Hugging Face's logo

furproxy
/
9b-61

Image-Text-to-Text
Transformers
Safetensors
qwen3_5
llama-factory
full
Generated from Trainer
conversational
Model card Files
xet
 Community
furproxy commited on
Commit
d28b21e
·
verified ·
1 Parent(s): 9eedb60

Upload folder using huggingface_hub

Browse files
Files changed (17) hide show
  1. .gitattributes +1 -0
  2. .ipynb_checkpoints/README-checkpoint.md +64 -0
  3. .ipynb_checkpoints/training_loss-checkpoint.png +0 -0
  4. README.md +64 -0
  5. all_results.json +9 -0
  6. chat_template.jinja +154 -0
  7. config.json +113 -0
  8. generation_config.json +10 -0
  9. model.safetensors +3 -0
  10. processor_config.json +60 -0
  11. tokenizer.json +3 -0
  12. tokenizer_config.json +33 -0
  13. train_results.json +9 -0
  14. trainer_log.jsonl +465 -0
  15. trainer_state.json +3291 -0
  16. training_args.bin +3 -0
  17. training_loss.png +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
.ipynb_checkpoints/README-checkpoint.md ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: other
4
+ base_model: Qwen3.5-9B
5
+ tags:
6
+ - llama-factory
7
+ - full
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: qwen35_caption_galore
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # qwen35_caption_galore
18
+
19
+ This model is a fine-tuned version of [/workspace/models/Qwen3.5-9B](https://huggingface.co//workspace/models/Qwen3.5-9B) on the my_caption dataset.
20
+ small 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
+ - family_to_adamw_lr = {
40
+ "language": _fallback(getattr(training_args, "language_adamw_lr", 4e-5), language_lr),
41
+ "vision": _fallback(getattr(training_args, "vision_adamw_lr", 5e-6), vision_lr),
42
+ "merger": _fallback(getattr(training_args, "merger_adamw_lr", 8e-5), merger_lr),
43
+ }
44
+ - train_batch_size: 1
45
+ - eval_batch_size: 8
46
+ - seed: 42
47
+ - distributed_type: multi-GPU
48
+ - gradient_accumulation_steps: 60
49
+ - total_train_batch_size: 60
50
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
+ - lr_scheduler_type: cosine_with_min_lr
52
+ - lr_scheduler_warmup_steps: 0.05
53
+ - num_epochs: 4
54
+
55
+ ### Training results
56
+
57
+
58
+
59
+ ### Framework versions
60
+
61
+ - Transformers 5.5.3
62
+ - Pytorch 2.11.0+cu128
63
+ - Datasets 4.0.0
64
+ - Tokenizers 0.22.2
.ipynb_checkpoints/training_loss-checkpoint.png ADDED
README.md ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: other
4
+ base_model: Qwen3.5-9B
5
+ tags:
6
+ - llama-factory
7
+ - full
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: qwen35_caption_galore
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # qwen35_caption_galore
18
+
19
+ This model is a fine-tuned version of [/workspace/models/Qwen3.5-9B](https://huggingface.co//workspace/models/Qwen3.5-9B) on the my_caption dataset.
20
+ small 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
+ - family_to_adamw_lr = {
40
+ "language": _fallback(getattr(training_args, "language_adamw_lr", 4e-5), language_lr),
41
+ "vision": _fallback(getattr(training_args, "vision_adamw_lr", 5e-6), vision_lr),
42
+ "merger": _fallback(getattr(training_args, "merger_adamw_lr", 8e-5), merger_lr),
43
+ }
44
+ - train_batch_size: 1
45
+ - eval_batch_size: 8
46
+ - seed: 42
47
+ - distributed_type: multi-GPU
48
+ - gradient_accumulation_steps: 60
49
+ - total_train_batch_size: 60
50
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
+ - lr_scheduler_type: cosine_with_min_lr
52
+ - lr_scheduler_warmup_steps: 0.05
53
+ - num_epochs: 4
54
+
55
+ ### Training results
56
+
57
+
58
+
59
+ ### Framework versions
60
+
61
+ - Transformers 5.5.3
62
+ - Pytorch 2.11.0+cu128
63
+ - Datasets 4.0.0
64
+ - Tokenizers 0.22.2
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "effective_tokens_per_sec": 2197.929095468877,
3
+ "epoch": 4.0,
4
+ "total_flos": 3.61934117404672e+18,
5
+ "train_loss": 0.7229323635552207,
6
+ "train_runtime": 32622.9814,
7
+ "train_samples_per_second": 1.707,
8
+ "train_steps_per_second": 0.028
9
+ }
chat_template.jinja ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- set image_count = namespace(value=0) %}
2
+ {%- set video_count = namespace(value=0) %}
3
+ {%- macro render_content(content, do_vision_count, is_system_content=false) %}
4
+ {%- if content is string %}
5
+ {{- content }}
6
+ {%- elif content is iterable and content is not mapping %}
7
+ {%- for item in content %}
8
+ {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
9
+ {%- if is_system_content %}
10
+ {{- raise_exception('System message cannot contain images.') }}
11
+ {%- endif %}
12
+ {%- if do_vision_count %}
13
+ {%- set image_count.value = image_count.value + 1 %}
14
+ {%- endif %}
15
+ {%- if add_vision_id %}
16
+ {{- 'Picture ' ~ image_count.value ~ ': ' }}
17
+ {%- endif %}
18
+ {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
19
+ {%- elif 'video' in item or item.type == 'video' %}
20
+ {%- if is_system_content %}
21
+ {{- raise_exception('System message cannot contain videos.') }}
22
+ {%- endif %}
23
+ {%- if do_vision_count %}
24
+ {%- set video_count.value = video_count.value + 1 %}
25
+ {%- endif %}
26
+ {%- if add_vision_id %}
27
+ {{- 'Video ' ~ video_count.value ~ ': ' }}
28
+ {%- endif %}
29
+ {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
30
+ {%- elif 'text' in item %}
31
+ {{- item.text }}
32
+ {%- else %}
33
+ {{- raise_exception('Unexpected item type in content.') }}
34
+ {%- endif %}
35
+ {%- endfor %}
36
+ {%- elif content is none or content is undefined %}
37
+ {{- '' }}
38
+ {%- else %}
39
+ {{- raise_exception('Unexpected content type.') }}
40
+ {%- endif %}
41
+ {%- endmacro %}
42
+ {%- if not messages %}
43
+ {{- raise_exception('No messages provided.') }}
44
+ {%- endif %}
45
+ {%- if tools and tools is iterable and tools is not mapping %}
46
+ {{- '<|im_start|>system\n' }}
47
+ {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
48
+ {%- for tool in tools %}
49
+ {{- "\n" }}
50
+ {{- tool | tojson }}
51
+ {%- endfor %}
52
+ {{- "\n</tools>" }}
53
+ {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
54
+ {%- if messages[0].role == 'system' %}
55
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
56
+ {%- if content %}
57
+ {{- '\n\n' + content }}
58
+ {%- endif %}
59
+ {%- endif %}
60
+ {{- '<|im_end|>\n' }}
61
+ {%- else %}
62
+ {%- if messages[0].role == 'system' %}
63
+ {%- set content = render_content(messages[0].content, false, true)|trim %}
64
+ {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
65
+ {%- endif %}
66
+ {%- endif %}
67
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
68
+ {%- for message in messages[::-1] %}
69
+ {%- set index = (messages|length - 1) - loop.index0 %}
70
+ {%- if ns.multi_step_tool and message.role == "user" %}
71
+ {%- set content = render_content(message.content, false)|trim %}
72
+ {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
73
+ {%- set ns.multi_step_tool = false %}
74
+ {%- set ns.last_query_index = index %}
75
+ {%- endif %}
76
+ {%- endif %}
77
+ {%- endfor %}
78
+ {%- if ns.multi_step_tool %}
79
+ {{- raise_exception('No user query found in messages.') }}
80
+ {%- endif %}
81
+ {%- for message in messages %}
82
+ {%- set content = render_content(message.content, true)|trim %}
83
+ {%- if message.role == "system" %}
84
+ {%- if not loop.first %}
85
+ {{- raise_exception('System message must be at the beginning.') }}
86
+ {%- endif %}
87
+ {%- elif message.role == "user" %}
88
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
89
+ {%- elif message.role == "assistant" %}
90
+ {%- set reasoning_content = '' %}
91
+ {%- if message.reasoning_content is string %}
92
+ {%- set reasoning_content = message.reasoning_content %}
93
+ {%- else %}
94
+ {%- if '</think>' in content %}
95
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
96
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
97
+ {%- endif %}
98
+ {%- endif %}
99
+ {%- set reasoning_content = reasoning_content|trim %}
100
+ {%- if loop.index0 > ns.last_query_index %}
101
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
102
+ {%- else %}
103
+ {{- '<|im_start|>' + message.role + '\n' + content }}
104
+ {%- endif %}
105
+ {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
106
+ {%- for tool_call in message.tool_calls %}
107
+ {%- if tool_call.function is defined %}
108
+ {%- set tool_call = tool_call.function %}
109
+ {%- endif %}
110
+ {%- if loop.first %}
111
+ {%- if content|trim %}
112
+ {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
113
+ {%- else %}
114
+ {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
115
+ {%- endif %}
116
+ {%- else %}
117
+ {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
118
+ {%- endif %}
119
+ {%- if tool_call.arguments is defined %}
120
+ {%- for args_name, args_value in tool_call.arguments|items %}
121
+ {{- '<parameter=' + args_name + '>\n' }}
122
+ {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
123
+ {{- args_value }}
124
+ {{- '\n</parameter>\n' }}
125
+ {%- endfor %}
126
+ {%- endif %}
127
+ {{- '</function>\n</tool_call>' }}
128
+ {%- endfor %}
129
+ {%- endif %}
130
+ {{- '<|im_end|>\n' }}
131
+ {%- elif message.role == "tool" %}
132
+ {%- if loop.previtem and loop.previtem.role != "tool" %}
133
+ {{- '<|im_start|>user' }}
134
+ {%- endif %}
135
+ {{- '\n<tool_response>\n' }}
136
+ {{- content }}
137
+ {{- '\n</tool_response>' }}
138
+ {%- if not loop.last and loop.nextitem.role != "tool" %}
139
+ {{- '<|im_end|>\n' }}
140
+ {%- elif loop.last %}
141
+ {{- '<|im_end|>\n' }}
142
+ {%- endif %}
143
+ {%- else %}
144
+ {{- raise_exception('Unexpected message role.') }}
145
+ {%- endif %}
146
+ {%- endfor %}
147
+ {%- if add_generation_prompt %}
148
+ {{- '<|im_start|>assistant\n' }}
149
+ {%- if enable_thinking is defined and enable_thinking is false %}
150
+ {{- '<think>\n\n</think>\n\n' }}
151
+ {%- else %}
152
+ {{- '<think>\n' }}
153
+ {%- endif %}
154
+ {%- endif %}
config.json ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3_5ForConditionalGeneration"
4
+ ],
5
+ "dtype": "bfloat16",
6
+ "eos_token_id": 248046,
7
+ "hidden_size": 4096,
8
+ "image_token_id": 248056,
9
+ "model_type": "qwen3_5",
10
+ "pad_token_id": 248044,
11
+ "text_config": {
12
+ "attention_bias": false,
13
+ "attention_dropout": 0.0,
14
+ "attn_output_gate": true,
15
+ "bos_token_id": null,
16
+ "dtype": "bfloat16",
17
+ "eos_token_id": 248044,
18
+ "full_attention_interval": 4,
19
+ "head_dim": 256,
20
+ "hidden_act": "silu",
21
+ "hidden_size": 4096,
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 12288,
24
+ "layer_types": [
25
+ "linear_attention",
26
+ "linear_attention",
27
+ "linear_attention",
28
+ "full_attention",
29
+ "linear_attention",
30
+ "linear_attention",
31
+ "linear_attention",
32
+ "full_attention",
33
+ "linear_attention",
34
+ "linear_attention",
35
+ "linear_attention",
36
+ "full_attention",
37
+ "linear_attention",
38
+ "linear_attention",
39
+ "linear_attention",
40
+ "full_attention",
41
+ "linear_attention",
42
+ "linear_attention",
43
+ "linear_attention",
44
+ "full_attention",
45
+ "linear_attention",
46
+ "linear_attention",
47
+ "linear_attention",
48
+ "full_attention",
49
+ "linear_attention",
50
+ "linear_attention",
51
+ "linear_attention",
52
+ "full_attention",
53
+ "linear_attention",
54
+ "linear_attention",
55
+ "linear_attention",
56
+ "full_attention"
57
+ ],
58
+ "linear_conv_kernel_dim": 4,
59
+ "linear_key_head_dim": 128,
60
+ "linear_num_key_heads": 16,
61
+ "linear_num_value_heads": 32,
62
+ "linear_value_head_dim": 128,
63
+ "mamba_ssm_dtype": "float32",
64
+ "max_position_embeddings": 262144,
65
+ "mlp_only_layers": [],
66
+ "model_type": "qwen3_5_text",
67
+ "mtp_num_hidden_layers": 1,
68
+ "mtp_use_dedicated_embeddings": false,
69
+ "num_attention_heads": 16,
70
+ "num_hidden_layers": 32,
71
+ "num_key_value_heads": 4,
72
+ "pad_token_id": null,
73
+ "partial_rotary_factor": 0.25,
74
+ "rms_norm_eps": 1e-06,
75
+ "rope_parameters": {
76
+ "mrope_interleaved": true,
77
+ "mrope_section": [
78
+ 11,
79
+ 11,
80
+ 10
81
+ ],
82
+ "partial_rotary_factor": 0.25,
83
+ "rope_theta": 10000000,
84
+ "rope_type": "default"
85
+ },
86
+ "tie_word_embeddings": false,
87
+ "use_cache": false,
88
+ "vocab_size": 248320
89
+ },
90
+ "tie_word_embeddings": false,
91
+ "transformers_version": "5.5.3",
92
+ "use_cache": false,
93
+ "video_token_id": 248057,
94
+ "vision_config": {
95
+ "deepstack_visual_indexes": [],
96
+ "depth": 27,
97
+ "dtype": "bfloat16",
98
+ "hidden_act": "gelu_pytorch_tanh",
99
+ "hidden_size": 1152,
100
+ "in_channels": 3,
101
+ "initializer_range": 0.02,
102
+ "intermediate_size": 4304,
103
+ "model_type": "qwen3_5",
104
+ "num_heads": 16,
105
+ "num_position_embeddings": 2304,
106
+ "out_hidden_size": 4096,
107
+ "patch_size": 16,
108
+ "spatial_merge_size": 2,
109
+ "temporal_patch_size": 2
110
+ },
111
+ "vision_end_token_id": 248054,
112
+ "vision_start_token_id": 248053
113
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "eos_token_id": [
4
+ 248046,
5
+ 248044
6
+ ],
7
+ "pad_token_id": 248044,
8
+ "transformers_version": "5.5.3",
9
+ "use_cache": true
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e755d0b3e2a684e3cb6fcb3dbc1af48847fdb89348fe8ed54592252914e1d58a
3
+ size 20859273368
processor_config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "image_processor": {
3
+ "do_convert_rgb": true,
4
+ "do_normalize": true,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.5,
9
+ 0.5,
10
+ 0.5
11
+ ],
12
+ "image_processor_type": "Qwen2VLImageProcessor",
13
+ "image_std": [
14
+ 0.5,
15
+ 0.5,
16
+ 0.5
17
+ ],
18
+ "merge_size": 2,
19
+ "patch_size": 16,
20
+ "resample": 3,
21
+ "rescale_factor": 0.00392156862745098,
22
+ "size": {
23
+ "longest_edge": 16777216,
24
+ "shortest_edge": 65536
25
+ },
26
+ "temporal_patch_size": 2
27
+ },
28
+ "processor_class": "Qwen3VLProcessor",
29
+ "video_processor": {
30
+ "do_convert_rgb": true,
31
+ "do_normalize": true,
32
+ "do_rescale": true,
33
+ "do_resize": true,
34
+ "do_sample_frames": true,
35
+ "fps": 2,
36
+ "image_mean": [
37
+ 0.5,
38
+ 0.5,
39
+ 0.5
40
+ ],
41
+ "image_std": [
42
+ 0.5,
43
+ 0.5,
44
+ 0.5
45
+ ],
46
+ "max_frames": 768,
47
+ "merge_size": 2,
48
+ "min_frames": 4,
49
+ "patch_size": 16,
50
+ "resample": 3,
51
+ "rescale_factor": 0.00392156862745098,
52
+ "return_metadata": false,
53
+ "size": {
54
+ "longest_edge": 25165824,
55
+ "shortest_edge": 4096
56
+ },
57
+ "temporal_patch_size": 2,
58
+ "video_processor_type": "Qwen3VLVideoProcessor"
59
+ }
60
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87a7830d63fcf43bf241c3c5242e96e62dd3fdc29224ca26fed8ea333db72de4
3
+ size 19989343
tokenizer_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "audio_bos_token": "<|audio_start|>",
4
+ "audio_eos_token": "<|audio_end|>",
5
+ "audio_token": "<|audio_pad|>",
6
+ "backend": "tokenizers",
7
+ "bos_token": null,
8
+ "clean_up_tokenization_spaces": false,
9
+ "eos_token": "<|im_end|>",
10
+ "errors": "replace",
11
+ "image_token": "<|image_pad|>",
12
+ "is_local": true,
13
+ "model_max_length": 262144,
14
+ "model_specific_special_tokens": {
15
+ "audio_bos_token": "<|audio_start|>",
16
+ "audio_eos_token": "<|audio_end|>",
17
+ "audio_token": "<|audio_pad|>",
18
+ "image_token": "<|image_pad|>",
19
+ "video_token": "<|video_pad|>",
20
+ "vision_bos_token": "<|vision_start|>",
21
+ "vision_eos_token": "<|vision_end|>"
22
+ },
23
+ "pad_token": "<|endoftext|>",
24
+ "padding_side": "right",
25
+ "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
26
+ "processor_class": "Qwen3VLProcessor",
27
+ "split_special_tokens": false,
28
+ "tokenizer_class": "TokenizersBackend",
29
+ "unk_token": null,
30
+ "video_token": "<|video_pad|>",
31
+ "vision_bos_token": "<|vision_start|>",
32
+ "vision_eos_token": "<|vision_end|>"
33
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "effective_tokens_per_sec": 2197.929095468877,
3
+ "epoch": 4.0,
4
+ "total_flos": 3.61934117404672e+18,
5
+ "train_loss": 0.7229323635552207,
6
+ "train_runtime": 32622.9814,
7
+ "train_samples_per_second": 1.707,
8
+ "train_steps_per_second": 0.028
9
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,465 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 2, "total_steps": 928, "loss": 3.7051703929901123, "lr": 8.510638297872341e-07, "epoch": 0.008620689655172414, "percentage": 0.22, "elapsed_time": "0:02:20", "remaining_time": "18:01:22"}
2
+ {"current_steps": 4, "total_steps": 928, "loss": 2.5432679653167725, "lr": 2.553191489361702e-06, "epoch": 0.017241379310344827, "percentage": 0.43, "elapsed_time": "0:03:41", "remaining_time": "14:11:05"}
3
