ {"current_steps": 2770, "total_steps": 4795, "loss": 0.0002, "lr": 4.518405914356865e-06, "epoch": 2.8855319703086337, "percentage": 57.77, "elapsed_time": "2:32:07", "remaining_time": "1:51:12"}
278
+ {"current_steps": 2780, "total_steps": 4795, "loss": 0.0003, "lr": 4.4821850943686164e-06, "epoch": 2.8959499934887356, "percentage": 57.98, "elapsed_time": "2:32:40", "remaining_time": "1:50:39"}
279
+ {"current_steps": 2790, "total_steps": 4795, "loss": 0.0002, "lr": 4.445991722384763e-06, "epoch": 2.906368016668837, "percentage": 58.19, "elapsed_time": "2:33:13", "remaining_time": "1:50:07"}
280
+ {"current_steps": 2800, "total_steps": 4795, "loss": 0.0001, "lr": 4.4098277169206315e-06, "epoch": 2.9167860398489385, "percentage": 58.39, "elapsed_time": "2:33:46", "remaining_time": "1:49:33"}
281
+ {"current_steps": 2810, "total_steps": 4795, "loss": 0.0194, "lr": 4.373694994934911e-06, "epoch": 2.9272040630290403, "percentage": 58.6, "elapsed_time": "2:34:19", "remaining_time": "1:49:00"}
282
+ {"current_steps": 2820, "total_steps": 4795, "loss": 0.0004, "lr": 4.337595471728029e-06, "epoch": 2.937622086209142, "percentage": 58.81, "elapsed_time": "2:34:51", "remaining_time": "1:48:27"}
283
+ {"current_steps": 2830, "total_steps": 4795, "loss": 0.0005, "lr": 4.30153106084064e-06, "epoch": 2.948040109389243, "percentage": 59.02, "elapsed_time": "2:35:23", "remaining_time": "1:47:53"}
284
+ {"current_steps": 2840, "total_steps": 4795, "loss": 0.0016, "lr": 4.2655036739521795e-06, "epoch": 2.958458132569345, "percentage": 59.23, "elapsed_time": "2:35:56", "remaining_time": "1:47:20"}
285
+ {"current_steps": 2850, "total_steps": 4795, "loss": 0.0001, "lr": 4.229515220779545e-06, "epoch": 2.9688761557494465, "percentage": 59.44, "elapsed_time": "2:36:28", "remaining_time": "1:46:47"}
286
+ {"current_steps": 2860, "total_steps": 4795, "loss": 0.0002, "lr": 4.193567608975856e-06, "epoch": 2.979294178929548, "percentage": 59.65, "elapsed_time": "2:37:01", "remaining_time": "1:46:14"}
287
+ {"current_steps": 2870, "total_steps": 4795, "loss": 0.0006, "lr": 4.1576627440293425e-06, "epoch": 2.98971220210965, "percentage": 59.85, "elapsed_time": "2:37:34", "remaining_time": "1:45:41"}
288
+ {"current_steps": 2880, "total_steps": 4795, "loss": 0.0, "lr": 4.121802529162331e-06, "epoch": 3.0, "percentage": 60.06, "elapsed_time": "2:38:06", "remaining_time": "1:45:07"}
289
+ {"current_steps": 2890, "total_steps": 4795, "loss": 0.0008, "lr": 4.08598886523037e-06, "epoch": 3.0104180231801014, "percentage": 60.27, "elapsed_time": "2:41:20", "remaining_time": "1:46:20"}
290
+ {"current_steps": 2900, "total_steps": 4795, "loss": 0.0003, "lr": 4.050223650621461e-06, "epoch": 3.0208360463602033, "percentage": 60.48, "elapsed_time": "2:41:52", "remaining_time": "1:45:46"}
291
+ {"current_steps": 2910, "total_steps": 4795, "loss": 0.0003, "lr": 4.014508781155441e-06, "epoch": 3.0312540695403047, "percentage": 60.69, "elapsed_time": "2:42:24", "remaining_time": "1:45:12"}
292
+ {"current_steps": 2920, "total_steps": 4795, "loss": 0.0002, "lr": 3.978846149983473e-06, "epoch": 3.041672092720406, "percentage": 60.9, "elapsed_time": "2:42:57", "remaining_time": "1:44:38"}
293
+ {"current_steps": 2930, "total_steps": 4795, "loss": 0.0014, "lr": 3.943237647487719e-06, "epoch": 3.052090115900508, "percentage": 61.11, "elapsed_time": "2:43:30", "remaining_time": "1:44:04"}
294
+ {"current_steps": 2940, "total_steps": 4795, "loss": 0.0001, "lr": 3.907685161181109e-06, "epoch": 3.0625081390806095, "percentage": 61.31, "elapsed_time": "2:44:02", "remaining_time": "1:43:30"}
295
+ {"current_steps": 2950, "total_steps": 4795, "loss": 0.0004, "lr": 3.8721905756073135e-06, "epoch": 3.072926162260711, "percentage": 61.52, "elapsed_time": "2:44:34", "remaining_time": "1:42:55"}
296
+ {"current_steps": 2960, "total_steps": 4795, "loss": 0.0005, "lr": 3.836755772240829e-06, "epoch": 3.083344185440813, "percentage": 61.73, "elapsed_time": "2:45:06", "remaining_time": "1:42:21"}
297
+ {"current_steps": 2970, "total_steps": 4795, "loss": 0.0, "lr": 3.8013826293872623e-06, "epoch": 3.0937622086209142, "percentage": 61.94, "elapsed_time": "2:45:39", "remaining_time": "1:41:47"}
298
+ {"current_steps": 2980, "total_steps": 4795, "loss": 0.0001, "lr": 3.7660730220837512e-06, "epoch": 3.1041802318010157, "percentage": 62.15, "elapsed_time": "2:46:12", "remaining_time": "1:41:13"}
299
+ {"current_steps": 2990, "total_steps": 4795, "loss": 0.0001, "lr": 3.7308288219995884e-06, "epoch": 3.1145982549811175, "percentage": 62.36, "elapsed_time": "2:46:46", "remaining_time": "1:40:40"}
300
+ {"current_steps": 3000, "total_steps": 4795, "loss": 0.0, "lr": 3.695651897336996e-06, "epoch": 3.125016278161219, "percentage": 62.57, "elapsed_time": "2:47:19", "remaining_time": "1:40:06"}
301
+ {"current_steps": 3010, "total_steps": 4795, "loss": 0.0001, "lr": 3.6605441127321074e-06, "epoch": 3.1354343013413204, "percentage": 62.77, "elapsed_time": "2:47:52", "remaining_time": "1:39:33"}
302
+ {"current_steps": 3020, "total_steps": 4795, "loss": 0.0, "lr": 3.6255073291561248e-06, "epoch": 3.145852324521422, "percentage": 62.98, "elapsed_time": "2:48:25", "remaining_time": "1:38:59"}
303
+ {"current_steps": 3030, "total_steps": 4795, "loss": 0.0001, "lr": 3.5905434038166653e-06, "epoch": 3.1562703477015237, "percentage": 63.19, "elapsed_time": "2:48:58", "remaining_time": "1:38:25"}
304
+ {"current_steps": 3040, "total_steps": 4795, "loss": 0.004, "lr": 3.555654190059333e-06, "epoch": 3.166688370881625, "percentage": 63.4, "elapsed_time": "2:49:30", "remaining_time": "1:37:51"}
305
+ {"current_steps": 3050, "total_steps": 4795, "loss": 0.0025, "lr": 3.520841537269458e-06, "epoch": 3.1771063940617266, "percentage": 63.61, "elapsed_time": "2:50:02", "remaining_time": "1:37:17"}
306
+ {"current_steps": 3060, "total_steps": 4795, "loss": 0.0004, "lr": 3.486107290774083e-06, "epoch": 3.1875244172418284, "percentage": 63.82, "elapsed_time": "2:50:35", "remaining_time": "1:36:43"}
307
+ {"current_steps": 3070, "total_steps": 4795, "loss": 0.0004, "lr": 3.4514532917441345e-06, "epoch": 3.19794244042193, "percentage": 64.03, "elapsed_time": "2:51:07", "remaining_time": "1:36:09"}
308
+ {"current_steps": 3080, "total_steps": 4795, "loss": 0.0001, "lr": 3.4168813770968357e-06, "epoch": 3.2083604636020313, "percentage": 64.23, "elapsed_time": "2:51:40", "remaining_time": "1:35:35"}
309
+ {"current_steps": 3090, "total_steps": 4795, "loss": 0.0003, "lr": 3.3823933793983298e-06, "epoch": 3.218778486782133, "percentage": 64.44, "elapsed_time": "2:52:13", "remaining_time": "1:35:01"}
310
+ {"current_steps": 3100, "total_steps": 4795, "loss": 0.0, "lr": 3.347991126766545e-06, "epoch": 3.2291965099622346, "percentage": 64.65, "elapsed_time": "2:52:46", "remaining_time": "1:34:27"}
311
+ {"current_steps": 3110, "total_steps": 4795, "loss": 0.0001, "lr": 3.3136764427742885e-06, "epoch": 3.239614533142336, "percentage": 64.86, "elapsed_time": "2:53:18", "remaining_time": "1:33:53"}
312
+ {"current_steps": 3120, "total_steps": 4795, "loss": 0.0001, "lr": 3.279451146352588e-06, "epoch": 3.250032556322438, "percentage": 65.07, "elapsed_time": "2:53:50", "remaining_time": "1:33:19"}
313
+ {"current_steps": 3130, "total_steps": 4795, "loss": 0.0001, "lr": 3.2453170516942657e-06, "epoch": 3.2604505795025394, "percentage": 65.28, "elapsed_time": "2:54:22", "remaining_time": "1:32:45"}
314
+ {"current_steps": 3140, "total_steps": 4795, "loss": 0.0001, "lr": 3.211275968157784e-06, "epoch": 3.270868602682641, "percentage": 65.48, "elapsed_time": "2:54:55", "remaining_time": "1:32:11"}
315
+ {"current_steps": 3150, "total_steps": 4795, "loss": 0.0002, "lr": 3.1773297001713266e-06, "epoch": 3.2812866258627427, "percentage": 65.69, "elapsed_time": "2:55:28", "remaining_time": "1:31:37"}
316
+ {"current_steps": 3160, "total_steps": 4795, "loss": 0.0, "lr": 3.1434800471371586e-06, "epoch": 3.291704649042844, "percentage": 65.9, "elapsed_time": "2:56:00", "remaining_time": "1:31:03"}
317
+ {"current_steps": 3170, "total_steps": 4795, "loss": 0.0, "lr": 3.109728803336234e-06, "epoch": 3.3021226722229455, "percentage": 66.11, "elapsed_time": "2:56:33", "remaining_time": "1:30:30"}
318
+ {"current_steps": 3180, "total_steps": 4795, "loss": 0.0001, "lr": 3.076077757833103e-06, "epoch": 3.3125406954030474, "percentage": 66.32, "elapsed_time": "2:57:05", "remaining_time": "1:29:56"}
319
+ {"current_steps": 3190, "total_steps": 4795, "loss": 0.0, "lr": 3.042528694381057e-06, "epoch": 3.322958718583149, "percentage": 66.53, "elapsed_time": "2:57:38", "remaining_time": "1:29:22"}
320
+ {"current_steps": 3200, "total_steps": 4795, "loss": 0.0001, "lr": 3.0090833913275965e-06, "epoch": 3.3333767417632503, "percentage": 66.74, "elapsed_time": "2:58:10", "remaining_time": "1:28:48"}
321
+ {"current_steps": 3210, "total_steps": 4795, "loss": 0.0, "lr": 2.9757436215201497e-06, "epoch": 3.343794764943352, "percentage": 66.94, "elapsed_time": "2:58:43", "remaining_time": "1:28:14"}
322
+ {"current_steps": 3220, "total_steps": 4795, "loss": 0.0009, "lr": 2.9425111522121085e-06, "epoch": 3.3542127881234536, "percentage": 67.15, "elapsed_time": "2:59:16", "remaining_time": "1:27:41"}
323
+ {"current_steps": 3230, "total_steps": 4795, "loss": 0.0, "lr": 2.9093877449691523e-06, "epoch": 3.364630811303555, "percentage": 67.36, "elapsed_time": "2:59:48", "remaining_time": "1:27:07"}
324
+ {"current_steps": 3240, "total_steps": 4795, "loss": 0.0011, "lr": 2.8763751555758544e-06, "epoch": 3.375048834483657, "percentage": 67.57, "elapsed_time": "3:00:21", "remaining_time": "1:26:33"}
325
+ {"current_steps": 3250, "total_steps": 4795, "loss": 0.0007, "lr": 2.843475133942642e-06, "epoch": 3.3854668576637583, "percentage": 67.78, "elapsed_time": "3:00:54", "remaining_time": "1:25:59"}
326
+ {"current_steps": 3260, "total_steps": 4795, "loss": 0.0, "lr": 2.810689424013011e-06, "epoch": 3.3958848808438598, "percentage": 67.99, "elapsed_time": "3:01:27", "remaining_time": "1:25:26"}
327
+ {"current_steps": 3270, "total_steps": 4795, "loss": 0.0, "lr": 2.7780197636710993e-06, "epoch": 3.4063029040239616, "percentage": 68.2, "elapsed_time": "3:01:59", "remaining_time": "1:24:52"}
328
+ {"current_steps": 3280, "total_steps": 4795, "loss": 0.0002, "lr": 2.7454678846495593e-06, "epoch": 3.416720927204063, "percentage": 68.4, "elapsed_time": "3:02:32", "remaining_time": "1:24:18"}
329
+ {"current_steps": 3290, "total_steps": 4795, "loss": 0.0001, "lr": 2.7130355124377684e-06, "epoch": 3.4271389503841645, "percentage": 68.61, "elapsed_time": "3:03:05", "remaining_time": "1:23:45"}
330
+ {"current_steps": 3300, "total_steps": 4795, "loss": 0.0, "lr": 2.6807243661903597e-06, "epoch": 3.4375569735642664, "percentage": 68.82, "elapsed_time": "3:03:38", "remaining_time": "1:23:11"}
331
+ {"current_steps": 3310, "total_steps": 4795, "loss": 0.0001, "lr": 2.6485361586361012e-06, "epoch": 3.447974996744368, "percentage": 69.03, "elapsed_time": "3:04:11", "remaining_time": "1:22:38"}
332
+ {"current_steps": 3320, "total_steps": 4795, "loss": 0.0008, "lr": 2.6164725959871005e-06, "epoch": 3.4583930199244692, "percentage": 69.24, "elapsed_time": "3:04:44", "remaining_time": "1:22:04"}
333
+ {"current_steps": 3330, "total_steps": 4795, "loss": 0.0, "lr": 2.5845353778483707e-06, "epoch": 3.468811043104571, "percentage": 69.45, "elapsed_time": "3:05:16", "remaining_time": "1:21:30"}
334
+ {"current_steps": 3340, "total_steps": 4795, "loss": 0.0001, "lr": 2.552726197127732e-06, "epoch": 3.4792290662846725, "percentage": 69.66, "elapsed_time": "3:05:49", "remaining_time": "1:20:56"}
335
+ {"current_steps": 3350, "total_steps": 4795, "loss": 0.0, "lr": 2.5210467399460794e-06, "epoch": 3.489647089464774, "percentage": 69.86, "elapsed_time": "3:06:21", "remaining_time": "1:20:22"}
336
+ {"current_steps": 3360, "total_steps": 4795, "loss": 0.0002, "lr": 2.489498685548005e-06, "epoch": 3.500065112644876, "percentage": 70.07, "elapsed_time": "3:06:54", "remaining_time": "1:19:49"}
337
+ {"current_steps": 3370, "total_steps": 4795, "loss": 0.0, "lr": 2.4580837062127867e-06, "epoch": 3.5104831358249773, "percentage": 70.28, "elapsed_time": "3:07:27", "remaining_time": "1:19:16"}
338
+ {"current_steps": 3380, "total_steps": 4795, "loss": 0.0002, "lr": 2.4268034671657412e-06, "epoch": 3.5209011590050787, "percentage": 70.49, "elapsed_time": "3:08:00", "remaining_time": "1:18:42"}
339
+ {"current_steps": 3390, "total_steps": 4795, "loss": 0.0, "lr": 2.39565962648996e-06, "epoch": 3.5313191821851806, "percentage": 70.7, "elapsed_time": "3:08:32", "remaining_time": "1:18:08"}
340
+ {"current_steps": 3400, "total_steps": 4795, "loss": 0.0001, "lr": 2.364653835038413e-06, "epoch": 3.541737205365282, "percentage": 70.91, "elapsed_time": "3:09:05", "remaining_time": "1:17:34"}
341
+ {"current_steps": 3410, "total_steps": 4795, "loss": 0.0008, "lr": 2.3337877363464475e-06, "epoch": 3.5521552285453835, "percentage": 71.12, "elapsed_time": "3:09:37", "remaining_time": "1:17:01"}
342
+ {"current_steps": 3420, "total_steps": 4795, "loss": 0.0001, "lr": 2.3030629665446635e-06, "epoch": 3.5625732517254853, "percentage": 71.32, "elapsed_time": "3:10:09", "remaining_time": "1:16:27"}
343
+ {"current_steps": 3430, "total_steps": 4795, "loss": 0.0062, "lr": 2.272481154272188e-06, "epoch": 3.5729912749055868, "percentage": 71.53, "elapsed_time": "3:10:42", "remaining_time": "1:15:53"}
344
+ {"current_steps": 3440, "total_steps": 4795, "loss": 0.0006, "lr": 2.2420439205903523e-06, "epoch": 3.583409298085688, "percentage": 71.74, "elapsed_time": "3:11:15", "remaining_time": "1:15:20"}
345
+ {"current_steps": 3450, "total_steps": 4795, "loss": 0.0131, "lr": 2.211752878896745e-06, "epoch": 3.59382732126579, "percentage": 71.95, "elapsed_time": "3:11:47", "remaining_time": "1:14:46"}
346
+ {"current_steps": 3460, "total_steps": 4795, "loss": 0.0, "lr": 2.1816096348397176e-06, "epoch": 3.6042453444458915, "percentage": 72.16, "elapsed_time": "3:12:19", "remaining_time": "1:14:12"}
347
+ {"current_steps": 3470, "total_steps": 4795, "loss": 0.0002, "lr": 2.1516157862332425e-06, "epoch": 3.614663367625993, "percentage": 72.37, "elapsed_time": "3:12:52", "remaining_time": "1:13:38"}
348
+ {"current_steps": 3480, "total_steps": 4795, "loss": 0.0001, "lr": 2.1217729229722485e-06, "epoch": 3.625081390806095, "percentage": 72.58, "elapsed_time": "3:13:26", "remaining_time": "1:13:05"}
349
+ {"current_steps": 3490, "total_steps": 4795, "loss": 0.0001, "lr": 2.092082626948313e-06, "epoch": 3.6354994139861962, "percentage": 72.78, "elapsed_time": "3:14:08", "remaining_time": "1:12:35"}
350
+ {"current_steps": 3500, "total_steps": 4795, "loss": 0.0017, "lr": 2.062546471965841e-06, "epoch": 3.6459174371662977, "percentage": 72.99, "elapsed_time": "3:14:49", "remaining_time": "1:12:05"}
351
+ {"current_steps": 3510, "total_steps": 4795, "loss": 0.0055, "lr": 2.033166023658613e-06, "epoch": 3.656335460346399, "percentage": 73.2, "elapsed_time": "3:15:31", "remaining_time": "1:11:34"}
352
+ {"current_steps": 3520, "total_steps": 4795, "loss": 0.0, "lr": 2.0039428394068224e-06, "epoch": 3.666753483526501, "percentage": 73.41, "elapsed_time": "3:16:13", "remaining_time": "1:11:04"}
353
+ {"current_steps": 3530, "total_steps": 4795, "loss": 0.0, "lr": 1.9748784682544955e-06, "epoch": 3.6771715067066024, "percentage": 73.62, "elapsed_time": "3:16:45", "remaining_time": "1:10:30"}
354
+ {"current_steps": 3540, "total_steps": 4795, "loss": 0.0014, "lr": 1.9459744508274076e-06, "epoch": 3.687589529886704, "percentage": 73.83, "elapsed_time": "3:17:18", "remaining_time": "1:09:56"}
355
+ {"current_steps": 3550, "total_steps": 4795, "loss": 0.0008, "lr": 1.917232319251392e-06, "epoch": 3.6980075530668057, "percentage": 74.04, "elapsed_time": "3:17:50", "remaining_time": "1:09:22"}
356
+ {"current_steps": 3560, "total_steps": 4795, "loss": 0.0062, "lr": 1.8886535970711522e-06, "epoch": 3.708425576246907, "percentage": 74.24, "elapsed_time": "3:18:22", "remaining_time": "1:08:49"}
357
+ {"current_steps": 3570, "total_steps": 4795, "loss": 0.0001, "lr": 1.860239799169482e-06, "epoch": 3.7188435994270086, "percentage": 74.45, "elapsed_time": "3:18:55", "remaining_time": "1:08:15"}