+ {"current_steps": 6, "total_steps": 928, "loss": 2.127363681793213, "lr": 4.255319148936171e-06, "epoch": 0.02586206896551724, "percentage": 0.65, "elapsed_time": "0:04:42", "remaining_time": "12:03:57"}
4
+ {"current_steps": 8, "total_steps": 928, "loss": 1.8431488275527954, "lr": 5.957446808510638e-06, "epoch": 0.034482758620689655, "percentage": 0.86, "elapsed_time": "0:06:04", "remaining_time": "11:38:13"}
5
+ {"current_steps": 10, "total_steps": 928, "loss": 1.2361581325531006, "lr": 7.659574468085107e-06, "epoch": 0.04310344827586207, "percentage": 1.08, "elapsed_time": "0:07:06", "remaining_time": "10:52:06"}
6
+ {"current_steps": 12, "total_steps": 928, "loss": 1.7195181846618652, "lr": 9.361702127659576e-06, "epoch": 0.05172413793103448, "percentage": 1.29, "elapsed_time": "0:08:29", "remaining_time": "10:48:12"}
7
+ {"current_steps": 14, "total_steps": 928, "loss": 1.7923121452331543, "lr": 1.1063829787234044e-05, "epoch": 0.0603448275862069, "percentage": 1.51, "elapsed_time": "0:09:49", "remaining_time": "10:40:59"}
8
+ {"current_steps": 16, "total_steps": 928, "loss": 0.7930053472518921, "lr": 1.2765957446808513e-05, "epoch": 0.06896551724137931, "percentage": 1.72, "elapsed_time": "0:10:32", "remaining_time": "10:00:53"}
9
+ {"current_steps": 18, "total_steps": 928, "loss": 0.8174023032188416, "lr": 1.4468085106382981e-05, "epoch": 0.07758620689655173, "percentage": 1.94, "elapsed_time": "0:11:19", "remaining_time": "9:32:08"}
10
+ {"current_steps": 20, "total_steps": 928, "loss": 1.5101033449172974, "lr": 1.6170212765957446e-05, "epoch": 0.08620689655172414, "percentage": 2.16, "elapsed_time": "0:12:45", "remaining_time": "9:39:00"}
11
+ {"current_steps": 22, "total_steps": 928, "loss": 1.6496508121490479, "lr": 1.7872340425531915e-05, "epoch": 0.09482758620689655, "percentage": 2.37, "elapsed_time": "0:13:57", "remaining_time": "9:34:40"}
12
+ {"current_steps": 24, "total_steps": 928, "loss": 1.5436862707138062, "lr": 1.9574468085106384e-05, "epoch": 0.10344827586206896, "percentage": 2.59, "elapsed_time": "0:15:18", "remaining_time": "9:36:34"}
13
+ {"current_steps": 26, "total_steps": 928, "loss": 1.258086919784546, "lr": 2.1276595744680852e-05, "epoch": 0.11206896551724138, "percentage": 2.8, "elapsed_time": "0:16:35", "remaining_time": "9:35:40"}
14
+ {"current_steps": 28, "total_steps": 928, "loss": 0.4653662145137787, "lr": 2.2978723404255324e-05, "epoch": 0.1206896551724138, "percentage": 3.02, "elapsed_time": "0:17:16", "remaining_time": "9:15:29"}
15
+ {"current_steps": 30, "total_steps": 928, "loss": 1.5492061376571655, "lr": 2.468085106382979e-05, "epoch": 0.12931034482758622, "percentage": 3.23, "elapsed_time": "0:18:42", "remaining_time": "9:19:54"}
16
+ {"current_steps": 32, "total_steps": 928, "loss": 1.1802101135253906, "lr": 2.6382978723404255e-05, "epoch": 0.13793103448275862, "percentage": 3.45, "elapsed_time": "0:19:46", "remaining_time": "9:13:41"}
17
+ {"current_steps": 34, "total_steps": 928, "loss": 1.5317002534866333, "lr": 2.8085106382978727e-05, "epoch": 0.14655172413793102, "percentage": 3.66, "elapsed_time": "0:20:57", "remaining_time": "9:10:54"}
18
+ {"current_steps": 36, "total_steps": 928, "loss": 1.4935221672058105, "lr": 2.9787234042553192e-05, "epoch": 0.15517241379310345, "percentage": 3.88, "elapsed_time": "0:22:19", "remaining_time": "9:13:19"}
19
+ {"current_steps": 38, "total_steps": 928, "loss": 1.4348247051239014, "lr": 3.1489361702127664e-05, "epoch": 0.16379310344827586, "percentage": 4.09, "elapsed_time": "0:23:11", "remaining_time": "9:03:03"}
20
+ {"current_steps": 40, "total_steps": 928, "loss": 1.6168015003204346, "lr": 3.319148936170213e-05, "epoch": 0.1724137931034483, "percentage": 4.31, "elapsed_time": "0:24:18", "remaining_time": "8:59:29"}
21
+ {"current_steps": 42, "total_steps": 928, "loss": 1.7296580076217651, "lr": 3.48936170212766e-05, "epoch": 0.1810344827586207, "percentage": 4.53, "elapsed_time": "0:25:38", "remaining_time": "9:01:04"}
22
+ {"current_steps": 44, "total_steps": 928, "loss": 1.0080050230026245, "lr": 3.6595744680851066e-05, "epoch": 0.1896551724137931, "percentage": 4.74, "elapsed_time": "0:26:39", "remaining_time": "8:55:34"}
23
+ {"current_steps": 46, "total_steps": 928, "loss": 1.2960833311080933, "lr": 3.829787234042554e-05, "epoch": 0.19827586206896552, "percentage": 4.96, "elapsed_time": "0:27:42", "remaining_time": "8:51:24"}
24
+ {"current_steps": 48, "total_steps": 928, "loss": 1.463219404220581, "lr": 4e-05, "epoch": 0.20689655172413793, "percentage": 5.17, "elapsed_time": "0:29:03", "remaining_time": "8:52:48"}
25
+ {"current_steps": 50, "total_steps": 928, "loss": 1.5820165872573853, "lr": 3.999954222867108e-05, "epoch": 0.21551724137931033, "percentage": 5.39, "elapsed_time": "0:30:16", "remaining_time": "8:51:42"}
26
+ {"current_steps": 52, "total_steps": 928, "loss": 0.7817078232765198, "lr": 3.999816893796815e-05, "epoch": 0.22413793103448276, "percentage": 5.6, "elapsed_time": "0:30:58", "remaining_time": "8:41:45"}
27
+ {"current_steps": 54, "total_steps": 928, "loss": 1.5489472150802612, "lr": 3.9995880197741576e-05, "epoch": 0.23275862068965517, "percentage": 5.82, "elapsed_time": "0:32:08", "remaining_time": "8:40:18"}
28
+ {"current_steps": 56, "total_steps": 928, "loss": 1.447698950767517, "lr": 3.999267612440463e-05, "epoch": 0.2413793103448276, "percentage": 6.03, "elapsed_time": "0:33:32", "remaining_time": "8:42:23"}
29
+ {"current_steps": 58, "total_steps": 928, "loss": 1.4835537672042847, "lr": 3.9988556880927647e-05, "epoch": 0.25, "percentage": 6.25, "elapsed_time": "0:34:44", "remaining_time": "8:41:10"}
30
+ {"current_steps": 60, "total_steps": 928, "loss": 1.4630577564239502, "lr": 3.998352267682969e-05, "epoch": 0.25862068965517243, "percentage": 6.47, "elapsed_time": "0:36:09", "remaining_time": "8:43:00"}
31
+ {"current_steps": 62, "total_steps": 928, "loss": 1.2140964269638062, "lr": 3.99775737681679e-05, "epoch": 0.2672413793103448, "percentage": 6.68, "elapsed_time": "0:37:00", "remaining_time": "8:36:52"}
32
+ {"current_steps": 64, "total_steps": 928, "loss": 1.4031065702438354, "lr": 3.9970710457524474e-05, "epoch": 0.27586206896551724, "percentage": 6.9, "elapsed_time": "0:38:24", "remaining_time": "8:38:33"}
33
+ {"current_steps": 66, "total_steps": 928, "loss": 1.3692433834075928, "lr": 3.9962933093991296e-05, "epoch": 0.28448275862068967, "percentage": 7.11, "elapsed_time": "0:39:39", "remaining_time": "8:37:59"}
34
+ {"current_steps": 68, "total_steps": 928, "loss": 1.3801196813583374, "lr": 3.995424207315214e-05, "epoch": 0.29310344827586204, "percentage": 7.33, "elapsed_time": "0:41:26", "remaining_time": "8:44:11"}
35
+ {"current_steps": 70, "total_steps": 928, "loss": 1.151434063911438, "lr": 3.994463783706259e-05, "epoch": 0.3017241379310345, "percentage": 7.54, "elapsed_time": "0:42:32", "remaining_time": "8:41:31"}
36
+ {"current_steps": 72, "total_steps": 928, "loss": 1.0419366359710693, "lr": 3.9934120874227505e-05, "epoch": 0.3103448275862069, "percentage": 7.76, "elapsed_time": "0:43:21", "remaining_time": "8:35:31"}
37
+ {"current_steps": 74, "total_steps": 928, "loss": 1.2281261682510376, "lr": 3.992269171957624e-05, "epoch": 0.31896551724137934, "percentage": 7.97, "elapsed_time": "0:44:30", "remaining_time": "8:33:43"}
38
+ {"current_steps": 76, "total_steps": 928, "loss": 1.3645248413085938, "lr": 3.991035095443538e-05, "epoch": 0.3275862068965517, "percentage": 8.19, "elapsed_time": "0:45:20", "remaining_time": "8:28:14"}
39
+ {"current_steps": 78, "total_steps": 928, "loss": 1.2278920412063599, "lr": 3.9897099206499204e-05, "epoch": 0.33620689655172414, "percentage": 8.41, "elapsed_time": "0:46:07", "remaining_time": "8:22:35"}
40
+ {"current_steps": 80, "total_steps": 928, "loss": 1.0160530805587769, "lr": 3.9882937149797735e-05, "epoch": 0.3448275862068966, "percentage": 8.62, "elapsed_time": "0:47:16", "remaining_time": "8:21:10"}
41
+ {"current_steps": 82, "total_steps": 928, "loss": 1.4256402254104614, "lr": 3.986786550466246e-05, "epoch": 0.35344827586206895, "percentage": 8.84, "elapsed_time": "0:48:43", "remaining_time": "8:22:36"}
42
+ {"current_steps": 84, "total_steps": 928, "loss": 1.318834900856018, "lr": 3.985188503768975e-05, "epoch": 0.3620689655172414, "percentage": 9.05, "elapsed_time": "0:49:51", "remaining_time": "8:20:52"}
43
+ {"current_steps": 86, "total_steps": 928, "loss": 1.474804162979126, "lr": 3.983499656170176e-05, "epoch": 0.3706896551724138, "percentage": 9.27, "elapsed_time": "0:51:16", "remaining_time": "8:22:03"}
44
+ {"current_steps": 88, "total_steps": 928, "loss": 1.3328633308410645, "lr": 3.981720093570517e-05, "epoch": 0.3793103448275862, "percentage": 9.48, "elapsed_time": "0:52:44", "remaining_time": "8:23:23"}
45
+ {"current_steps": 90, "total_steps": 928, "loss": 1.1703399419784546, "lr": 3.9798499064847466e-05, "epoch": 0.3879310344827586, "percentage": 9.7, "elapsed_time": "0:53:48", "remaining_time": "8:20:59"}
46
+ {"current_steps": 92, "total_steps": 928, "loss": 1.0334569215774536, "lr": 3.9778891900370905e-05, "epoch": 0.39655172413793105, "percentage": 9.91, "elapsed_time": "0:54:30", "remaining_time": "8:15:21"}
47
+ {"current_steps": 94, "total_steps": 928, "loss": 1.1891283988952637, "lr": 3.9758380439564117e-05, "epoch": 0.4051724137931034, "percentage": 10.13, "elapsed_time": "0:55:41", "remaining_time": "8:14:02"}
48
+ {"current_steps": 96, "total_steps": 928, "loss": 1.4113942384719849, "lr": 3.97369657257114e-05, "epoch": 0.41379310344827586, "percentage": 10.34, "elapsed_time": "0:56:48", "remaining_time": "8:12:16"}
49
+ {"current_steps": 98, "total_steps": 928, "loss": 1.2900447845458984, "lr": 3.9714648848039655e-05, "epoch": 0.4224137931034483, "percentage": 10.56, "elapsed_time": "0:58:09", "remaining_time": "8:12:34"}
50
+ {"current_steps": 100, "total_steps": 928, "loss": 1.270521640777588, "lr": 3.969143094166295e-05, "epoch": 0.43103448275862066, "percentage": 10.78, "elapsed_time": "0:59:13", "remaining_time": "8:10:25"}
51
+ {"current_steps": 102, "total_steps": 928, "loss": 1.2587134838104248, "lr": 3.966731318752484e-05, "epoch": 0.4396551724137931, "percentage": 10.99, "elapsed_time": "1:00:17", "remaining_time": "8:08:18"}
52
+ {"current_steps": 104, "total_steps": 928, "loss": 1.3480840921401978, "lr": 3.964229681233825e-05, "epoch": 0.4482758620689655, "percentage": 11.21, "elapsed_time": "1:01:43", "remaining_time": "8:09:04"}
53
+ {"current_steps": 106, "total_steps": 928, "loss": 1.0994645357131958, "lr": 3.961638308852309e-05, "epoch": 0.45689655172413796, "percentage": 11.42, "elapsed_time": "1:02:48", "remaining_time": "8:07:05"}
54
+ {"current_steps": 108, "total_steps": 928, "loss": 1.233306646347046, "lr": 3.958957333414157e-05, "epoch": 0.46551724137931033, "percentage": 11.64, "elapsed_time": "1:04:12", "remaining_time": "8:07:26"}
55
+ {"current_steps": 110, "total_steps": 928, "loss": 1.2420070171356201, "lr": 3.9561868912831135e-05, "epoch": 0.47413793103448276, "percentage": 11.85, "elapsed_time": "1:05:25", "remaining_time": "8:06:33"}
56
+ {"current_steps": 112, "total_steps": 928, "loss": 1.5031483173370361, "lr": 3.953327123373506e-05, "epoch": 0.4827586206896552, "percentage": 12.07, "elapsed_time": "1:06:42", "remaining_time": "8:05:59"}
57
+ {"current_steps": 114, "total_steps": 928, "loss": 1.2730351686477661, "lr": 3.950378175143088e-05, "epoch": 0.49137931034482757, "percentage": 12.28, "elapsed_time": "1:07:44", "remaining_time": "8:03:45"}
58
+ {"current_steps": 116, "total_steps": 928, "loss": 1.0227445363998413, "lr": 3.947340196585631e-05, "epoch": 0.5, "percentage": 12.5, "elapsed_time": "1:08:48", "remaining_time": "8:01:41"}
59
+ {"current_steps": 118, "total_steps": 928, "loss": 1.3545396327972412, "lr": 3.944213342223299e-05, "epoch": 0.5086206896551724, "percentage": 12.72, "elapsed_time": "1:10:14", "remaining_time": "8:02:09"}
60
+ {"current_steps": 120, "total_steps": 928, "loss": 1.332112431526184, "lr": 3.9409977710987896e-05, "epoch": 0.5172413793103449, "percentage": 12.93, "elapsed_time": "1:11:26", "remaining_time": "8:01:00"}
61
+ {"current_steps": 122, "total_steps": 928, "loss": 1.3639230728149414, "lr": 3.937693646767245e-05, "epoch": 0.5258620689655172, "percentage": 13.15, "elapsed_time": "1:12:52", "remaining_time": "8:01:25"}
62
+ {"current_steps": 124, "total_steps": 928, "loss": 1.5418974161148071, "lr": 3.9343011372879275e-05, "epoch": 0.5344827586206896, "percentage": 13.36, "elapsed_time": "1:14:08", "remaining_time": "8:00:41"}
63
+ {"current_steps": 126, "total_steps": 928, "loss": 1.6416376829147339, "lr": 3.930820415215681e-05, "epoch": 0.5431034482758621, "percentage": 13.58, "elapsed_time": "1:15:31", "remaining_time": "8:00:42"}
64
+ {"current_steps": 128, "total_steps": 928, "loss": 1.1251301765441895, "lr": 3.927251657592146e-05, "epoch": 0.5517241379310345, "percentage": 13.79, "elapsed_time": "1:16:25", "remaining_time": "7:57:39"}
65
+ {"current_steps": 130, "total_steps": 928, "loss": 1.3253697156906128, "lr": 3.923595045936757e-05, "epoch": 0.5603448275862069, "percentage": 14.01, "elapsed_time": "1:17:40", "remaining_time": "7:56:45"}
66
+ {"current_steps": 132, "total_steps": 928, "loss": 1.263968586921692, "lr": 3.919850766237512e-05, "epoch": 0.5689655172413793, "percentage": 14.22, "elapsed_time": "1:19:00", "remaining_time": "7:56:27"}
67
+ {"current_steps": 134, "total_steps": 928, "loss": 0.7878425717353821, "lr": 3.9160190089415106e-05, "epoch": 0.5775862068965517, "percentage": 14.44, "elapsed_time": "1:20:01", "remaining_time": "7:54:08"}
68
+ {"current_steps": 136, "total_steps": 928, "loss": 1.0626349449157715, "lr": 3.912099968945268e-05, "epoch": 0.5862068965517241, "percentage": 14.66, "elapsed_time": "1:21:05", "remaining_time": "7:52:16"}
69
+ {"current_steps": 138, "total_steps": 928, "loss": 1.05846107006073, "lr": 3.908093845584798e-05, "epoch": 0.5948275862068966, "percentage": 14.87, "elapsed_time": "1:21:51", "remaining_time": "7:48:35"}
70
+ {"current_steps": 140, "total_steps": 928, "loss": 1.0136967897415161, "lr": 3.9040008426254824e-05, "epoch": 0.603448275862069, "percentage": 15.09, "elapsed_time": "1:22:53", "remaining_time": "7:46:32"}
71
+ {"current_steps": 142, "total_steps": 928, "loss": 1.0979063510894775, "lr": 3.8998211682516976e-05, "epoch": 0.6120689655172413, "percentage": 15.3, "elapsed_time": "1:23:59", "remaining_time": "7:44:53"}
72
+ {"current_steps": 144, "total_steps": 928, "loss": 1.2308785915374756, "lr": 3.895555035056233e-05, "epoch": 0.6206896551724138, "percentage": 15.52, "elapsed_time": "1:25:04", "remaining_time": "7:43:13"}
73
+ {"current_steps": 146, "total_steps": 928, "loss": 1.3009754419326782, "lr": 3.891202660029474e-05, "epoch": 0.6293103448275862, "percentage": 15.73, "elapsed_time": "1:26:10", "remaining_time": "7:41:34"}
74
+ {"current_steps": 148, "total_steps": 928, "loss": 1.3296358585357666, "lr": 3.886764264548363e-05, "epoch": 0.6379310344827587, "percentage": 15.95, "elapsed_time": "1:27:10", "remaining_time": "7:39:26"}
75
+ {"current_steps": 150, "total_steps": 928, "loss": 1.2398273944854736, "lr": 3.882240074365145e-05, "epoch": 0.646551724137931, "percentage": 16.16, "elapsed_time": "1:28:10", "remaining_time": "7:37:20"}
76
+ {"current_steps": 152, "total_steps": 928, "loss": 0.9495888948440552, "lr": 3.8776303195958814e-05, "epoch": 0.6551724137931034, "percentage": 16.38, "elapsed_time": "1:29:12", "remaining_time": "7:35:27"}
77
+ {"current_steps": 154, "total_steps": 928, "loss": 1.1250660419464111, "lr": 3.872935234708747e-05, "epoch": 0.6637931034482759, "percentage": 16.59, "elapsed_time": "1:30:24", "remaining_time": "7:34:22"}
78
+ {"current_steps": 156, "total_steps": 928, "loss": 1.2095718383789062, "lr": 3.868155058512102e-05, "epoch": 0.6724137931034483, "percentage": 16.81, "elapsed_time": "1:31:38", "remaining_time": "7:33:30"}
79
+ {"current_steps": 158, "total_steps": 928, "loss": 1.295078992843628, "lr": 3.8632900341423464e-05, "epoch": 0.6810344827586207, "percentage": 17.03, "elapsed_time": "1:33:02", "remaining_time": "7:33:24"}
80
+ {"current_steps": 160, "total_steps": 928, "loss": 1.338647723197937, "lr": 3.858340409051558e-05, "epoch": 0.6896551724137931, "percentage": 17.24, "elapsed_time": "1:34:16", "remaining_time": "7:32:31"}
81
+ {"current_steps": 162, "total_steps": 928, "loss": 1.6283466815948486, "lr": 3.853306434994895e-05, "epoch": 0.6982758620689655, "percentage": 17.46, "elapsed_time": "1:35:35", "remaining_time": "7:32:00"}
82
+ {"current_steps": 164, "total_steps": 928, "loss": 1.055685043334961, "lr": 3.848188368017803e-05, "epoch": 0.7068965517241379, "percentage": 17.67, "elapsed_time": "1:36:41", "remaining_time": "7:30:25"}