358
+ {"current_steps": 3580, "total_steps": 4795, "loss": 0.0003, "lr": 1.8319924316869763e-06, "epoch": 3.72926162260711, "percentage": 74.66, "elapsed_time": "3:19:27", "remaining_time": "1:07:41"}
359
+ {"current_steps": 3590, "total_steps": 4795, "loss": 0.0036, "lr": 1.8039129919421905e-06, "epoch": 3.739679645787212, "percentage": 74.87, "elapsed_time": "3:20:00", "remaining_time": "1:07:08"}
360
+ {"current_steps": 3600, "total_steps": 4795, "loss": 0.0, "lr": 1.7760029683522734e-06, "epoch": 3.7500976689673133, "percentage": 75.08, "elapsed_time": "3:20:33", "remaining_time": "1:06:34"}
361
+ {"current_steps": 3610, "total_steps": 4795, "loss": 0.0001, "lr": 1.7482638403540703e-06, "epoch": 3.7605156921474148, "percentage": 75.29, "elapsed_time": "3:21:05", "remaining_time": "1:06:00"}
362
+ {"current_steps": 3620, "total_steps": 4795, "loss": 0.0005, "lr": 1.720697078325701e-06, "epoch": 3.7709337153275166, "percentage": 75.5, "elapsed_time": "3:21:38", "remaining_time": "1:05:26"}
363
+ {"current_steps": 3630, "total_steps": 4795, "loss": 0.0, "lr": 1.693304143508618e-06, "epoch": 3.781351738507618, "percentage": 75.7, "elapsed_time": "3:22:10", "remaining_time": "1:04:53"}
364
+ {"current_steps": 3640, "total_steps": 4795, "loss": 0.0002, "lr": 1.666086487930153e-06, "epoch": 3.7917697616877195, "percentage": 75.91, "elapsed_time": "3:22:42", "remaining_time": "1:04:19"}
365
+ {"current_steps": 3650, "total_steps": 4795, "loss": 0.0001, "lr": 1.6390455543265454e-06, "epoch": 3.8021877848678214, "percentage": 76.12, "elapsed_time": "3:23:14", "remaining_time": "1:03:45"}
366
+ {"current_steps": 3660, "total_steps": 4795, "loss": 0.0001, "lr": 1.6121827760664677e-06, "epoch": 3.812605808047923, "percentage": 76.33, "elapsed_time": "3:23:47", "remaining_time": "1:03:11"}
367
+ {"current_steps": 3670, "total_steps": 4795, "loss": 0.0001, "lr": 1.5854995770750526e-06, "epoch": 3.8230238312280242, "percentage": 76.54, "elapsed_time": "3:24:19", "remaining_time": "1:02:38"}
368
+ {"current_steps": 3680, "total_steps": 4795, "loss": 0.0, "lr": 1.5589973717583968e-06, "epoch": 3.833441854408126, "percentage": 76.75, "elapsed_time": "3:24:52", "remaining_time": "1:02:04"}
369
+ {"current_steps": 3690, "total_steps": 4795, "loss": 0.0001, "lr": 1.5326775649286135e-06, "epoch": 3.8438598775882276, "percentage": 76.96, "elapsed_time": "3:25:25", "remaining_time": "1:01:30"}
370
+ {"current_steps": 3700, "total_steps": 4795, "loss": 0.0001, "lr": 1.506541551729338e-06, "epoch": 3.854277900768329, "percentage": 77.16, "elapsed_time": "3:25:57", "remaining_time": "1:00:57"}
371
+ {"current_steps": 3710, "total_steps": 4795, "loss": 0.0001, "lr": 1.4805907175618034e-06, "epoch": 3.864695923948431, "percentage": 77.37, "elapsed_time": "3:26:30", "remaining_time": "1:00:23"}
372
+ {"current_steps": 3720, "total_steps": 4795, "loss": 0.0, "lr": 1.4548264380113763e-06, "epoch": 3.8751139471285323, "percentage": 77.58, "elapsed_time": "3:27:03", "remaining_time": "0:59:50"}
373
+ {"current_steps": 3730, "total_steps": 4795, "loss": 0.0, "lr": 1.4292500787746677e-06, "epoch": 3.8855319703086337, "percentage": 77.79, "elapsed_time": "3:27:26", "remaining_time": "0:59:13"}
374
+ {"current_steps": 3740, "total_steps": 4795, "loss": 0.0, "lr": 1.4038629955871147e-06, "epoch": 3.8959499934887356, "percentage": 78.0, "elapsed_time": "3:27:48", "remaining_time": "0:58:37"}
375
+ {"current_steps": 3750, "total_steps": 4795, "loss": 0.0002, "lr": 1.3786665341511424e-06, "epoch": 3.906368016668837, "percentage": 78.21, "elapsed_time": "3:28:10", "remaining_time": "0:58:00"}
376
+ {"current_steps": 3760, "total_steps": 4795, "loss": 0.0001, "lr": 1.3536620300648062e-06, "epoch": 3.9167860398489385, "percentage": 78.42, "elapsed_time": "3:28:32", "remaining_time": "0:57:24"}
377
+ {"current_steps": 3770, "total_steps": 4795, "loss": 0.0, "lr": 1.3288508087510198e-06, "epoch": 3.9272040630290403, "percentage": 78.62, "elapsed_time": "3:28:54", "remaining_time": "0:56:47"}
378
+ {"current_steps": 3780, "total_steps": 4795, "loss": 0.0001, "lr": 1.3042341853872753e-06, "epoch": 3.937622086209142, "percentage": 78.83, "elapsed_time": "3:29:15", "remaining_time": "0:56:11"}
379
+ {"current_steps": 3790, "total_steps": 4795, "loss": 0.0, "lr": 1.2798134648359512e-06, "epoch": 3.948040109389243, "percentage": 79.04, "elapsed_time": "3:29:37", "remaining_time": "0:55:35"}
380
+ {"current_steps": 3800, "total_steps": 4795, "loss": 0.0002, "lr": 1.2555899415751211e-06, "epoch": 3.958458132569345, "percentage": 79.25, "elapsed_time": "3:29:58", "remaining_time": "0:54:58"}
381
+ {"current_steps": 3810, "total_steps": 4795, "loss": 0.0, "lr": 1.2315648996299628e-06, "epoch": 3.9688761557494465, "percentage": 79.46, "elapsed_time": "3:30:20", "remaining_time": "0:54:22"}
382
+ {"current_steps": 3820, "total_steps": 4795, "loss": 0.0002, "lr": 1.2077396125046703e-06, "epoch": 3.979294178929548, "percentage": 79.67, "elapsed_time": "3:30:42", "remaining_time": "0:53:46"}
383
+ {"current_steps": 3830, "total_steps": 4795, "loss": 0.0, "lr": 1.1841153431149715e-06, "epoch": 3.98971220210965, "percentage": 79.87, "elapsed_time": "3:31:03", "remaining_time": "0:53:10"}
384
+ {"current_steps": 3840, "total_steps": 4795, "loss": 0.0, "lr": 1.160693343721163e-06, "epoch": 4.0, "percentage": 80.08, "elapsed_time": "3:31:25", "remaining_time": "0:52:34"}
385
+ {"current_steps": 3850, "total_steps": 4795, "loss": 0.0001, "lr": 1.137474855861751e-06, "epoch": 4.010418023180102, "percentage": 80.29, "elapsed_time": "3:33:05", "remaining_time": "0:52:18"}
386
+ {"current_steps": 3860, "total_steps": 4795, "loss": 0.0, "lr": 1.1144611102876251e-06, "epoch": 4.020836046360203, "percentage": 80.5, "elapsed_time": "3:33:27", "remaining_time": "0:51:42"}
387
+ {"current_steps": 3870, "total_steps": 4795, "loss": 0.0, "lr": 1.0916533268968293e-06, "epoch": 4.031254069540305, "percentage": 80.71, "elapsed_time": "3:33:49", "remaining_time": "0:51:06"}
388
+ {"current_steps": 3880, "total_steps": 4795, "loss": 0.0001, "lr": 1.0690527146698915e-06, "epoch": 4.041672092720407, "percentage": 80.92, "elapsed_time": "3:34:10", "remaining_time": "0:50:30"}
389
+ {"current_steps": 3890, "total_steps": 4795, "loss": 0.0, "lr": 1.046660471605744e-06, "epoch": 4.052090115900508, "percentage": 81.13, "elapsed_time": "3:34:32", "remaining_time": "0:49:54"}
390
+ {"current_steps": 3900, "total_steps": 4795, "loss": 0.0001, "lr": 1.0244777846582172e-06, "epoch": 4.0625081390806095, "percentage": 81.33, "elapsed_time": "3:34:54", "remaining_time": "0:49:19"}
391
+ {"current_steps": 3910, "total_steps": 4795, "loss": 0.0004, "lr": 1.0025058296731254e-06, "epoch": 4.072926162260711, "percentage": 81.54, "elapsed_time": "3:35:15", "remaining_time": "0:48:43"}
392
+ {"current_steps": 3920, "total_steps": 4795, "loss": 0.0025, "lr": 9.807457713259354e-07, "epoch": 4.083344185440812, "percentage": 81.75, "elapsed_time": "3:35:37", "remaining_time": "0:48:07"}
393
+ {"current_steps": 3930, "total_steps": 4795, "loss": 0.0002, "lr": 9.591987630600313e-07, "epoch": 4.093762208620914, "percentage": 81.96, "elapsed_time": "3:35:58", "remaining_time": "0:47:32"}
394
+ {"current_steps": 3940, "total_steps": 4795, "loss": 0.0, "lr": 9.37865947025573e-07, "epoch": 4.104180231801016, "percentage": 82.17, "elapsed_time": "3:36:20", "remaining_time": "0:46:56"}
395
+ {"current_steps": 3950, "total_steps": 4795, "loss": 0.0, "lr": 9.167484540189558e-07, "epoch": 4.114598254981117, "percentage": 82.38, "elapsed_time": "3:36:42", "remaining_time": "0:46:21"}
396
+ {"current_steps": 3960, "total_steps": 4795, "loss": 0.0002, "lr": 8.958474034228676e-07, "epoch": 4.125016278161219, "percentage": 82.59, "elapsed_time": "3:37:03", "remaining_time": "0:45:46"}
397
+ {"current_steps": 3970, "total_steps": 4795, "loss": 0.0001, "lr": 8.751639031469522e-07, "epoch": 4.135434301341321, "percentage": 82.79, "elapsed_time": "3:37:25", "remaining_time": "0:45:10"}
398
+ {"current_steps": 3980, "total_steps": 4795, "loss": 0.0, "lr": 8.546990495690893e-07, "epoch": 4.145852324521422, "percentage": 83.0, "elapsed_time": "3:37:47", "remaining_time": "0:44:35"}
399
+ {"current_steps": 3990, "total_steps": 4795, "loss": 0.0, "lr": 8.344539274772657e-07, "epoch": 4.156270347701524, "percentage": 83.21, "elapsed_time": "3:38:09", "remaining_time": "0:44:00"}
400
+ {"current_steps": 4000, "total_steps": 4795, "loss": 0.0001, "lr": 8.144296100120886e-07, "epoch": 4.166688370881626, "percentage": 83.42, "elapsed_time": "3:38:31", "remaining_time": "0:43:25"}
401
+ {"current_steps": 4010, "total_steps": 4795, "loss": 0.0001, "lr": 7.946271586098847e-07, "epoch": 4.177106394061727, "percentage": 83.63, "elapsed_time": "3:38:54", "remaining_time": "0:42:51"}
402
+ {"current_steps": 4020, "total_steps": 4795, "loss": 0.0, "lr": 7.750476229464543e-07, "epoch": 4.1875244172418284, "percentage": 83.84, "elapsed_time": "3:39:31", "remaining_time": "0:42:19"}
403
+ {"current_steps": 4030, "total_steps": 4795, "loss": 0.0, "lr": 7.556920408814133e-07, "epoch": 4.19794244042193, "percentage": 84.05, "elapsed_time": "3:40:13", "remaining_time": "0:41:48"}
404
+ {"current_steps": 4040, "total_steps": 4795, "loss": 0.0002, "lr": 7.365614384031944e-07, "epoch": 4.208360463602031, "percentage": 84.25, "elapsed_time": "3:40:54", "remaining_time": "0:41:17"}
405
+ {"current_steps": 4050, "total_steps": 4795, "loss": 0.0, "lr": 7.176568295746467e-07, "epoch": 4.218778486782133, "percentage": 84.46, "elapsed_time": "3:41:36", "remaining_time": "0:40:45"}
406
+ {"current_steps": 4060, "total_steps": 4795, "loss": 0.0, "lr": 6.989792164793003e-07, "epoch": 4.229196509962235, "percentage": 84.67, "elapsed_time": "3:42:17", "remaining_time": "0:40:14"}
407
+ {"current_steps": 4070, "total_steps": 4795, "loss": 0.0, "lr": 6.805295891682323e-07, "epoch": 4.239614533142336, "percentage": 84.88, "elapsed_time": "3:42:59", "remaining_time": "0:39:43"}
408
+ {"current_steps": 4080, "total_steps": 4795, "loss": 0.0, "lr": 6.62308925607597e-07, "epoch": 4.250032556322438, "percentage": 85.09, "elapsed_time": "3:43:42", "remaining_time": "0:39:12"}
409
+ {"current_steps": 4090, "total_steps": 4795, "loss": 0.0, "lr": 6.443181916267826e-07, "epoch": 4.26045057950254, "percentage": 85.3, "elapsed_time": "3:44:24", "remaining_time": "0:38:40"}
410
+ {"current_steps": 4100, "total_steps": 4795, "loss": 0.0, "lr": 6.26558340867216e-07, "epoch": 4.270868602682641, "percentage": 85.51, "elapsed_time": "3:45:06", "remaining_time": "0:38:09"}
411
+ {"current_steps": 4110, "total_steps": 4795, "loss": 0.0, "lr": 6.090303147318122e-07, "epoch": 4.281286625862743, "percentage": 85.71, "elapsed_time": "3:45:47", "remaining_time": "0:37:37"}
412
+ {"current_steps": 4120, "total_steps": 4795, "loss": 0.0, "lr": 5.917350423350731e-07, "epoch": 4.291704649042844, "percentage": 85.92, "elapsed_time": "3:46:30", "remaining_time": "0:37:06"}
413
+ {"current_steps": 4130, "total_steps": 4795, "loss": 0.0001, "lr": 5.74673440453839e-07, "epoch": 4.3021226722229455, "percentage": 86.13, "elapsed_time": "3:47:11", "remaining_time": "0:36:34"}
414
+ {"current_steps": 4140, "total_steps": 4795, "loss": 0.0, "lr": 5.578464134786916e-07, "epoch": 4.312540695403047, "percentage": 86.34, "elapsed_time": "3:47:54", "remaining_time": "0:36:03"}
415
+ {"current_steps": 4150, "total_steps": 4795, "loss": 0.0, "lr": 5.412548533660133e-07, "epoch": 4.322958718583148, "percentage": 86.55, "elapsed_time": "3:48:36", "remaining_time": "0:35:31"}
416
+ {"current_steps": 4160, "total_steps": 4795, "loss": 0.0, "lr": 5.248996395907091e-07, "epoch": 4.33337674176325, "percentage": 86.76, "elapsed_time": "3:49:18", "remaining_time": "0:35:00"}
417
+ {"current_steps": 4170, "total_steps": 4795, "loss": 0.0, "lr": 5.08781639099587e-07, "epoch": 4.343794764943352, "percentage": 86.97, "elapsed_time": "3:50:00", "remaining_time": "0:34:28"}
418
+ {"current_steps": 4180, "total_steps": 4795, "loss": 0.0, "lr": 4.92901706265404e-07, "epoch": 4.354212788123453, "percentage": 87.17, "elapsed_time": "3:50:42", "remaining_time": "0:33:56"}
419
+ {"current_steps": 4190, "total_steps": 4795, "loss": 0.001, "lr": 4.772606828415772e-07, "epoch": 4.364630811303555, "percentage": 87.38, "elapsed_time": "3:51:24", "remaining_time": "0:33:24"}
420
+ {"current_steps": 4200, "total_steps": 4795, "loss": 0.0, "lr": 4.6185939791756397e-07, "epoch": 4.375048834483657, "percentage": 87.59, "elapsed_time": "3:52:05", "remaining_time": "0:32:52"}
421
+ {"current_steps": 4210, "total_steps": 4795, "loss": 0.0025, "lr": 4.46698667874918e-07, "epoch": 4.385466857663758, "percentage": 87.8, "elapsed_time": "3:52:47", "remaining_time": "0:32:20"}
422
+ {"current_steps": 4220, "total_steps": 4795, "loss": 0.0, "lr": 4.3177929634400926e-07, "epoch": 4.39588488084386, "percentage": 88.01, "elapsed_time": "3:53:30", "remaining_time": "0:31:48"}
423
+ {"current_steps": 4230, "total_steps": 4795, "loss": 0.0, "lr": 4.1710207416143066e-07, "epoch": 4.406302904023962, "percentage": 88.22, "elapsed_time": "3:54:12", "remaining_time": "0:31:16"}
424
+ {"current_steps": 4240, "total_steps": 4795, "loss": 0.0001, "lr": 4.026677793280748e-07, "epoch": 4.416720927204063, "percentage": 88.43, "elapsed_time": "3:54:54", "remaining_time": "0:30:44"}
425
+ {"current_steps": 4250, "total_steps": 4795, "loss": 0.0002, "lr": 3.884771769678952e-07, "epoch": 4.4271389503841645, "percentage": 88.63, "elapsed_time": "3:55:37", "remaining_time": "0:30:12"}
426
+ {"current_steps": 4260, "total_steps": 4795, "loss": 0.0, "lr": 3.745310192873508e-07, "epoch": 4.437556973564266, "percentage": 88.84, "elapsed_time": "3:56:19", "remaining_time": "0:29:40"}
427
+ {"current_steps": 4270, "total_steps": 4795, "loss": 0.0001, "lr": 3.6083004553552905e-07, "epoch": 4.447974996744367, "percentage": 89.05, "elapsed_time": "3:57:01", "remaining_time": "0:29:08"}
428
+ {"current_steps": 4280, "total_steps": 4795, "loss": 0.0, "lr": 3.4737498196496686e-07, "epoch": 4.458393019924469, "percentage": 89.26, "elapsed_time": "3:57:43", "remaining_time": "0:28:36"}
429
+ {"current_steps": 4290, "total_steps": 4795, "loss": 0.0, "lr": 3.3416654179314665e-07, "epoch": 4.468811043104571, "percentage": 89.47, "elapsed_time": "3:58:25", "remaining_time": "0:28:04"}
430
+ {"current_steps": 4300, "total_steps": 4795, "loss": 0.0, "lr": 3.212054251646962e-07, "epoch": 4.479229066284672, "percentage": 89.68, "elapsed_time": "3:59:07", "remaining_time": "0:27:31"}
431
+ {"current_steps": 4310, "total_steps": 4795, "loss": 0.0, "lr": 3.0849231911427134e-07, "epoch": 4.489647089464774, "percentage": 89.89, "elapsed_time": "3:59:49", "remaining_time": "0:26:59"}
432
+ {"current_steps": 4320, "total_steps": 4795, "loss": 0.0, "lr": 2.960278975301428e-07, "epoch": 4.500065112644876, "percentage": 90.09, "elapsed_time": "4:00:31", "remaining_time": "0:26:26"}
433
+ {"current_steps": 4330, "total_steps": 4795, "loss": 0.0, "lr": 2.8381282111847097e-07, "epoch": 4.510483135824977, "percentage": 90.3, "elapsed_time": "4:01:13", "remaining_time": "0:25:54"}
434
+ {"current_steps": 4340, "total_steps": 4795, "loss": 0.0001, "lr": 2.718477373682865e-07, "epoch": 4.520901159005079, "percentage": 90.51, "elapsed_time": "4:01:54", "remaining_time": "0:25:21"}
435
+ {"current_steps": 4350, "total_steps": 4795, "loss": 0.0001, "lr": 2.601332805171669e-07, "epoch": 4.531319182185181, "percentage": 90.72, "elapsed_time": "4:02:36", "remaining_time": "0:24:49"}
436
+ {"current_steps": 4360, "total_steps": 4795, "loss": 0.0002, "lr": 2.4867007151761766e-07, "epoch": 4.541737205365282, "percentage": 90.93, "elapsed_time": "4:03:18", "remaining_time": "0:24:16"}
437
+ {"current_steps": 4370, "total_steps": 4795, "loss": 0.0049, "lr": 2.3745871800415765e-07, "epoch": 4.5521552285453835, "percentage": 91.14, "elapsed_time": "4:04:00", "remaining_time": "0:23:43"}
438
+ {"current_steps": 4380, "total_steps": 4795, "loss": 0.0002, "lr": 2.2649981426110967e-07, "epoch": 4.562573251725485, "percentage": 91.35, "elapsed_time": "4:04:42", "remaining_time": "0:23:11"}
439