83
+ {"current_steps": 166, "total_steps": 928, "loss": 1.3115266561508179, "lr": 3.8429864684429846e-05, "epoch": 0.7155172413793104, "percentage": 17.89, "elapsed_time": "1:37:46", "remaining_time": "7:28:47"}
84
+ {"current_steps": 168, "total_steps": 928, "loss": 1.3648704290390015, "lr": 3.837701000857159e-05, "epoch": 0.7241379310344828, "percentage": 18.1, "elapsed_time": "1:39:11", "remaining_time": "7:28:43"}
85
+ {"current_steps": 170, "total_steps": 928, "loss": 1.2736293077468872, "lr": 3.832332234097606e-05, "epoch": 0.7327586206896551, "percentage": 18.32, "elapsed_time": "1:40:37", "remaining_time": "7:28:38"}
86
+ {"current_steps": 172, "total_steps": 928, "loss": 1.2950979471206665, "lr": 3.8268804412384936e-05, "epoch": 0.7413793103448276, "percentage": 18.53, "elapsed_time": "1:41:59", "remaining_time": "7:28:17"}
87
+ {"current_steps": 174, "total_steps": 928, "loss": 0.9605190753936768, "lr": 3.821345899576982e-05, "epoch": 0.75, "percentage": 18.75, "elapsed_time": "1:42:52", "remaining_time": "7:25:46"}
88
+ {"current_steps": 176, "total_steps": 928, "loss": 0.7583910226821899, "lr": 3.815728890619127e-05, "epoch": 0.7586206896551724, "percentage": 18.97, "elapsed_time": "1:43:38", "remaining_time": "7:22:50"}
89
+ {"current_steps": 178, "total_steps": 928, "loss": 1.3208308219909668, "lr": 3.8100297000655566e-05, "epoch": 0.7672413793103449, "percentage": 19.18, "elapsed_time": "1:44:54", "remaining_time": "7:22:00"}
90
+ {"current_steps": 180, "total_steps": 928, "loss": 1.3078787326812744, "lr": 3.804248617796941e-05, "epoch": 0.7758620689655172, "percentage": 19.4, "elapsed_time": "1:46:00", "remaining_time": "7:20:33"}
91
+ {"current_steps": 182, "total_steps": 928, "loss": 1.0150703191757202, "lr": 3.798385937859249e-05, "epoch": 0.7844827586206896, "percentage": 19.61, "elapsed_time": "1:47:03", "remaining_time": "7:18:51"}
92
+ {"current_steps": 184, "total_steps": 928, "loss": 0.8019794225692749, "lr": 3.79244195844879e-05, "epoch": 0.7931034482758621, "percentage": 19.83, "elapsed_time": "1:48:06", "remaining_time": "7:17:07"}
93
+ {"current_steps": 186, "total_steps": 928, "loss": 1.423434853553772, "lr": 3.7864169818970465e-05, "epoch": 0.8017241379310345, "percentage": 20.04, "elapsed_time": "1:49:19", "remaining_time": "7:16:08"}
94
+ {"current_steps": 188, "total_steps": 928, "loss": 1.5573757886886597, "lr": 3.7803113146553e-05, "epoch": 0.8103448275862069, "percentage": 20.26, "elapsed_time": "1:50:32", "remaining_time": "7:15:05"}
95
+ {"current_steps": 190, "total_steps": 928, "loss": 1.2926381826400757, "lr": 3.774125267279041e-05, "epoch": 0.8189655172413793, "percentage": 20.47, "elapsed_time": "1:51:58", "remaining_time": "7:14:56"}
96
+ {"current_steps": 192, "total_steps": 928, "loss": 1.2611286640167236, "lr": 3.767859154412171e-05, "epoch": 0.8275862068965517, "percentage": 20.69, "elapsed_time": "1:53:25", "remaining_time": "7:14:45"}
97
+ {"current_steps": 194, "total_steps": 928, "loss": 1.3005847930908203, "lr": 3.7615132947710036e-05, "epoch": 0.8362068965517241, "percentage": 20.91, "elapsed_time": "1:54:35", "remaining_time": "7:13:35"}
98
+ {"current_steps": 196, "total_steps": 928, "loss": 1.3176685571670532, "lr": 3.755088011128049e-05, "epoch": 0.8448275862068966, "percentage": 21.12, "elapsed_time": "1:55:58", "remaining_time": "7:13:08"}
99
+ {"current_steps": 198, "total_steps": 928, "loss": 1.2926079034805298, "lr": 3.7485836302956016e-05, "epoch": 0.853448275862069, "percentage": 21.34, "elapsed_time": "1:57:02", "remaining_time": "7:11:30"}
100
+ {"current_steps": 200, "total_steps": 928, "loss": 1.290125846862793, "lr": 3.7420004831091105e-05, "epoch": 0.8620689655172413, "percentage": 21.55, "elapsed_time": "1:58:08", "remaining_time": "7:10:02"}
101
+ {"current_steps": 202, "total_steps": 928, "loss": 1.333167552947998, "lr": 3.735338904410358e-05, "epoch": 0.8706896551724138, "percentage": 21.77, "elapsed_time": "1:59:27", "remaining_time": "7:09:20"}
102
+ {"current_steps": 204, "total_steps": 928, "loss": 0.894460916519165, "lr": 3.728599233030425e-05, "epoch": 0.8793103448275862, "percentage": 21.98, "elapsed_time": "2:00:31", "remaining_time": "7:07:46"}
103
+ {"current_steps": 206, "total_steps": 928, "loss": 1.2725472450256348, "lr": 3.72178181177246e-05, "epoch": 0.8879310344827587, "percentage": 22.2, "elapsed_time": "2:01:46", "remaining_time": "7:06:48"}
104
+ {"current_steps": 208, "total_steps": 928, "loss": 1.3160998821258545, "lr": 3.714886987394238e-05, "epoch": 0.896551724137931, "percentage": 22.41, "elapsed_time": "2:03:12", "remaining_time": "7:06:30"}
105
+ {"current_steps": 210, "total_steps": 928, "loss": 0.6522892117500305, "lr": 3.70791511059053e-05, "epoch": 0.9051724137931034, "percentage": 22.63, "elapsed_time": "2:04:13", "remaining_time": "7:04:45"}
106
+ {"current_steps": 212, "total_steps": 928, "loss": 0.9885504245758057, "lr": 3.700866535975256e-05, "epoch": 0.9137931034482759, "percentage": 22.84, "elapsed_time": "2:05:15", "remaining_time": "7:03:01"}
107
+ {"current_steps": 214, "total_steps": 928, "loss": 1.2210713624954224, "lr": 3.69374162206346e-05, "epoch": 0.9224137931034483, "percentage": 23.06, "elapsed_time": "2:06:26", "remaining_time": "7:01:51"}
108
+ {"current_steps": 216, "total_steps": 928, "loss": 1.2765154838562012, "lr": 3.6865407312530635e-05, "epoch": 0.9310344827586207, "percentage": 23.28, "elapsed_time": "2:07:39", "remaining_time": "7:00:47"}
109
+ {"current_steps": 218, "total_steps": 928, "loss": 1.0038096904754639, "lr": 3.67926422980644e-05, "epoch": 0.9396551724137931, "percentage": 23.49, "elapsed_time": "2:08:43", "remaining_time": "6:59:15"}
110
+ {"current_steps": 220, "total_steps": 928, "loss": 0.9949377179145813, "lr": 3.671912487831783e-05, "epoch": 0.9482758620689655, "percentage": 23.71, "elapsed_time": "2:09:44", "remaining_time": "6:57:33"}
111
+ {"current_steps": 222, "total_steps": 928, "loss": 1.3989291191101074, "lr": 3.664485879264279e-05, "epoch": 0.9568965517241379, "percentage": 23.92, "elapsed_time": "2:11:06", "remaining_time": "6:56:58"}
112
+ {"current_steps": 224, "total_steps": 928, "loss": 0.9785476922988892, "lr": 3.656984781847094e-05, "epoch": 0.9655172413793104, "percentage": 24.14, "elapsed_time": "2:12:09", "remaining_time": "6:55:20"}
113
+ {"current_steps": 226, "total_steps": 928, "loss": 0.9384239315986633, "lr": 3.649409577112152e-05, "epoch": 0.9741379310344828, "percentage": 24.35, "elapsed_time": "2:13:32", "remaining_time": "6:54:47"}
114
+ {"current_steps": 228, "total_steps": 928, "loss": 1.2639554738998413, "lr": 3.641760650360736e-05, "epoch": 0.9827586206896551, "percentage": 24.57, "elapsed_time": "2:14:57", "remaining_time": "6:54:21"}
115
+ {"current_steps": 230, "total_steps": 928, "loss": 0.9682251811027527, "lr": 3.634038390643886e-05, "epoch": 0.9913793103448276, "percentage": 24.78, "elapsed_time": "2:15:47", "remaining_time": "6:52:04"}
116
+ {"current_steps": 232, "total_steps": 928, "loss": 0.870396614074707, "lr": 3.626243190742613e-05, "epoch": 1.0, "percentage": 25.0, "elapsed_time": "2:16:49", "remaining_time": "6:50:28"}
117
+ {"current_steps": 234, "total_steps": 928, "loss": 0.9028921723365784, "lr": 3.618375447147918e-05, "epoch": 1.0086206896551724, "percentage": 25.22, "elapsed_time": "2:18:02", "remaining_time": "6:49:24"}
118
+ {"current_steps": 236, "total_steps": 928, "loss": 0.7830209136009216, "lr": 3.6104355600406284e-05, "epoch": 1.0172413793103448, "percentage": 25.43, "elapsed_time": "2:19:06", "remaining_time": "6:47:52"}
119
+ {"current_steps": 238, "total_steps": 928, "loss": 0.8674835562705994, "lr": 3.6024239332710415e-05, "epoch": 1.0258620689655173, "percentage": 25.65, "elapsed_time": "2:20:18", "remaining_time": "6:46:45"}
120
+ {"current_steps": 240, "total_steps": 928, "loss": 0.6074855327606201, "lr": 3.5943409743383826e-05, "epoch": 1.0344827586206897, "percentage": 25.86, "elapsed_time": "2:21:07", "remaining_time": "6:44:32"}
121
+ {"current_steps": 242, "total_steps": 928, "loss": 1.2056649923324585, "lr": 3.586187094370079e-05, "epoch": 1.043103448275862, "percentage": 26.08, "elapsed_time": "2:22:22", "remaining_time": "6:43:36"}
122
+ {"current_steps": 244, "total_steps": 928, "loss": 0.39286842942237854, "lr": 3.577962708100851e-05, "epoch": 1.0517241379310345, "percentage": 26.29, "elapsed_time": "2:23:07", "remaining_time": "6:41:13"}
123
+ {"current_steps": 246, "total_steps": 928, "loss": 0.9662060141563416, "lr": 3.569668233851613e-05, "epoch": 1.0603448275862069, "percentage": 26.51, "elapsed_time": "2:24:31", "remaining_time": "6:40:39"}
124
+ {"current_steps": 248, "total_steps": 928, "loss": 0.9487460851669312, "lr": 3.561304093508198e-05, "epoch": 1.0689655172413792, "percentage": 26.72, "elapsed_time": "2:25:54", "remaining_time": "6:40:05"}
125
+ {"current_steps": 250, "total_steps": 928, "loss": 0.8610017895698547, "lr": 3.552870712499898e-05, "epoch": 1.0775862068965518, "percentage": 26.94, "elapsed_time": "2:27:03", "remaining_time": "6:38:49"}
126
+ {"current_steps": 252, "total_steps": 928, "loss": 0.448039710521698, "lr": 3.54436851977783e-05, "epoch": 1.0862068965517242, "percentage": 27.16, "elapsed_time": "2:27:43", "remaining_time": "6:36:17"}
127
+ {"current_steps": 254, "total_steps": 928, "loss": 0.8761284351348877, "lr": 3.535797947793111e-05, "epoch": 1.0948275862068966, "percentage": 27.37, "elapsed_time": "2:28:57", "remaining_time": "6:35:15"}
128
+ {"current_steps": 256, "total_steps": 928, "loss": 0.8125666975975037, "lr": 3.527159432474865e-05, "epoch": 1.103448275862069, "percentage": 27.59, "elapsed_time": "2:29:58", "remaining_time": "6:33:41"}
129
+ {"current_steps": 258, "total_steps": 928, "loss": 0.6569501757621765, "lr": 3.518453413208053e-05, "epoch": 1.1120689655172413, "percentage": 27.8, "elapsed_time": "2:30:45", "remaining_time": "6:31:30"}
130
+ {"current_steps": 260, "total_steps": 928, "loss": 0.7028253674507141, "lr": 3.509680332811121e-05, "epoch": 1.1206896551724137, "percentage": 28.02, "elapsed_time": "2:31:48", "remaining_time": "6:30:02"}
131
+ {"current_steps": 262, "total_steps": 928, "loss": 1.1632713079452515, "lr": 3.5008406375134756e-05, "epoch": 1.1293103448275863, "percentage": 28.23, "elapsed_time": "2:33:14", "remaining_time": "6:29:32"}
132
+ {"current_steps": 264, "total_steps": 928, "loss": 0.8371788263320923, "lr": 3.491934776932791e-05, "epoch": 1.1379310344827587, "percentage": 28.45, "elapsed_time": "2:34:26", "remaining_time": "6:28:25"}
133
+ {"current_steps": 266, "total_steps": 928, "loss": 1.0003291368484497, "lr": 3.482963204052139e-05, "epoch": 1.146551724137931, "percentage": 28.66, "elapsed_time": "2:35:41", "remaining_time": "6:27:27"}
134
+ {"current_steps": 268, "total_steps": 928, "loss": 0.9226457476615906, "lr": 3.473926375196943e-05, "epoch": 1.1551724137931034, "percentage": 28.88, "elapsed_time": "2:36:53", "remaining_time": "6:26:22"}
135
+ {"current_steps": 270, "total_steps": 928, "loss": 1.1149680614471436, "lr": 3.464824750011779e-05, "epoch": 1.1637931034482758, "percentage": 29.09, "elapsed_time": "2:38:16", "remaining_time": "6:25:42"}
136
+ {"current_steps": 272, "total_steps": 928, "loss": 1.0302170515060425, "lr": 3.455658791436985e-05, "epoch": 1.1724137931034484, "percentage": 29.31, "elapsed_time": "2:39:38", "remaining_time": "6:25:01"}
137
+ {"current_steps": 274, "total_steps": 928, "loss": 0.7569156289100647, "lr": 3.446428965685121e-05, "epoch": 1.1810344827586208, "percentage": 29.53, "elapsed_time": "2:40:39", "remaining_time": "6:23:29"}
138
+ {"current_steps": 276, "total_steps": 928, "loss": 0.3794441223144531, "lr": 3.437135742217254e-05, "epoch": 1.1896551724137931, "percentage": 29.74, "elapsed_time": "2:41:22", "remaining_time": "6:21:13"}
139
+ {"current_steps": 278, "total_steps": 928, "loss": 1.1280944347381592, "lr": 3.427779593719079e-05, "epoch": 1.1982758620689655, "percentage": 29.96, "elapsed_time": "2:42:47", "remaining_time": "6:20:38"}
140
+ {"current_steps": 280, "total_steps": 928, "loss": 0.9517163634300232, "lr": 3.4183609960768764e-05, "epoch": 1.206896551724138, "percentage": 30.17, "elapsed_time": "2:44:07", "remaining_time": "6:19:49"}
141
+ {"current_steps": 282, "total_steps": 928, "loss": 0.6671708822250366, "lr": 3.4088804283533094e-05, "epoch": 1.2155172413793103, "percentage": 30.39, "elapsed_time": "2:45:16", "remaining_time": "6:18:35"}
142
+ {"current_steps": 284, "total_steps": 928, "loss": 0.7694864869117737, "lr": 3.399338372763055e-05, "epoch": 1.2241379310344827, "percentage": 30.6, "elapsed_time": "2:46:20", "remaining_time": "6:17:11"}
143
+ {"current_steps": 286, "total_steps": 928, "loss": 0.8068587183952332, "lr": 3.389735314648274e-05, "epoch": 1.2327586206896552, "percentage": 30.82, "elapsed_time": "2:47:23", "remaining_time": "6:15:45"}
144
+ {"current_steps": 288, "total_steps": 928, "loss": 0.40287792682647705, "lr": 3.380071742453931e-05, "epoch": 1.2413793103448276, "percentage": 31.03, "elapsed_time": "2:48:07", "remaining_time": "6:13:37"}
145
+ {"current_steps": 290, "total_steps": 928, "loss": 1.0401684045791626, "lr": 3.370348147702949e-05, "epoch": 1.25, "percentage": 31.25, "elapsed_time": "2:49:14", "remaining_time": "6:12:19"}
146
+ {"current_steps": 292, "total_steps": 928, "loss": 0.8889655470848083, "lr": 3.360565024971202e-05, "epoch": 1.2586206896551724, "percentage": 31.47, "elapsed_time": "2:50:26", "remaining_time": "6:11:14"}
147
+ {"current_steps": 294, "total_steps": 928, "loss": 0.9757481813430786, "lr": 3.350722871862368e-05, "epoch": 1.2672413793103448, "percentage": 31.68, "elapsed_time": "2:51:34", "remaining_time": "6:09:59"}
148
+ {"current_steps": 296, "total_steps": 928, "loss": 0.7736673355102539, "lr": 3.340822188982616e-05, "epoch": 1.2758620689655173, "percentage": 31.9, "elapsed_time": "2:52:39", "remaining_time": "6:08:37"}
149
+ {"current_steps": 298, "total_steps": 928, "loss": 1.0164954662322998, "lr": 3.330863479915138e-05, "epoch": 1.2844827586206897, "percentage": 32.11, "elapsed_time": "2:54:08", "remaining_time": "6:08:08"}
150
+ {"current_steps": 300, "total_steps": 928, "loss": 0.9475960731506348, "lr": 3.320847251194546e-05, "epoch": 1.293103448275862, "percentage": 32.33, "elapsed_time": "2:55:31", "remaining_time": "6:07:25"}
151
+ {"current_steps": 302, "total_steps": 928, "loss": 0.6825069785118103, "lr": 3.310774012281099e-05, "epoch": 1.3017241379310345, "percentage": 32.54, "elapsed_time": "2:56:36", "remaining_time": "6:06:04"}
152
+ {"current_steps": 304, "total_steps": 928, "loss": 0.5717735290527344, "lr": 3.300644275534793e-05, "epoch": 1.3103448275862069, "percentage": 32.76, "elapsed_time": "2:57:20", "remaining_time": "6:04:01"}
153
+ {"current_steps": 306, "total_steps": 928, "loss": 1.3711295127868652, "lr": 3.290458556189299e-05, "epoch": 1.3189655172413794, "percentage": 32.97, "elapsed_time": "2:58:43", "remaining_time": "6:03:16"}
154
+ {"current_steps": 308, "total_steps": 928, "loss": 0.6421374082565308, "lr": 3.2802173723257604e-05, "epoch": 1.3275862068965516, "percentage": 33.19, "elapsed_time": "2:59:45", "remaining_time": "6:01:51"}
155
+ {"current_steps": 310, "total_steps": 928, "loss": 0.9758880734443665, "lr": 3.2699212448464385e-05, "epoch": 1.3362068965517242, "percentage": 33.41, "elapsed_time": "3:00:55", "remaining_time": "6:00:40"}
156
+ {"current_steps": 312, "total_steps": 928, "loss": 0.9329778552055359, "lr": 3.259570697448217e-05, "epoch": 1.3448275862068966, "percentage": 33.62, "elapsed_time": "3:01:54", "remaining_time": "5:59:08"}
157
+ {"current_steps": 314, "total_steps": 928, "loss": 1.2179062366485596, "lr": 3.249166256595967e-05, "epoch": 1.353448275862069, "percentage": 33.84, "elapsed_time": "3:02:58", "remaining_time": "5:57:47"}
158
+ {"current_steps": 316, "total_steps": 928, "loss": 1.3471888303756714, "lr": 3.2387084514957675e-05, "epoch": 1.3620689655172413, "percentage": 34.05, "elapsed_time": "3:04:17", "remaining_time": "5:56:54"}
159
+ {"current_steps": 318, "total_steps": 928, "loss": 0.9441757202148438, "lr": 3.2281978140679894e-05, "epoch": 1.3706896551724137, "percentage": 34.27, "elapsed_time": "3:05:33", "remaining_time": "5:55:57"}
160
+ {"current_steps": 320, "total_steps": 928, "loss": 0.7498874664306641, "lr": 3.21763487892024e-05, "epoch": 1.3793103448275863, "percentage": 34.48, "elapsed_time": "3:06:23", "remaining_time": "5:54:08"}
161
+ {"current_steps": 322, "total_steps": 928, "loss": 1.4824918508529663, "lr": 3.207020183320171e-05, "epoch": 1.3879310344827587, "percentage": 34.7, "elapsed_time": "3:07:46", "remaining_time": "5:53:24"}