+ {"current_steps": 4390, "total_steps": 4795, "loss": 0.0, "lr": 2.1579394119109863e-07, "epoch": 4.572991274905586, "percentage": 91.55, "elapsed_time": "4:05:23", "remaining_time": "0:22:38"}
440
+ {"current_steps": 4400, "total_steps": 4795, "loss": 0.014, "lr": 2.053416662842611e-07, "epoch": 4.583409298085688, "percentage": 91.76, "elapsed_time": "4:06:05", "remaining_time": "0:22:05"}
441
+ {"current_steps": 4410, "total_steps": 4795, "loss": 0.0005, "lr": 1.9514354358816102e-07, "epoch": 4.59382732126579, "percentage": 91.97, "elapsed_time": "4:06:47", "remaining_time": "0:21:32"}
442
+ {"current_steps": 4420, "total_steps": 4795, "loss": 0.0, "lr": 1.8520011367842538e-07, "epoch": 4.604245344445891, "percentage": 92.18, "elapsed_time": "4:07:30", "remaining_time": "0:20:59"}
443
+ {"current_steps": 4430, "total_steps": 4795, "loss": 0.0001, "lr": 1.7551190363008453e-07, "epoch": 4.614663367625993, "percentage": 92.39, "elapsed_time": "4:08:12", "remaining_time": "0:20:27"}
444
+ {"current_steps": 4440, "total_steps": 4795, "loss": 0.0, "lr": 1.6607942698963875e-07, "epoch": 4.625081390806095, "percentage": 92.6, "elapsed_time": "4:08:54", "remaining_time": "0:19:54"}
445
+ {"current_steps": 4450, "total_steps": 4795, "loss": 0.0, "lr": 1.5690318374783243e-07, "epoch": 4.635499413986196, "percentage": 92.81, "elapsed_time": "4:09:36", "remaining_time": "0:19:21"}
446
+ {"current_steps": 4460, "total_steps": 4795, "loss": 0.0026, "lr": 1.4798366031315292e-07, "epoch": 4.645917437166298, "percentage": 93.01, "elapsed_time": "4:10:18", "remaining_time": "0:18:48"}
447
+ {"current_steps": 4470, "total_steps": 4795, "loss": 0.0, "lr": 1.393213294860457e-07, "epoch": 4.6563354603464, "percentage": 93.22, "elapsed_time": "4:11:00", "remaining_time": "0:18:15"}
448
+ {"current_steps": 4480, "total_steps": 4795, "loss": 0.0001, "lr": 1.3091665043385383e-07, "epoch": 4.6667534835265005, "percentage": 93.43, "elapsed_time": "4:11:42", "remaining_time": "0:17:41"}
449
+ {"current_steps": 4490, "total_steps": 4795, "loss": 0.0003, "lr": 1.2277006866648122e-07, "epoch": 4.677171506706602, "percentage": 93.64, "elapsed_time": "4:12:24", "remaining_time": "0:17:08"}
450
+ {"current_steps": 4500, "total_steps": 4795, "loss": 0.0, "lr": 1.1488201601276894e-07, "epoch": 4.687589529886704, "percentage": 93.85, "elapsed_time": "4:13:07", "remaining_time": "0:16:35"}
451
+ {"current_steps": 4510, "total_steps": 4795, "loss": 0.0001, "lr": 1.0725291059761611e-07, "epoch": 4.698007553066805, "percentage": 94.06, "elapsed_time": "4:13:48", "remaining_time": "0:16:02"}
452
+ {"current_steps": 4520, "total_steps": 4795, "loss": 0.0, "lr": 9.98831568198061e-08, "epoch": 4.708425576246907, "percentage": 94.26, "elapsed_time": "4:14:30", "remaining_time": "0:15:29"}
453
+ {"current_steps": 4530, "total_steps": 4795, "loss": 0.0, "lr": 9.277314533057913e-08, "epoch": 4.718843599427009, "percentage": 94.47, "elapsed_time": "4:15:11", "remaining_time": "0:14:55"}
454
+ {"current_steps": 4540, "total_steps": 4795, "loss": 0.0, "lr": 8.592325301291782e-08, "epoch": 4.72926162260711, "percentage": 94.68, "elapsed_time": "4:15:54", "remaining_time": "0:14:22"}
455
+ {"current_steps": 4550, "total_steps": 4795, "loss": 0.0, "lr": 7.933384296157365e-08, "epoch": 4.739679645787212, "percentage": 94.89, "elapsed_time": "4:16:36", "remaining_time": "0:13:49"}
456
+ {"current_steps": 4560, "total_steps": 4795, "loss": 0.0002, "lr": 7.300526446381906e-08, "epoch": 4.750097668967314, "percentage": 95.1, "elapsed_time": "4:17:17", "remaining_time": "0:13:15"}
457
+ {"current_steps": 4570, "total_steps": 4795, "loss": 0.0, "lr": 6.693785298093336e-08, "epoch": 4.760515692147415, "percentage": 95.31, "elapsed_time": "4:17:59", "remaining_time": "0:12:42"}
458
+ {"current_steps": 4580, "total_steps": 4795, "loss": 0.0, "lr": 6.113193013041918e-08, "epoch": 4.770933715327517, "percentage": 95.52, "elapsed_time": "4:18:42", "remaining_time": "0:12:08"}
459
+ {"current_steps": 4590, "total_steps": 4795, "loss": 0.0, "lr": 5.558780366895611e-08, "epoch": 4.7813517385076185, "percentage": 95.72, "elapsed_time": "4:19:23", "remaining_time": "0:11:35"}
460
+ {"current_steps": 4600, "total_steps": 4795, "loss": 0.0001, "lr": 5.0305767476087066e-08, "epoch": 4.7917697616877195, "percentage": 95.93, "elapsed_time": "4:20:05", "remaining_time": "0:11:01"}
461
+ {"current_steps": 4610, "total_steps": 4795, "loss": 0.0001, "lr": 4.528610153863855e-08, "epoch": 4.802187784867821, "percentage": 96.14, "elapsed_time": "4:20:48", "remaining_time": "0:10:27"}
462
+ {"current_steps": 4620, "total_steps": 4795, "loss": 0.0, "lr": 4.052907193588251e-08, "epoch": 4.812605808047923, "percentage": 96.35, "elapsed_time": "4:21:30", "remaining_time": "0:09:54"}
463
+ {"current_steps": 4630, "total_steps": 4795, "loss": 0.0001, "lr": 3.60349308254293e-08, "epoch": 4.823023831228024, "percentage": 96.56, "elapsed_time": "4:22:12", "remaining_time": "0:09:20"}
464
+ {"current_steps": 4640, "total_steps": 4795, "loss": 0.0, "lr": 3.1803916429863355e-08, "epoch": 4.833441854408126, "percentage": 96.77, "elapsed_time": "4:22:54", "remaining_time": "0:08:46"}
465
+ {"current_steps": 4650, "total_steps": 4795, "loss": 0.0003, "lr": 2.7836253024114412e-08, "epoch": 4.843859877588228, "percentage": 96.98, "elapsed_time": "4:23:37", "remaining_time": "0:08:13"}
466
+ {"current_steps": 4660, "total_steps": 4795, "loss": 0.0003, "lr": 2.4132150923570353e-08, "epoch": 4.854277900768329, "percentage": 97.18, "elapsed_time": "4:24:19", "remaining_time": "0:07:39"}
467
+ {"current_steps": 4670, "total_steps": 4795, "loss": 0.0001, "lr": 2.0691806472928344e-08, "epoch": 4.864695923948431, "percentage": 97.39, "elapsed_time": "4:25:00", "remaining_time": "0:07:05"}
468
+ {"current_steps": 4680, "total_steps": 4795, "loss": 0.0, "lr": 1.7515402035787053e-08, "epoch": 4.875113947128533, "percentage": 97.6, "elapsed_time": "4:25:42", "remaining_time": "0:06:31"}
469
+ {"current_steps": 4690, "total_steps": 4795, "loss": 0.0001, "lr": 1.4603105984979382e-08, "epoch": 4.885531970308634, "percentage": 97.81, "elapsed_time": "4:26:24", "remaining_time": "0:05:57"}
470
+ {"current_steps": 4700, "total_steps": 4795, "loss": 0.0, "lr": 1.1955072693649594e-08, "epoch": 4.895949993488736, "percentage": 98.02, "elapsed_time": "4:27:06", "remaining_time": "0:05:23"}
471
+ {"current_steps": 4710, "total_steps": 4795, "loss": 0.0, "lr": 9.571442527068209e-09, "epoch": 4.9063680166688375, "percentage": 98.23, "elapsed_time": "4:27:48", "remaining_time": "0:04:49"}
472
+ {"current_steps": 4720, "total_steps": 4795, "loss": 0.0, "lr": 7.452341835192388e-09, "epoch": 4.9167860398489385, "percentage": 98.44, "elapsed_time": "4:28:31", "remaining_time": "0:04:16"}
473
+ {"current_steps": 4730, "total_steps": 4795, "loss": 0.0001, "lr": 5.597882945969635e-09, "epoch": 4.92720406302904, "percentage": 98.64, "elapsed_time": "4:29:13", "remaining_time": "0:03:41"}
474
+ {"current_steps": 4740, "total_steps": 4795, "loss": 0.0001, "lr": 4.008164159380879e-09, "epoch": 4.937622086209142, "percentage": 98.85, "elapsed_time": "4:29:55", "remaining_time": "0:03:07"}
475
+ {"current_steps": 4750, "total_steps": 4795, "loss": 0.0002, "lr": 2.6832697422324307e-09, "epoch": 4.948040109389243, "percentage": 99.06, "elapsed_time": "4:30:37", "remaining_time": "0:02:33"}
476
+ {"current_steps": 4760, "total_steps": 4795, "loss": 0.0001, "lr": 1.623269923688442e-09, "epoch": 4.958458132569345, "percentage": 99.27, "elapsed_time": "4:31:20", "remaining_time": "0:01:59"}
477
+ {"current_steps": 4770, "total_steps": 4795, "loss": 0.0001, "lr": 8.282208915466605e-10, "epoch": 4.968876155749447, "percentage": 99.48, "elapsed_time": "4:32:02", "remaining_time": "0:01:25"}
478
+ {"current_steps": 4780, "total_steps": 4795, "loss": 0.0, "lr": 2.981647892635886e-10, "epoch": 4.979294178929548, "percentage": 99.69, "elapsed_time": "4:32:44", "remaining_time": "0:00:51"}
479
+ {"current_steps": 4790, "total_steps": 4795, "loss": 0.0001, "lr": 3.312971371627338e-11, "epoch": 4.98971220210965, "percentage": 99.9, "elapsed_time": "4:33:25", "remaining_time": "0:00:17"}
480
+ {"current_steps": 4795, "total_steps": 4795, "epoch": 4.9949212136997, "percentage": 100.0, "elapsed_time": "4:33:49", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,3395 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.9949212136997,
5
+ "eval_steps": 500,
6
+ "global_step": 4795,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.010418023180101576,
13
+ "grad_norm": 11.121468544006348,
14
+ "learning_rate": 2.0833333333333333e-07,
15
+ "loss": 1.0155,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.020836046360203152,
20
+ "grad_norm": 10.899005889892578,
21
+ "learning_rate": 4.1666666666666667e-07,
22
+ "loss": 1.0247,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.03125406954030473,
27
+ "grad_norm": 11.787379264831543,
28
+ "learning_rate": 6.25e-07,
29
+ "loss": 0.9569,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.041672092720406305,
34
+ "grad_norm": 10.255136489868164,
35
+ "learning_rate": 8.333333333333333e-07,
36
+ "loss": 0.9531,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.052090115900507876,
41
+ "grad_norm": 11.36585807800293,
42
+ "learning_rate": 1.0416666666666667e-06,
43
+ "loss": 0.913,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.06250813908060945,
48
+ "grad_norm": 8.4786958694458,
49
+ "learning_rate": 1.25e-06,
50
+ "loss": 0.9508,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.07292616226071102,
55
+ "grad_norm": 13.468831062316895,
56
+ "learning_rate": 1.4583333333333335e-06,
57
+ "loss": 0.9788,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.08334418544081261,
62
+ "grad_norm": 13.184584617614746,
63
+ "learning_rate": 1.6666666666666667e-06,
64
+ "loss": 0.9664,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.09376220862091418,
69
+ "grad_norm": 8.695037841796875,
70
+ "learning_rate": 1.8750000000000003e-06,
71
+ "loss": 0.9333,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.10418023180101575,
76
+ "grad_norm": 10.278554916381836,
77
+ "learning_rate": 2.0833333333333334e-06,
78
+ "loss": 0.8748,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.11459825498111734,
83
+ "grad_norm": 10.934048652648926,
84
+ "learning_rate": 2.2916666666666666e-06,
85
+ "loss": 0.8221,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.1250162781612189,
90
+ "grad_norm": 6.949201583862305,
91
+ "learning_rate": 2.5e-06,
92
+ "loss": 0.7869,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.13543430134132048,
97
+ "grad_norm": 9.884735107421875,
98
+ "learning_rate": 2.7083333333333334e-06,
99
+ "loss": 0.7208,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.14585232452142205,
104
+ "grad_norm": 8.795683860778809,
105
+ "learning_rate": 2.916666666666667e-06,
106
+ "loss": 0.7022,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.15627034770152365,
111
+ "grad_norm": 7.725055694580078,
112
+ "learning_rate": 3.125e-06,
113
+ "loss": 0.6499,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.16668837088162522,
118
+ "grad_norm": 5.934634208679199,
119
+ "learning_rate": 3.3333333333333333e-06,
120
+ "loss": 0.5223,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.1771063940617268,
125
+ "grad_norm": 5.791957378387451,
126
+ "learning_rate": 3.5416666666666673e-06,
127
+ "loss": 0.426,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.18752441724182836,
132
+ "grad_norm": 4.230670928955078,
133
+ "learning_rate": 3.7500000000000005e-06,
134
+ "loss": 0.2739,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.19794244042192993,
139
+ "grad_norm": 3.5117030143737793,
140
+ "learning_rate": 3.958333333333333e-06,
141
+ "loss": 0.2038,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.2083604636020315,
146
+ "grad_norm": 1.7673850059509277,
147
+ "learning_rate": 4.166666666666667e-06,
148
+ "loss": 0.1047,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.2187784867821331,
153
+ "grad_norm": 2.710550308227539,
154
+ "learning_rate": 4.3750000000000005e-06,
155
+ "loss": 0.0695,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.22919650996223467,
160
+ "grad_norm": 1.8703516721725464,
161
+ "learning_rate": 4.583333333333333e-06,
162
+ "loss": 0.0448,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.23961453314233624,
167
+ "grad_norm": 2.233571767807007,
168
+ "learning_rate": 4.791666666666668e-06,
169
+ "loss": 0.0482,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.2500325563224378,
174
+ "grad_norm": 2.937638282775879,
175
+ "learning_rate": 5e-06,
176
+ "loss": 0.0464,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 0.2604505795025394,
181
+ "grad_norm": 3.5171966552734375,
182
+ "learning_rate": 5.208333333333334e-06,
183
+ "loss": 0.026,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 0.27086860268264096,
188
+ "grad_norm": 3.3302793502807617,
189
+ "learning_rate": 5.416666666666667e-06,
190
+ "loss": 0.0131,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 0.28128662586274256,
195
+ "grad_norm": 4.264321804046631,
196
+ "learning_rate": 5.625e-06,
197
+ "loss": 0.0267,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 0.2917046490428441,
202
+ "grad_norm": 1.3102387189865112,
203
+ "learning_rate": 5.833333333333334e-06,
204
+ "loss": 0.0192,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 0.3021226722229457,
209
+ "grad_norm": 2.629667043685913,
210
+ "learning_rate": 6.041666666666667e-06,
211
+ "loss": 0.0221,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 0.3125406954030473,
216
+ "grad_norm": 5.036803245544434,
217
+ "learning_rate": 6.25e-06,
218
+ "loss": 0.043,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 0.32295871858314884,
223
+ "grad_norm": 1.3269509077072144,
224
+ "learning_rate": 6.458333333333334e-06,
225
+ "loss": 0.049,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 0.33337674176325044,
230
+ "grad_norm": 3.0598790645599365,
231
+ "learning_rate": 6.666666666666667e-06,
232
+ "loss": 0.0131,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 0.343794764943352,
237
+ "grad_norm": 0.758718729019165,
238
+ "learning_rate": 6.875e-06,
239
+ "loss": 0.0244,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 0.3542127881234536,
244
+ "grad_norm": 1.2521103620529175,
245
+ "learning_rate": 7.083333333333335e-06,
246
+ "loss": 0.0149,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 0.3646308113035551,
251
+ "grad_norm": 0.35801389813423157,
252
+ "learning_rate": 7.291666666666667e-06,
253
+ "loss": 0.0086,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 0.3750488344836567,
258
+ "grad_norm": 1.5886716842651367,
259
+ "learning_rate": 7.500000000000001e-06,
260
+ "loss": 0.0158,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 0.3854668576637583,
265
+ "grad_norm": 4.816068649291992,
266
+ "learning_rate": 7.708333333333334e-06,
267
+ "loss": 0.0331,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 0.39588488084385987,
272
+ "grad_norm": 1.0439949035644531,
273
+ "learning_rate": 7.916666666666667e-06,
274
+ "loss": 0.0109,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 0.40630290402396146,
279
+ "grad_norm": 0.6839566826820374,
280
+ "learning_rate": 8.125000000000001e-06,
281
+ "loss": 0.0116,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 0.416720927204063,
286
+ "grad_norm": 2.3680825233459473,
287
+ "learning_rate": 8.333333333333334e-06,
288
+ "loss": 0.0086,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 0.4271389503841646,
293
+ "grad_norm": 2.888700485229492,
294
+ "learning_rate": 8.541666666666666e-06,
295
+ "loss": 0.0257,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 0.4375569735642662,
300
+ "grad_norm": 1.5104542970657349,
301
+ "learning_rate": 8.750000000000001e-06,
302
+ "loss": 0.0211,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 0.44797499674436775,
307
+ "grad_norm": 3.3087830543518066,
308
+ "learning_rate": 8.958333333333334e-06,
309
+ "loss": 0.0195,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 0.45839301992446935,
314
+ "grad_norm": 0.7269104719161987,
315
+ "learning_rate": 9.166666666666666e-06,
316
+ "loss": 0.0042,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 0.4688110431045709,
321
+ "grad_norm": 0.03662832826375961,
322
+ "learning_rate": 9.375000000000001e-06,
323
+ "loss": 0.0225,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 0.4792290662846725,
328
+ "grad_norm": 0.027355490252375603,
329
+ "learning_rate": 9.583333333333335e-06,
330
+ "loss": 0.0207,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 0.48964708946477403,
335
+ "grad_norm": 1.296669602394104,
336
+ "learning_rate": 9.791666666666666e-06,
337
+ "loss": 0.0184,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 0.5000651126448756,
342
+ "grad_norm": 0.7718325853347778,
343
+ "learning_rate": 1e-05,
344
+ "loss": 0.0082,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 0.5104831358249772,
349
+ "grad_norm": 0.6721673607826233,
350
+ "learning_rate": 9.999867481584167e-06,
351
+ "loss": 0.0201,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 0.5209011590050788,
356
+ "grad_norm": 2.800171136856079,
357
+ "learning_rate": 9.99946993336112e-06,
358
+ "loss": 0.0118,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 0.5313191821851804,
363
+ "grad_norm": 0.13001485168933868,
364
+ "learning_rate": 9.998807376403843e-06,
365
+ "loss": 0.0028,
366
+ "step": 510
367
+ },
368
+ {
369
+ "epoch": 0.5417372053652819,
370
+ "grad_norm": 0.11868428438901901,
371
+ "learning_rate": 9.997879845832736e-06,
372
+ "loss": 0.0248,
373
+ "step": 520
374
+ },
375
+ {
376
+ "epoch": 0.5521552285453835,
377
+ "grad_norm": 0.1077343001961708,
378
+ "learning_rate": 9.996687390813751e-06,
379
+ "loss": 0.0026,
380
+ "step": 530
381
+ },
382
+ {
383
+ "epoch": 0.5625732517254851,
384
+ "grad_norm": 0.004652015864849091,
385
+ "learning_rate": 9.995230074555788e-06,
386
+ "loss": 0.0216,
387
+ "step": 540
388
+ },
389
+ {
390
+ "epoch": 0.5729912749055867,
391
+ "grad_norm": 2.4979023933410645,
392
+ "learning_rate": 9.993507974307346e-06,
393
+ "loss": 0.0169,
394
+ "step": 550
395
+ },
396
+ {
397
+ "epoch": 0.5834092980856882,
398
+ "grad_norm": 2.996638059616089,
399
+ "learning_rate": 9.991521181352419e-06,
400
+ "loss": 0.0111,
401
+ "step": 560
402
+ },
403
+ {
404
+ "epoch": 0.5938273212657899,
405
+ "grad_norm": 1.996310830116272,
406
+ "learning_rate": 9.989269801005675e-06,
407
+ "loss": 0.0115,
408
+ "step": 570
409
+ },
410
+ {
411
+ "epoch": 0.6042453444458914,
412
+ "grad_norm": 0.1705595701932907,
413
+ "learning_rate": 9.986753952606851e-06,
414
+ "loss": 0.0072,
415
+ "step": 580
416
+ },
417
+ {
418
+ "epoch": 0.6146633676259929,
419
+ "grad_norm": 1.5530083179473877,
420
+ "learning_rate": 9.983973769514448e-06,
421
+ "loss": 0.0056,
422
+ "step": 590
423
+ },
424
+ {
425
+ "epoch": 0.6250813908060946,
426
+ "grad_norm": 0.45140549540519714,
427
+ "learning_rate": 9.98092939909865e-06,
428
+ "loss": 0.0153,
429
+ "step": 600
430
+ },
431
+ {
432
+ "epoch": 0.6354994139861961,
433
+ "grad_norm": 0.16771870851516724,
434
+ "learning_rate": 9.977621002733512e-06,
435
+ "loss": 0.0186,
436
+ "step": 610
437
+ },
438
+ {
439
+ "epoch": 0.6459174371662977,
440
+ "grad_norm": 1.3226218223571777,
441
+ "learning_rate": 9.974048755788416e-06,
442
+ "loss": 0.0042,
443
+ "step": 620
444
+ },
445
+ {
446
+ "epoch": 0.6563354603463992,
447
+ "grad_norm": 0.0040002502501010895,
448
+ "learning_rate": 9.970212847618761e-06,
449
+ "loss": 0.0136,
450
+ "step": 630
451
+ },
452
+ {
453
+ "epoch": 0.6667534835265009,
454
+ "grad_norm": 0.26981475949287415,
455
+ "learning_rate": 9.96611348155594e-06,
456
+ "loss": 0.0068,
457
+ "step": 640
458
+ },
459
+ {
460
+ "epoch": 0.6771715067066024,
461
+ "grad_norm": 0.12803907692432404,
462
+ "learning_rate": 9.961750874896548e-06,
463
+ "loss": 0.0075,
464
+ "step": 650
465
+ },
466
+ {
467
+ "epoch": 0.687589529886704,
468
+ "grad_norm": 0.15524764358997345,
469
+ "learning_rate": 9.957125258890877e-06,
470
+ "loss": 0.0154,
471
+ "step": 660
472
+ },
473
+ {
474
+ "epoch": 0.6980075530668056,
475
+ "grad_norm": 1.0714681148529053,
476
+ "learning_rate": 9.952236878730648e-06,
477
+ "loss": 0.008,
478
+ "step": 670
479
+ },
480
+ {
481
+ "epoch": 0.7084255762469072,
482
+ "grad_norm": 5.858039379119873,
483
+ "learning_rate": 9.947085993536019e-06,
484
+ "loss": 0.0293,
485
+ "step": 680
486
+ },
487
+ {
488
+ "epoch": 0.7188435994270087,
489
+ "grad_norm": 0.6457405090332031,
490
+ "learning_rate": 9.941672876341848e-06,
491
+ "loss": 0.0225,
492
+ "step": 690
493
+ },
494
+ {
495
+ "epoch": 0.7292616226071102,
496
+ "grad_norm": 0.08024393022060394,
497
+ "learning_rate": 9.935997814083221e-06,
498
+ "loss": 0.003,
499
+ "step": 700
500
+ },
501
+ {
502
+ "epoch": 0.7396796457872119,
503
+ "grad_norm": 2.8071186542510986,
504
+ "learning_rate": 9.930061107580245e-06,
505
+ "loss": 0.0134,
506
+ "step": 710
507
+ },
508
+ {
509
+ "epoch": 0.7500976689673134,
510
+ "grad_norm": 1.7718358039855957,
511
+ "learning_rate": 9.923863071522092e-06,
512
+ "loss": 0.0228,
513
+ "step": 720
514
+ },
515
+ {
516
+ "epoch": 0.760515692147415,
517
+ "grad_norm": 0.015169305726885796,
518
+ "learning_rate": 9.917404034450333e-06,
519
+ "loss": 0.0066,
520
+ "step": 730
521
+ },
522
+ {
523
+ "epoch": 0.7709337153275166,
524
+ "grad_norm": 0.020366037264466286,
525
+ "learning_rate": 9.91068433874151e-06,
526
+ "loss": 0.0034,
527
+ "step": 740
528
+ },
529
+ {
530
+ "epoch": 0.7813517385076182,
531
+ "grad_norm": 0.4552544951438904,
532
+ "learning_rate": 9.903704340589e-06,
533
+ "loss": 0.0172,
534
+ "step": 750
535
+ },
536
+ {
537
+ "epoch": 0.7917697616877197,
538
+ "grad_norm": 0.586904764175415,
539
+ "learning_rate": 9.896464409984115e-06,
540
+ "loss": 0.0158,
541
+ "step": 760
542
+ },
543
+ {
544
+ "epoch": 0.8021877848678214,
545
+ "grad_norm": 0.003931673243641853,
546
+ "learning_rate": 9.888964930696514e-06,
547
+ "loss": 0.0118,
548
+ "step": 770
549
+ },
550
+ {
551
+ "epoch": 0.8126058080479229,
552
+ "grad_norm": 0.14547383785247803,
553
+ "learning_rate": 9.881206300253838e-06,
554
+ "loss": 0.0091,
555
+ "step": 780
556
+ },
557
+ {
558
+ "epoch": 0.8230238312280245,
559
+ "grad_norm": 3.7878336906433105,
560
+ "learning_rate": 9.87318892992066e-06,
561
+ "loss": 0.0195,
562
+ "step": 790
563
+ },
564
+ {
565
+ "epoch": 0.833441854408126,
566
+ "grad_norm": 0.4627358913421631,
567
+ "learning_rate": 9.864913244676661e-06,
568
+ "loss": 0.0072,
569
+ "step": 800
570
+ },
571
+ {
572
+ "epoch": 0.8438598775882277,
573
+ "grad_norm": 4.011754035949707,
574
+ "learning_rate": 9.856379683194122e-06,
575
+ "loss": 0.0102,
576
+ "step": 810
577
+ },
578
+ {
579
+ "epoch": 0.8542779007683292,
580
+ "grad_norm": 0.15841282904148102,
581
+ "learning_rate": 9.847588697814662e-06,
582
+ "loss": 0.0037,
583
+ "step": 820
584
+ },
585
+ {
586
+ "epoch": 0.8646959239484308,
587
+ "grad_norm": 0.011311404407024384,
588
+ "learning_rate": 9.838540754525266e-06,
589
+ "loss": 0.0057,
590
+ "step": 830
591
+ },
592
+ {
593
+ "epoch": 0.8751139471285324,
594
+ "grad_norm": 2.975158929824829,
595
+ "learning_rate": 9.829236332933573e-06,
596
+ "loss": 0.0094,
597
+ "step": 840
598
+ },
599
+ {
600
+ "epoch": 0.885531970308634,
601
+ "grad_norm": 0.0023622314911335707,
602
+ "learning_rate": 9.819675926242473e-06,
603
+ "loss": 0.0051,
604
+ "step": 850
605
+ },
606
+ {
607
+ "epoch": 0.8959499934887355,
608
+ "grad_norm": 1.6916306018829346,
609
+ "learning_rate": 9.809860041223945e-06,
610
+ "loss": 0.0179,
611
+ "step": 860
612
+ },
613
+ {
614
+ "epoch": 0.906368016668837,
615
+ "grad_norm": 0.002486775629222393,
616
+ "learning_rate": 9.799789198192197e-06,
617
+ "loss": 0.0257,
618
+ "step": 870
619
+ },
620
+ {
621
+ "epoch": 0.9167860398489387,
622
+ "grad_norm": 0.03367191180586815,
623
+ "learning_rate": 9.789463930976098e-06,
624
+ "loss": 0.0044,
625
+ "step": 880
626
+ },
627
+ {
628
+ "epoch": 0.9272040630290402,
629
+ "grad_norm": 3.230694055557251,
630
+ "learning_rate": 9.77888478689087e-06,
631
+ "loss": 0.0096,
632
+ "step": 890
633
+ },
634
+ {
635
+ "epoch": 0.9376220862091418,
636
+ "grad_norm": 0.7497555017471313,
637
+ "learning_rate": 9.76805232670908e-06,
638
+ "loss": 0.0132,
639
+ "step": 900
640
+ },
641
+ {
642
+ "epoch": 0.9480401093892434,
643
+ "grad_norm": 0.5059885382652283,
644
+ "learning_rate": 9.756967124630911e-06,
645
+ "loss": 0.0021,
646
+ "step": 910
647
+ },
648
+ {
649
+ "epoch": 0.958458132569345,
650
+ "grad_norm": 0.031383853405714035,
651
+ "learning_rate": 9.745629768253735e-06,
652
+ "loss": 0.0027,
653
+ "step": 920
654
+ },
655
+ {
656
+ "epoch": 0.9688761557494465,
657
+ "grad_norm": 0.2179183065891266,
658
+ "learning_rate": 9.734040858540947e-06,
659
+ "loss": 0.0081,
660
+ "step": 930
661
+ },
662
+ {
663
+ "epoch": 0.9792941789295481,
664
+ "grad_norm": 0.02150609716773033,
665
+ "learning_rate": 9.722201009790135e-06,
666
+ "loss": 0.0013,
667
+ "step": 940
668
+ },
669
+ {
670
+ "epoch": 0.9897122021096497,
671
+ "grad_norm": 0.7268407344818115,
672
+ "learning_rate": 9.710110849600498e-06,
673
+ "loss": 0.0154,
674
+ "step": 950
675
+ },
676
+ {
677
+ "epoch": 1.0,
678
+ "grad_norm": 4.076342582702637,
679
+ "learning_rate": 9.697771018839586e-06,
680
+ "loss": 0.0182,
681
+ "step": 960
682
+ },
683
+ {
684
+ "epoch": 1.0104180231801017,
685
+ "grad_norm": 0.0026843964587897062,
686
+ "learning_rate": 9.685182171609328e-06,
687
+ "loss": 0.0078,
688
+ "step": 970
689
+ },
690
+ {
691
+ "epoch": 1.020836046360203,
692
+ "grad_norm": 1.928469181060791,
693
+ "learning_rate": 9.672344975211361e-06,
694
+ "loss": 0.0026,
695
+ "step": 980
696
+ },
697
+ {
698
+ "epoch": 1.0312540695403047,
699
+ "grad_norm": 0.19345267117023468,
700
+ "learning_rate": 9.659260110111659e-06,
701
+ "loss": 0.0013,
702
+ "step": 990
703
+ },
704
+ {
705
+ "epoch": 1.0416720927204064,
706
+ "grad_norm": 0.10904184728860855,
707
+ "learning_rate": 9.645928269904459e-06,
708
+ "loss": 0.0006,
709
+ "step": 1000
710
+ },
711
+ {
712
+ "epoch": 1.0520901159005078,
713
+ "grad_norm": 0.0044245533645153046,
714
+ "learning_rate": 9.632350161275496e-06,
715
+ "loss": 0.0078,
716
+ "step": 1010
717
+ },
718
+ {
719
+ "epoch": 1.0625081390806095,
720
+ "grad_norm": 0.052690137177705765,
721
+ "learning_rate": 9.618526503964552e-06,
722
+ "loss": 0.0033,
723
+ "step": 1020
724
+ },
725
+ {
726
+ "epoch": 1.072926162260711,
727
+ "grad_norm": 8.081074338406324e-05,
728
+ "learning_rate": 9.604458030727292e-06,
729
+ "loss": 0.0041,
730
+ "step": 1030
731
+ },
732
+ {
733
+ "epoch": 1.0833441854408126,
734
+ "grad_norm": 0.9879348278045654,
735
+ "learning_rate": 9.590145487296431e-06,
736
+ "loss": 0.0015,
737
+ "step": 1040
738
+ },
739
+ {
740
+ "epoch": 1.0937622086209142,
741
+ "grad_norm": 0.1349502056837082,
742
+ "learning_rate": 9.575589632342201e-06,
743
+ "loss": 0.0003,
744
+ "step": 1050
745
+ },
746
+ {
747
+ "epoch": 1.1041802318010157,
748
+ "grad_norm": 0.0031191352754831314,
749
+ "learning_rate": 9.560791237432141e-06,
750
+ "loss": 0.0157,
751
+ "step": 1060
752
+ },
753
+ {
754
+ "epoch": 1.1145982549811173,
755
+ "grad_norm": 0.003794798394665122,
756
+ "learning_rate": 9.545751086990187e-06,
757
+ "loss": 0.0074,
758
+ "step": 1070
759
+ },
760
+ {
761
+ "epoch": 1.125016278161219,
762
+ "grad_norm": 0.04094767943024635,
763
+ "learning_rate": 9.530469978255105e-06,
764
+ "loss": 0.0001,
765
+ "step": 1080
766
+ },
767
+ {
768
+ "epoch": 1.1354343013413204,
769
+ "grad_norm": 1.0790066880872473e-05,
770
+ "learning_rate": 9.514948721238227e-06,
771
+ "loss": 0.0074,
772
+ "step": 1090
773
+ },
774
+ {
775
+ "epoch": 1.145852324521422,
776
+ "grad_norm": 3.5826034545898438,
777
+ "learning_rate": 9.499188138680504e-06,
778
+ "loss": 0.0108,
779
+ "step": 1100
780
+ },
781
+ {
782
+ "epoch": 1.1562703477015237,
783
+ "grad_norm": 0.39376917481422424,
784
+ "learning_rate": 9.48318906600891e-06,
785
+ "loss": 0.002,
786
+ "step": 1110
787
+ },
788
+ {
789
+ "epoch": 1.1666883708816251,
790
+ "grad_norm": 0.008306131698191166,
791
+ "learning_rate": 9.466952351292158e-06,
792
+ "loss": 0.0033,
793
+ "step": 1120
794
+ },
795
+ {
796
+ "epoch": 1.1771063940617268,
797
+ "grad_norm": 0.08543845266103745,
798
+ "learning_rate": 9.450478855195724e-06,
799
+ "loss": 0.0002,
800
+ "step": 1130
801
+ },
802
+ {
803
+ "epoch": 1.1875244172418284,
804
+ "grad_norm": 1.0472006797790527,
805
+ "learning_rate": 9.433769450936254e-06,
806
+ "loss": 0.0033,
807
+ "step": 1140
808
+ },
809
+ {
810
+ "epoch": 1.1979424404219299,
811
+ "grad_norm": 0.0014151977375149727,
812
+ "learning_rate": 9.416825024235262e-06,
813
+ "loss": 0.0005,
814
+ "step": 1150
815
+ },
816
+ {
817
+ "epoch": 1.2083604636020315,
818
+ "grad_norm": 3.7921109199523926,
819
+ "learning_rate": 9.399646473272181e-06,
820
+ "loss": 0.0062,
821
+ "step": 1160
822
+ },
823
+ {
824
+ "epoch": 1.2187784867821332,
825
+ "grad_norm": 0.0031763967126607895,
826
+ "learning_rate": 9.382234708636753e-06,
827
+ "loss": 0.0018,
828
+ "step": 1170
829
+ },
830
+ {
831
+ "epoch": 1.2291965099622346,
832
+ "grad_norm": 0.00694266939535737,
833
+ "learning_rate": 9.364590653280767e-06,
834
+ "loss": 0.0023,
835
+ "step": 1180
836
+ },
837
+ {
838
+ "epoch": 1.2396145331423363,
839
+ "grad_norm": 0.0004896274767816067,
840
+ "learning_rate": 9.346715242469128e-06,
841
+ "loss": 0.0065,
842
+ "step": 1190
843
+ },
844
+ {
845
+ "epoch": 1.250032556322438,
846
+ "grad_norm": 0.008612466044723988,
847
+ "learning_rate": 9.328609423730285e-06,
848
+ "loss": 0.0004,
849
+ "step": 1200
850
+ },
851
+ {
852
+ "epoch": 1.2604505795025394,
853
+ "grad_norm": 0.00032048547291196883,
854
+ "learning_rate": 9.310274156806006e-06,
855
+ "loss": 0.0122,
856
+ "step": 1210
857
+ },
858
+ {
859
+ "epoch": 1.270868602682641,
860
+ "grad_norm": 0.024016475304961205,
861
+ "learning_rate": 9.291710413600498e-06,
862
+ "loss": 0.0105,
863
+ "step": 1220
864
+ },
865
+ {
866
+ "epoch": 1.2812866258627427,
867
+ "grad_norm": 0.059294044971466064,
868
+ "learning_rate": 9.272919178128902e-06,
869
+ "loss": 0.0005,
870
+ "step": 1230
871
+ },
872
+ {
873
+ "epoch": 1.291704649042844,
874
+ "grad_norm": 0.0428699254989624,
875
+ "learning_rate": 9.253901446465116e-06,
876
+ "loss": 0.0032,
877
+ "step": 1240
878
+ },
879
+ {
880
+ "epoch": 1.3021226722229458,
881
+ "grad_norm": 0.1531277745962143,
882
+ "learning_rate": 9.234658226689015e-06,
883
+ "loss": 0.0003,
884
+ "step": 1250
885
+ },
886
+ {
887
+ "epoch": 1.3125406954030474,
888
+ "grad_norm": 2.18913197517395,
889
+ "learning_rate": 9.215190538832995e-06,
890
+ "loss": 0.0046,
891
+ "step": 1260
892
+ },
893
+ {
894
+ "epoch": 1.3229587185831488,
895
+ "grad_norm": 0.017851542681455612,
896
+ "learning_rate": 9.195499414827917e-06,
897
+ "loss": 0.0001,
898
+ "step": 1270
899
+ },
900
+ {
901
+ "epoch": 1.3333767417632505,
902
+ "grad_norm": 9.223073959350586,
903
+ "learning_rate": 9.175585898448408e-06,
904
+ "loss": 0.0267,
905
+ "step": 1280
906
+ },
907
+ {
908
+ "epoch": 1.343794764943352,
909
+ "grad_norm": 0.0561683289706707,
910
+ "learning_rate": 9.155451045257523e-06,
911
+ "loss": 0.0157,
912
+ "step": 1290
913
+ },
914
+ {
915
+ "epoch": 1.3542127881234536,
916
+ "grad_norm": 0.02189466543495655,
917
+ "learning_rate": 9.135095922550801e-06,
918
+ "loss": 0.002,
919
+ "step": 1300
920
+ },
921
+ {
922
+ "epoch": 1.364630811303555,
923
+ "grad_norm": 3.704211950302124,
924
+ "learning_rate": 9.114521609299691e-06,
925
+ "loss": 0.008,
926
+ "step": 1310
927
+ },
928
+ {
929
+ "epoch": 1.3750488344836567,
930
+ "grad_norm": 0.09731532633304596,
931
+ "learning_rate": 9.09372919609435e-06,
932
+ "loss": 0.0023,
933
+ "step": 1320
934
+ },
935
+ {
936
+ "epoch": 1.3854668576637583,
937
+ "grad_norm": 0.18077053129673004,
938
+ "learning_rate": 9.072719785085842e-06,
939
+ "loss": 0.013,
940
+ "step": 1330
941
+ },
942
+ {
943
+ "epoch": 1.3958848808438598,
944
+ "grad_norm": 0.026519620791077614,
945
+ "learning_rate": 9.051494489927714e-06,
946
+ "loss": 0.0098,
947
+ "step": 1340
948
+ },
949
+ {
950
+ "epoch": 1.4063029040239614,
951
+ "grad_norm": 3.807239294052124,
952
+ "learning_rate": 9.030054435716961e-06,
953
+ "loss": 0.0152,
954
+ "step": 1350
955
+ },
956
+ {
957
+ "epoch": 1.416720927204063,
958
+ "grad_norm": 1.5597429275512695,
959