162
+ {"current_steps": 324, "total_steps": 928, "loss": 0.49022743105888367, "lr": 3.196354267168149e-05, "epoch": 1.396551724137931, "percentage": 34.91, "elapsed_time": "3:09:06", "remaining_time": "5:52:31"}
163
+ {"current_steps": 326, "total_steps": 928, "loss": 0.6543675661087036, "lr": 3.185637672969799e-05, "epoch": 1.4051724137931034, "percentage": 35.13, "elapsed_time": "3:09:47", "remaining_time": "5:50:28"}
164
+ {"current_steps": 328, "total_steps": 928, "loss": 0.8541685342788696, "lr": 3.1748709458084045e-05, "epoch": 1.4137931034482758, "percentage": 35.34, "elapsed_time": "3:10:50", "remaining_time": "5:49:06"}
165
+ {"current_steps": 330, "total_steps": 928, "loss": 0.7656717300415039, "lr": 3.1640546333171894e-05, "epoch": 1.4224137931034484, "percentage": 35.56, "elapsed_time": "3:11:56", "remaining_time": "5:47:49"}
166
+ {"current_steps": 332, "total_steps": 928, "loss": 0.6957482695579529, "lr": 3.153189285651458e-05, "epoch": 1.4310344827586206, "percentage": 35.78, "elapsed_time": "3:13:00", "remaining_time": "5:46:29"}
167
+ {"current_steps": 334, "total_steps": 928, "loss": 0.6638420224189758, "lr": 3.142275455460614e-05, "epoch": 1.4396551724137931, "percentage": 35.99, "elapsed_time": "3:14:05", "remaining_time": "5:45:10"}
168
+ {"current_steps": 336, "total_steps": 928, "loss": 0.7661845088005066, "lr": 3.131313697860053e-05, "epoch": 1.4482758620689655, "percentage": 36.21, "elapsed_time": "3:15:06", "remaining_time": "5:43:45"}
169
+ {"current_steps": 338, "total_steps": 928, "loss": 1.1160173416137695, "lr": 3.120304570402924e-05, "epoch": 1.456896551724138, "percentage": 36.42, "elapsed_time": "3:16:26", "remaining_time": "5:42:54"}
170
+ {"current_steps": 340, "total_steps": 928, "loss": 1.384441614151001, "lr": 3.1092486330517714e-05, "epoch": 1.4655172413793103, "percentage": 36.64, "elapsed_time": "3:17:49", "remaining_time": "5:42:06"}
171
+ {"current_steps": 342, "total_steps": 928, "loss": 1.145660638809204, "lr": 3.098146448150055e-05, "epoch": 1.4741379310344827, "percentage": 36.85, "elapsed_time": "3:19:18", "remaining_time": "5:41:31"}
172
+ {"current_steps": 344, "total_steps": 928, "loss": 0.9891381859779358, "lr": 3.086998580393547e-05, "epoch": 1.4827586206896552, "percentage": 37.07, "elapsed_time": "3:20:38", "remaining_time": "5:40:37"}
173
+ {"current_steps": 346, "total_steps": 928, "loss": 0.6093174815177917, "lr": 3.075805596801605e-05, "epoch": 1.4913793103448276, "percentage": 37.28, "elapsed_time": "3:21:20", "remaining_time": "5:38:40"}
174
+ {"current_steps": 348, "total_steps": 928, "loss": 0.9298641681671143, "lr": 3.0645680666883374e-05, "epoch": 1.5, "percentage": 37.5, "elapsed_time": "3:22:43", "remaining_time": "5:37:52"}
175
+ {"current_steps": 350, "total_steps": 928, "loss": 0.9974504113197327, "lr": 3.053286561633644e-05, "epoch": 1.5086206896551724, "percentage": 37.72, "elapsed_time": "3:24:07", "remaining_time": "5:37:06"}
176
+ {"current_steps": 352, "total_steps": 928, "loss": 0.9739059209823608, "lr": 3.041961655454143e-05, "epoch": 1.5172413793103448, "percentage": 37.93, "elapsed_time": "3:25:21", "remaining_time": "5:36:02"}
177
+ {"current_steps": 354, "total_steps": 928, "loss": 1.133984088897705, "lr": 3.030593924173984e-05, "epoch": 1.5258620689655173, "percentage": 38.15, "elapsed_time": "3:26:44", "remaining_time": "5:35:14"}
178
+ {"current_steps": 356, "total_steps": 928, "loss": 0.8807175755500793, "lr": 3.0191839459955514e-05, "epoch": 1.5344827586206895, "percentage": 38.36, "elapsed_time": "3:28:07", "remaining_time": "5:34:23"}
179
+ {"current_steps": 358, "total_steps": 928, "loss": 0.8361281156539917, "lr": 3.0077323012700534e-05, "epoch": 1.543103448275862, "percentage": 38.58, "elapsed_time": "3:29:02", "remaining_time": "5:32:49"}
180
+ {"current_steps": 360, "total_steps": 928, "loss": 1.2387166023254395, "lr": 2.996239572468003e-05, "epoch": 1.5517241379310345, "percentage": 38.79, "elapsed_time": "3:30:15", "remaining_time": "5:31:43"}
181
+ {"current_steps": 362, "total_steps": 928, "loss": 1.0467900037765503, "lr": 2.984706344149595e-05, "epoch": 1.5603448275862069, "percentage": 39.01, "elapsed_time": "3:31:31", "remaining_time": "5:30:42"}
182
+ {"current_steps": 364, "total_steps": 928, "loss": 1.0626113414764404, "lr": 2.9731332029349667e-05, "epoch": 1.5689655172413794, "percentage": 39.22, "elapsed_time": "3:32:57", "remaining_time": "5:29:57"}
183
+ {"current_steps": 366, "total_steps": 928, "loss": 1.0128107070922852, "lr": 2.961520737474367e-05, "epoch": 1.5775862068965516, "percentage": 39.44, "elapsed_time": "3:34:01", "remaining_time": "5:28:38"}
184
+ {"current_steps": 368, "total_steps": 928, "loss": 0.9877223968505859, "lr": 2.9498695384182123e-05, "epoch": 1.5862068965517242, "percentage": 39.66, "elapsed_time": "3:35:27", "remaining_time": "5:27:52"}
185
+ {"current_steps": 370, "total_steps": 928, "loss": 0.9603118300437927, "lr": 2.9381801983870435e-05, "epoch": 1.5948275862068966, "percentage": 39.87, "elapsed_time": "3:36:41", "remaining_time": "5:26:47"}
186
+ {"current_steps": 372, "total_steps": 928, "loss": 1.081476092338562, "lr": 2.9264533119413866e-05, "epoch": 1.603448275862069, "percentage": 40.09, "elapsed_time": "3:38:02", "remaining_time": "5:25:52"}
187
+ {"current_steps": 374, "total_steps": 928, "loss": 0.7714329957962036, "lr": 2.914689475551506e-05, "epoch": 1.6120689655172413, "percentage": 40.3, "elapsed_time": "3:39:06", "remaining_time": "5:24:33"}
188
+ {"current_steps": 376, "total_steps": 928, "loss": 0.9913143515586853, "lr": 2.902889287567072e-05, "epoch": 1.6206896551724137, "percentage": 40.52, "elapsed_time": "3:40:32", "remaining_time": "5:23:46"}
189
+ {"current_steps": 378, "total_steps": 928, "loss": 1.1765313148498535, "lr": 2.8910533481867195e-05, "epoch": 1.6293103448275863, "percentage": 40.73, "elapsed_time": "3:41:57", "remaining_time": "5:22:56"}
190
+ {"current_steps": 380, "total_steps": 928, "loss": 0.7655573487281799, "lr": 2.879182259427528e-05, "epoch": 1.6379310344827587, "percentage": 40.95, "elapsed_time": "3:42:47", "remaining_time": "5:21:18"}
191
+ {"current_steps": 382, "total_steps": 928, "loss": 1.3452657461166382, "lr": 2.8672766250943947e-05, "epoch": 1.646551724137931, "percentage": 41.16, "elapsed_time": "3:44:11", "remaining_time": "5:20:26"}
192
+ {"current_steps": 384, "total_steps": 928, "loss": 0.9972445964813232, "lr": 2.8553370507493246e-05, "epoch": 1.6551724137931034, "percentage": 41.38, "elapsed_time": "3:45:25", "remaining_time": "5:19:20"}
193
+ {"current_steps": 386, "total_steps": 928, "loss": 0.8845785856246948, "lr": 2.8433641436806306e-05, "epoch": 1.6637931034482758, "percentage": 41.59, "elapsed_time": "3:46:38", "remaining_time": "5:18:13"}
194
+ {"current_steps": 388, "total_steps": 928, "loss": 1.3110713958740234, "lr": 2.8313585128720444e-05, "epoch": 1.6724137931034484, "percentage": 41.81, "elapsed_time": "3:47:49", "remaining_time": "5:17:04"}
195
+ {"current_steps": 390, "total_steps": 928, "loss": 0.8128502368927002, "lr": 2.8193207689717393e-05, "epoch": 1.6810344827586206, "percentage": 42.03, "elapsed_time": "3:49:01", "remaining_time": "5:15:56"}
196
+ {"current_steps": 392, "total_steps": 928, "loss": 0.6244351863861084, "lr": 2.807251524261275e-05, "epoch": 1.6896551724137931, "percentage": 42.24, "elapsed_time": "3:49:45", "remaining_time": "5:14:10"}
197
+ {"current_steps": 394, "total_steps": 928, "loss": 1.127506136894226, "lr": 2.7951513926244484e-05, "epoch": 1.6982758620689655, "percentage": 42.46, "elapsed_time": "3:51:07", "remaining_time": "5:13:15"}
198
+ {"current_steps": 396, "total_steps": 928, "loss": 1.042289137840271, "lr": 2.7830209895160764e-05, "epoch": 1.706896551724138, "percentage": 42.67, "elapsed_time": "3:52:34", "remaining_time": "5:12:26"}
199
+ {"current_steps": 398, "total_steps": 928, "loss": 1.0570330619812012, "lr": 2.770860931930687e-05, "epoch": 1.7155172413793105, "percentage": 42.89, "elapsed_time": "3:53:45", "remaining_time": "5:11:16"}
200
+ {"current_steps": 400, "total_steps": 928, "loss": 0.9959380626678467, "lr": 2.7586718383711367e-05, "epoch": 1.7241379310344827, "percentage": 43.1, "elapsed_time": "3:54:54", "remaining_time": "5:10:04"}
201
+ {"current_steps": 402, "total_steps": 928, "loss": 1.0200254917144775, "lr": 2.7464543288171558e-05, "epoch": 1.7327586206896552, "percentage": 43.32, "elapsed_time": "3:55:57", "remaining_time": "5:08:44"}
202
+ {"current_steps": 404, "total_steps": 928, "loss": 0.6205574870109558, "lr": 2.7342090246938076e-05, "epoch": 1.7413793103448276, "percentage": 43.53, "elapsed_time": "3:56:59", "remaining_time": "5:07:23"}
203
+ {"current_steps": 406, "total_steps": 928, "loss": 0.8922735452651978, "lr": 2.721936548839887e-05, "epoch": 1.75, "percentage": 43.75, "elapsed_time": "3:58:04", "remaining_time": "5:06:05"}
204
+ {"current_steps": 408, "total_steps": 928, "loss": 0.6991145014762878, "lr": 2.709637525476236e-05, "epoch": 1.7586206896551724, "percentage": 43.97, "elapsed_time": "3:58:59", "remaining_time": "5:04:35"}
205
+ {"current_steps": 410, "total_steps": 928, "loss": 0.8093492984771729, "lr": 2.697312580173995e-05, "epoch": 1.7672413793103448, "percentage": 44.18, "elapsed_time": "4:00:03", "remaining_time": "5:03:17"}
206
+ {"current_steps": 412, "total_steps": 928, "loss": 0.7507970929145813, "lr": 2.684962339822785e-05, "epoch": 1.7758620689655173, "percentage": 44.4, "elapsed_time": "4:01:24", "remaining_time": "5:02:21"}
207
+ {"current_steps": 414, "total_steps": 928, "loss": 0.5883830189704895, "lr": 2.672587432598823e-05, "epoch": 1.7844827586206895, "percentage": 44.61, "elapsed_time": "4:02:29", "remaining_time": "5:01:03"}
208
+ {"current_steps": 416, "total_steps": 928, "loss": 0.7915773391723633, "lr": 2.6601884879329653e-05, "epoch": 1.793103448275862, "percentage": 44.83, "elapsed_time": "4:03:32", "remaining_time": "4:59:44"}
209
+ {"current_steps": 418, "total_steps": 928, "loss": 1.0269769430160522, "lr": 2.6477661364786996e-05, "epoch": 1.8017241379310345, "percentage": 45.04, "elapsed_time": "4:04:54", "remaining_time": "4:58:48"}
210
+ {"current_steps": 420, "total_steps": 928, "loss": 1.053789496421814, "lr": 2.635321010080062e-05, "epoch": 1.8103448275862069, "percentage": 45.26, "elapsed_time": "4:06:16", "remaining_time": "4:57:52"}
211
+ {"current_steps": 422, "total_steps": 928, "loss": 1.158492088317871, "lr": 2.6228537417395034e-05, "epoch": 1.8189655172413794, "percentage": 45.47, "elapsed_time": "4:07:42", "remaining_time": "4:57:00"}
212
+ {"current_steps": 424, "total_steps": 928, "loss": 0.9592758417129517, "lr": 2.61036496558569e-05, "epoch": 1.8275862068965516, "percentage": 45.69, "elapsed_time": "4:08:47", "remaining_time": "4:55:44"}
213
+ {"current_steps": 426, "total_steps": 928, "loss": 0.6086317300796509, "lr": 2.59785531684125e-05, "epoch": 1.8362068965517242, "percentage": 45.91, "elapsed_time": "4:09:33", "remaining_time": "4:54:04"}
214
+ {"current_steps": 428, "total_steps": 928, "loss": 1.0528879165649414, "lr": 2.585325431790464e-05, "epoch": 1.8448275862068966, "percentage": 46.12, "elapsed_time": "4:10:59", "remaining_time": "4:53:12"}
215
+ {"current_steps": 430, "total_steps": 928, "loss": 1.0576783418655396, "lr": 2.572775947746903e-05, "epoch": 1.853448275862069, "percentage": 46.34, "elapsed_time": "4:12:24", "remaining_time": "4:52:19"}
216
+ {"current_steps": 432, "total_steps": 928, "loss": 0.9204137325286865, "lr": 2.5602075030210096e-05, "epoch": 1.8620689655172413, "percentage": 46.55, "elapsed_time": "4:13:35", "remaining_time": "4:51:09"}
217
+ {"current_steps": 434, "total_steps": 928, "loss": 1.114011287689209, "lr": 2.5476207368876334e-05, "epoch": 1.8706896551724137, "percentage": 46.77, "elapsed_time": "4:14:50", "remaining_time": "4:50:04"}
218
+ {"current_steps": 436, "total_steps": 928, "loss": 0.7933326363563538, "lr": 2.535016289553514e-05, "epoch": 1.8793103448275863, "percentage": 46.98, "elapsed_time": "4:15:35", "remaining_time": "4:48:25"}
219
+ {"current_steps": 438, "total_steps": 928, "loss": 0.9807726144790649, "lr": 2.5223948021247197e-05, "epoch": 1.8879310344827587, "percentage": 47.2, "elapsed_time": "4:16:51", "remaining_time": "4:47:20"}
220
+ {"current_steps": 440, "total_steps": 928, "loss": 1.0345503091812134, "lr": 2.509756916574035e-05, "epoch": 1.896551724137931, "percentage": 47.41, "elapsed_time": "4:18:01", "remaining_time": "4:46:10"}
221
+ {"current_steps": 442, "total_steps": 928, "loss": 1.1201728582382202, "lr": 2.4971032757083123e-05, "epoch": 1.9051724137931034, "percentage": 47.63, "elapsed_time": "4:19:20", "remaining_time": "4:45:09"}
222
+ {"current_steps": 444, "total_steps": 928, "loss": 0.28341731429100037, "lr": 2.4844345231357734e-05, "epoch": 1.9137931034482758, "percentage": 47.84, "elapsed_time": "4:20:04", "remaining_time": "4:43:30"}
223
+ {"current_steps": 446, "total_steps": 928, "loss": 0.7789583206176758, "lr": 2.4717513032332736e-05, "epoch": 1.9224137931034484, "percentage": 48.06, "elapsed_time": "4:21:12", "remaining_time": "4:42:17"}
224
+ {"current_steps": 448, "total_steps": 928, "loss": 1.012285590171814, "lr": 2.4590542611135274e-05, "epoch": 1.9310344827586206, "percentage": 48.28, "elapsed_time": "4:22:28", "remaining_time": "4:41:13"}
225
+ {"current_steps": 450, "total_steps": 928, "loss": 1.0196033716201782, "lr": 2.446344042592295e-05, "epoch": 1.9396551724137931, "percentage": 48.49, "elapsed_time": "4:23:54", "remaining_time": "4:40:19"}
226
+ {"current_steps": 452, "total_steps": 928, "loss": 0.8052966594696045, "lr": 2.433621294155535e-05, "epoch": 1.9482758620689655, "percentage": 48.71, "elapsed_time": "4:25:02", "remaining_time": "4:39:06"}
227
+ {"current_steps": 454, "total_steps": 928, "loss": 0.9915321469306946, "lr": 2.420886662926521e-05, "epoch": 1.956896551724138, "percentage": 48.92, "elapsed_time": "4:26:17", "remaining_time": "4:38:01"}
228
+ {"current_steps": 456, "total_steps": 928, "loss": 0.9689676761627197, "lr": 2.4081407966329256e-05, "epoch": 1.9655172413793105, "percentage": 49.14, "elapsed_time": "4:27:43", "remaining_time": "4:37:07"}
229
+ {"current_steps": 458, "total_steps": 928, "loss": 0.41972166299819946, "lr": 2.3953843435738775e-05, "epoch": 1.9741379310344827, "percentage": 49.35, "elapsed_time": "4:28:28", "remaining_time": "4:35:30"}
230
+ {"current_steps": 460, "total_steps": 928, "loss": 1.1633706092834473, "lr": 2.3826179525869836e-05, "epoch": 1.9827586206896552, "percentage": 49.57, "elapsed_time": "4:29:51", "remaining_time": "4:34:33"}
231
+ {"current_steps": 462, "total_steps": 928, "loss": 0.7487952709197998, "lr": 2.36984227301533e-05, "epoch": 1.9913793103448276, "percentage": 49.78, "elapsed_time": "4:30:52", "remaining_time": "4:33:12"}
232
+ {"current_steps": 464, "total_steps": 928, "loss": 0.8847077488899231, "lr": 2.3570579546744504e-05, "epoch": 2.0, "percentage": 50.0, "elapsed_time": "4:32:09", "remaining_time": "4:32:09"}
233
+ {"current_steps": 466, "total_steps": 928, "loss": 0.484560489654541, "lr": 2.3442656478192794e-05, "epoch": 2.0086206896551726, "percentage": 50.22, "elapsed_time": "4:33:13", "remaining_time": "4:30:52"}
234
+ {"current_steps": 468, "total_steps": 928, "loss": 0.6984850764274597, "lr": 2.331466003111073e-05, "epoch": 2.0172413793103448, "percentage": 50.43, "elapsed_time": "4:34:26", "remaining_time": "4:29:44"}
235
+ {"current_steps": 470, "total_steps": 928, "loss": 0.4863373935222626, "lr": 2.318659671584316e-05, "epoch": 2.0258620689655173, "percentage": 50.65, "elapsed_time": "4:35:32", "remaining_time": "4:28:30"}
236
+ {"current_steps": 472, "total_steps": 928, "loss": 0.4588513970375061, "lr": 2.305847304613609e-05, "epoch": 2.0344827586206895, "percentage": 50.86, "elapsed_time": "4:36:26", "remaining_time": "4:27:04"}
237
+ {"current_steps": 474, "total_steps": 928, "loss": 0.4486234486103058, "lr": 2.293029553880536e-05, "epoch": 2.043103448275862, "percentage": 51.08, "elapsed_time": "4:37:32", "remaining_time": "4:25:49"}
238
+ {"current_steps": 476, "total_steps": 928, "loss": 0.5298870205879211, "lr": 2.280207071340517e-05, "epoch": 2.0517241379310347, "percentage": 51.29, "elapsed_time": "4:38:36", "remaining_time": "4:24:33"}
239
+ {"current_steps": 478, "total_steps": 928, "loss": 0.4382156729698181, "lr": 2.26738050918965e-05, "epoch": 2.060344827586207, "percentage": 51.51, "elapsed_time": "4:39:40", "remaining_time": "4:23:17"}
240
+ {"current_steps": 480, "total_steps": 928, "loss": 0.5762298107147217, "lr": 2.2545505198315346e-05, "epoch": 2.0689655172413794, "percentage": 51.72, "elapsed_time": "4:41:06", "remaining_time": "4:22:22"}
241
+ {"current_steps": 482, "total_steps": 928, "loss": 0.36859992146492004, "lr": 2.2417177558440907e-05, "epoch": 2.0775862068965516, "percentage": 51.94, "elapsed_time": "4:42:10", "remaining_time": "4:21:06"}
242
+ {"current_steps": 484, "total_steps": 928, "loss": 0.5293700098991394, "lr": 2.2288828699463652e-05, "epoch": 2.086206896551724, "percentage": 52.16, "elapsed_time": "4:43:32", "remaining_time": "4:20:06"}
243
+ {"current_steps": 486, "total_steps": 928, "loss": 0.49147215485572815, "lr": 2.2160465149653337e-05, "epoch": 2.0948275862068964, "percentage": 52.37, "elapsed_time": "4:44:43", "remaining_time": "4:18:57"}
244
+ {"current_steps": 488, "total_steps": 928, "loss": 0.5180780291557312, "lr": 2.203209343802692e-05, "epoch": 2.103448275862069, "percentage": 52.59, "elapsed_time": "4:46:04", "remaining_time": "4:17:56"}
245
+ {"current_steps": 490, "total_steps": 928, "loss": 0.581203818321228, "lr": 2.1903720094016537e-05, "epoch": 2.1120689655172415, "percentage": 52.8, "elapsed_time": "4:47:29", "remaining_time": "4:16:59"}
246
+ {"current_steps": 492, "total_steps": 928, "loss": 0.4889185130596161, "lr": 2.1775351647137323e-05, "epoch": 2.1206896551724137, "percentage": 53.02, "elapsed_time": "4:48:45", "remaining_time": "4:15:53"}
247
+ {"current_steps": 494, "total_steps": 928, "loss": 0.6391059756278992, "lr": 2.1646994626655332e-05, "epoch": 2.1293103448275863, "percentage": 53.23, "elapsed_time": "4:50:09", "remaining_time": "4:14:55"}
248
+ {"current_steps": 496, "total_steps": 928, "loss": 0.6237853169441223, "lr": 2.151865556125544e-05, "epoch": 2.1379310344827585, "percentage": 53.45, "elapsed_time": "4:51:18", "remaining_time": "4:13:43"}
249
+ {"current_steps": 498, "total_steps": 928, "loss": 0.36577755212783813, "lr": 2.1390340978709254e-05, "epoch": 2.146551724137931, "percentage": 53.66, "elapsed_time": "4:52:07", "remaining_time": "4:12:14"}
250
+ {"current_steps": 500, "total_steps": 928, "loss": 0.49633127450942993, "lr": 2.1262057405543115e-05, "epoch": 2.1551724137931036, "percentage": 53.88, "elapsed_time": "4:53:08", "remaining_time": "4:10:56"}
251
+ {"current_steps": 502, "total_steps": 928, "loss": 0.38259175419807434, "lr": 2.1133811366706097e-05, "epoch": 2.163793103448276, "percentage": 54.09, "elapsed_time": "4:54:13", "remaining_time": "4:09:40"}
252
+ {"current_steps": 504, "total_steps": 928, "loss": 0.4427034258842468, "lr": 2.100560938523817e-05, "epoch": 2.1724137931034484, "percentage": 54.31, "elapsed_time": "4:55:05", "remaining_time": "4:08:15"}
253
+ {"current_steps": 506, "total_steps": 928, "loss": 0.6942803263664246, "lr": 2.0877457981938364e-05, "epoch": 2.1810344827586206, "percentage": 54.53, "elapsed_time": "4:56:27", "remaining_time": "4:07:14"}
254
+ {"current_steps": 508, "total_steps": 928, "loss": 0.5671365261077881, "lr": 2.074936367503317e-05, "epoch": 2.189655172413793, "percentage": 54.74, "elapsed_time": "4:57:32", "remaining_time": "4:05:59"}
255
+ {"current_steps": 510, "total_steps": 928, "loss": 0.6063480377197266, "lr": 2.0621332979844904e-05, "epoch": 2.1982758620689653, "percentage": 54.96, "elapsed_time": "4:58:45", "remaining_time": "4:04:52"}
256
+ {"current_steps": 512, "total_steps": 928, "loss": 0.6027957201004028, "lr": 2.0493372408460425e-05, "epoch": 2.206896551724138, "percentage": 55.17, "elapsed_time": "4:59:36", "remaining_time": "4:03:25"}
257
+ {"current_steps": 514, "total_steps": 928, "loss": 0.6078309416770935, "lr": 2.0365488469399795e-05, "epoch": 2.2155172413793105, "percentage": 55.39, "elapsed_time": "5:00:39", "remaining_time": "4:02:10"}
258
+ {"current_steps": 516, "total_steps": 928, "loss": 0.3304949402809143, "lr": 2.0237687667285345e-05, "epoch": 2.2241379310344827, "percentage": 55.6, "elapsed_time": "5:01:44", "remaining_time": "4:00:55"}
259
+ {"current_steps": 518, "total_steps": 928, "loss": 0.12825970351696014, "lr": 2.010997650251072e-05, "epoch": 2.2327586206896552, "percentage": 55.82, "elapsed_time": "5:02:28", "remaining_time": "3:59:24"}
260
+ {"current_steps": 520, "total_steps": 928, "loss": 0.1828547865152359, "lr": 1.9982361470910342e-05, "epoch": 2.2413793103448274, "percentage": 56.03, "elapsed_time": "5:03:27", "remaining_time": "3:58:06"}
261
+ {"current_steps": 522, "total_steps": 928, "loss": 0.6522985696792603, "lr": 1.9854849063428926e-05, "epoch": 2.25, "percentage": 56.25, "elapsed_time": "5:04:52", "remaining_time": "3:57:07"}
262
+ {"current_steps": 524, "total_steps": 928, "loss": 0.47932472825050354, "lr": 1.9727445765791405e-05, "epoch": 2.2586206896551726, "percentage": 56.47, "elapsed_time": "5:05:49", "remaining_time": "3:55:46"}
263
+ {"current_steps": 526, "total_steps": 928, "loss": 0.4181676208972931, "lr": 1.9600158058172974e-05, "epoch": 2.2672413793103448, "percentage": 56.68, "elapsed_time": "5:06:51", "remaining_time": "3:54:31"}
264
+ {"current_steps": 528, "total_steps": 928, "loss": 0.46739447116851807, "lr": 1.9472992414869534e-05, "epoch": 2.2758620689655173, "percentage": 56.9, "elapsed_time": "5:07:54", "remaining_time": "3:53:16"}
265
+ {"current_steps": 530, "total_steps": 928, "loss": 0.38251054286956787, "lr": 1.9345955303968365e-05, "epoch": 2.2844827586206895, "percentage": 57.11, "elapsed_time": "5:08:59", "remaining_time": "3:52:02"}
266
+ {"current_steps": 532, "total_steps": 928, "loss": 0.4366922080516815, "lr": 1.9219053187019144e-05, "epoch": 2.293103448275862, "percentage": 57.33, "elapsed_time": "5:10:03", "remaining_time": "3:50:48"}
267
+ {"current_steps": 534, "total_steps": 928, "loss": 0.4965798556804657, "lr": 1.909229251870528e-05, "epoch": 2.3017241379310347, "percentage": 57.54, "elapsed_time": "5:11:08", "remaining_time": "3:49:34"}
268
+ {"current_steps": 536, "total_steps": 928, "loss": 0.43146276473999023, "lr": 1.8965679746515628e-05, "epoch": 2.310344827586207, "percentage": 57.76, "elapsed_time": "5:12:19", "remaining_time": "3:48:24"}
269
+ {"current_steps": 538, "total_steps": 928, "loss": 0.2771337330341339, "lr": 1.88392213104165e-05, "epoch": 2.3189655172413794, "percentage": 57.97, "elapsed_time": "5:13:10", "remaining_time": "3:47:01"}
270
+ {"current_steps": 540, "total_steps": 928, "loss": 0.36878013610839844, "lr": 1.8712923642524175e-05, "epoch": 2.3275862068965516, "percentage": 58.19, "elapsed_time": "5:14:13", "remaining_time": "3:45:46"}
271
+ {"current_steps": 542, "total_steps": 928, "loss": 0.6058629751205444, "lr": 1.858679316677767e-05, "epoch": 2.336206896551724, "percentage": 58.41, "elapsed_time": "5:15:32", "remaining_time": "3:44:43"}
272
+ {"current_steps": 544, "total_steps": 928, "loss": 0.6428977251052856, "lr": 1.8460836298612056e-05, "epoch": 2.344827586206897, "percentage": 58.62, "elapsed_time": "5:16:52", "remaining_time": "3:43:40"}
273
+ {"current_steps": 546, "total_steps": 928, "loss": 0.2821408212184906, "lr": 1.8335059444632078e-05, "epoch": 2.353448275862069, "percentage": 58.84, "elapsed_time": "5:17:33", "remaining_time": "3:42:10"}
274
+ {"current_steps": 548, "total_steps": 928, "loss": 0.8290093541145325, "lr": 1.820946900228639e-05, "epoch": 2.3620689655172415, "percentage": 59.05, "elapsed_time": "5:18:48", "remaining_time": "3:41:04"}
275
+ {"current_steps": 550, "total_steps": 928, "loss": 0.5475698709487915, "lr": 1.808407135954204e-05, "epoch": 2.3706896551724137, "percentage": 59.27, "elapsed_time": "5:19:49", "remaining_time": "3:39:48"}
276
+ {"current_steps": 552, "total_steps": 928, "loss": 0.6245191693305969, "lr": 1.7958872894559666e-05, "epoch": 2.3793103448275863, "percentage": 59.48, "elapsed_time": "5:21:10", "remaining_time": "3:38:46"}
277
+ {"current_steps": 554, "total_steps": 928, "loss": 0.5108689665794373, "lr": 1.7833879975368994e-05, "epoch": 2.3879310344827585, "percentage": 59.7, "elapsed_time": "5:22:21", "remaining_time": "3:37:37"}
278
+ {"current_steps": 556, "total_steps": 928, "loss": 0.5519805550575256, "lr": 1.7709098959545015e-05, "epoch": 2.396551724137931, "percentage": 59.91, "elapsed_time": "5:23:42", "remaining_time": "3:36:35"}
279
+ {"current_steps": 558, "total_steps": 928, "loss": 0.4397192597389221, "lr": 1.758453619388453e-05, "epoch": 2.405172413793103, "percentage": 60.13, "elapsed_time": "5:24:39", "remaining_time": "3:35:16"}
280
+ {"current_steps": 560, "total_steps": 928, "loss": 0.38739266991615295, "lr": 1.7460198014083424e-05, "epoch": 2.413793103448276, "percentage": 60.34, "elapsed_time": "5:25:44", "remaining_time": "3:34:03"}
281
+ {"current_steps": 562, "total_steps": 928, "loss": 0.5064358711242676, "lr": 1.733609074441433e-05, "epoch": 2.4224137931034484, "percentage": 60.56, "elapsed_time": "5:27:07", "remaining_time": "3:33:02"}
282
+ {"current_steps": 564, "total_steps": 928, "loss": 0.540324330329895, "lr": 1.7212220697405003e-05, "epoch": 2.4310344827586206, "percentage": 60.78, "elapsed_time": "5:28:09", "remaining_time": "3:31:47"}
283
+ {"current_steps": 566, "total_steps": 928, "loss": 0.5431786179542542, "lr": 1.7088594173517225e-05, "epoch": 2.439655172413793, "percentage": 60.99, "elapsed_time": "5:29:31", "remaining_time": "3:30:45"}
284
+ {"current_steps": 568, "total_steps": 928, "loss": 0.3365917205810547, "lr": 1.6965217460826345e-05, "epoch": 2.4482758620689653, "percentage": 61.21, "elapsed_time": "5:30:17", "remaining_time": "3:29:20"}
285
+ {"current_steps": 570, "total_steps": 928, "loss": 0.6636412739753723, "lr": 1.6842096834701443e-05, "epoch": 2.456896551724138, "percentage": 61.42, "elapsed_time": "5:31:41", "remaining_time": "3:28:19"}
286
+ {"current_steps": 572, "total_steps": 928, "loss": 0.3930183947086334, "lr": 1.6719238557486143e-05, "epoch": 2.4655172413793105, "percentage": 61.64, "elapsed_time": "5:32:47", "remaining_time": "3:27:07"}
287
+ {"current_steps": 574, "total_steps": 928, "loss": 0.4772527813911438, "lr": 1.6596648878180088e-05, "epoch": 2.4741379310344827, "percentage": 61.85, "elapsed_time": "5:33:57", "remaining_time": "3:25:57"}
288
+ {"current_steps": 576, "total_steps": 928, "loss": 0.6689369082450867, "lr": 1.647433403212112e-05, "epoch": 2.4827586206896552, "percentage": 62.07, "elapsed_time": "5:35:21", "remaining_time": "3:24:56"}
289
+ {"current_steps": 578, "total_steps": 928, "loss": 0.5050515532493591, "lr": 1.635230024066807e-05, "epoch": 2.4913793103448274, "percentage": 62.28, "elapsed_time": "5:36:24", "remaining_time": "3:23:42"}
290
+ {"current_steps": 580, "total_steps": 928, "loss": 0.6936325430870056, "lr": 1.6230553710884373e-05, "epoch": 2.5, "percentage": 62.5, "elapsed_time": "5:37:50", "remaining_time": "3:22:42"}
291
+ {"current_steps": 582, "total_steps": 928, "loss": 0.5566367506980896, "lr": 1.610910063522233e-05, "epoch": 2.5086206896551726, "percentage": 62.72, "elapsed_time": "5:39:16", "remaining_time": "3:21:41"}
292
+ {"current_steps": 584, "total_steps": 928, "loss": 0.5264196991920471, "lr": 1.598794719120816e-05, "epoch": 2.5172413793103448, "percentage": 62.93, "elapsed_time": "5:40:16", "remaining_time": "3:20:25"}
293
+ {"current_steps": 586, "total_steps": 928, "loss": 0.4999127686023712, "lr": 1.5867099541127737e-05, "epoch": 2.5258620689655173, "percentage": 63.15, "elapsed_time": "5:41:38", "remaining_time": "3:19:23"}
294
+ {"current_steps": 588, "total_steps": 928, "loss": 0.5660111308097839, "lr": 1.5746563831713236e-05, "epoch": 2.5344827586206895, "percentage": 63.36, "elapsed_time": "5:43:02", "remaining_time": "3:18:21"}
295
+ {"current_steps": 590, "total_steps": 928, "loss": 0.7160353064537048, "lr": 1.56263461938304e-05, "epoch": 2.543103448275862, "percentage": 63.58, "elapsed_time": "5:44:21", "remaining_time": "3:17:16"}
296
+ {"current_steps": 592, "total_steps": 928, "loss": 0.575738251209259, "lr": 1.5506452742166796e-05, "epoch": 2.5517241379310347, "percentage": 63.79, "elapsed_time": "5:45:33", "remaining_time": "3:16:07"}
297
+ {"current_steps": 594, "total_steps": 928, "loss": 0.35511380434036255, "lr": 1.5386889574920692e-05, "epoch": 2.560344827586207, "percentage": 64.01, "elapsed_time": "5:46:37", "remaining_time": "3:14:54"}
298
+ {"current_steps": 596, "total_steps": 928, "loss": 0.40391749143600464, "lr": 1.5267662773491e-05, "epoch": 2.5689655172413794, "percentage": 64.22, "elapsed_time": "5:47:34", "remaining_time": "3:13:36"}
299
+ {"current_steps": 598, "total_steps": 928, "loss": 0.5037810802459717, "lr": 1.514877840216785e-05, "epoch": 2.5775862068965516, "percentage": 64.44, "elapsed_time": "5:48:56", "remaining_time": "3:12:33"}
300
+ {"current_steps": 600, "total_steps": 928, "loss": 0.6312216520309448, "lr": 1.5030242507824215e-05, "epoch": 2.586206896551724, "percentage": 64.66, "elapsed_time": "5:50:16", "remaining_time": "3:11:29"}
301
+ {"current_steps": 602, "total_steps": 928, "loss": 0.39456382393836975, "lr": 1.4912061119608292e-05, "epoch": 2.594827586206897, "percentage": 64.87, "elapsed_time": "5:51:22", "remaining_time": "3:10:16"}
302
+ {"current_steps": 604, "total_steps": 928, "loss": 0.5595788359642029, "lr": 1.4794240248636885e-05, "epoch": 2.603448275862069, "percentage": 65.09, "elapsed_time": "5:52:44", "remaining_time": "3:09:13"}
303
+ {"current_steps": 606, "total_steps": 928, "loss": 0.41876575350761414, "lr": 1.4676785887689614e-05, "epoch": 2.612068965517241, "percentage": 65.3, "elapsed_time": "5:53:49", "remaining_time": "3:08:00"}
304
+ {"current_steps": 608, "total_steps": 928, "loss": 0.6346225142478943, "lr": 1.4559704010904145e-05, "epoch": 2.6206896551724137, "percentage": 65.52, "elapsed_time": "5:55:15", "remaining_time": "3:06:58"}
305
+ {"current_steps": 610, "total_steps": 928, "loss": 0.5777739882469177, "lr": 1.444300057347229e-05, "epoch": 2.6293103448275863, "percentage": 65.73, "elapsed_time": "5:56:16", "remaining_time": "3:05:43"}
306
+ {"current_steps": 612, "total_steps": 928, "loss": 0.5916672945022583, "lr": 1.432668151133712e-05, "epoch": 2.637931034482759, "percentage": 65.95, "elapsed_time": "5:57:30", "remaining_time": "3:04:35"}
307
+ {"current_steps": 614, "total_steps": 928, "loss": 0.5175487995147705, "lr": 1.4210752740891032e-05, "epoch": 2.646551724137931, "percentage": 66.16, "elapsed_time": "5:58:52", "remaining_time": "3:03:31"}
308
+ {"current_steps": 616, "total_steps": 928, "loss": 0.37486380338668823, "lr": 1.4095220158674851e-05, "epoch": 2.655172413793103, "percentage": 66.38, "elapsed_time": "5:59:58", "remaining_time": "3:02:19"}
309
+ {"current_steps": 618, "total_steps": 928, "loss": 0.5902385115623474, "lr": 1.3980089641077864e-05, "epoch": 2.663793103448276, "percentage": 66.59, "elapsed_time": "6:01:22", "remaining_time": "3:01:16"}
310
+ {"current_steps": 620, "total_steps": 928, "loss": 0.3626130223274231, "lr": 1.3865367044038972e-05, "epoch": 2.6724137931034484, "percentage": 66.81, "elapsed_time": "6:02:26", "remaining_time": "3:00:02"}
311
+ {"current_steps": 622, "total_steps": 928, "loss": 0.6260622143745422, "lr": 1.3751058202748815e-05, "epoch": 2.6810344827586206, "percentage": 67.03, "elapsed_time": "6:03:49", "remaining_time": "2:58:59"}
312
+ {"current_steps": 624, "total_steps": 928, "loss": 0.3847617506980896, "lr": 1.3637168931352952e-05, "epoch": 2.689655172413793, "percentage": 67.24, "elapsed_time": "6:04:52", "remaining_time": "2:57:45"}
313
+ {"current_steps": 626, "total_steps": 928, "loss": 0.5213911533355713, "lr": 1.3523705022656194e-05, "epoch": 2.6982758620689653, "percentage": 67.46, "elapsed_time": "6:06:05", "remaining_time": "2:56:36"}
314
+ {"current_steps": 628, "total_steps": 928, "loss": 0.3843521475791931, "lr": 1.3410672247827887e-05, "epoch": 2.706896551724138, "percentage": 67.67, "elapsed_time": "6:07:11", "remaining_time": "2:55:24"}
315
+ {"current_steps": 630, "total_steps": 928, "loss": 0.7390468716621399, "lr": 1.3298076356108431e-05, "epoch": 2.7155172413793105, "percentage": 67.89, "elapsed_time": "6:08:38", "remaining_time": "2:54:22"}
316
+ {"current_steps": 632, "total_steps": 928, "loss": 0.3152429461479187, "lr": 1.318592307451683e-05, "epoch": 2.7241379310344827, "percentage": 68.1, "elapsed_time": "6:09:39", "remaining_time": "2:53:08"}
317
+ {"current_steps": 634, "total_steps": 928, "loss": 0.5903550982475281, "lr": 1.307421810755938e-05, "epoch": 2.7327586206896552, "percentage": 68.32, "elapsed_time": "6:10:52", "remaining_time": "2:51:58"}
318
+ {"current_steps": 636, "total_steps": 928, "loss": 0.4196533262729645, "lr": 1.296296713693956e-05, "epoch": 2.7413793103448274, "percentage": 68.53, "elapsed_time": "6:11:53", "remaining_time": "2:50:44"}
319