+ "learning_rate": 9.008400758934392e-06,
960
+ "loss": 0.0014,
961
+ "step": 1360
962
+ },
963
+ {
964
+ "epoch": 1.4271389503841645,
965
+ "grad_norm": 0.00023376860190182924,
966
+ "learning_rate": 8.986534607384383e-06,
967
+ "loss": 0.0089,
968
+ "step": 1370
969
+ },
970
+ {
971
+ "epoch": 1.4375569735642661,
972
+ "grad_norm": 0.13520453870296478,
973
+ "learning_rate": 8.96445714013404e-06,
974
+ "loss": 0.0008,
975
+ "step": 1380
976
+ },
977
+ {
978
+ "epoch": 1.4479749967443678,
979
+ "grad_norm": 3.3357255458831787,
980
+ "learning_rate": 8.942169527451756e-06,
981
+ "loss": 0.0177,
982
+ "step": 1390
983
+ },
984
+ {
985
+ "epoch": 1.4583930199244692,
986
+ "grad_norm": 0.13861112296581268,
987
+ "learning_rate": 8.919672950745185e-06,
988
+ "loss": 0.0177,
989
+ "step": 1400
990
+ },
991
+ {
992
+ "epoch": 1.468811043104571,
993
+ "grad_norm": 0.0005054327775724232,
994
+ "learning_rate": 8.896968602498605e-06,
995
+ "loss": 0.0087,
996
+ "step": 1410
997
+ },
998
+ {
999
+ "epoch": 1.4792290662846725,
1000
+ "grad_norm": 0.188633993268013,
1001
+ "learning_rate": 8.874057686209727e-06,
1002
+ "loss": 0.0023,
1003
+ "step": 1420
1004
+ },
1005
+ {
1006
+ "epoch": 1.489647089464774,
1007
+ "grad_norm": 4.461554527282715,
1008
+ "learning_rate": 8.850941416325876e-06,
1009
+ "loss": 0.01,
1010
+ "step": 1430
1011
+ },
1012
+ {
1013
+ "epoch": 1.5000651126448756,
1014
+ "grad_norm": 0.0012838880065828562,
1015
+ "learning_rate": 8.827621018179644e-06,
1016
+ "loss": 0.0129,
1017
+ "step": 1440
1018
+ },
1019
+ {
1020
+ "epoch": 1.5104831358249773,
1021
+ "grad_norm": 0.40561285614967346,
1022
+ "learning_rate": 8.804097727923916e-06,
1023
+ "loss": 0.0049,
1024
+ "step": 1450
1025
+ },
1026
+ {
1027
+ "epoch": 1.5209011590050787,
1028
+ "grad_norm": 0.038120679557323456,
1029
+ "learning_rate": 8.780372792466356e-06,
1030
+ "loss": 0.0104,
1031
+ "step": 1460
1032
+ },
1033
+ {
1034
+ "epoch": 1.5313191821851804,
1035
+ "grad_norm": 2.760854959487915,
1036
+ "learning_rate": 8.75644746940331e-06,
1037
+ "loss": 0.0032,
1038
+ "step": 1470
1039
+ },
1040
+ {
1041
+ "epoch": 1.541737205365282,
1042
+ "grad_norm": 0.5340994596481323,
1043
+ "learning_rate": 8.732323026953141e-06,
1044
+ "loss": 0.008,
1045
+ "step": 1480
1046
+ },
1047
+ {
1048
+ "epoch": 1.5521552285453835,
1049
+ "grad_norm": 2.0399329662323,
1050
+ "learning_rate": 8.708000743889009e-06,
1051
+ "loss": 0.0052,
1052
+ "step": 1490
1053
+ },
1054
+ {
1055
+ "epoch": 1.5625732517254851,
1056
+ "grad_norm": 2.4984633922576904,
1057
+ "learning_rate": 8.68348190947108e-06,
1058
+ "loss": 0.0151,
1059
+ "step": 1500
1060
+ },
1061
+ {
1062
+ "epoch": 1.5729912749055868,
1063
+ "grad_norm": 0.07911086082458496,
1064
+ "learning_rate": 8.658767823378198e-06,
1065
+ "loss": 0.0012,
1066
+ "step": 1510
1067
+ },
1068
+ {
1069
+ "epoch": 1.5834092980856882,
1070
+ "grad_norm": 0.12755373120307922,
1071
+ "learning_rate": 8.63385979563897e-06,
1072
+ "loss": 0.0039,
1073
+ "step": 1520
1074
+ },
1075
+ {
1076
+ "epoch": 1.5938273212657899,
1077
+ "grad_norm": 0.03218844160437584,
1078
+ "learning_rate": 8.608759146562352e-06,
1079
+ "loss": 0.0013,
1080
+ "step": 1530
1081
+ },
1082
+ {
1083
+ "epoch": 1.6042453444458915,
1084
+ "grad_norm": 0.00013593579933512956,
1085
+ "learning_rate": 8.583467206667643e-06,
1086
+ "loss": 0.0007,
1087
+ "step": 1540
1088
+ },
1089
+ {
1090
+ "epoch": 1.614663367625993,
1091
+ "grad_norm": 0.00014157673285808414,
1092
+ "learning_rate": 8.557985316613967e-06,
1093
+ "loss": 0.0005,
1094
+ "step": 1550
1095
+ },
1096
+ {
1097
+ "epoch": 1.6250813908060946,
1098
+ "grad_norm": 0.001298804534599185,
1099
+ "learning_rate": 8.532314827129207e-06,
1100
+ "loss": 0.0003,
1101
+ "step": 1560
1102
+ },
1103
+ {
1104
+ "epoch": 1.6354994139861962,
1105
+ "grad_norm": 0.00196313438937068,
1106
+ "learning_rate": 8.5064570989384e-06,
1107
+ "loss": 0.0042,
1108
+ "step": 1570
1109
+ },
1110
+ {
1111
+ "epoch": 1.6459174371662977,
1112
+ "grad_norm": 0.05907834693789482,
1113
+ "learning_rate": 8.480413502691618e-06,
1114
+ "loss": 0.0042,
1115
+ "step": 1580
1116
+ },
1117
+ {
1118
+ "epoch": 1.6563354603463991,
1119
+ "grad_norm": 2.9666049480438232,
1120
+ "learning_rate": 8.454185418891305e-06,
1121
+ "loss": 0.0009,
1122
+ "step": 1590
1123
+ },
1124
+ {
1125
+ "epoch": 1.666753483526501,
1126
+ "grad_norm": 0.04065735638141632,
1127
+ "learning_rate": 8.427774237819113e-06,
1128
+ "loss": 0.0071,
1129
+ "step": 1600
1130
+ },
1131
+ {
1132
+ "epoch": 1.6771715067066024,
1133
+ "grad_norm": 2.4857747554779053,
1134
+ "learning_rate": 8.401181359462187e-06,
1135
+ "loss": 0.0078,
1136
+ "step": 1610
1137
+ },
1138
+ {
1139
+ "epoch": 1.6875895298867039,
1140
+ "grad_norm": 0.0004068031266797334,
1141
+ "learning_rate": 8.374408193438977e-06,
1142
+ "loss": 0.0003,
1143
+ "step": 1620
1144
+ },
1145
+ {
1146
+ "epoch": 1.6980075530668057,
1147
+ "grad_norm": 0.000535597384441644,
1148
+ "learning_rate": 8.347456158924496e-06,
1149
+ "loss": 0.0054,
1150
+ "step": 1630
1151
+ },
1152
+ {
1153
+ "epoch": 1.7084255762469072,
1154
+ "grad_norm": 0.06767778843641281,
1155
+ "learning_rate": 8.320326684575116e-06,
1156
+ "loss": 0.0016,
1157
+ "step": 1640
1158
+ },
1159
+ {
1160
+ "epoch": 1.7188435994270086,
1161
+ "grad_norm": 0.00011920960969291627,
1162
+ "learning_rate": 8.29302120845282e-06,
1163
+ "loss": 0.0008,
1164
+ "step": 1650
1165
+ },
1166
+ {
1167
+ "epoch": 1.7292616226071102,
1168
+ "grad_norm": 0.1632106900215149,
1169
+ "learning_rate": 8.265541177948986e-06,
1170
+ "loss": 0.0011,
1171
+ "step": 1660
1172
+ },
1173
+ {
1174
+ "epoch": 1.739679645787212,
1175
+ "grad_norm": 0.03428950533270836,
1176
+ "learning_rate": 8.237888049707656e-06,
1177
+ "loss": 0.0087,
1178
+ "step": 1670
1179
+ },
1180
+ {
1181
+ "epoch": 1.7500976689673133,
1182
+ "grad_norm": 0.0007083571399562061,
1183
+ "learning_rate": 8.210063289548328e-06,
1184
+ "loss": 0.0025,
1185
+ "step": 1680
1186
+ },
1187
+ {
1188
+ "epoch": 1.760515692147415,
1189
+ "grad_norm": 1.6527312254766002e-05,
1190
+ "learning_rate": 8.182068372388259e-06,
1191
+ "loss": 0.0044,
1192
+ "step": 1690
1193
+ },
1194
+ {
1195
+ "epoch": 1.7709337153275166,
1196
+ "grad_norm": 0.0002037805679719895,
1197
+ "learning_rate": 8.153904782164275e-06,
1198
+ "loss": 0.0011,
1199
+ "step": 1700
1200
+ },
1201
+ {
1202
+ "epoch": 1.781351738507618,
1203
+ "grad_norm": 0.0011550731724128127,
1204
+ "learning_rate": 8.125574011754125e-06,
1205
+ "loss": 0.0028,
1206
+ "step": 1710
1207
+ },
1208
+ {
1209
+ "epoch": 1.7917697616877197,
1210
+ "grad_norm": 0.058955416083335876,
1211
+ "learning_rate": 8.097077562897332e-06,
1212
+ "loss": 0.0003,
1213
+ "step": 1720
1214
+ },
1215
+ {
1216
+ "epoch": 1.8021877848678214,
1217
+ "grad_norm": 0.6716377139091492,
1218
+ "learning_rate": 8.0684169461156e-06,
1219
+ "loss": 0.0082,
1220
+ "step": 1730
1221
+ },
1222
+ {
1223
+ "epoch": 1.8126058080479228,
1224
+ "grad_norm": 0.0014718943275511265,
1225
+ "learning_rate": 8.03959368063274e-06,
1226
+ "loss": 0.0069,
1227
+ "step": 1740
1228
+ },
1229
+ {
1230
+ "epoch": 1.8230238312280245,
1231
+ "grad_norm": 0.00023426729603670537,
1232
+ "learning_rate": 8.01060929429415e-06,
1233
+ "loss": 0.0107,
1234
+ "step": 1750
1235
+ },
1236
+ {
1237
+ "epoch": 1.8334418544081261,
1238
+ "grad_norm": 0.002175110625103116,
1239
+ "learning_rate": 7.981465323485808e-06,
1240
+ "loss": 0.0082,
1241
+ "step": 1760
1242
+ },
1243
+ {
1244
+ "epoch": 1.8438598775882276,
1245
+ "grad_norm": 0.719397783279419,
1246
+ "learning_rate": 7.952163313052856e-06,
1247
+ "loss": 0.0009,
1248
+ "step": 1770
1249
+ },
1250
+ {
1251
+ "epoch": 1.8542779007683292,
1252
+ "grad_norm": 0.976002037525177,
1253
+ "learning_rate": 7.922704816217693e-06,
1254
+ "loss": 0.0082,
1255
+ "step": 1780
1256
+ },
1257
+ {
1258
+ "epoch": 1.8646959239484309,
1259
+ "grad_norm": 3.530712604522705,
1260
+ "learning_rate": 7.893091394497651e-06,
1261
+ "loss": 0.0239,
1262
+ "step": 1790
1263
+ },
1264
+ {
1265
+ "epoch": 1.8751139471285323,
1266
+ "grad_norm": 1.4467989206314087,
1267
+ "learning_rate": 7.863324617622227e-06,
1268
+ "loss": 0.0019,
1269
+ "step": 1800
1270
+ },
1271
+ {
1272
+ "epoch": 1.885531970308634,
1273
+ "grad_norm": 0.0016262925928458571,
1274
+ "learning_rate": 7.833406063449866e-06,
1275
+ "loss": 0.0004,
1276
+ "step": 1810
1277
+ },
1278
+ {
1279
+ "epoch": 1.8959499934887356,
1280
+ "grad_norm": 0.007764583453536034,
1281
+ "learning_rate": 7.803337317884328e-06,
1282
+ "loss": 0.0025,
1283
+ "step": 1820
1284
+ },
1285
+ {
1286
+ "epoch": 1.906368016668837,
1287
+ "grad_norm": 4.686193278757855e-06,
1288
+ "learning_rate": 7.773119974790626e-06,
1289
+ "loss": 0.0008,
1290
+ "step": 1830
1291
+ },
1292
+ {
1293
+ "epoch": 1.9167860398489387,
1294
+ "grad_norm": 0.02094154804944992,
1295
+ "learning_rate": 7.742755635910532e-06,
1296
+ "loss": 0.0,
1297
+ "step": 1840
1298
+ },
1299
+ {
1300
+ "epoch": 1.9272040630290403,
1301
+ "grad_norm": 0.026232223957777023,
1302
+ "learning_rate": 7.712245910777684e-06,
1303
+ "loss": 0.0004,
1304
+ "step": 1850
1305
+ },
1306
+ {
1307
+ "epoch": 1.9376220862091418,
1308
+ "grad_norm": 0.07305476814508438,
1309
+ "learning_rate": 7.68159241663226e-06,
1310
+ "loss": 0.0108,
1311
+ "step": 1860
1312
+ },
1313
+ {
1314
+ "epoch": 1.9480401093892434,
1315
+ "grad_norm": 0.11340963840484619,
1316
+ "learning_rate": 7.650796778335248e-06,
1317
+ "loss": 0.0006,
1318
+ "step": 1870
1319
+ },
1320
+ {
1321
+ "epoch": 1.958458132569345,
1322
+ "grad_norm": 2.027043592534028e-05,
1323
+ "learning_rate": 7.619860628282333e-06,
1324
+ "loss": 0.0178,
1325
+ "step": 1880
1326
+ },
1327
+ {
1328
+ "epoch": 1.9688761557494465,
1329
+ "grad_norm": 0.14668358862400055,
1330
+ "learning_rate": 7.5887856063173525e-06,
1331
+ "loss": 0.0031,
1332
+ "step": 1890
1333
+ },
1334
+ {
1335
+ "epoch": 1.979294178929548,
1336
+ "grad_norm": 0.056070610880851746,
1337
+ "learning_rate": 7.5575733596453805e-06,
1338
+ "loss": 0.0006,
1339
+ "step": 1900
1340
+ },
1341
+ {
1342
+ "epoch": 1.9897122021096498,
1343
+ "grad_norm": 0.01226119790226221,
1344
+ "learning_rate": 7.526225542745408e-06,
1345
+ "loss": 0.0086,
1346
+ "step": 1910
1347
+ },
1348
+ {
1349
+ "epoch": 2.0,
1350
+ "grad_norm": 0.026876511052250862,
1351
+ "learning_rate": 7.494743817282651e-06,
1352
+ "loss": 0.0021,
1353
+ "step": 1920
1354
+ },
1355
+ {
1356
+ "epoch": 2.0104180231801014,
1357
+ "grad_norm": 1.010679006576538,
1358
+ "learning_rate": 7.463129852020465e-06,
1359
+ "loss": 0.001,
1360
+ "step": 1930
1361
+ },
1362
+ {
1363
+ "epoch": 2.0208360463602033,
1364
+ "grad_norm": 0.05851694196462631,
1365
+ "learning_rate": 7.431385322731885e-06,
1366
+ "loss": 0.0002,
1367
+ "step": 1940
1368
+ },
1369
+ {
1370
+ "epoch": 2.0312540695403047,
1371
+ "grad_norm": 0.0013823857298120856,
1372
+ "learning_rate": 7.399511912110807e-06,
1373
+ "loss": 0.0003,
1374
+ "step": 1950
1375
+ },
1376
+ {
1377
+ "epoch": 2.041672092720406,
1378
+ "grad_norm": 0.0012095741694793105,
1379
+ "learning_rate": 7.367511309682782e-06,
1380
+ "loss": 0.0009,
1381
+ "step": 1960
1382
+ },
1383
+ {
1384
+ "epoch": 2.052090115900508,
1385
+ "grad_norm": 5.620197296142578,
1386
+ "learning_rate": 7.335385211715467e-06,
1387
+ "loss": 0.0064,
1388
+ "step": 1970
1389
+ },
1390
+ {
1391
+ "epoch": 2.0625081390806095,
1392
+ "grad_norm": 0.0062335277907550335,
1393
+ "learning_rate": 7.303135321128705e-06,
1394
+ "loss": 0.0002,
1395
+ "step": 1980
1396
+ },
1397
+ {
1398
+ "epoch": 2.072926162260711,
1399
+ "grad_norm": 0.001201906125061214,
1400
+ "learning_rate": 7.270763347404262e-06,
1401
+ "loss": 0.0017,
1402
+ "step": 1990
1403
+ },
1404
+ {
1405
+ "epoch": 2.083344185440813,
1406
+ "grad_norm": 0.016548430547118187,
1407
+ "learning_rate": 7.238271006495206e-06,
1408
+ "loss": 0.0015,
1409
+ "step": 2000
1410
+ },
1411
+ {
1412
+ "epoch": 2.0937622086209142,
1413
+ "grad_norm": 0.0003294723283033818,
1414
+ "learning_rate": 7.205660020734955e-06,
1415
+ "loss": 0.0002,
1416
+ "step": 2010
1417
+ },
1418
+ {
1419
+ "epoch": 2.1041802318010157,
1420
+ "grad_norm": 0.02431473508477211,
1421
+ "learning_rate": 7.172932118745978e-06,
1422
+ "loss": 0.0001,
1423
+ "step": 2020
1424
+ },
1425
+ {
1426
+ "epoch": 2.1145982549811175,
1427
+ "grad_norm": 0.006785488221794367,
1428
+ "learning_rate": 7.140089035348166e-06,
1429
+ "loss": 0.0018,
1430
+ "step": 2030
1431
+ },
1432
+ {
1433
+ "epoch": 2.125016278161219,
1434
+ "grad_norm": 0.06791669875383377,
1435
+ "learning_rate": 7.107132511466872e-06,
1436
+ "loss": 0.0005,
1437
+ "step": 2040
1438
+ },
1439
+ {
1440
+ "epoch": 2.1354343013413204,
1441
+ "grad_norm": 0.0033805356360971928,
1442
+ "learning_rate": 7.074064294040629e-06,
1443
+ "loss": 0.0026,
1444
+ "step": 2050
1445
+ },
1446
+ {
1447
+ "epoch": 2.145852324521422,
1448
+ "grad_norm": 0.0271758995950222,
1449
+ "learning_rate": 7.040886135928554e-06,
1450
+ "loss": 0.0004,
1451
+ "step": 2060
1452
+ },
1453
+ {
1454
+ "epoch": 2.1562703477015237,
1455
+ "grad_norm": 0.07952375710010529,
1456
+ "learning_rate": 7.007599795817426e-06,
1457
+ "loss": 0.0,
1458
+ "step": 2070
1459
+ },
1460
+ {
1461
+ "epoch": 2.166688370881625,
1462
+ "grad_norm": 0.12942220270633698,
1463
+ "learning_rate": 6.9742070381284724e-06,
1464
+ "loss": 0.0006,
1465
+ "step": 2080
1466
+ },
1467
+ {
1468
+ "epoch": 2.1771063940617266,
1469
+ "grad_norm": 5.3233266953611746e-05,
1470
+ "learning_rate": 6.94070963292383e-06,
1471
+ "loss": 0.0001,
1472
+ "step": 2090
1473
+ },
1474
+ {
1475
+ "epoch": 2.1875244172418284,
1476
+ "grad_norm": 1.8653046254257788e-06,
1477
+ "learning_rate": 6.907109355812729e-06,
1478
+ "loss": 0.0014,
1479
+ "step": 2100
1480
+ },
1481
+ {
1482
+ "epoch": 2.19794244042193,
1483
+ "grad_norm": 0.2578833997249603,
1484
+ "learning_rate": 6.873407987857367e-06,
1485
+ "loss": 0.0005,
1486
+ "step": 2110
1487
+ },
1488
+ {
1489
+ "epoch": 2.2083604636020313,
1490
+ "grad_norm": 4.3816239667648915e-06,
1491
+ "learning_rate": 6.8396073154785e-06,
1492
+ "loss": 0.0,
1493
+ "step": 2120
1494
+ },
1495
+ {
1496
+ "epoch": 2.218778486782133,
1497
+ "grad_norm": 0.4164877235889435,
1498
+ "learning_rate": 6.805709130360754e-06,
1499
+ "loss": 0.0017,
1500
+ "step": 2130
1501
+ },
1502
+ {
1503
+ "epoch": 2.2291965099622346,
1504
+ "grad_norm": 7.609223303006729e-06,
1505
+ "learning_rate": 6.771715229357643e-06,
1506
+ "loss": 0.0002,
1507
+ "step": 2140
1508
+ },
1509
+ {
1510
+ "epoch": 2.239614533142336,
1511
+ "grad_norm": 0.17523327469825745,
1512
+ "learning_rate": 6.73762741439633e-06,
1513
+ "loss": 0.0018,
1514
+ "step": 2150
1515
+ },
1516
+ {
1517
+ "epoch": 2.250032556322438,
1518
+ "grad_norm": 9.692726135253906,
1519
+ "learning_rate": 6.7034474923821135e-06,
1520
+ "loss": 0.0034,
1521
+ "step": 2160
1522
+ },
1523
+ {
1524
+ "epoch": 2.2604505795025394,
1525
+ "grad_norm": 3.903836841345765e-05,
1526
+ "learning_rate": 6.669177275102637e-06,
1527
+ "loss": 0.0004,
1528
+ "step": 2170
1529
+ },
1530
+ {
1531
+ "epoch": 2.270868602682641,
1532
+ "grad_norm": 0.26724299788475037,
1533
+ "learning_rate": 6.634818579131865e-06,
1534
+ "loss": 0.0007,
1535
+ "step": 2180
1536
+ },
1537
+ {
1538
+ "epoch": 2.2812866258627427,
1539
+ "grad_norm": 0.002805814379826188,
1540
+ "learning_rate": 6.6003732257337805e-06,
1541
+ "loss": 0.0009,
1542
+ "step": 2190
1543
+ },
1544
+ {
1545
+ "epoch": 2.291704649042844,
1546
+ "grad_norm": 0.00600125128403306,
1547
+ "learning_rate": 6.565843040765849e-06,
1548
+ "loss": 0.0003,
1549
+ "step": 2200
1550
+ },
1551
+ {
1552
+ "epoch": 2.3021226722229455,
1553
+ "grad_norm": 0.07256772369146347,
1554
+ "learning_rate": 6.531229854582235e-06,
1555
+ "loss": 0.0002,
1556
+ "step": 2210
1557
+ },
1558
+ {
1559
+ "epoch": 2.3125406954030474,
1560
+ "grad_norm": 0.05456492677330971,
1561
+ "learning_rate": 6.496535501936781e-06,
1562
+ "loss": 0.0,
1563
+ "step": 2220
1564
+ },
1565
+ {
1566
+ "epoch": 2.322958718583149,
1567
+ "grad_norm": 0.002974656643345952,
1568
+ "learning_rate": 6.461761821885742e-06,
1569
+ "loss": 0.0008,
1570
+ "step": 2230
1571
+ },
1572
+ {
1573
+ "epoch": 2.3333767417632503,
1574