+ {"current_steps": 638, "total_steps": 928, "loss": 0.5849826335906982, "lr": 1.2852175821268977e-05, "epoch": 2.75, "percentage": 68.75, "elapsed_time": "6:13:09", "remaining_time": "2:49:37"}
320
+ {"current_steps": 640, "total_steps": 928, "loss": 0.40721848607063293, "lr": 1.274184979577963e-05, "epoch": 2.7586206896551726, "percentage": 68.97, "elapsed_time": "6:14:15", "remaining_time": "2:48:25"}
321
+ {"current_steps": 642, "total_steps": 928, "loss": 0.5138668417930603, "lr": 1.2631994672037205e-05, "epoch": 2.7672413793103448, "percentage": 69.18, "elapsed_time": "6:15:18", "remaining_time": "2:47:11"}
322
+ {"current_steps": 644, "total_steps": 928, "loss": 0.6097421646118164, "lr": 1.2522616037655713e-05, "epoch": 2.7758620689655173, "percentage": 69.4, "elapsed_time": "6:16:44", "remaining_time": "2:46:08"}
323
+ {"current_steps": 646, "total_steps": 928, "loss": 0.5522211194038391, "lr": 1.2413719456013231e-05, "epoch": 2.7844827586206895, "percentage": 69.61, "elapsed_time": "6:17:59", "remaining_time": "2:45:00"}
324
+ {"current_steps": 648, "total_steps": 928, "loss": 0.3453619182109833, "lr": 1.2305310465968985e-05, "epoch": 2.793103448275862, "percentage": 69.83, "elapsed_time": "6:19:03", "remaining_time": "2:43:47"}
325
+ {"current_steps": 650, "total_steps": 928, "loss": 0.7121859788894653, "lr": 1.2197394581581561e-05, "epoch": 2.8017241379310347, "percentage": 70.04, "elapsed_time": "6:20:16", "remaining_time": "2:42:38"}
326
+ {"current_steps": 652, "total_steps": 928, "loss": 0.7990239262580872, "lr": 1.2089977291828512e-05, "epoch": 2.810344827586207, "percentage": 70.26, "elapsed_time": "6:21:36", "remaining_time": "2:41:32"}
327
+ {"current_steps": 654, "total_steps": 928, "loss": 0.6081220507621765, "lr": 1.1983064060327098e-05, "epoch": 2.8189655172413794, "percentage": 70.47, "elapsed_time": "6:23:02", "remaining_time": "2:40:28"}
328
+ {"current_steps": 656, "total_steps": 928, "loss": 0.43975335359573364, "lr": 1.187666032505645e-05, "epoch": 2.8275862068965516, "percentage": 70.69, "elapsed_time": "6:24:07", "remaining_time": "2:39:16"}
329
+ {"current_steps": 658, "total_steps": 928, "loss": 0.6137202978134155, "lr": 1.1770771498080921e-05, "epoch": 2.836206896551724, "percentage": 70.91, "elapsed_time": "6:25:33", "remaining_time": "2:38:12"}
330
+ {"current_steps": 660, "total_steps": 928, "loss": 0.20562584698200226, "lr": 1.1665402965274866e-05, "epoch": 2.844827586206897, "percentage": 71.12, "elapsed_time": "6:26:24", "remaining_time": "2:36:54"}
331
+ {"current_steps": 662, "total_steps": 928, "loss": 0.42035165429115295, "lr": 1.1560560086048632e-05, "epoch": 2.853448275862069, "percentage": 71.34, "elapsed_time": "6:27:14", "remaining_time": "2:35:35"}
332
+ {"current_steps": 664, "total_steps": 928, "loss": 0.6786882877349854, "lr": 1.1456248193076027e-05, "epoch": 2.862068965517241, "percentage": 71.55, "elapsed_time": "6:28:26", "remaining_time": "2:34:26"}
333
+ {"current_steps": 666, "total_steps": 928, "loss": 0.34481775760650635, "lr": 1.1352472592023026e-05, "epoch": 2.8706896551724137, "percentage": 71.77, "elapsed_time": "6:29:30", "remaining_time": "2:33:13"}
334
+ {"current_steps": 668, "total_steps": 928, "loss": 0.37077146768569946, "lr": 1.1249238561277957e-05, "epoch": 2.8793103448275863, "percentage": 71.98, "elapsed_time": "6:30:29", "remaining_time": "2:31:59"}
335
+ {"current_steps": 670, "total_steps": 928, "loss": 0.6234573125839233, "lr": 1.1146551351682962e-05, "epoch": 2.887931034482759, "percentage": 72.2, "elapsed_time": "6:31:53", "remaining_time": "2:30:54"}
336
+ {"current_steps": 672, "total_steps": 928, "loss": 0.43646591901779175, "lr": 1.1044416186266985e-05, "epoch": 2.896551724137931, "percentage": 72.41, "elapsed_time": "6:32:50", "remaining_time": "2:29:39"}
337
+ {"current_steps": 674, "total_steps": 928, "loss": 0.6099374890327454, "lr": 1.0942838259980065e-05, "epoch": 2.905172413793103, "percentage": 72.63, "elapsed_time": "6:33:52", "remaining_time": "2:28:25"}
338
+ {"current_steps": 676, "total_steps": 928, "loss": 0.5961918830871582, "lr": 1.0841822739429131e-05, "epoch": 2.913793103448276, "percentage": 72.84, "elapsed_time": "6:35:18", "remaining_time": "2:27:21"}
339
+ {"current_steps": 678, "total_steps": 928, "loss": 0.5247670412063599, "lr": 1.0741374762615181e-05, "epoch": 2.9224137931034484, "percentage": 73.06, "elapsed_time": "6:36:33", "remaining_time": "2:26:13"}
340
+ {"current_steps": 680, "total_steps": 928, "loss": 0.4245750606060028, "lr": 1.0641499438671994e-05, "epoch": 2.9310344827586206, "percentage": 73.28, "elapsed_time": "6:37:38", "remaining_time": "2:25:01"}
341
+ {"current_steps": 682, "total_steps": 928, "loss": 0.21983936429023743, "lr": 1.054220184760619e-05, "epoch": 2.939655172413793, "percentage": 73.49, "elapsed_time": "6:38:21", "remaining_time": "2:23:41"}
342
+ {"current_steps": 684, "total_steps": 928, "loss": 0.3854738771915436, "lr": 1.0443487040038919e-05, "epoch": 2.9482758620689653, "percentage": 73.71, "elapsed_time": "6:39:23", "remaining_time": "2:22:28"}
343
+ {"current_steps": 686, "total_steps": 928, "loss": 0.6782163381576538, "lr": 1.0345360036948912e-05, "epoch": 2.956896551724138, "percentage": 73.92, "elapsed_time": "6:40:37", "remaining_time": "2:21:19"}
344
+ {"current_steps": 688, "total_steps": 928, "loss": 0.5401458740234375, "lr": 1.0247825829417132e-05, "epoch": 2.9655172413793105, "percentage": 74.14, "elapsed_time": "6:41:51", "remaining_time": "2:20:10"}
345
+ {"current_steps": 690, "total_steps": 928, "loss": 0.7114209532737732, "lr": 1.0150889378372878e-05, "epoch": 2.9741379310344827, "percentage": 74.35, "elapsed_time": "6:43:14", "remaining_time": "2:19:05"}
346
+ {"current_steps": 692, "total_steps": 928, "loss": 0.660466730594635, "lr": 1.00545556143415e-05, "epoch": 2.9827586206896552, "percentage": 74.57, "elapsed_time": "6:44:39", "remaining_time": "2:18:00"}
347
+ {"current_steps": 694, "total_steps": 928, "loss": 0.4320341944694519, "lr": 9.958829437193558e-06, "epoch": 2.9913793103448274, "percentage": 74.78, "elapsed_time": "6:45:37", "remaining_time": "2:16:45"}
348
+ {"current_steps": 696, "total_steps": 928, "loss": 0.6856396198272705, "lr": 9.863715715895658e-06, "epoch": 3.0, "percentage": 75.0, "elapsed_time": "6:47:02", "remaining_time": "2:15:40"}
349
+ {"current_steps": 698, "total_steps": 928, "loss": 0.16509434580802917, "lr": 9.769219288262745e-06, "epoch": 3.0086206896551726, "percentage": 75.22, "elapsed_time": "6:48:05", "remaining_time": "2:14:28"}
350
+ {"current_steps": 700, "total_steps": 928, "loss": 0.29928964376449585, "lr": 9.675344960712074e-06, "epoch": 3.0172413793103448, "percentage": 75.43, "elapsed_time": "6:49:30", "remaining_time": "2:13:22"}
351
+ {"current_steps": 702, "total_steps": 928, "loss": 0.25162428617477417, "lr": 9.582097508018724e-06, "epoch": 3.0258620689655173, "percentage": 75.65, "elapsed_time": "6:50:35", "remaining_time": "2:12:11"}
352
+ {"current_steps": 704, "total_steps": 928, "loss": 0.3514169454574585, "lr": 9.489481673072723e-06, "epoch": 3.0344827586206895, "percentage": 75.86, "elapsed_time": "6:51:56", "remaining_time": "2:11:04"}
353
+ {"current_steps": 706, "total_steps": 928, "loss": 0.07074951380491257, "lr": 9.397502166637837e-06, "epoch": 3.043103448275862, "percentage": 76.08, "elapsed_time": "6:52:51", "remaining_time": "2:09:49"}
354
+ {"current_steps": 708, "total_steps": 928, "loss": 0.20599356293678284, "lr": 9.30616366711195e-06, "epoch": 3.0517241379310347, "percentage": 76.29, "elapsed_time": "6:54:18", "remaining_time": "2:08:44"}
355
+ {"current_steps": 710, "total_steps": 928, "loss": 0.11480194330215454, "lr": 9.21547082028908e-06, "epoch": 3.060344827586207, "percentage": 76.51, "elapsed_time": "6:55:19", "remaining_time": "2:07:31"}
356
+ {"current_steps": 712, "total_steps": 928, "loss": 0.26979854702949524, "lr": 9.125428239123133e-06, "epoch": 3.0689655172413794, "percentage": 76.72, "elapsed_time": "6:56:27", "remaining_time": "2:06:20"}
357
+ {"current_steps": 714, "total_steps": 928, "loss": 0.42210009694099426, "lr": 9.036040503493213e-06, "epoch": 3.0775862068965516, "percentage": 76.94, "elapsed_time": "6:57:53", "remaining_time": "2:05:15"}
358
+ {"current_steps": 716, "total_steps": 928, "loss": 0.1675470620393753, "lr": 8.947312159970725e-06, "epoch": 3.086206896551724, "percentage": 77.16, "elapsed_time": "6:58:51", "remaining_time": "2:04:01"}
359
+ {"current_steps": 718, "total_steps": 928, "loss": 0.20833522081375122, "lr": 8.859247721588064e-06, "epoch": 3.0948275862068964, "percentage": 77.37, "elapsed_time": "7:00:12", "remaining_time": "2:02:54"}
360
+ {"current_steps": 720, "total_steps": 928, "loss": 0.16950953006744385, "lr": 8.77185166760914e-06, "epoch": 3.103448275862069, "percentage": 77.59, "elapsed_time": "7:00:53", "remaining_time": "2:01:35"}
361
+ {"current_steps": 722, "total_steps": 928, "loss": 0.12641456723213196, "lr": 8.685128443301465e-06, "epoch": 3.1120689655172415, "percentage": 77.8, "elapsed_time": "7:02:00", "remaining_time": "2:00:24"}
362
+ {"current_steps": 724, "total_steps": 928, "loss": 0.13568329811096191, "lr": 8.599082459710125e-06, "epoch": 3.1206896551724137, "percentage": 78.02, "elapsed_time": "7:03:23", "remaining_time": "1:59:17"}
363
+ {"current_steps": 726, "total_steps": 928, "loss": 0.21239104866981506, "lr": 8.513718093433354e-06, "epoch": 3.1293103448275863, "percentage": 78.23, "elapsed_time": "7:04:43", "remaining_time": "1:58:10"}
364
+ {"current_steps": 728, "total_steps": 928, "loss": 0.2763456702232361, "lr": 8.42903968639999e-06, "epoch": 3.1379310344827585, "percentage": 78.45, "elapsed_time": "7:06:06", "remaining_time": "1:57:03"}
365
+ {"current_steps": 730, "total_steps": 928, "loss": 0.12836386263370514, "lr": 8.345051545648565e-06, "epoch": 3.146551724137931, "percentage": 78.66, "elapsed_time": "7:07:22", "remaining_time": "1:55:54"}
366
+ {"current_steps": 732, "total_steps": 928, "loss": 0.16560682654380798, "lr": 8.261757943108296e-06, "epoch": 3.1551724137931036, "percentage": 78.88, "elapsed_time": "7:08:23", "remaining_time": "1:54:42"}
367
+ {"current_steps": 734, "total_steps": 928, "loss": 0.2081730216741562, "lr": 8.179163115381737e-06, "epoch": 3.163793103448276, "percentage": 79.09, "elapsed_time": "7:09:49", "remaining_time": "1:53:36"}
368
+ {"current_steps": 736, "total_steps": 928, "loss": 0.14392191171646118, "lr": 8.097271263529346e-06, "epoch": 3.1724137931034484, "percentage": 79.31, "elapsed_time": "7:10:54", "remaining_time": "1:52:24"}
369
+ {"current_steps": 738, "total_steps": 928, "loss": 0.15577132999897003, "lr": 8.016086552855771e-06, "epoch": 3.1810344827586206, "percentage": 79.53, "elapsed_time": "7:12:18", "remaining_time": "1:51:17"}
370
+ {"current_steps": 740, "total_steps": 928, "loss": 0.0789552852511406, "lr": 7.935613112698003e-06, "epoch": 3.189655172413793, "percentage": 79.74, "elapsed_time": "7:13:03", "remaining_time": "1:50:01"}
371
+ {"current_steps": 742, "total_steps": 928, "loss": 0.10101716220378876, "lr": 7.855855036215328e-06, "epoch": 3.1982758620689653, "percentage": 79.96, "elapsed_time": "7:14:07", "remaining_time": "1:48:49"}
372
+ {"current_steps": 744, "total_steps": 928, "loss": 0.1658182144165039, "lr": 7.776816380181165e-06, "epoch": 3.206896551724138, "percentage": 80.17, "elapsed_time": "7:15:17", "remaining_time": "1:47:39"}
373
+ {"current_steps": 746, "total_steps": 928, "loss": 0.19248032569885254, "lr": 7.698501164776679e-06, "epoch": 3.2155172413793105, "percentage": 80.39, "elapsed_time": "7:16:30", "remaining_time": "1:46:29"}
374
+ {"current_steps": 748, "total_steps": 928, "loss": 0.21470694243907928, "lr": 7.620913373386356e-06, "epoch": 3.2241379310344827, "percentage": 80.6, "elapsed_time": "7:17:45", "remaining_time": "1:45:20"}
375
+ {"current_steps": 750, "total_steps": 928, "loss": 0.15740104019641876, "lr": 7.5440569523953315e-06, "epoch": 3.2327586206896552, "percentage": 80.82, "elapsed_time": "7:19:10", "remaining_time": "1:44:13"}
376
+ {"current_steps": 752, "total_steps": 928, "loss": 0.18646365404129028, "lr": 7.467935810988729e-06, "epoch": 3.2413793103448274, "percentage": 81.03, "elapsed_time": "7:20:12", "remaining_time": "1:43:01"}
377
+ {"current_steps": 754, "total_steps": 928, "loss": 0.22709967195987701, "lr": 7.392553820952764e-06, "epoch": 3.25, "percentage": 81.25, "elapsed_time": "7:21:36", "remaining_time": "1:41:54"}
378
+ {"current_steps": 756, "total_steps": 928, "loss": 0.1782107949256897, "lr": 7.317914816477865e-06, "epoch": 3.2586206896551726, "percentage": 81.47, "elapsed_time": "7:22:44", "remaining_time": "1:40:43"}
379
+ {"current_steps": 758, "total_steps": 928, "loss": 0.19192323088645935, "lr": 7.244022593963609e-06, "epoch": 3.2672413793103448, "percentage": 81.68, "elapsed_time": "7:24:07", "remaining_time": "1:39:36"}
380
+ {"current_steps": 760, "total_steps": 928, "loss": 0.13779321312904358, "lr": 7.170880911825657e-06, "epoch": 3.2758620689655173, "percentage": 81.9, "elapsed_time": "7:25:15", "remaining_time": "1:38:25"}
381
+ {"current_steps": 762, "total_steps": 928, "loss": 0.24427469074726105, "lr": 7.098493490304566e-06, "epoch": 3.2844827586206895, "percentage": 82.11, "elapsed_time": "7:26:38", "remaining_time": "1:37:18"}
382
+ {"current_steps": 764, "total_steps": 928, "loss": 0.32002437114715576, "lr": 7.026864011276575e-06, "epoch": 3.293103448275862, "percentage": 82.33, "elapsed_time": "7:27:53", "remaining_time": "1:36:08"}
383
+ {"current_steps": 766, "total_steps": 928, "loss": 0.11214806139469147, "lr": 6.955996118066326e-06, "epoch": 3.3017241379310347, "percentage": 82.54, "elapsed_time": "7:28:57", "remaining_time": "1:34:56"}
384
+ {"current_steps": 768, "total_steps": 928, "loss": 0.1987936794757843, "lr": 6.8858934152615646e-06, "epoch": 3.310344827586207, "percentage": 82.76, "elapsed_time": "7:30:12", "remaining_time": "1:33:47"}
385
+ {"current_steps": 770, "total_steps": 928, "loss": 0.058321211487054825, "lr": 6.816559468529773e-06, "epoch": 3.3189655172413794, "percentage": 82.97, "elapsed_time": "7:31:15", "remaining_time": "1:32:35"}
386
+ {"current_steps": 772, "total_steps": 928, "loss": 0.08903615176677704, "lr": 6.747997804436846e-06, "epoch": 3.3275862068965516, "percentage": 83.19, "elapsed_time": "7:32:24", "remaining_time": "1:31:25"}
387
+ {"current_steps": 774, "total_steps": 928, "loss": 0.35364535450935364, "lr": 6.680211910267665e-06, "epoch": 3.336206896551724, "percentage": 83.41, "elapsed_time": "7:33:36", "remaining_time": "1:30:15"}
388
+ {"current_steps": 776, "total_steps": 928, "loss": 0.20209553837776184, "lr": 6.613205233848783e-06, "epoch": 3.344827586206897, "percentage": 83.62, "elapsed_time": "7:35:02", "remaining_time": "1:29:07"}
389
+ {"current_steps": 778, "total_steps": 928, "loss": 0.19359779357910156, "lr": 6.546981183373009e-06, "epoch": 3.353448275862069, "percentage": 83.84, "elapsed_time": "7:36:16", "remaining_time": "1:27:58"}
390
+ {"current_steps": 780, "total_steps": 928, "loss": 0.28171947598457336, "lr": 6.481543127226073e-06, "epoch": 3.3620689655172415, "percentage": 84.05, "elapsed_time": "7:37:38", "remaining_time": "1:26:50"}
391
+ {"current_steps": 782, "total_steps": 928, "loss": 0.1644493192434311, "lr": 6.4168943938153e-06, "epoch": 3.3706896551724137, "percentage": 84.27, "elapsed_time": "7:38:32", "remaining_time": "1:25:36"}
392
+ {"current_steps": 784, "total_steps": 928, "loss": 0.17818251252174377, "lr": 6.353038271400319e-06, "epoch": 3.3793103448275863, "percentage": 84.48, "elapsed_time": "7:39:47", "remaining_time": "1:24:27"}
393
+ {"current_steps": 786, "total_steps": 928, "loss": 0.08893375098705292, "lr": 6.289978007925791e-06, "epoch": 3.3879310344827585, "percentage": 84.7, "elapsed_time": "7:40:51", "remaining_time": "1:23:15"}
394
+ {"current_steps": 788, "total_steps": 928, "loss": 0.16159863770008087, "lr": 6.227716810856235e-06, "epoch": 3.396551724137931, "percentage": 84.91, "elapsed_time": "7:42:16", "remaining_time": "1:22:07"}
395
+ {"current_steps": 790, "total_steps": 928, "loss": 0.19914919137954712, "lr": 6.1662578470128595e-06, "epoch": 3.405172413793103, "percentage": 85.13, "elapsed_time": "7:43:31", "remaining_time": "1:20:58"}
396
+ {"current_steps": 792, "total_steps": 928, "loss": 0.1833517998456955, "lr": 6.105604242412507e-06, "epoch": 3.413793103448276, "percentage": 85.34, "elapsed_time": "7:44:22", "remaining_time": "1:19:44"}
397
+ {"current_steps": 794, "total_steps": 928, "loss": 0.1568892002105713, "lr": 6.0457590821086364e-06, "epoch": 3.4224137931034484, "percentage": 85.56, "elapsed_time": "7:45:30", "remaining_time": "1:18:33"}