+ "grad_norm": 0.32170239090919495,
1575
+ "learning_rate": 6.42691065769032e-06,
1576
+ "loss": 0.0002,
1577
+ "step": 2240
1578
+ },
1579
+ {
1580
+ "epoch": 2.343794764943352,
1581
+ "grad_norm": 0.0577409565448761,
1582
+ "learning_rate": 6.391983856718941e-06,
1583
+ "loss": 0.0138,
1584
+ "step": 2250
1585
+ },
1586
+ {
1587
+ "epoch": 2.3542127881234536,
1588
+ "grad_norm": 0.5012153387069702,
1589
+ "learning_rate": 6.356983270349339e-06,
1590
+ "loss": 0.0047,
1591
+ "step": 2260
1592
+ },
1593
+ {
1594
+ "epoch": 2.364630811303555,
1595
+ "grad_norm": 0.0003676996275316924,
1596
+ "learning_rate": 6.321910753870416e-06,
1597
+ "loss": 0.0003,
1598
+ "step": 2270
1599
+ },
1600
+ {
1601
+ "epoch": 2.375048834483657,
1602
+ "grad_norm": 0.14126235246658325,
1603
+ "learning_rate": 6.286768166383905e-06,
1604
+ "loss": 0.0001,
1605
+ "step": 2280
1606
+ },
1607
+ {
1608
+ "epoch": 2.3854668576637583,
1609
+ "grad_norm": 0.0022616020869463682,
1610
+ "learning_rate": 6.251557370705809e-06,
1611
+ "loss": 0.0002,
1612
+ "step": 2290
1613
+ },
1614
+ {
1615
+ "epoch": 2.3958848808438598,
1616
+ "grad_norm": 6.679229736328125,
1617
+ "learning_rate": 6.216280233267681e-06,
1618
+ "loss": 0.0135,
1619
+ "step": 2300
1620
+ },
1621
+ {
1622
+ "epoch": 2.4063029040239616,
1623
+ "grad_norm": 0.006057079881429672,
1624
+ "learning_rate": 6.180938624017663e-06,
1625
+ "loss": 0.0067,
1626
+ "step": 2310
1627
+ },
1628
+ {
1629
+ "epoch": 2.416720927204063,
1630
+ "grad_norm": 0.058568984270095825,
1631
+ "learning_rate": 6.145534416321384e-06,
1632
+ "loss": 0.0002,
1633
+ "step": 2320
1634
+ },
1635
+ {
1636
+ "epoch": 2.4271389503841645,
1637
+ "grad_norm": 0.001173496013507247,
1638
+ "learning_rate": 6.110069486862652e-06,
1639
+ "loss": 0.0083,
1640
+ "step": 2330
1641
+ },
1642
+ {
1643
+ "epoch": 2.4375569735642664,
1644
+ "grad_norm": 7.219270706176758,
1645
+ "learning_rate": 6.074545715543975e-06,
1646
+ "loss": 0.0173,
1647
+ "step": 2340
1648
+ },
1649
+ {
1650
+ "epoch": 2.447974996744368,
1651
+ "grad_norm": 0.1928091198205948,
1652
+ "learning_rate": 6.0389649853869116e-06,
1653
+ "loss": 0.0001,
1654
+ "step": 2350
1655
+ },
1656
+ {
1657
+ "epoch": 2.4583930199244692,
1658
+ "grad_norm": 0.2435365468263626,
1659
+ "learning_rate": 6.00332918243226e-06,
1660
+ "loss": 0.0003,
1661
+ "step": 2360
1662
+ },
1663
+ {
1664
+ "epoch": 2.468811043104571,
1665
+ "grad_norm": 1.0856934785842896,
1666
+ "learning_rate": 5.967640195640083e-06,
1667
+ "loss": 0.0016,
1668
+ "step": 2370
1669
+ },
1670
+ {
1671
+ "epoch": 2.4792290662846725,
1672
+ "grad_norm": 0.0007231564377434552,
1673
+ "learning_rate": 5.931899916789576e-06,
1674
+ "loss": 0.001,
1675
+ "step": 2380
1676
+ },
1677
+ {
1678
+ "epoch": 2.489647089464774,
1679
+ "grad_norm": 0.0067249564453959465,
1680
+ "learning_rate": 5.896110240378794e-06,
1681
+ "loss": 0.001,
1682
+ "step": 2390
1683
+ },
1684
+ {
1685
+ "epoch": 2.500065112644876,
1686
+ "grad_norm": 0.00024194463912863284,
1687
+ "learning_rate": 5.860273063524227e-06,
1688
+ "loss": 0.0019,
1689
+ "step": 2400
1690
+ },
1691
+ {
1692
+ "epoch": 2.5104831358249773,
1693
+ "grad_norm": 0.011660280637443066,
1694
+ "learning_rate": 5.824390285860233e-06,
1695
+ "loss": 0.0,
1696
+ "step": 2410
1697
+ },
1698
+ {
1699
+ "epoch": 2.5209011590050787,
1700
+ "grad_norm": 7.278788689291105e-05,
1701
+ "learning_rate": 5.788463809438356e-06,
1702
+ "loss": 0.0,
1703
+ "step": 2420
1704
+ },
1705
+ {
1706
+ "epoch": 2.5313191821851806,
1707
+ "grad_norm": 0.0003818434779532254,
1708
+ "learning_rate": 5.752495538626493e-06,
1709
+ "loss": 0.0042,
1710
+ "step": 2430
1711
+ },
1712
+ {
1713
+ "epoch": 2.541737205365282,
1714
+ "grad_norm": 4.750408172607422,
1715
+ "learning_rate": 5.7164873800079475e-06,
1716
+ "loss": 0.009,
1717
+ "step": 2440
1718
+ },
1719
+ {
1720
+ "epoch": 2.5521552285453835,
1721
+ "grad_norm": 9.329643830824352e-07,
1722
+ "learning_rate": 5.680441242280378e-06,
1723
+ "loss": 0.0067,
1724
+ "step": 2450
1725
+ },
1726
+ {
1727
+ "epoch": 2.5625732517254853,
1728
+ "grad_norm": 3.5650442441692576e-05,
1729
+ "learning_rate": 5.6443590361546095e-06,
1730
+ "loss": 0.0017,
1731
+ "step": 2460
1732
+ },
1733
+ {
1734
+ "epoch": 2.5729912749055868,
1735
+ "grad_norm": 0.030909501016139984,
1736
+ "learning_rate": 5.608242674253362e-06,
1737
+ "loss": 0.0,
1738
+ "step": 2470
1739
+ },
1740
+ {
1741
+ "epoch": 2.583409298085688,
1742
+ "grad_norm": 0.011211927980184555,
1743
+ "learning_rate": 5.57209407100986e-06,
1744
+ "loss": 0.0101,
1745
+ "step": 2480
1746
+ },
1747
+ {
1748
+ "epoch": 2.59382732126579,
1749
+ "grad_norm": 0.0002943580038845539,
1750
+ "learning_rate": 5.535915142566361e-06,
1751
+ "loss": 0.0003,
1752
+ "step": 2490
1753
+ },
1754
+ {
1755
+ "epoch": 2.6042453444458915,
1756
+ "grad_norm": 6.406533066183329e-05,
1757
+ "learning_rate": 5.499707806672575e-06,
1758
+ "loss": 0.0077,
1759
+ "step": 2500
1760
+ },
1761
+ {
1762
+ "epoch": 2.614663367625993,
1763
+ "grad_norm": 0.0030457451939582825,
1764
+ "learning_rate": 5.463473982584023e-06,
1765
+ "loss": 0.0019,
1766
+ "step": 2510
1767
+ },
1768
+ {
1769
+ "epoch": 2.625081390806095,
1770
+ "grad_norm": 0.00026603075093589723,
1771
+ "learning_rate": 5.4272155909602875e-06,
1772
+ "loss": 0.0161,
1773
+ "step": 2520
1774
+ },
1775
+ {
1776
+ "epoch": 2.6354994139861962,
1777
+ "grad_norm": 0.03001578524708748,
1778
+ "learning_rate": 5.3909345537632205e-06,
1779
+ "loss": 0.0153,
1780
+ "step": 2530
1781
+ },
1782
+ {
1783
+ "epoch": 2.6459174371662977,
1784
+ "grad_norm": 0.5489205121994019,
1785
+ "learning_rate": 5.354632794155049e-06,
1786
+ "loss": 0.0008,
1787
+ "step": 2540
1788
+ },
1789
+ {
1790
+ "epoch": 2.656335460346399,
1791
+ "grad_norm": 0.00020121924171689898,
1792
+ "learning_rate": 5.318312236396445e-06,
1793
+ "loss": 0.0025,
1794
+ "step": 2550
1795
+ },
1796
+ {
1797
+ "epoch": 2.666753483526501,
1798
+ "grad_norm": 0.16872790455818176,
1799
+ "learning_rate": 5.281974805744516e-06,
1800
+ "loss": 0.0009,
1801
+ "step": 2560
1802
+ },
1803
+ {
1804
+ "epoch": 2.6771715067066024,
1805
+ "grad_norm": 1.6018518635974033e-06,
1806
+ "learning_rate": 5.245622428350764e-06,
1807
+ "loss": 0.0019,
1808
+ "step": 2570
1809
+ },
1810
+ {
1811
+ "epoch": 2.687589529886704,
1812
+ "grad_norm": 0.15927599370479584,
1813
+ "learning_rate": 5.209257031158972e-06,
1814
+ "loss": 0.0042,
1815
+ "step": 2580
1816
+ },
1817
+ {
1818
+ "epoch": 2.6980075530668057,
1819
+ "grad_norm": 0.002104927320033312,
1820
+ "learning_rate": 5.1728805418030725e-06,
1821
+ "loss": 0.002,
1822
+ "step": 2590
1823
+ },
1824
+ {
1825
+ "epoch": 2.708425576246907,
1826
+ "grad_norm": 0.00043586574611254036,
1827
+ "learning_rate": 5.13649488850496e-06,
1828
+ "loss": 0.011,
1829
+ "step": 2600
1830
+ },
1831
+ {
1832
+ "epoch": 2.7188435994270086,
1833
+ "grad_norm": 0.5086974501609802,
1834
+ "learning_rate": 5.100101999972291e-06,
1835
+ "loss": 0.0004,
1836
+ "step": 2610
1837
+ },
1838
+ {
1839
+ "epoch": 2.72926162260711,
1840
+ "grad_norm": 0.04724502190947533,
1841
+ "learning_rate": 5.063703805296239e-06,
1842
+ "loss": 0.0001,
1843
+ "step": 2620
1844
+ },
1845
+ {
1846
+ "epoch": 2.739679645787212,
1847
+ "grad_norm": 2.4284263417939655e-06,
1848
+ "learning_rate": 5.027302233849243e-06,
1849
+ "loss": 0.0028,
1850
+ "step": 2630
1851
+ },
1852
+ {
1853
+ "epoch": 2.7500976689673133,
1854
+ "grad_norm": 0.24547390639781952,
1855
+ "learning_rate": 4.990899215182735e-06,
1856
+ "loss": 0.0032,
1857
+ "step": 2640
1858
+ },
1859
+ {
1860
+ "epoch": 2.7605156921474148,
1861
+ "grad_norm": 0.001153081888332963,
1862
+ "learning_rate": 4.954496678924861e-06,
1863
+ "loss": 0.0004,
1864
+ "step": 2650
1865
+ },
1866
+ {
1867
+ "epoch": 2.7709337153275166,
1868
+ "grad_norm": 0.005142812617123127,
1869
+ "learning_rate": 4.9180965546781985e-06,
1870
+ "loss": 0.0001,
1871
+ "step": 2660
1872
+ },
1873
+ {
1874
+ "epoch": 2.781351738507618,
1875
+ "grad_norm": 0.0020258312579244375,
1876
+ "learning_rate": 4.8817007719174635e-06,
1877
+ "loss": 0.0014,
1878
+ "step": 2670
1879
+ },
1880
+ {
1881
+ "epoch": 2.7917697616877195,
1882
+ "grad_norm": 0.09598887711763382,
1883
+ "learning_rate": 4.845311259887251e-06,
1884
+ "loss": 0.0017,
1885
+ "step": 2680
1886
+ },
1887
+ {
1888
+ "epoch": 2.8021877848678214,
1889
+ "grad_norm": 5.523069921764545e-05,
1890
+ "learning_rate": 4.808929947499751e-06,
1891
+ "loss": 0.0001,
1892
+ "step": 2690
1893
+ },
1894
+ {
1895
+ "epoch": 2.812605808047923,
1896
+ "grad_norm": 4.3237105273874477e-05,
1897
+ "learning_rate": 4.772558763232521e-06,
1898
+ "loss": 0.0,
1899
+ "step": 2700
1900
+ },
1901
+ {
1902
+ "epoch": 2.8230238312280242,
1903
+ "grad_norm": 0.017797382548451424,
1904
+ "learning_rate": 4.736199635026247e-06,
1905
+ "loss": 0.0004,
1906
+ "step": 2710
1907
+ },
1908
+ {
1909
+ "epoch": 2.833441854408126,
1910
+ "grad_norm": 0.0372743122279644,
1911
+ "learning_rate": 4.699854490182558e-06,
1912
+ "loss": 0.0001,
1913
+ "step": 2720
1914
+ },
1915
+ {
1916
+ "epoch": 2.8438598775882276,
1917
+ "grad_norm": 0.017943061888217926,
1918
+ "learning_rate": 4.663525255261861e-06,
1919
+ "loss": 0.0026,
1920
+ "step": 2730
1921
+ },
1922
+ {
1923
+ "epoch": 2.854277900768329,
1924
+ "grad_norm": 0.0005965960444882512,
1925
+ "learning_rate": 4.627213855981221e-06,
1926
+ "loss": 0.0002,
1927
+ "step": 2740
1928
+ },
1929
+ {
1930
+ "epoch": 2.864695923948431,
1931
+ "grad_norm": 8.808985148789361e-05,
1932
+ "learning_rate": 4.590922217112279e-06,
1933
+ "loss": 0.0005,
1934
+ "step": 2750
1935
+ },
1936
+ {
1937
+ "epoch": 2.8751139471285323,
1938
+ "grad_norm": 0.010554715059697628,
1939
+ "learning_rate": 4.554652262379236e-06,
1940
+ "loss": 0.0031,
1941
+ "step": 2760
1942
+ },
1943
+ {
1944
+ "epoch": 2.8855319703086337,
1945
+ "grad_norm": 0.030894100666046143,
1946
+ "learning_rate": 4.518405914356865e-06,
1947
+ "loss": 0.0002,
1948
+ "step": 2770
1949
+ },
1950
+ {
1951
+ "epoch": 2.8959499934887356,
1952
+ "grad_norm": 0.13782520592212677,
1953
+ "learning_rate": 4.4821850943686164e-06,
1954
+ "loss": 0.0003,
1955
+ "step": 2780
1956
+ },
1957
+ {
1958
+ "epoch": 2.906368016668837,
1959
+ "grad_norm": 0.0024414719082415104,
1960
+ "learning_rate": 4.445991722384763e-06,
1961
+ "loss": 0.0002,
1962
+ "step": 2790
1963
+ },
1964
+ {
1965
+ "epoch": 2.9167860398489385,
1966
+ "grad_norm": 0.0004072992014698684,
1967
+ "learning_rate": 4.4098277169206315e-06,
1968
+ "loss": 0.0001,
1969
+ "step": 2800
1970
+ },
1971
+ {
1972
+ "epoch": 2.9272040630290403,
1973
+ "grad_norm": 0.0007056115427985787,
1974
+ "learning_rate": 4.373694994934911e-06,
1975
+ "loss": 0.0194,
1976
+ "step": 2810
1977
+ },
1978
+ {
1979
+ "epoch": 2.937622086209142,
1980
+ "grad_norm": 0.005580561701208353,
1981
+ "learning_rate": 4.337595471728029e-06,
1982
+ "loss": 0.0004,
1983
+ "step": 2820
1984
+ },
1985
+ {
1986
+ "epoch": 2.948040109389243,
1987
+ "grad_norm": 1.4271708726882935,
1988
+ "learning_rate": 4.30153106084064e-06,
1989
+ "loss": 0.0005,
1990
+ "step": 2830
1991
+ },
1992
+ {
1993
+ "epoch": 2.958458132569345,
1994
+ "grad_norm": 0.6616150736808777,
1995
+ "learning_rate": 4.2655036739521795e-06,
1996
+ "loss": 0.0016,
1997
+ "step": 2840
1998
+ },
1999
+ {
2000
+ "epoch": 2.9688761557494465,
2001
+ "grad_norm": 0.0038947018329054117,
2002
+ "learning_rate": 4.229515220779545e-06,
2003
+ "loss": 0.0001,
2004
+ "step": 2850
2005
+ },
2006
+ {
2007
+ "epoch": 2.979294178929548,
2008
+ "grad_norm": 0.049666061997413635,
2009
+ "learning_rate": 4.193567608975856e-06,
2010
+ "loss": 0.0002,
2011
+ "step": 2860
2012
+ },
2013
+ {
2014
+ "epoch": 2.98971220210965,
2015
+ "grad_norm": 0.001238349243067205,
2016
+ "learning_rate": 4.1576627440293425e-06,
2017
+ "loss": 0.0006,
2018
+ "step": 2870
2019
+ },
2020
+ {
2021
+ "epoch": 3.0,
2022
+ "grad_norm": 0.0001485509128542617,
2023
+ "learning_rate": 4.121802529162331e-06,
2024
+ "loss": 0.0,
2025
+ "step": 2880
2026
+ },
2027
+ {
2028
+ "epoch": 3.0104180231801014,
2029
+ "grad_norm": 0.0052583469077944756,
2030
+ "learning_rate": 4.08598886523037e-06,
2031
+ "loss": 0.0008,
2032
+ "step": 2890
2033
+ },
2034
+ {
2035
+ "epoch": 3.0208360463602033,
2036
+ "grad_norm": 0.0016122941160574555,
2037
+ "learning_rate": 4.050223650621461e-06,
2038
+ "loss": 0.0003,
2039
+ "step": 2900
2040
+ },
2041
+ {
2042
+ "epoch": 3.0312540695403047,
2043
+ "grad_norm": 3.2575666409684345e-05,
2044
+ "learning_rate": 4.014508781155441e-06,
2045
+ "loss": 0.0003,
2046
+ "step": 2910
2047
+ },
2048
+ {
2049
+ "epoch": 3.041672092720406,
2050
+ "grad_norm": 0.056297626346349716,
2051
+ "learning_rate": 3.978846149983473e-06,
2052
+ "loss": 0.0002,
2053
+ "step": 2920
2054
+ },
2055
+ {
2056
+ "epoch": 3.052090115900508,
2057
+ "grad_norm": 2.323356420674827e-05,
2058
+ "learning_rate": 3.943237647487719e-06,
2059
+ "loss": 0.0014,
2060
+ "step": 2930
2061
+ },
2062
+ {
2063
+ "epoch": 3.0625081390806095,
2064
+ "grad_norm": 0.05314822122454643,
2065
+ "learning_rate": 3.907685161181109e-06,
2066
+ "loss": 0.0001,
2067
+ "step": 2940
2068
+ },
2069
+ {
2070
+ "epoch": 3.072926162260711,
2071
+ "grad_norm": 1.262199878692627,
2072
+ "learning_rate": 3.8721905756073135e-06,
2073
+ "loss": 0.0004,
2074
+ "step": 2950
2075
+ },
2076
+ {
2077
+ "epoch": 3.083344185440813,
2078
+ "grad_norm": 0.012801315635442734,
2079
+ "learning_rate": 3.836755772240829e-06,
2080
+ "loss": 0.0005,
2081
+ "step": 2960
2082
+ },
2083
+ {
2084
+ "epoch": 3.0937622086209142,
2085
+ "grad_norm": 5.073728516435949e-06,
2086
+ "learning_rate": 3.8013826293872623e-06,
2087
+ "loss": 0.0,
2088
+ "step": 2970
2089
+ },
2090
+ {
2091
+ "epoch": 3.1041802318010157,
2092
+ "grad_norm": 0.0036370500456541777,
2093
+ "learning_rate": 3.7660730220837512e-06,
2094
+ "loss": 0.0001,
2095
+ "step": 2980
2096
+ },
2097
+ {
2098
+ "epoch": 3.1145982549811175,
2099
+ "grad_norm": 0.006650363560765982,
2100
+ "learning_rate": 3.7308288219995884e-06,
2101
+ "loss": 0.0001,
2102
+ "step": 2990
2103
+ },
2104
+ {
2105
+ "epoch": 3.125016278161219,
2106
+ "grad_norm": 0.00013867947563994676,
2107
+ "learning_rate": 3.695651897336996e-06,
2108
+ "loss": 0.0,
2109
+ "step": 3000
2110
+ },
2111
+ {
2112
+ "epoch": 3.1354343013413204,
2113
+ "grad_norm": 4.687119508162141e-05,
2114
+ "learning_rate": 3.6605441127321074e-06,
2115
+ "loss": 0.0001,
2116
+ "step": 3010
2117
+ },
2118
+ {
2119
+ "epoch": 3.145852324521422,
2120
+ "grad_norm": 0.0005026389262638986,
2121
+ "learning_rate": 3.6255073291561248e-06,
2122
+ "loss": 0.0,
2123
+ "step": 3020
2124
+ },
2125
+ {
2126
+ "epoch": 3.1562703477015237,
2127
+ "grad_norm": 0.01530701294541359,
2128
+ "learning_rate": 3.5905434038166653e-06,
2129
+ "loss": 0.0001,
2130
+ "step": 3030
2131
+ },
2132
+ {
2133
+ "epoch": 3.166688370881625,
2134
+ "grad_norm": 0.15575547516345978,
2135
+ "learning_rate": 3.555654190059333e-06,
2136
+ "loss": 0.004,
2137
+ "step": 3040
2138
+ },
2139
+ {
2140
+ "epoch": 3.1771063940617266,
2141
+ "grad_norm": 0.0063483756966888905,
2142
+ "learning_rate": 3.520841537269458e-06,
2143
+ "loss": 0.0025,
2144
+ "step": 3050
2145
+ },
2146
+ {
2147
+ "epoch": 3.1875244172418284,
2148
+ "grad_norm": 0.0002552935038693249,
2149
+ "learning_rate": 3.486107290774083e-06,
2150
+ "loss": 0.0004,
2151
+ "step": 3060
2152
+ },
2153
+ {
2154
+ "epoch": 3.19794244042193,
2155
+ "grad_norm": 1.8458106296748156e-06,
2156
+ "learning_rate": 3.4514532917441345e-06,
2157
+ "loss": 0.0004,
2158
+ "step": 3070
2159
+ },
2160
+ {
2161
+ "epoch": 3.2083604636020313,
2162
+ "grad_norm": 3.37355140800355e-06,
2163
+ "learning_rate": 3.4168813770968357e-06,
2164
+ "loss": 0.0001,
2165
+ "step": 3080
2166
+ },
2167
+ {
2168
+ "epoch": 3.218778486782133,
2169
+ "grad_norm": 0.00028720268164761364,
2170
+ "learning_rate": 3.3823933793983298e-06,
2171
+ "loss": 0.0003,
2172
+ "step": 3090
2173
+ },
2174
+ {
2175
+ "epoch": 3.2291965099622346,
2176
+ "grad_norm": 4.652118150261231e-06,
2177
+ "learning_rate": 3.347991126766545e-06,
2178
+ "loss": 0.0,
2179
+ "step": 3100
2180
+ },
2181
+ {
2182
+ "epoch": 3.239614533142336,
2183
+ "grad_norm": 0.0003041624731849879,
2184