398
+ {"current_steps": 796, "total_steps": 928, "loss": 0.5605343580245972, "lr": 5.9867254100344305e-06, "epoch": 3.4310344827586206, "percentage": 85.78, "elapsed_time": "7:46:57", "remaining_time": "1:17:26"}
399
+ {"current_steps": 798, "total_steps": 928, "loss": 0.2542985677719116, "lr": 5.92850622884794e-06, "epoch": 3.439655172413793, "percentage": 85.99, "elapsed_time": "7:48:20", "remaining_time": "1:16:17"}
400
+ {"current_steps": 800, "total_steps": 928, "loss": 0.3042844533920288, "lr": 5.871104499779383e-06, "epoch": 3.4482758620689653, "percentage": 86.21, "elapsed_time": "7:49:47", "remaining_time": "1:15:09"}
401
+ {"current_steps": 802, "total_steps": 928, "loss": 0.23688863217830658, "lr": 5.814523142480514e-06, "epoch": 3.456896551724138, "percentage": 86.42, "elapsed_time": "7:51:10", "remaining_time": "1:14:01"}
402
+ {"current_steps": 804, "total_steps": 928, "loss": 0.1413598358631134, "lr": 5.758765034876124e-06, "epoch": 3.4655172413793105, "percentage": 86.64, "elapsed_time": "7:52:21", "remaining_time": "1:12:51"}
403
+ {"current_steps": 806, "total_steps": 928, "loss": 0.26621344685554504, "lr": 5.703833013017659e-06, "epoch": 3.4741379310344827, "percentage": 86.85, "elapsed_time": "7:53:34", "remaining_time": "1:11:40"}
404
+ {"current_steps": 808, "total_steps": 928, "loss": 0.1856929361820221, "lr": 5.649729870938974e-06, "epoch": 3.4827586206896552, "percentage": 87.07, "elapsed_time": "7:54:42", "remaining_time": "1:10:30"}
405
+ {"current_steps": 810, "total_steps": 928, "loss": 0.11116787791252136, "lr": 5.596458360514197e-06, "epoch": 3.4913793103448274, "percentage": 87.28, "elapsed_time": "7:55:33", "remaining_time": "1:09:16"}
406
+ {"current_steps": 812, "total_steps": 928, "loss": 0.2585083842277527, "lr": 5.544021191317797e-06, "epoch": 3.5, "percentage": 87.5, "elapsed_time": "7:56:55", "remaining_time": "1:08:07"}
407
+ {"current_steps": 814, "total_steps": 928, "loss": 0.24390508234500885, "lr": 5.492421030486723e-06, "epoch": 3.5086206896551726, "percentage": 87.72, "elapsed_time": "7:58:08", "remaining_time": "1:06:57"}
408
+ {"current_steps": 816, "total_steps": 928, "loss": 0.19690856337547302, "lr": 5.441660502584782e-06, "epoch": 3.5172413793103448, "percentage": 87.93, "elapsed_time": "7:59:21", "remaining_time": "1:05:47"}
409
+ {"current_steps": 818, "total_steps": 928, "loss": 0.18222372233867645, "lr": 5.391742189469118e-06, "epoch": 3.5258620689655173, "percentage": 88.15, "elapsed_time": "8:00:48", "remaining_time": "1:04:39"}
410
+ {"current_steps": 820, "total_steps": 928, "loss": 0.14991328120231628, "lr": 5.342668630158901e-06, "epoch": 3.5344827586206895, "percentage": 88.36, "elapsed_time": "8:02:01", "remaining_time": "1:03:29"}
411
+ {"current_steps": 822, "total_steps": 928, "loss": 0.12025367468595505, "lr": 5.294442320706179e-06, "epoch": 3.543103448275862, "percentage": 88.58, "elapsed_time": "8:03:24", "remaining_time": "1:02:20"}
412
+ {"current_steps": 824, "total_steps": 928, "loss": 0.0922561064362526, "lr": 5.247065714068933e-06, "epoch": 3.5517241379310347, "percentage": 88.79, "elapsed_time": "8:04:23", "remaining_time": "1:01:08"}
413
+ {"current_steps": 826, "total_steps": 928, "loss": 0.03818206116557121, "lr": 5.200541219986286e-06, "epoch": 3.560344827586207, "percentage": 89.01, "elapsed_time": "8:05:09", "remaining_time": "0:59:54"}
414
+ {"current_steps": 828, "total_steps": 928, "loss": 0.2238304615020752, "lr": 5.1548712048559655e-06, "epoch": 3.5689655172413794, "percentage": 89.22, "elapsed_time": "8:05:53", "remaining_time": "0:58:40"}
415
+ {"current_steps": 830, "total_steps": 928, "loss": 0.1549633890390396, "lr": 5.110057991613912e-06, "epoch": 3.5775862068965516, "percentage": 89.44, "elapsed_time": "8:07:05", "remaining_time": "0:57:30"}
416
+ {"current_steps": 832, "total_steps": 928, "loss": 0.14927032589912415, "lr": 5.0661038596161515e-06, "epoch": 3.586206896551724, "percentage": 89.66, "elapsed_time": "8:08:07", "remaining_time": "0:56:19"}
417
+ {"current_steps": 834, "total_steps": 928, "loss": 0.2999204397201538, "lr": 5.023011044522834e-06, "epoch": 3.594827586206897, "percentage": 89.87, "elapsed_time": "8:09:23", "remaining_time": "0:55:09"}
418
+ {"current_steps": 836, "total_steps": 928, "loss": 0.20024727284908295, "lr": 4.980781738184549e-06, "epoch": 3.603448275862069, "percentage": 90.09, "elapsed_time": "8:10:25", "remaining_time": "0:53:58"}
419
+ {"current_steps": 838, "total_steps": 928, "loss": 0.13863810896873474, "lr": 4.939418088530811e-06, "epoch": 3.612068965517241, "percentage": 90.3, "elapsed_time": "8:11:50", "remaining_time": "0:52:49"}
420
+ {"current_steps": 840, "total_steps": 928, "loss": 0.18534965813159943, "lr": 4.898922199460831e-06, "epoch": 3.6206896551724137, "percentage": 90.52, "elapsed_time": "8:12:51", "remaining_time": "0:51:37"}
421
+ {"current_steps": 842, "total_steps": 928, "loss": 0.15518294274806976, "lr": 4.859296130736489e-06, "epoch": 3.6293103448275863, "percentage": 90.73, "elapsed_time": "8:13:32", "remaining_time": "0:50:24"}
422
+ {"current_steps": 844, "total_steps": 928, "loss": 0.23298737406730652, "lr": 4.820541897877585e-06, "epoch": 3.637931034482759, "percentage": 90.95, "elapsed_time": "8:14:54", "remaining_time": "0:49:15"}
423
+ {"current_steps": 846, "total_steps": 928, "loss": 0.264419823884964, "lr": 4.782661472059298e-06, "epoch": 3.646551724137931, "percentage": 91.16, "elapsed_time": "8:16:16", "remaining_time": "0:48:06"}
424
+ {"current_steps": 848, "total_steps": 928, "loss": 0.15973711013793945, "lr": 4.745656780011951e-06, "epoch": 3.655172413793103, "percentage": 91.38, "elapsed_time": "8:17:05", "remaining_time": "0:46:53"}
425
+ {"current_steps": 850, "total_steps": 928, "loss": 0.17208503186702728, "lr": 4.709529703922993e-06, "epoch": 3.663793103448276, "percentage": 91.59, "elapsed_time": "8:18:15", "remaining_time": "0:45:43"}
426
+ {"current_steps": 852, "total_steps": 928, "loss": 0.19475609064102173, "lr": 4.674282081341271e-06, "epoch": 3.6724137931034484, "percentage": 91.81, "elapsed_time": "8:19:42", "remaining_time": "0:44:34"}
427
+ {"current_steps": 854, "total_steps": 928, "loss": 0.1562570333480835, "lr": 4.639915705083572e-06, "epoch": 3.6810344827586206, "percentage": 92.03, "elapsed_time": "8:20:46", "remaining_time": "0:43:23"}
428
+ {"current_steps": 856, "total_steps": 928, "loss": 0.15900962054729462, "lr": 4.606432323143412e-06, "epoch": 3.689655172413793, "percentage": 92.24, "elapsed_time": "8:22:11", "remaining_time": "0:42:14"}
429
+ {"current_steps": 858, "total_steps": 928, "loss": 0.22381483018398285, "lr": 4.573833638602159e-06, "epoch": 3.6982758620689653, "percentage": 92.46, "elapsed_time": "8:23:39", "remaining_time": "0:41:05"}
430
+ {"current_steps": 860, "total_steps": 928, "loss": 0.09598782658576965, "lr": 4.542121309542383e-06, "epoch": 3.706896551724138, "percentage": 92.67, "elapsed_time": "8:24:42", "remaining_time": "0:39:54"}
431
+ {"current_steps": 862, "total_steps": 928, "loss": 0.19943147897720337, "lr": 4.511296948963527e-06, "epoch": 3.7155172413793105, "percentage": 92.89, "elapsed_time": "8:25:58", "remaining_time": "0:38:44"}
432
+ {"current_steps": 864, "total_steps": 928, "loss": 0.07272744178771973, "lr": 4.4813621246998765e-06, "epoch": 3.7241379310344827, "percentage": 93.1, "elapsed_time": "8:27:00", "remaining_time": "0:37:33"}
433
+ {"current_steps": 866, "total_steps": 928, "loss": 0.1303609311580658, "lr": 4.45231835934079e-06, "epoch": 3.7327586206896552, "percentage": 93.32, "elapsed_time": "8:27:41", "remaining_time": "0:36:20"}
434
+ {"current_steps": 868, "total_steps": 928, "loss": 0.18073761463165283, "lr": 4.424167130153277e-06, "epoch": 3.7413793103448274, "percentage": 93.53, "elapsed_time": "8:28:43", "remaining_time": "0:35:09"}
435
+ {"current_steps": 870, "total_steps": 928, "loss": 0.12275875359773636, "lr": 4.396909869006847e-06, "epoch": 3.75, "percentage": 93.75, "elapsed_time": "8:29:47", "remaining_time": "0:33:59"}
436
+ {"current_steps": 872, "total_steps": 928, "loss": 0.06019383668899536, "lr": 4.3705479623006866e-06, "epoch": 3.7586206896551726, "percentage": 93.97, "elapsed_time": "8:30:52", "remaining_time": "0:32:48"}
437
+ {"current_steps": 874, "total_steps": 928, "loss": 0.10059908032417297, "lr": 4.345082750893132e-06, "epoch": 3.7672413793103448, "percentage": 94.18, "elapsed_time": "8:31:57", "remaining_time": "0:31:37"}
438
+ {"current_steps": 876, "total_steps": 928, "loss": 0.30081015825271606, "lr": 4.320515530033487e-06, "epoch": 3.7758620689655173, "percentage": 94.4, "elapsed_time": "8:33:17", "remaining_time": "0:30:28"}
439
+ {"current_steps": 878, "total_steps": 928, "loss": 0.30149415135383606, "lr": 4.296847549296115e-06, "epoch": 3.7844827586206895, "percentage": 94.61, "elapsed_time": "8:34:39", "remaining_time": "0:29:18"}
440
+ {"current_steps": 880, "total_steps": 928, "loss": 0.13996456563472748, "lr": 4.274080012516909e-06, "epoch": 3.793103448275862, "percentage": 94.83, "elapsed_time": "8:35:32", "remaining_time": "0:28:07"}
441
+ {"current_steps": 882, "total_steps": 928, "loss": 0.09511252492666245, "lr": 4.2522140777320404e-06, "epoch": 3.8017241379310347, "percentage": 95.04, "elapsed_time": "8:36:35", "remaining_time": "0:26:56"}
442
+ {"current_steps": 884, "total_steps": 928, "loss": 0.23412549495697021, "lr": 4.23125085711907e-06, "epoch": 3.810344827586207, "percentage": 95.26, "elapsed_time": "8:38:00", "remaining_time": "0:25:47"}
443
+ {"current_steps": 886, "total_steps": 928, "loss": 0.160780131816864, "lr": 4.21119141694037e-06, "epoch": 3.8189655172413794, "percentage": 95.47, "elapsed_time": "8:39:05", "remaining_time": "0:24:36"}
444
+ {"current_steps": 888, "total_steps": 928, "loss": 0.11835036426782608, "lr": 4.192036777488896e-06, "epoch": 3.8275862068965516, "percentage": 95.69, "elapsed_time": "8:40:28", "remaining_time": "0:23:26"}
445
+ {"current_steps": 890, "total_steps": 928, "loss": 0.11370360106229782, "lr": 4.173787913036284e-06, "epoch": 3.836206896551724, "percentage": 95.91, "elapsed_time": "8:41:17", "remaining_time": "0:22:15"}
446
+ {"current_steps": 892, "total_steps": 928, "loss": 0.17437399923801422, "lr": 4.156445751783308e-06, "epoch": 3.844827586206897, "percentage": 96.12, "elapsed_time": "8:42:39", "remaining_time": "0:21:05"}
447
+ {"current_steps": 894, "total_steps": 928, "loss": 0.15946733951568604, "lr": 4.140011175812656e-06, "epoch": 3.853448275862069, "percentage": 96.34, "elapsed_time": "8:43:45", "remaining_time": "0:19:55"}
448
+ {"current_steps": 896, "total_steps": 928, "loss": 0.16160649061203003, "lr": 4.124485021044069e-06, "epoch": 3.862068965517241, "percentage": 96.55, "elapsed_time": "8:44:57", "remaining_time": "0:18:44"}
449
+ {"current_steps": 898, "total_steps": 928, "loss": 0.13039463758468628, "lr": 4.1098680771918245e-06, "epoch": 3.8706896551724137, "percentage": 96.77, "elapsed_time": "8:45:59", "remaining_time": "0:17:34"}
450
+ {"current_steps": 900, "total_steps": 928, "loss": 0.16959071159362793, "lr": 4.096161087724573e-06, "epoch": 3.8793103448275863, "percentage": 96.98, "elapsed_time": "8:47:26", "remaining_time": "0:16:24"}
451
+ {"current_steps": 902, "total_steps": 928, "loss": 0.20945216715335846, "lr": 4.0833647498275085e-06, "epoch": 3.887931034482759, "percentage": 97.2, "elapsed_time": "8:48:31", "remaining_time": "0:15:14"}
452
+ {"current_steps": 904, "total_steps": 928, "loss": 0.22212789952754974, "lr": 4.07147971436692e-06, "epoch": 3.896551724137931, "percentage": 97.41, "elapsed_time": "8:49:36", "remaining_time": "0:14:03"}
453
+ {"current_steps": 906, "total_steps": 928, "loss": 0.21481694281101227, "lr": 4.060506585857085e-06, "epoch": 3.905172413793103, "percentage": 97.63, "elapsed_time": "8:51:01", "remaining_time": "0:12:53"}
454
+ {"current_steps": 908, "total_steps": 928, "loss": 0.16184020042419434, "lr": 4.0504459224295174e-06, "epoch": 3.913793103448276, "percentage": 97.84, "elapsed_time": "8:52:04", "remaining_time": "0:11:43"}
455
+ {"current_steps": 910, "total_steps": 928, "loss": 0.1316578984260559, "lr": 4.041298235804577e-06, "epoch": 3.9224137931034484, "percentage": 98.06, "elapsed_time": "8:52:50", "remaining_time": "0:10:32"}
456
+ {"current_steps": 912, "total_steps": 928, "loss": 0.11852114647626877, "lr": 4.0330639912654516e-06, "epoch": 3.9310344827586206, "percentage": 98.28, "elapsed_time": "8:53:56", "remaining_time": "0:09:22"}
457
+ {"current_steps": 914, "total_steps": 928, "loss": 0.16175302863121033, "lr": 4.02574360763448e-06, "epoch": 3.939655172413793, "percentage": 98.49, "elapsed_time": "8:55:22", "remaining_time": "0:08:12"}
458
+ {"current_steps": 916, "total_steps": 928, "loss": 0.16411718726158142, "lr": 4.019337457251857e-06, "epoch": 3.9482758620689653, "percentage": 98.71, "elapsed_time": "8:56:40", "remaining_time": "0:07:01"}
459
+ {"current_steps": 918, "total_steps": 928, "loss": 0.22733992338180542, "lr": 4.013845865956692e-06, "epoch": 3.956896551724138, "percentage": 98.92, "elapsed_time": "8:57:32", "remaining_time": "0:05:51"}
460
+ {"current_steps": 920, "total_steps": 928, "loss": 0.1820860654115677, "lr": 4.00926911307043e-06, "epoch": 3.9655172413793105, "percentage": 99.14, "elapsed_time": "8:58:36", "remaining_time": "0:04:41"}
461
+ {"current_steps": 922, "total_steps": 928, "loss": 0.15438126027584076, "lr": 4.005607431382659e-06, "epoch": 3.9741379310344827, "percentage": 99.35, "elapsed_time": "8:59:47", "remaining_time": "0:03:30"}
462
+ {"current_steps": 924, "total_steps": 928, "loss": 0.17508190870285034, "lr": 4.002861007139253e-06, "epoch": 3.9827586206896552, "percentage": 99.57, "elapsed_time": "9:01:02", "remaining_time": "0:02:20"}
463
+ {"current_steps": 926, "total_steps": 928, "loss": 0.1996261328458786, "lr": 4.001029980032909e-06, "epoch": 3.9913793103448274, "percentage": 99.78, "elapsed_time": "9:02:17", "remaining_time": "0:01:10"}
464
+ {"current_steps": 928, "total_steps": 928, "loss": 0.2806675136089325, "lr": 4.000114443196044e-06, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "9:03:42", "remaining_time": "0:00:00"}
465
+ {"current_steps": 928, "total_steps": 928, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "9:03:42", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,3291 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 928,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.008620689655172414,
14
+ "grad_norm": 0.2490028440952301,
15
+ "learning_rate": 8.510638297872341e-07,
16
+ "loss": 3.7051703929901123,
17
+ "step": 2
18
+ },
19
+ {
20
+ "epoch": 0.017241379310344827,
21
+ "grad_norm": 0.9354268908500671,
22
+ "learning_rate": 2.553191489361702e-06,
23
+ "loss": 2.5432679653167725,
24
+ "step": 4
25
+ },
26
+ {
27
+ "epoch": 0.02586206896551724,
28
+ "grad_norm": 1.1884268522262573,
29
+ "learning_rate": 4.255319148936171e-06,
30
+ "loss": 2.127363681793213,
31
+ "step": 6
32
+ },
33
+ {
34
+ "epoch": 0.034482758620689655,
35
+ "grad_norm": 0.21074028313159943,
36
+ "learning_rate": 5.957446808510638e-06,
37
+ "loss": 1.8431488275527954,
38
+ "step": 8
39
+ },
40
+ {
41
+ "epoch": 0.04310344827586207,
42
+ "grad_norm": 1.0560111999511719,
43
+ "learning_rate": 7.659574468085107e-06,
44
+ "loss": 1.2361581325531006,
45
+ "step": 10
46
+ },
47
+ {
48
+ "epoch": 0.05172413793103448,
49
+ "grad_norm": 0.1471565067768097,
50
+ "learning_rate": 9.361702127659576e-06,
51
+ "loss": 1.7195181846618652,
52
+ "step": 12
53
+ },
54
+ {
55
+ "epoch": 0.0603448275862069,
56
+ "grad_norm": 0.28380081057548523,
57
+ "learning_rate": 1.1063829787234044e-05,
58
+ "loss": 1.7923121452331543,
59
+ "step": 14
60
+ },
61
+ {
62
+ "epoch": 0.06896551724137931,
63
+ "grad_norm": 0.16648316383361816,
64
+ "learning_rate": 1.2765957446808513e-05,
65
+ "loss": 0.7930053472518921,
66
+ "step": 16
67
+ },
68
+ {
69
+ "epoch": 0.07758620689655173,
70
+ "grad_norm": 0.662397027015686,
71
+ "learning_rate": 1.4468085106382981e-05,
72
+ "loss": 0.8174023032188416,
73
+ "step": 18
74
+ },
75
+ {
76
+ "epoch": 0.08620689655172414,
77
+ "grad_norm": 0.13494855165481567,
78
+ "learning_rate": 1.6170212765957446e-05,
79
+ "loss": 1.5101033449172974,
80
+ "step": 20
81
+ },
82
+ {
83
+ "epoch": 0.09482758620689655,
84
+ "grad_norm": 0.1281285285949707,
85
+ "learning_rate": 1.7872340425531915e-05,
86
+ "loss": 1.6496508121490479,
87
+ "step": 22
88
+ },
89
+ {
90
+ "epoch": 0.10344827586206896,
91
+ "grad_norm": 0.19361676275730133,
92
+ "learning_rate": 1.9574468085106384e-05,
93
+ "loss": 1.5436862707138062,