+ "learning_rate": 3.3136764427742885e-06,
2185
+ "loss": 0.0001,
2186
+ "step": 3110
2187
+ },
2188
+ {
2189
+ "epoch": 3.250032556322438,
2190
+ "grad_norm": 0.00011942172568524256,
2191
+ "learning_rate": 3.279451146352588e-06,
2192
+ "loss": 0.0001,
2193
+ "step": 3120
2194
+ },
2195
+ {
2196
+ "epoch": 3.2604505795025394,
2197
+ "grad_norm": 0.044789642095565796,
2198
+ "learning_rate": 3.2453170516942657e-06,
2199
+ "loss": 0.0001,
2200
+ "step": 3130
2201
+ },
2202
+ {
2203
+ "epoch": 3.270868602682641,
2204
+ "grad_norm": 0.02395668812096119,
2205
+ "learning_rate": 3.211275968157784e-06,
2206
+ "loss": 0.0001,
2207
+ "step": 3140
2208
+ },
2209
+ {
2210
+ "epoch": 3.2812866258627427,
2211
+ "grad_norm": 0.02200975827872753,
2212
+ "learning_rate": 3.1773297001713266e-06,
2213
+ "loss": 0.0002,
2214
+ "step": 3150
2215
+ },
2216
+ {
2217
+ "epoch": 3.291704649042844,
2218
+ "grad_norm": 0.0002672448754310608,
2219
+ "learning_rate": 3.1434800471371586e-06,
2220
+ "loss": 0.0,
2221
+ "step": 3160
2222
+ },
2223
+ {
2224
+ "epoch": 3.3021226722229455,
2225
+ "grad_norm": 1.4595940228900872e-05,
2226
+ "learning_rate": 3.109728803336234e-06,
2227
+ "loss": 0.0,
2228
+ "step": 3170
2229
+ },
2230
+ {
2231
+ "epoch": 3.3125406954030474,
2232
+ "grad_norm": 0.0023994040675461292,
2233
+ "learning_rate": 3.076077757833103e-06,
2234
+ "loss": 0.0001,
2235
+ "step": 3180
2236
+ },
2237
+ {
2238
+ "epoch": 3.322958718583149,
2239
+ "grad_norm": 0.009379798546433449,
2240
+ "learning_rate": 3.042528694381057e-06,
2241
+ "loss": 0.0,
2242
+ "step": 3190
2243
+ },
2244
+ {
2245
+ "epoch": 3.3333767417632503,
2246
+ "grad_norm": 0.008108630776405334,
2247
+ "learning_rate": 3.0090833913275965e-06,
2248
+ "loss": 0.0001,
2249
+ "step": 3200
2250
+ },
2251
+ {
2252
+ "epoch": 3.343794764943352,
2253
+ "grad_norm": 0.03200116753578186,
2254
+ "learning_rate": 2.9757436215201497e-06,
2255
+ "loss": 0.0,
2256
+ "step": 3210
2257
+ },
2258
+ {
2259
+ "epoch": 3.3542127881234536,
2260
+ "grad_norm": 7.638196279913245e-07,
2261
+ "learning_rate": 2.9425111522121085e-06,
2262
+ "loss": 0.0009,
2263
+ "step": 3220
2264
+ },
2265
+ {
2266
+ "epoch": 3.364630811303555,
2267
+ "grad_norm": 3.822772123385221e-05,
2268
+ "learning_rate": 2.9093877449691523e-06,
2269
+ "loss": 0.0,
2270
+ "step": 3230
2271
+ },
2272
+ {
2273
+ "epoch": 3.375048834483657,
2274
+ "grad_norm": 1.9638941287994385,
2275
+ "learning_rate": 2.8763751555758544e-06,
2276
+ "loss": 0.0011,
2277
+ "step": 3240
2278
+ },
2279
+ {
2280
+ "epoch": 3.3854668576637583,
2281
+ "grad_norm": 8.83774831095252e-08,
2282
+ "learning_rate": 2.843475133942642e-06,
2283
+ "loss": 0.0007,
2284
+ "step": 3250
2285
+ },
2286
+ {
2287
+ "epoch": 3.3958848808438598,
2288
+ "grad_norm": 5.0836853915825486e-05,
2289
+ "learning_rate": 2.810689424013011e-06,
2290
+ "loss": 0.0,
2291
+ "step": 3260
2292
+ },
2293
+ {
2294
+ "epoch": 3.4063029040239616,
2295
+ "grad_norm": 3.2860607461770996e-05,
2296
+ "learning_rate": 2.7780197636710993e-06,
2297
+ "loss": 0.0,
2298
+ "step": 3270
2299
+ },
2300
+ {
2301
+ "epoch": 3.416720927204063,
2302
+ "grad_norm": 0.1506471037864685,
2303
+ "learning_rate": 2.7454678846495593e-06,
2304
+ "loss": 0.0002,
2305
+ "step": 3280
2306
+ },
2307
+ {
2308
+ "epoch": 3.4271389503841645,
2309
+ "grad_norm": 0.0003242450475227088,
2310
+ "learning_rate": 2.7130355124377684e-06,
2311
+ "loss": 0.0001,
2312
+ "step": 3290
2313
+ },
2314
+ {
2315
+ "epoch": 3.4375569735642664,
2316
+ "grad_norm": 4.1059661270992365e-06,
2317
+ "learning_rate": 2.6807243661903597e-06,
2318
+ "loss": 0.0,
2319
+ "step": 3300
2320
+ },
2321
+ {
2322
+ "epoch": 3.447974996744368,
2323
+ "grad_norm": 0.007434395141899586,
2324
+ "learning_rate": 2.6485361586361012e-06,
2325
+ "loss": 0.0001,
2326
+ "step": 3310
2327
+ },
2328
+ {
2329
+ "epoch": 3.4583930199244692,
2330
+ "grad_norm": 0.2851642966270447,
2331
+ "learning_rate": 2.6164725959871005e-06,
2332
+ "loss": 0.0008,
2333
+ "step": 3320
2334
+ },
2335
+ {
2336
+ "epoch": 3.468811043104571,
2337
+ "grad_norm": 0.001458955928683281,
2338
+ "learning_rate": 2.5845353778483707e-06,
2339
+ "loss": 0.0,
2340
+ "step": 3330
2341
+ },
2342
+ {
2343
+ "epoch": 3.4792290662846725,
2344
+ "grad_norm": 3.6944745716027683e-06,
2345
+ "learning_rate": 2.552726197127732e-06,
2346
+ "loss": 0.0001,
2347
+ "step": 3340
2348
+ },
2349
+ {
2350
+ "epoch": 3.489647089464774,
2351
+ "grad_norm": 0.00017191942606586963,
2352
+ "learning_rate": 2.5210467399460794e-06,
2353
+ "loss": 0.0,
2354
+ "step": 3350
2355
+ },
2356
+ {
2357
+ "epoch": 3.500065112644876,
2358
+ "grad_norm": 8.9817083789967e-05,
2359
+ "learning_rate": 2.489498685548005e-06,
2360
+ "loss": 0.0002,
2361
+ "step": 3360
2362
+ },
2363
+ {
2364
+ "epoch": 3.5104831358249773,
2365
+ "grad_norm": 0.0007733273669146001,
2366
+ "learning_rate": 2.4580837062127867e-06,
2367
+ "loss": 0.0,
2368
+ "step": 3370
2369
+ },
2370
+ {
2371
+ "epoch": 3.5209011590050787,
2372
+ "grad_norm": 0.05479538068175316,
2373
+ "learning_rate": 2.4268034671657412e-06,
2374
+ "loss": 0.0002,
2375
+ "step": 3380
2376
+ },
2377
+ {
2378
+ "epoch": 3.5313191821851806,
2379
+ "grad_norm": 0.02038147673010826,
2380
+ "learning_rate": 2.39565962648996e-06,
2381
+ "loss": 0.0,
2382
+ "step": 3390
2383
+ },
2384
+ {
2385
+ "epoch": 3.541737205365282,
2386
+ "grad_norm": 1.4380566426552832e-06,
2387
+ "learning_rate": 2.364653835038413e-06,
2388
+ "loss": 0.0001,
2389
+ "step": 3400
2390
+ },
2391
+ {
2392
+ "epoch": 3.5521552285453835,
2393
+ "grad_norm": 0.00023164517187979072,
2394
+ "learning_rate": 2.3337877363464475e-06,
2395
+ "loss": 0.0008,
2396
+ "step": 3410
2397
+ },
2398
+ {
2399
+ "epoch": 3.5625732517254853,
2400
+ "grad_norm": 0.00041943430551327765,
2401
+ "learning_rate": 2.3030629665446635e-06,
2402
+ "loss": 0.0001,
2403
+ "step": 3420
2404
+ },
2405
+ {
2406
+ "epoch": 3.5729912749055868,
2407
+ "grad_norm": 5.017073154449463,
2408
+ "learning_rate": 2.272481154272188e-06,
2409
+ "loss": 0.0062,
2410
+ "step": 3430
2411
+ },
2412
+ {
2413
+ "epoch": 3.583409298085688,
2414
+ "grad_norm": 0.13844065368175507,
2415
+ "learning_rate": 2.2420439205903523e-06,
2416
+ "loss": 0.0006,
2417
+ "step": 3440
2418
+ },
2419
+ {
2420
+ "epoch": 3.59382732126579,
2421
+ "grad_norm": 5.108922004699707,
2422
+ "learning_rate": 2.211752878896745e-06,
2423
+ "loss": 0.0131,
2424
+ "step": 3450
2425
+ },
2426
+ {
2427
+ "epoch": 3.6042453444458915,
2428
+ "grad_norm": 0.012104752473533154,
2429
+ "learning_rate": 2.1816096348397176e-06,
2430
+ "loss": 0.0,
2431
+ "step": 3460
2432
+ },
2433
+ {
2434
+ "epoch": 3.614663367625993,
2435
+ "grad_norm": 0.31581804156303406,
2436
+ "learning_rate": 2.1516157862332425e-06,
2437
+ "loss": 0.0002,
2438
+ "step": 3470
2439
+ },
2440
+ {
2441
+ "epoch": 3.625081390806095,
2442
+ "grad_norm": 0.0012989668175578117,
2443
+ "learning_rate": 2.1217729229722485e-06,
2444
+ "loss": 0.0001,
2445
+ "step": 3480
2446
+ },
2447
+ {
2448
+ "epoch": 3.6354994139861962,
2449
+ "grad_norm": 5.394697291194461e-05,
2450
+ "learning_rate": 2.092082626948313e-06,
2451
+ "loss": 0.0001,
2452
+ "step": 3490
2453
+ },
2454
+ {
2455
+ "epoch": 3.6459174371662977,
2456
+ "grad_norm": 0.0002785645192489028,
2457
+ "learning_rate": 2.062546471965841e-06,
2458
+ "loss": 0.0017,
2459
+ "step": 3500
2460
+ },
2461
+ {
2462
+ "epoch": 3.656335460346399,
2463
+ "grad_norm": 0.04909211024641991,
2464
+ "learning_rate": 2.033166023658613e-06,
2465
+ "loss": 0.0055,
2466
+ "step": 3510
2467
+ },
2468
+ {
2469
+ "epoch": 3.666753483526501,
2470
+ "grad_norm": 0.007108056452125311,
2471
+ "learning_rate": 2.0039428394068224e-06,
2472
+ "loss": 0.0,
2473
+ "step": 3520
2474
+ },
2475
+ {
2476
+ "epoch": 3.6771715067066024,
2477
+ "grad_norm": 0.0002842152607627213,
2478
+ "learning_rate": 1.9748784682544955e-06,
2479
+ "loss": 0.0,
2480
+ "step": 3530
2481
+ },
2482
+ {
2483
+ "epoch": 3.687589529886704,
2484
+ "grad_norm": 0.037801098078489304,
2485
+ "learning_rate": 1.9459744508274076e-06,
2486
+ "loss": 0.0014,
2487
+ "step": 3540
2488
+ },
2489
+ {
2490
+ "epoch": 3.6980075530668057,
2491
+ "grad_norm": 0.004841256886720657,
2492
+ "learning_rate": 1.917232319251392e-06,
2493
+ "loss": 0.0008,
2494
+ "step": 3550
2495
+ },
2496
+ {
2497
+ "epoch": 3.708425576246907,
2498
+ "grad_norm": 0.3873371481895447,
2499
+ "learning_rate": 1.8886535970711522e-06,
2500
+ "loss": 0.0062,
2501
+ "step": 3560
2502
+ },
2503
+ {
2504
+ "epoch": 3.7188435994270086,
2505
+ "grad_norm": 1.5684008758398704e-07,
2506
+ "learning_rate": 1.860239799169482e-06,
2507
+ "loss": 0.0001,
2508
+ "step": 3570
2509
+ },
2510
+ {
2511
+ "epoch": 3.72926162260711,
2512
+ "grad_norm": 0.0005140849971212447,
2513
+ "learning_rate": 1.8319924316869763e-06,
2514
+ "loss": 0.0003,
2515
+ "step": 3580
2516
+ },
2517
+ {
2518
+ "epoch": 3.739679645787212,
2519
+ "grad_norm": 8.463976882921997e-06,
2520
+ "learning_rate": 1.8039129919421905e-06,
2521
+ "loss": 0.0036,
2522
+ "step": 3590
2523
+ },
2524
+ {
2525
+ "epoch": 3.7500976689673133,
2526
+ "grad_norm": 0.0017335577867925167,
2527
+ "learning_rate": 1.7760029683522734e-06,
2528
+ "loss": 0.0,
2529
+ "step": 3600
2530
+ },
2531
+ {
2532
+ "epoch": 3.7605156921474148,
2533
+ "grad_norm": 6.008242417010479e-05,
2534
+ "learning_rate": 1.7482638403540703e-06,
2535
+ "loss": 0.0001,
2536
+ "step": 3610
2537
+ },
2538
+ {
2539
+ "epoch": 3.7709337153275166,
2540
+ "grad_norm": 0.0020610829815268517,
2541
+ "learning_rate": 1.720697078325701e-06,
2542
+ "loss": 0.0005,
2543
+ "step": 3620
2544
+ },
2545
+ {
2546
+ "epoch": 3.781351738507618,
2547
+ "grad_norm": 0.009846694767475128,
2548
+ "learning_rate": 1.693304143508618e-06,
2549
+ "loss": 0.0,
2550
+ "step": 3630
2551
+ },
2552
+ {
2553
+ "epoch": 3.7917697616877195,
2554
+ "grad_norm": 0.00021927843044977635,
2555
+ "learning_rate": 1.666086487930153e-06,
2556
+ "loss": 0.0002,
2557
+ "step": 3640
2558
+ },
2559
+ {
2560
+ "epoch": 3.8021877848678214,
2561
+ "grad_norm": 6.052812750567682e-05,
2562
+ "learning_rate": 1.6390455543265454e-06,
2563
+ "loss": 0.0001,
2564
+ "step": 3650
2565
+ },
2566
+ {
2567
+ "epoch": 3.812605808047923,
2568
+ "grad_norm": 0.0021989853121340275,
2569
+ "learning_rate": 1.6121827760664677e-06,
2570
+ "loss": 0.0001,
2571
+ "step": 3660
2572
+ },
2573
+ {
2574
+ "epoch": 3.8230238312280242,
2575
+ "grad_norm": 0.00010219099931418896,
2576
+ "learning_rate": 1.5854995770750526e-06,
2577
+ "loss": 0.0001,
2578
+ "step": 3670
2579
+ },
2580
+ {
2581
+ "epoch": 3.833441854408126,
2582
+ "grad_norm": 0.00012348932796157897,
2583
+ "learning_rate": 1.5589973717583968e-06,
2584
+ "loss": 0.0,
2585
+ "step": 3680
2586
+ },
2587
+ {
2588
+ "epoch": 3.8438598775882276,
2589
+ "grad_norm": 0.008015728555619717,
2590
+ "learning_rate": 1.5326775649286135e-06,
2591
+ "loss": 0.0001,
2592
+ "step": 3690
2593
+ },
2594
+ {
2595
+ "epoch": 3.854277900768329,
2596
+ "grad_norm": 3.6400449516804656e-06,
2597
+ "learning_rate": 1.506541551729338e-06,
2598
+ "loss": 0.0001,
2599
+ "step": 3700
2600
+ },
2601
+ {
2602
+ "epoch": 3.864695923948431,
2603
+ "grad_norm": 0.05682307854294777,
2604
+ "learning_rate": 1.4805907175618034e-06,
2605
+ "loss": 0.0001,
2606
+ "step": 3710
2607
+ },
2608
+ {
2609
+ "epoch": 3.8751139471285323,
2610
+ "grad_norm": 2.3464983314624988e-05,
2611
+ "learning_rate": 1.4548264380113763e-06,
2612
+ "loss": 0.0,
2613
+ "step": 3720
2614
+ },
2615
+ {
2616
+ "epoch": 3.8855319703086337,
2617
+ "grad_norm": 3.6688052205136046e-05,
2618
+ "learning_rate": 1.4292500787746677e-06,
2619
+ "loss": 0.0,
2620
+ "step": 3730
2621
+ },
2622
+ {
2623
+ "epoch": 3.8959499934887356,
2624
+ "grad_norm": 6.737604053341784e-06,
2625
+ "learning_rate": 1.4038629955871147e-06,
2626
+ "loss": 0.0,
2627
+ "step": 3740
2628
+ },
2629
+ {
2630
+ "epoch": 3.906368016668837,
2631
+ "grad_norm": 0.3771498501300812,
2632
+ "learning_rate": 1.3786665341511424e-06,
2633
+ "loss": 0.0002,
2634
+ "step": 3750
2635
+ },
2636
+ {
2637
+ "epoch": 3.9167860398489385,
2638
+ "grad_norm": 0.21408240497112274,
2639
+ "learning_rate": 1.3536620300648062e-06,
2640
+ "loss": 0.0001,
2641
+ "step": 3760
2642
+ },
2643
+ {
2644
+ "epoch": 3.9272040630290403,
2645
+ "grad_norm": 0.020450890064239502,
2646
+ "learning_rate": 1.3288508087510198e-06,
2647
+ "loss": 0.0,
2648
+ "step": 3770
2649
+ },
2650
+ {
2651
+ "epoch": 3.937622086209142,
2652
+ "grad_norm": 0.0006295586354099214,
2653
+ "learning_rate": 1.3042341853872753e-06,
2654
+ "loss": 0.0001,
2655
+ "step": 3780
2656
+ },
2657
+ {
2658
+ "epoch": 3.948040109389243,
2659
+ "grad_norm": 0.018750308081507683,
2660
+ "learning_rate": 1.2798134648359512e-06,
2661
+ "loss": 0.0,
2662
+ "step": 3790
2663
+ },
2664
+ {
2665
+ "epoch": 3.958458132569345,
2666
+ "grad_norm": 0.00036703175283037126,
2667
+ "learning_rate": 1.2555899415751211e-06,
2668
+ "loss": 0.0002,
2669
+ "step": 3800
2670
+ },
2671
+ {
2672
+ "epoch": 3.9688761557494465,
2673
+ "grad_norm": 0.005201701074838638,
2674
+ "learning_rate": 1.2315648996299628e-06,
2675
+ "loss": 0.0,
2676
+ "step": 3810
2677
+ },
2678
+ {
2679
+ "epoch": 3.979294178929548,
2680
+ "grad_norm": 1.7224958355654962e-05,
2681
+ "learning_rate": 1.2077396125046703e-06,
2682
+ "loss": 0.0002,
2683
+ "step": 3820
2684
+ },
2685
+ {
2686
+ "epoch": 3.98971220210965,
2687
+ "grad_norm": 0.0006623083609156311,
2688
+ "learning_rate": 1.1841153431149715e-06,
2689
+ "loss": 0.0,
2690
+ "step": 3830
2691
+ },
2692
+ {
2693
+ "epoch": 4.0,
2694
+ "grad_norm": 0.05106138437986374,
2695
+ "learning_rate": 1.160693343721163e-06,
2696
+ "loss": 0.0,
2697
+ "step": 3840
2698
+ },
2699
+ {
2700
+ "epoch": 4.010418023180102,
2701
+ "grad_norm": 0.13067518174648285,
2702
+ "learning_rate": 1.137474855861751e-06,
2703
+ "loss": 0.0001,
2704
+ "step": 3850
2705
+ },
2706
+ {
2707
+ "epoch": 4.020836046360203,
2708
+ "grad_norm": 0.024557417258620262,
2709
+ "learning_rate": 1.1144611102876251e-06,
2710
+ "loss": 0.0,
2711
+ "step": 3860
2712
+ },
2713
+ {
2714
+ "epoch": 4.031254069540305,
2715
+ "grad_norm": 0.00017925056454259902,
2716
+ "learning_rate": 1.0916533268968293e-06,
2717
+ "loss": 0.0,
2718
+ "step": 3870
2719
+ },
2720
+ {
2721
+ "epoch": 4.041672092720407,
2722
+ "grad_norm": 0.0006559378234669566,
2723
+ "learning_rate": 1.0690527146698915e-06,
2724
+ "loss": 0.0001,
2725
+ "step": 3880
2726
+ },
2727
+ {
2728
+ "epoch": 4.052090115900508,
2729
+ "grad_norm": 0.0031467515509575605,
2730
+ "learning_rate": 1.046660471605744e-06,
2731
+ "loss": 0.0,
2732
+ "step": 3890
2733
+ },
2734
+ {
2735
+ "epoch": 4.0625081390806095,
2736
+ "grad_norm": 1.6517557241968461e-06,
2737
+ "learning_rate": 1.0244777846582172e-06,
2738
+ "loss": 0.0001,
2739
+ "step": 3900
2740
+ },
2741
+ {
2742
+ "epoch": 4.072926162260711,
2743
+ "grad_norm": 3.2786699648568174e-06,
2744
+ "learning_rate": 1.0025058296731254e-06,
2745
+ "loss": 0.0004,
2746
+ "step": 3910
2747
+ },
2748
+ {
2749
+ "epoch": 4.083344185440812,
2750
+ "grad_norm": 2.993703674292192e-05,
2751
+ "learning_rate": 9.807457713259354e-07,
2752
+ "loss": 0.0025,
2753
+ "step": 3920
2754
+ },
2755
+ {
2756
+ "epoch": 4.093762208620914,
2757
+ "grad_norm": 0.00212846789509058,
2758
+ "learning_rate": 9.591987630600313e-07,
2759
+ "loss": 0.0002,
2760
+ "step": 3930
2761
+ },
2762
+ {
2763
+ "epoch": 4.104180231801016,
2764
+ "grad_norm": 0.0008644104236736894,
2765
+ "learning_rate": 9.37865947025573e-07,
2766
+ "loss": 0.0,
2767
+ "step": 3940
2768
+ },
2769
+ {
2770
+ "epoch": 4.114598254981117,
2771
+ "grad_norm": 4.824515417567454e-05,
2772
+ "learning_rate": 9.167484540189558e-07,
2773
+ "loss": 0.0,
2774
+ "step": 3950
2775
+ },
2776
+ {
2777
+ "epoch": 4.125016278161219,
2778
+ "grad_norm": 0.010090484283864498,
2779
+ "learning_rate": 8.958474034228676e-07,
2780
+ "loss": 0.0002,
2781
+ "step": 3960
2782
+ },
2783
+ {
2784
+ "epoch": 4.135434301341321,
2785
+ "grad_norm": 0.0014343432849273086,
2786
+ "learning_rate": 8.751639031469522e-07,
2787
+ "loss": 0.0001,
2788
+ "step": 3970
2789
+ },
2790
+ {
2791
+ "epoch": 4.145852324521422,
2792
+ "grad_norm": 0.0008025519782677293,
2793
+ "learning_rate": 8.546990495690893e-07,
2794
+ "loss": 0.0,
2795
+ "step": 3980
2796
+ },
2797
+ {
2798
+ "epoch": 4.156270347701524,
2799
+ "grad_norm": 0.0038316240534186363,
2800
+ "learning_rate": 8.344539274772657e-07,
2801
+ "loss": 0.0,
2802
+ "step": 3990
2803
+ },
2804
+ {
2805
+ "epoch": 4.166688370881626,
2806
+ "grad_norm": 0.0011604970786720514,
2807
+ "learning_rate": 8.144296100120886e-07,
2808
+ "loss": 0.0001,
2809
+ "step": 4000
2810
+ },
2811
+ {
2812
+ "epoch": 4.177106394061727,
2813
+ "grad_norm": 0.0030226227827370167,
2814
+ "learning_rate": 7.946271586098847e-07,
2815
+ "loss": 0.0001,
2816
+ "step": 4010
2817
+ },
2818
+ {
2819
+ "epoch": 4.1875244172418284,
2820
+ "grad_norm": 0.024601083248853683,
2821
+ "learning_rate": 7.750476229464543e-07,
2822
+ "loss": 0.0,
2823
+ "step": 4020
2824
+ },
2825
+ {
2826
+ "epoch": 4.19794244042193,
2827
+ "grad_norm": 0.00018696175538934767,
2828
+ "learning_rate": 7.556920408814133e-07,
2829
+ "loss": 0.0,
2830
+ "step": 4030
2831
+ },
2832
+ {
2833
+ "epoch": 4.208360463602031,
2834
+ "grad_norm": 2.6345056539867073e-05,
2835
+ "learning_rate": 7.365614384031944e-07,
2836
+ "loss": 0.0002,
2837
+ "step": 4040
2838
+ },
2839
+ {
2840
+ "epoch": 4.218778486782133,
2841
+ "grad_norm": 0.00012209890701342374,
2842
+ "learning_rate": 7.176568295746467e-07,
2843
+ "loss": 0.0,
2844
+ "step": 4050
2845
+ },
2846
+ {
2847
+ "epoch": 4.229196509962235,
2848
+ "grad_norm": 0.0035056450869888067,
2849
+ "learning_rate": 6.989792164793003e-07,
2850
+ "loss": 0.0,
2851
+ "step": 4060
2852
+ },
2853
+ {
2854
+ "epoch": 4.239614533142336,
2855
+ "grad_norm": 0.00014033827756065875,
2856
+ "learning_rate": 6.805295891682323e-07,
2857
+ "loss": 0.0,
2858
+ "step": 4070
2859
+ },
2860
+ {
2861
+ "epoch": 4.250032556322438,
2862
+ "grad_norm": 8.16184274299303e-06,
2863
+ "learning_rate": 6.62308925607597e-07,
2864
+ "loss": 0.0,
2865
+ "step": 4080
2866
+ },
2867
+ {
2868
+ "epoch": 4.26045057950254,
2869
+ "grad_norm": 0.0005164833273738623,
2870
+ "learning_rate": 6.443181916267826e-07,
2871
+ "loss": 0.0,
2872
+ "step": 4090
2873
+ },
2874
+ {
2875
+ "epoch": 4.270868602682641,
2876
+ "grad_norm": 0.01795101910829544,
2877
+ "learning_rate": 6.26558340867216e-07,
2878
+ "loss": 0.0,
2879
+ "step": 4100
2880
+ },
2881
+ {
2882
+ "epoch": 4.281286625862743,
2883
+ "grad_norm": 0.01263391226530075,
2884
+ "learning_rate": 6.090303147318122e-07,
2885
+ "loss": 0.0,
2886
+ "step": 4110
2887
+ },
2888
+ {
2889
+ "epoch": 4.291704649042844,
2890
+ "grad_norm": 0.0007271830691024661,
2891
+ "learning_rate": 5.917350423350731e-07,
2892
+ "loss": 0.0,
2893
+ "step": 4120
2894
+ },
2895
+ {
2896
+ "epoch": 4.3021226722229455,
2897
+ "grad_norm": 5.166072696738411e-06,
2898
+ "learning_rate": 5.74673440453839e-07,
2899
+ "loss": 0.0001,
2900
+ "step": 4130
2901
+ },
2902
+ {
2903
+ "epoch": 4.312540695403047,
2904
+ "grad_norm": 0.005551379173994064,
2905
+ "learning_rate": 5.578464134786916e-07,
2906
+ "loss": 0.0,
2907
+ "step": 4140
2908
+ },
2909
+ {
2910
+ "epoch": 4.322958718583148,
2911
+ "grad_norm": 0.0028216512873768806,
2912
+ "learning_rate": 5.412548533660133e-07,
2913
+ "loss": 0.0,
2914
+ "step": 4150
2915
+ },
2916
+ {
2917
+ "epoch": 4.33337674176325,
2918
+ "grad_norm": 0.0007700317073613405,
2919
+ "learning_rate": 5.248996395907091e-07,
2920
+ "loss": 0.0,
2921
+ "step": 4160
2922
+ },
2923
+ {
2924
+ "epoch": 4.343794764943352,
2925
+ "grad_norm": 0.01867193728685379,
2926
+ "learning_rate": 5.08781639099587e-07,
2927
+ "loss": 0.0,
2928
+ "step": 4170
2929
+ },
2930
+ {
2931
+ "epoch": 4.354212788123453,
2932
+ "grad_norm": 0.0008056411170400679,
2933
+ "learning_rate": 4.92901706265404e-07,
2934
+ "loss": 0.0,
2935
+ "step": 4180
2936
+ },
2937
+ {
2938
+ "epoch": 4.364630811303555,
2939
+ "grad_norm": 0.008770433254539967,
2940
+ "learning_rate": 4.772606828415772e-07,
2941
+ "loss": 0.001,
2942
+ "step": 4190
2943
+ },
2944
+ {
2945
+ "epoch": 4.375048834483657,
2946
+ "grad_norm": 0.005740197841078043,
2947
+ "learning_rate": 4.6185939791756397e-07,
2948
+ "loss": 0.0,
2949
+ "step": 4200
2950
+ },
2951
+ {
2952
+ "epoch": 4.385466857663758,
2953
+ "grad_norm": 0.03286215290427208,
2954
+ "learning_rate": 4.46698667874918e-07,
2955
+ "loss": 0.0025,
2956
+ "step": 4210
2957
+ },
2958
+ {
2959
+ "epoch": 4.39588488084386,
2960
+ "grad_norm": 0.000286923284875229,
2961
+ "learning_rate": 4.3177929634400926e-07,
2962
+ "loss": 0.0,
2963
+ "step": 4220
2964
+ },
2965
+ {
2966
+ "epoch": 4.406302904023962,
2967
+ "grad_norm": 0.00014024133270140737,
2968
+ "learning_rate": 4.1710207416143066e-07,
2969
+ "loss": 0.0,
2970
+ "step": 4230
2971
+ },
2972
+ {
2973
+ "epoch": 4.416720927204063,
2974
+ "grad_norm": 0.004977494943886995,
2975
+ "learning_rate": 4.026677793280748e-07,
2976
+ "loss": 0.0001,
2977
+ "step": 4240
2978
+ },
2979
+ {
2980
+ "epoch": 4.4271389503841645,
2981
+ "grad_norm": 6.539112655445933e-05,
2982
+ "learning_rate": 3.884771769678952e-07,
2983
+ "loss": 0.0002,
2984
+ "step": 4250
2985
+ },
2986
+ {
2987
+ "epoch": 4.437556973564266,
2988
+ "grad_norm": 0.009294900111854076,
2989
+ "learning_rate": 3.745310192873508e-07,
2990
+ "loss": 0.0,
2991
+ "step": 4260
2992
+ },
2993
+ {
2994
+ "epoch": 4.447974996744367,
2995
+ "grad_norm": 5.570728262682678e-06,
2996
+ "learning_rate": 3.6083004553552905e-07,
2997
+ "loss": 0.0001,
2998
+ "step": 4270
2999
+ },
3000
+ {
3001
+ "epoch": 4.458393019924469,
3002
+ "grad_norm": 0.0003835258539766073,
3003
+ "learning_rate": 3.4737498196496686e-07,
3004
+ "loss": 0.0,
3005
+ "step": 4280
3006
+ },
3007
+ {
3008
+ "epoch": 4.468811043104571,
3009
+ "grad_norm": 2.6702812405687837e-08,
3010
+ "learning_rate": 3.3416654179314665e-07,
3011
+ "loss": 0.0,
3012
+ "step": 4290
3013
+ },
3014
+ {
3015
+ "epoch": 4.479229066284672,
3016
+ "grad_norm": 0.00016556118498556316,
3017
+ "learning_rate": 3.212054251646962e-07,
3018
+ "loss": 0.0,
3019
+ "step": 4300
3020
+ },
3021
+ {
3022
+ "epoch": 4.489647089464774,
3023
+ "grad_norm": 3.6678444303106517e-05,
3024
+ "learning_rate": 3.0849231911427134e-07,
3025
+ "loss": 0.0,
3026
+ "step": 4310
3027
+ },
3028
+ {
3029
+ "epoch": 4.500065112644876,
3030
+ "grad_norm": 0.00038032321026548743,
3031
+ "learning_rate": 2.960278975301428e-07,
3032
+ "loss": 0.0,
3033
+ "step": 4320
3034
+ },
3035
+ {
3036
+ "epoch": 4.510483135824977,
3037
+ "grad_norm": 0.00020073131599929184,
3038
+ "learning_rate": 2.8381282111847097e-07,
3039
+ "loss": 0.0,
3040
+ "step": 4330
3041
+ },
3042
+ {
3043
+ "epoch": 4.520901159005079,
3044
+ "grad_norm": 4.448674735613167e-05,
3045
+ "learning_rate": 2.718477373682865e-07,
3046
+ "loss": 0.0001,
3047
+ "step": 4340
3048
+ },
3049
+ {
3050
+ "epoch": 4.531319182185181,
3051
+ "grad_norm": 0.17879758775234222,
3052
+ "learning_rate": 2.601332805171669e-07,
3053
+ "loss": 0.0001,
3054
+ "step": 4350
3055
+ },
3056
+ {
3057
+ "epoch": 4.541737205365282,
3058
+ "grad_norm": 0.0016016721492633224,
3059
+ "learning_rate": 2.4867007151761766e-07,
3060
+ "loss": 0.0002,
3061
+ "step": 4360
3062
+ },
3063
+ {
3064
+ "epoch": 4.5521552285453835,
3065
+ "grad_norm": 0.011347103863954544,
3066
+ "learning_rate": 2.3745871800415765e-07,
3067
+ "loss": 0.0049,
3068
+ "step": 4370
3069
+ },
3070
+ {
3071
+ "epoch": 4.562573251725485,
3072
+ "grad_norm": 0.000959846016485244,
3073
+ "learning_rate": 2.2649981426110967e-07,
3074
+ "loss": 0.0002,
3075
+ "step": 4380
3076
+ },
3077
+ {
3078
+ "epoch": 4.572991274905586,
3079
+ "grad_norm": 0.00030386244179680943,
3080
+ "learning_rate": 2.1579394119109863e-07,
3081
+ "loss": 0.0,
3082
+ "step": 4390
3083
+ },
3084
+ {
3085
+ "epoch": 4.583409298085688,
3086
+ "grad_norm": 0.000289548363070935,
3087
+ "learning_rate": 2.053416662842611e-07,
3088
+ "loss": 0.014,
3089
+ "step": 4400
3090
+ },
3091
+ {
3092
+ "epoch": 4.59382732126579,
3093
+ "grad_norm": 0.009739086031913757,
3094
+ "learning_rate": 1.9514354358816102e-07,
3095
+ "loss": 0.0005,
3096
+ "step": 4410
3097
+ },
3098
+ {
3099
+ "epoch": 4.604245344445891,
3100
+ "grad_norm": 0.0015085875056684017,
3101
+ "learning_rate": 1.8520011367842538e-07,
3102
+ "loss": 0.0,
3103
+ "step": 4420
3104
+ },
3105
+ {
3106
+ "epoch": 4.614663367625993,
3107
+ "grad_norm": 8.586205694882665e-06,
3108
+ "learning_rate": 1.7551190363008453e-07,
3109
+ "loss": 0.0001,
3110
+ "step": 4430
3111
+ },
3112
+ {
3113
+ "epoch": 4.625081390806095,
3114
+ "grad_norm": 5.23238668392878e-07,
3115
+ "learning_rate": 1.6607942698963875e-07,
3116
+ "loss": 0.0,
3117
+ "step": 4440
3118
+ },
3119
+ {
3120
+ "epoch": 4.635499413986196,
3121
+ "grad_norm": 0.006870610639452934,
3122
+ "learning_rate": 1.5690318374783243e-07,
3123
+ "loss": 0.0,
3124
+ "step": 4450
3125
+ },
3126
+ {
3127
+ "epoch": 4.645917437166298,
3128
+ "grad_norm": 0.005947893485426903,
3129
+ "learning_rate": 1.4798366031315292e-07,
3130
+ "loss": 0.0026,
3131
+ "step": 4460
3132
+ },
3133
+ {
3134
+ "epoch": 4.6563354603464,
3135
+ "grad_norm": 1.6190052747333539e-06,
3136
+ "learning_rate": 1.393213294860457e-07,
3137
+ "loss": 0.0,
3138
+ "step": 4470
3139
+ },
3140
+ {
3141
+ "epoch": 4.6667534835265005,
3142
+ "grad_norm": 0.11864887177944183,
3143
+ "learning_rate": 1.3091665043385383e-07,
3144
+ "loss": 0.0001,
3145
+ "step": 4480
3146
+ },
3147
+ {
3148
+ "epoch": 4.677171506706602,
3149
+ "grad_norm": 0.002785899443551898,
3150
+ "learning_rate": 1.2277006866648122e-07,
3151
+ "loss": 0.0003,
3152
+ "step": 4490
3153
+ },
3154
+ {
3155
+ "epoch": 4.687589529886704,
3156
+ "grad_norm": 2.433152758385404e-06,
3157
+ "learning_rate": 1.1488201601276894e-07,
3158
+ "loss": 0.0,
3159
+ "step": 4500
3160
+ },
3161
+ {
3162
+ "epoch": 4.698007553066805,
3163
+ "grad_norm": 0.0006944002816453576,
3164
+ "learning_rate": 1.0725291059761611e-07,
3165
+ "loss": 0.0001,
3166
+ "step": 4510
3167
+ },
3168
+ {
3169
+ "epoch": 4.708425576246907,
3170
+ "grad_norm": 4.231273487675935e-05,
3171
+ "learning_rate": 9.98831568198061e-08,
3172
+ "loss": 0.0,
3173
+ "step": 4520
3174
+ },
3175
+ {
3176
+ "epoch": 4.718843599427009,
3177
+ "grad_norm": 1.5037702496556449e-06,
3178
+ "learning_rate": 9.277314533057913e-08,
3179
+ "loss": 0.0,
3180
+ "step": 4530
3181
+ },
3182
+ {
3183
+ "epoch": 4.72926162260711,
3184
+ "grad_norm": 9.992829745897325e-07,
3185
+ "learning_rate": 8.592325301291782e-08,
3186
+ "loss": 0.0,
3187
+ "step": 4540
3188
+ },
3189
+ {
3190
+ "epoch": 4.739679645787212,
3191
+ "grad_norm": 0.0003942627226933837,
3192
+ "learning_rate": 7.933384296157365e-08,
3193
+ "loss": 0.0,
3194
+ "step": 4550
3195
+ },
3196
+ {
3197
+ "epoch": 4.750097668967314,
3198
+ "grad_norm": 0.09215450286865234,
3199
+ "learning_rate": 7.300526446381906e-08,
3200
+ "loss": 0.0002,
3201
+ "step": 4560
3202
+ },
3203
+ {
3204
+ "epoch": 4.760515692147415,
3205
+ "grad_norm": 2.6581383281154558e-05,
3206
+ "learning_rate": 6.693785298093336e-08,
3207
+ "loss": 0.0,
3208
+ "step": 4570
3209
+ },
3210
+ {
3211
+ "epoch": 4.770933715327517,
3212
+ "grad_norm": 9.696443470375016e-08,
3213
+ "learning_rate": 6.113193013041918e-08,
3214
+ "loss": 0.0,
3215
+ "step": 4580
3216
+ },
3217
+ {
3218
+ "epoch": 4.7813517385076185,
3219
+ "grad_norm": 0.0001541363453725353,
3220
+ "learning_rate": 5.558780366895611e-08,
3221
+ "loss": 0.0,
3222
+ "step": 4590
3223
+ },
3224
+ {
3225
+ "epoch": 4.7917697616877195,
3226
+ "grad_norm": 0.001118164393119514,
3227
+ "learning_rate": 5.0305767476087066e-08,
3228
+ "loss": 0.0001,
3229
+ "step": 4600
3230
+ },
3231
+ {
3232
+ "epoch": 4.802187784867821,
3233
+ "grad_norm": 5.372874511522241e-06,
3234
+ "learning_rate": 4.528610153863855e-08,
3235
+ "loss": 0.0001,
3236
+ "step": 4610
3237
+ },
3238
+ {
3239
+ "epoch": 4.812605808047923,
3240
+ "grad_norm": 4.5697615860262886e-05,
3241
+ "learning_rate": 4.052907193588251e-08,
3242
+ "loss": 0.0,
3243
+ "step": 4620
3244
+ },
3245
+ {
3246
+ "epoch": 4.823023831228024,
3247
+ "grad_norm": 1.0059036867460236e-05,
3248
+ "learning_rate": 3.60349308254293e-08,
3249
+ "loss": 0.0001,
3250
+ "step": 4630
3251
+ },
3252
+ {
3253
+ "epoch": 4.833441854408126,
3254
+ "grad_norm": 0.00015976910071913153,
3255
+ "learning_rate": 3.1803916429863355e-08,
3256
+ "loss": 0.0,
3257
+ "step": 4640
3258
+ },
3259
+ {
3260
+ "epoch": 4.843859877588228,
3261
+ "grad_norm": 1.2175183296203613,
3262
+ "learning_rate": 2.7836253024114412e-08,
3263
+ "loss": 0.0003,
3264
+ "step": 4650
3265
+ },
3266
+ {
3267
+ "epoch": 4.854277900768329,
3268
+ "grad_norm": 0.0008595722611062229,
3269
+ "learning_rate": 2.4132150923570353e-08,
3270
+ "loss": 0.0003,
3271
+ "step": 4660
3272
+ },
3273
+ {
3274
+ "epoch": 4.864695923948431,
3275
+ "grad_norm": 1.9864066871377872e-06,
3276
+ "learning_rate": 2.0691806472928344e-08,
3277
+ "loss": 0.0001,
3278
+ "step": 4670
3279
+ },
3280
+ {
3281
+ "epoch": 4.875113947128533,
3282
+ "grad_norm": 1.0369881238148082e-06,
3283
+ "learning_rate": 1.7515402035787053e-08,
3284
+ "loss": 0.0,
3285
+ "step": 4680
3286
+ },
3287
+ {
3288
+ "epoch": 4.885531970308634,
3289
+ "grad_norm": 0.08182670176029205,
3290
+ "learning_rate": 1.4603105984979382e-08,
3291
+ "loss": 0.0001,
3292
+ "step": 4690
3293
+ },
3294
+ {
3295
+ "epoch": 4.895949993488736,
3296
+ "grad_norm": 1.3140484043105971e-06,
3297
+ "learning_rate": 1.1955072693649594e-08,
3298
+ "loss": 0.0,
3299
+ "step": 4700
3300
+ },
3301
+ {
3302
+ "epoch": 4.9063680166688375,
3303
+ "grad_norm": 0.008761771954596043,
3304
+ "learning_rate": 9.571442527068209e-09,
3305
+ "loss": 0.0,
3306
+ "step": 4710
3307
+ },
3308
+ {
3309
+ "epoch": 4.9167860398489385,
3310
+ "grad_norm": 9.459521243115887e-05,
3311
+ "learning_rate": 7.452341835192388e-09,
3312
+ "loss": 0.0,
3313
+ "step": 4720
3314
+ },
3315
+ {
3316
+ "epoch": 4.92720406302904,
3317
+ "grad_norm": 0.00031195214251056314,
3318
+ "learning_rate": 5.597882945969635e-09,
3319
+ "loss": 0.0001,
3320
+ "step": 4730
3321
+ },
3322
+ {
3323
+ "epoch": 4.937622086209142,
3324
+ "grad_norm": 0.00505842175334692,
3325
+ "learning_rate": 4.008164159380879e-09,
3326
+ "loss": 0.0001,
3327
+ "step": 4740
3328
+ },
3329
+ {
3330
+ "epoch": 4.948040109389243,
3331
+ "grad_norm": 0.028236793354153633,
3332
+ "learning_rate": 2.6832697422324307e-09,
3333
+ "loss": 0.0002,
3334
+ "step": 4750
3335
+ },
3336
+ {
3337
+ "epoch": 4.958458132569345,
3338
+ "grad_norm": 3.2701405871193856e-05,
3339
+ "learning_rate": 1.623269923688442e-09,
3340
+ "loss": 0.0001,
3341
+ "step": 4760
3342
+ },
3343
+ {
3344
+ "epoch": 4.968876155749447,
3345
+ "grad_norm": 0.0009896554984152317,
3346
+ "learning_rate": 8.282208915466605e-10,
3347
+ "loss": 0.0001,
3348
+ "step": 4770
3349
+ },
3350
+ {
3351
+ "epoch": 4.979294178929548,
3352
+ "grad_norm": 4.405805157148279e-06,
3353
+ "learning_rate": 2.981647892635886e-10,
3354
+ "loss": 0.0,
3355
+ "step": 4780
3356
+ },
3357
+ {
3358
+ "epoch": 4.98971220210965,
3359
+ "grad_norm": 2.589948280729004e-06,
3360
+ "learning_rate": 3.312971371627338e-11,
3361
+ "loss": 0.0001,
3362
+ "step": 4790
3363
+ },
3364
+ {
3365
+ "epoch": 4.9949212136997,
3366
+ "step": 4795,
3367
+ "total_flos": 0.0,
3368
+ "train_loss": 0.03522291941339332,
3369
+ "train_runtime": 16429.2929,
3370
+ "train_samples_per_second": 18.696,
3371
+ "train_steps_per_second": 0.292
3372
+ }
3373
+ ],
3374
+ "logging_steps": 10,
3375
+ "max_steps": 4795,
3376
+ "num_input_tokens_seen": 0,
3377
+ "num_train_epochs": 5,
3378
+ "save_steps": 1,
3379
+ "stateful_callbacks": {
3380
+ "TrainerControl": {
3381
+ "args": {
3382
+ "should_epoch_stop": false,
3383
+ "should_evaluate": false,
3384
+ "should_log": false,
3385
+ "should_save": true,
3386
+ "should_training_stop": true
3387
+ },
3388
+ "attributes": {}
3389
+ }
3390
+ },
3391
+ "total_flos": 0.0,
3392
+ "train_batch_size": 2,
3393
+ "trial_name": null,
3394
+ "trial_params": null
3395
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b11b7696084292d630700903a03d62ad08ba25af719f2d22719d20f7e2ddd3fb
3
+ size 5880
training_loss.png ADDED
value_head.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4d8080ced01a59cf132b37fa1c2f855492b1d2928914eb0beb072ee6f494e0a
3
+ size 8388
vocab.json ADDED
The diff for this file is too large to render. See raw diff