94
+ "step": 24
95
+ },
96
+ {
97
+ "epoch": 0.11206896551724138,
98
+ "grad_norm": 0.26801514625549316,
99
+ "learning_rate": 2.1276595744680852e-05,
100
+ "loss": 1.258086919784546,
101
+ "step": 26
102
+ },
103
+ {
104
+ "epoch": 0.1206896551724138,
105
+ "grad_norm": 0.6263072490692139,
106
+ "learning_rate": 2.2978723404255324e-05,
107
+ "loss": 0.4653662145137787,
108
+ "step": 28
109
+ },
110
+ {
111
+ "epoch": 0.12931034482758622,
112
+ "grad_norm": 0.09440683573484421,
113
+ "learning_rate": 2.468085106382979e-05,
114
+ "loss": 1.5492061376571655,
115
+ "step": 30
116
+ },
117
+ {
118
+ "epoch": 0.13793103448275862,
119
+ "grad_norm": 0.18837958574295044,
120
+ "learning_rate": 2.6382978723404255e-05,
121
+ "loss": 1.1802101135253906,
122
+ "step": 32
123
+ },
124
+ {
125
+ "epoch": 0.14655172413793102,
126
+ "grad_norm": 0.13221898674964905,
127
+ "learning_rate": 2.8085106382978727e-05,
128
+ "loss": 1.5317002534866333,
129
+ "step": 34
130
+ },
131
+ {
132
+ "epoch": 0.15517241379310345,
133
+ "grad_norm": 0.09329993277788162,
134
+ "learning_rate": 2.9787234042553192e-05,
135
+ "loss": 1.4935221672058105,
136
+ "step": 36
137
+ },
138
+ {
139
+ "epoch": 0.16379310344827586,
140
+ "grad_norm": 0.1460646688938141,
141
+ "learning_rate": 3.1489361702127664e-05,
142
+ "loss": 1.4348247051239014,
143
+ "step": 38
144
+ },
145
+ {
146
+ "epoch": 0.1724137931034483,
147
+ "grad_norm": 0.11409169435501099,
148
+ "learning_rate": 3.319148936170213e-05,
149
+ "loss": 1.6168015003204346,
150
+ "step": 40
151
+ },
152
+ {
153
+ "epoch": 0.1810344827586207,
154
+ "grad_norm": 0.212503582239151,
155
+ "learning_rate": 3.48936170212766e-05,
156
+ "loss": 1.7296580076217651,
157
+ "step": 42
158
+ },
159
+ {
160
+ "epoch": 0.1896551724137931,
161
+ "grad_norm": 0.2593461275100708,
162
+ "learning_rate": 3.6595744680851066e-05,
163
+ "loss": 1.0080050230026245,
164
+ "step": 44
165
+ },
166
+ {
167
+ "epoch": 0.19827586206896552,
168
+ "grad_norm": 0.15509171783924103,
169
+ "learning_rate": 3.829787234042554e-05,
170
+ "loss": 1.2960833311080933,
171
+ "step": 46
172
+ },
173
+ {
174
+ "epoch": 0.20689655172413793,
175
+ "grad_norm": 0.11303484439849854,
176
+ "learning_rate": 4e-05,
177
+ "loss": 1.463219404220581,
178
+ "step": 48
179
+ },
180
+ {
181
+ "epoch": 0.21551724137931033,
182
+ "grad_norm": 0.10671091824769974,
183
+ "learning_rate": 3.999954222867108e-05,
184
+ "loss": 1.5820165872573853,
185
+ "step": 50
186
+ },
187
+ {
188
+ "epoch": 0.22413793103448276,
189
+ "grad_norm": 0.2280566394329071,
190
+ "learning_rate": 3.999816893796815e-05,
191
+ "loss": 0.7817078232765198,
192
+ "step": 52
193
+ },
194
+ {
195
+ "epoch": 0.23275862068965517,
196
+ "grad_norm": 0.0907350480556488,
197
+ "learning_rate": 3.9995880197741576e-05,
198
+ "loss": 1.5489472150802612,
199
+ "step": 54
200
+ },
201
+ {
202
+ "epoch": 0.2413793103448276,
203
+ "grad_norm": 0.11344298720359802,
204
+ "learning_rate": 3.999267612440463e-05,
205
+ "loss": 1.447698950767517,
206
+ "step": 56
207
+ },
208
+ {
209
+ "epoch": 0.25,
210
+ "grad_norm": 0.14231091737747192,
211
+ "learning_rate": 3.9988556880927647e-05,
212
+ "loss": 1.4835537672042847,
213
+ "step": 58
214
+ },
215
+ {
216
+ "epoch": 0.25862068965517243,
217
+ "grad_norm": 0.10744752734899521,
218
+ "learning_rate": 3.998352267682969e-05,
219
+ "loss": 1.4630577564239502,
220
+ "step": 60
221
+ },
222
+ {
223
+ "epoch": 0.2672413793103448,
224
+ "grad_norm": 0.33425694704055786,
225
+ "learning_rate": 3.99775737681679e-05,
226
+ "loss": 1.2140964269638062,
227
+ "step": 62
228
+ },
229
+ {
230
+ "epoch": 0.27586206896551724,
231
+ "grad_norm": 0.0645914226770401,
232
+ "learning_rate": 3.9970710457524474e-05,
233
+ "loss": 1.4031065702438354,
234
+ "step": 64
235
+ },
236
+ {
237
+ "epoch": 0.28448275862068967,
238
+ "grad_norm": 0.07510636001825333,
239
+ "learning_rate": 3.9962933093991296e-05,
240
+ "loss": 1.3692433834075928,
241
+ "step": 66
242
+ },
243
+ {
244
+ "epoch": 0.29310344827586204,
245
+ "grad_norm": 0.06527955085039139,
246
+ "learning_rate": 3.995424207315214e-05,
247
+ "loss": 1.3801196813583374,
248
+ "step": 68
249
+ },
250
+ {
251
+ "epoch": 0.3017241379310345,
252
+ "grad_norm": 0.12786391377449036,
253
+ "learning_rate": 3.994463783706259e-05,
254
+ "loss": 1.151434063911438,
255
+ "step": 70
256
+ },
257
+ {
258
+ "epoch": 0.3103448275862069,
259
+ "grad_norm": 0.19329607486724854,
260
+ "learning_rate": 3.9934120874227505e-05,
261
+ "loss": 1.0419366359710693,
262
+ "step": 72
263
+ },
264
+ {
265
+ "epoch": 0.31896551724137934,
266
+ "grad_norm": 0.08128409832715988,
267
+ "learning_rate": 3.992269171957624e-05,
268
+ "loss": 1.2281261682510376,
269
+ "step": 74
270
+ },
271
+ {
272
+ "epoch": 0.3275862068965517,
273
+ "grad_norm": 0.1377028524875641,
274
+ "learning_rate": 3.991035095443538e-05,
275
+ "loss": 1.3645248413085938,
276
+ "step": 76
277
+ },
278
+ {
279
+ "epoch": 0.33620689655172414,
280
+ "grad_norm": 0.05881645902991295,
281
+ "learning_rate": 3.9897099206499204e-05,
282
+ "loss": 1.2278920412063599,
283
+ "step": 78
284
+ },
285
+ {
286
+ "epoch": 0.3448275862068966,
287
+ "grad_norm": 0.15875652432441711,
288
+ "learning_rate": 3.9882937149797735e-05,
289
+ "loss": 1.0160530805587769,
290
+ "step": 80
291
+ },
292
+ {
293
+ "epoch": 0.35344827586206895,
294
+ "grad_norm": 0.09680726379156113,
295
+ "learning_rate": 3.986786550466246e-05,
296
+ "loss": 1.4256402254104614,
297
+ "step": 82
298
+ },
299
+ {
300
+ "epoch": 0.3620689655172414,
301
+ "grad_norm": 0.30841922760009766,
302
+ "learning_rate": 3.985188503768975e-05,
303
+ "loss": 1.318834900856018,
304
+ "step": 84
305
+ },
306
+ {
307
+ "epoch": 0.3706896551724138,
308
+ "grad_norm": 0.14286209642887115,
309
+ "learning_rate": 3.983499656170176e-05,
310
+ "loss": 1.474804162979126,
311
+ "step": 86
312
+ },
313
+ {
314
+ "epoch": 0.3793103448275862,
315
+ "grad_norm": 0.11407710611820221,
316
+ "learning_rate": 3.981720093570517e-05,
317
+ "loss": 1.3328633308410645,
318
+ "step": 88
319
+ },
320
+ {
321
+ "epoch": 0.3879310344827586,
322
+ "grad_norm": 0.4702969193458557,
323
+ "learning_rate": 3.9798499064847466e-05,
324
+ "loss": 1.1703399419784546,
325
+ "step": 90
326
+ },
327
+ {
328
+ "epoch": 0.39655172413793105,
329
+ "grad_norm": 0.13214579224586487,
330
+ "learning_rate": 3.9778891900370905e-05,
331
+ "loss": 1.0334569215774536,
332
+ "step": 92
333
+ },
334
+ {
335
+ "epoch": 0.4051724137931034,
336
+ "grad_norm": 0.17201748490333557,
337
+ "learning_rate": 3.9758380439564117e-05,
338
+ "loss": 1.1891283988952637,
339
+ "step": 94
340
+ },
341
+ {
342
+ "epoch": 0.41379310344827586,
343
+ "grad_norm": 0.2221606969833374,
344
+ "learning_rate": 3.97369657257114e-05,
345
+ "loss": 1.4113942384719849,
346
+ "step": 96
347
+ },
348
+ {
349
+ "epoch": 0.4224137931034483,
350
+ "grad_norm": 0.39266011118888855,
351
+ "learning_rate": 3.9714648848039655e-05,
352
+ "loss": 1.2900447845458984,
353
+ "step": 98
354
+ },
355
+ {
356
+ "epoch": 0.43103448275862066,
357
+ "grad_norm": 0.09964156895875931,
358
+ "learning_rate": 3.969143094166295e-05,
359
+ "loss": 1.270521640777588,
360
+ "step": 100
361
+ },
362
+ {
363
+ "epoch": 0.4396551724137931,
364
+ "grad_norm": 0.5467281937599182,
365
+ "learning_rate": 3.966731318752484e-05,
366
+ "loss": 1.2587134838104248,
367
+ "step": 102
368
+ },
369
+ {
370
+ "epoch": 0.4482758620689655,
371
+ "grad_norm": 0.10468795150518417,
372
+ "learning_rate": 3.964229681233825e-05,
373
+ "loss": 1.3480840921401978,
374
+ "step": 104
375
+ },
376
+ {
377
+ "epoch": 0.45689655172413796,
378
+ "grad_norm": 0.14980582892894745,
379
+ "learning_rate": 3.961638308852309e-05,
380
+ "loss": 1.0994645357131958,
381
+ "step": 106
382
+ },
383
+ {
384
+ "epoch": 0.46551724137931033,
385
+ "grad_norm": 0.22398428618907928,
386
+ "learning_rate": 3.958957333414157e-05,
387
+ "loss": 1.233306646347046,
388
+ "step": 108
389
+ },
390
+ {
391
+ "epoch": 0.47413793103448276,
392
+ "grad_norm": 0.20820066332817078,
393
+ "learning_rate": 3.9561868912831135e-05,
394
+ "loss": 1.2420070171356201,
395
+ "step": 110
396
+ },
397
+ {
398
+ "epoch": 0.4827586206896552,
399
+ "grad_norm": 0.07252290099859238,
400
+ "learning_rate": 3.953327123373506e-05,
401
+ "loss": 1.5031483173370361,
402
+ "step": 112
403
+ },
404
+ {
405
+ "epoch": 0.49137931034482757,
406
+ "grad_norm": 0.30114564299583435,
407
+ "learning_rate": 3.950378175143088e-05,
408
+ "loss": 1.2730351686477661,
409
+ "step": 114
410
+ },
411
+ {
412
+ "epoch": 0.5,
413
+ "grad_norm": 0.5594906806945801,
414
+ "learning_rate": 3.947340196585631e-05,
415
+ "loss": 1.0227445363998413,
416
+ "step": 116
417
+ },
418
+ {
419
+ "epoch": 0.5086206896551724,
420
+ "grad_norm": 0.06979751586914062,
421
+ "learning_rate": 3.944213342223299e-05,
422
+ "loss": 1.3545396327972412,
423
+ "step": 118
424
+ },
425
+ {
426
+ "epoch": 0.5172413793103449,
427
+ "grad_norm": 0.13030360639095306,
428
+ "learning_rate": 3.9409977710987896e-05,
429
+ "loss": 1.332112431526184,
430
+ "step": 120
431
+ },
432
+ {
433
+ "epoch": 0.5258620689655172,
434
+ "grad_norm": 0.0672382041811943,
435
+ "learning_rate": 3.937693646767245e-05,
436
+ "loss": 1.3639230728149414,
437
+ "step": 122
438
+ },
439
+ {
440
+ "epoch": 0.5344827586206896,
441
+ "grad_norm": 0.13222964107990265,
442
+ "learning_rate": 3.9343011372879275e-05,
443
+ "loss": 1.5418974161148071,
444
+ "step": 124
445
+ },
446
+ {
447
+ "epoch": 0.5431034482758621,
448
+ "grad_norm": 0.09776262193918228,
449
+ "learning_rate": 3.930820415215681e-05,
450
+ "loss": 1.6416376829147339,
451
+ "step": 126
452
+ },
453
+ {
454
+ "epoch": 0.5517241379310345,
455
+ "grad_norm": 0.058448825031518936,
456
+ "learning_rate": 3.927251657592146e-05,
457
+ "loss": 1.1251301765441895,
458
+ "step": 128
459
+ },
460
+ {
461
+ "epoch": 0.5603448275862069,
462
+ "grad_norm": 0.052552610635757446,
463
+ "learning_rate": 3.923595045936757e-05,
464
+ "loss": 1.3253697156906128,
465
+ "step": 130
466
+ },
467
+ {
468
+ "epoch": 0.5689655172413793,
469
+ "grad_norm": 0.19469662010669708,
470
+ "learning_rate": 3.919850766237512e-05,
471
+ "loss": 1.263968586921692,
472
+ "step": 132
473
+ },
474
+ {
475
+ "epoch": 0.5775862068965517,
476
+ "grad_norm": 0.17169688642024994,
477
+ "learning_rate": 3.9160190089415106e-05,
478
+ "loss": 0.7878425717353821,
479
+ "step": 134
480
+ },
481
+ {
482
+ "epoch": 0.5862068965517241,
483
+ "grad_norm": 0.07661473006010056,
484
+ "learning_rate": 3.912099968945268e-05,
485
+ "loss": 1.0626349449157715,
486
+ "step": 136
487
+ },
488
+ {
489
+ "epoch": 0.5948275862068966,
490
+ "grad_norm": 0.04046183452010155,
491
+ "learning_rate": 3.908093845584798e-05,
492
+ "loss": 1.05846107006073,
493
+ "step": 138
494
+ },
495
+ {
496
+ "epoch": 0.603448275862069,
497
+ "grad_norm": 0.09312810003757477,
498
+ "learning_rate": 3.9040008426254824e-05,
499
+ "loss": 1.0136967897415161,
500
+ "step": 140
501
+ },
502
+ {
503
+ "epoch": 0.6120689655172413,
504
+ "grad_norm": 0.06341353058815002,
505
+ "learning_rate": 3.8998211682516976e-05,
506
+ "loss": 1.0979063510894775,
507
+ "step": 142
508
+ },
509
+ {
510
+ "epoch": 0.6206896551724138,
511
+ "grad_norm": 0.07724281400442123,
512
+ "learning_rate": 3.895555035056233e-05,
513
+ "loss": 1.2308785915374756,
514
+ "step": 144
515
+ },
516
+ {
517
+ "epoch": 0.6293103448275862,
518
+ "grad_norm": 0.06727743148803711,
519
+ "learning_rate": 3.891202660029474e-05,
520
+ "loss": 1.3009754419326782,
521
+ "step": 146
522
+ },
523
+ {
524
+ "epoch": 0.6379310344827587,
525
+ "grad_norm": 0.06898898631334305,
526
+ "learning_rate": 3.886764264548363e-05,
527
+ "loss": 1.3296358585357666,
528
+ "step": 148
529
+ },
530
+ {
531
+ "epoch": 0.646551724137931,
532
+ "grad_norm": 0.09349878877401352,
533
+ "learning_rate": 3.882240074365145e-05,
534
+ "loss": 1.2398273944854736,
535
+ "step": 150
536
+ },
537
+ {
538
+ "epoch": 0.6551724137931034,
539
+ "grad_norm": 0.17715542018413544,
540
+ "learning_rate": 3.8776303195958814e-05,
541
+ "loss": 0.9495888948440552,
542
+ "step": 152
543
+ },
544
+ {
545
+ "epoch": 0.6637931034482759,
546
+ "grad_norm": 0.7314029932022095,
547
+ "learning_rate": 3.872935234708747e-05,
548
+ "loss": 1.1250660419464111,
549
+ "step": 154
550
+ },
551
+ {
552
+ "epoch": 0.6724137931034483,
553
+ "grad_norm": 0.24987009167671204,
554
+ "learning_rate": 3.868155058512102e-05,
555
+ "loss": 1.2095718383789062,
556
+ "step": 156
557
+ },
558
+ {
559
+ "epoch": 0.6810344827586207,
560
+ "grad_norm": 0.14576859772205353,
561
+ "learning_rate": 3.8632900341423464e-05,
562
+ "loss": 1.295078992843628,
563
+ "step": 158
564
+ },
565
+ {
566
+ "epoch": 0.6896551724137931,
567
+ "grad_norm": 0.11293840408325195,
568
+ "learning_rate": 3.858340409051558e-05,
569
+ "loss": 1.338647723197937,
570
+ "step": 160
571
+ },
572
+ {
573
+ "epoch": 0.6982758620689655,
574
+ "grad_norm": 0.14143511652946472,
575
+ "learning_rate": 3.853306434994895e-05,
576
+ "loss": 1.6283466815948486,
577
+ "step": 162
578
+ },
579
+ {
580
+ "epoch": 0.7068965517241379,
581
+ "grad_norm": 0.06157878786325455,
582
+ "learning_rate": 3.848188368017803e-05,
583
+ "loss": 1.055685043334961,
584
+ "step": 164
585
+ },
586
+ {
587
+ "epoch": 0.7155172413793104,
588
+ "grad_norm": 0.10017464309930801,
589
+ "learning_rate": 3.8429864684429846e-05,
590
+ "loss": 1.3115266561508179,
591
+ "step": 166
592
+ },
593
+ {
594
+ "epoch": 0.7241379310344828,
595
+ "grad_norm": 0.1262677162885666,
596
+ "learning_rate": 3.837701000857159e-05,
597
+ "loss": 1.3648704290390015,
598
+ "step": 168
599
+ },
600
+ {
601
+ "epoch": 0.7327586206896551,
602
+ "grad_norm": 0.1431019902229309,
603
+ "learning_rate": 3.832332234097606e-05,
604
+ "loss": 1.2736293077468872,
605
+ "step": 170
606
+ },
607
+ {
608
+ "epoch": 0.7413793103448276,
609
+ "grad_norm": 0.3130134046077728,
610
+ "learning_rate": 3.8268804412384936e-05,
611
+ "loss": 1.2950979471206665,
612
+ "step": 172
613
+ },
614
+ {
615
+ "epoch": 0.75,
616
+ "grad_norm": 0.10647283494472504,
617
+ "learning_rate": 3.821345899576982e-05,
618
+ "loss": 0.9605190753936768,
619
+ "step": 174
620
+ },
621
+ {
622
+ "epoch": 0.7586206896551724,
623
+ "grad_norm": 0.16715826094150543,
624
+ "learning_rate": 3.815728890619127e-05,
625
+ "loss": 0.7583910226821899,
626
+ "step": 176
627
+ },
628
+ {
629
+ "epoch": 0.7672413793103449,
630
+ "grad_norm": 0.21642223000526428,
631
+ "learning_rate": 3.8100297000655566e-05,
632
+ "loss": 1.3208308219909668,
633
+ "step": 178
634
+ },
635
+ {
636
+ "epoch": 0.7758620689655172,
637
+ "grad_norm": 0.09079600870609283,
638
+ "learning_rate": 3.804248617796941e-05,
639
+ "loss": 1.3078787326812744,
640
+ "step": 180
641
+ },
642
+ {
643
+ "epoch": 0.7844827586206896,
644
+ "grad_norm": 0.08445187658071518,
645
+ "learning_rate": 3.798385937859249e-05,
646
+ "loss": 1.0150703191757202,
647
+ "step": 182
648
+ },
649
+ {
650
+ "epoch": 0.7931034482758621,
651
+ "grad_norm": 0.07777632772922516,
652
+ "learning_rate": 3.79244195844879e-05,
653
+ "loss": 0.8019794225692749,
654
+ "step": 184
655
+ },
656
+ {
657
+ "epoch": 0.8017241379310345,
658
+ "grad_norm": 0.0645889863371849,
659
+ "learning_rate": 3.7864169818970465e-05,
660
+ "loss": 1.423434853553772,
661
+ "step": 186
662
+ },
663
+ {
664
+ "epoch": 0.8103448275862069,
665
+ "grad_norm": 0.14517554640769958,
666
+ "learning_rate": 3.7803113146553e-05,
667
+ "loss": 1.5573757886886597,
668
+ "step": 188
669
+ },
670
+ {
671
+ "epoch": 0.8189655172413793,
672
+ "grad_norm": 0.08607929199934006,
673
+ "learning_rate": 3.774125267279041e-05,
674
+ "loss": 1.2926381826400757,
675
+ "step": 190
676
+ },
677
+ {
678
+ "epoch": 0.8275862068965517,
679
+ "grad_norm": 0.07077501714229584,
680
+ "learning_rate": 3.767859154412171e-05,
681
+ "loss": 1.2611286640167236,
682
+ "step": 192
683
+ },
684
+ {
685
+ "epoch": 0.8362068965517241,
686
+ "grad_norm": 0.08681857585906982,
687
+ "learning_rate": 3.7615132947710036e-05,
688
+ "loss": 1.3005847930908203,
689
+ "step": 194
690
+ },
691
+ {
692
+ "epoch": 0.8448275862068966,
693
+ "grad_norm": 0.15100322663784027,
694
+ "learning_rate": 3.755088011128049e-05,
695
+ "loss": 1.3176685571670532,
696
+ "step": 196
697
+ },
698
+ {
699
+ "epoch": 0.853448275862069,
700
+ "grad_norm": 0.1908247470855713,
701
+ "learning_rate": 3.7485836302956016e-05,
702
+ "loss": 1.2926079034805298,
703
+ "step": 198
704
+ },
705
+ {
706
+ "epoch": 0.8620689655172413,
707
+ "grad_norm": 0.11072229593992233,
708
+ "learning_rate": 3.7420004831091105e-05,
709
+ "loss": 1.290125846862793,
710
+ "step": 200
711
+ },
712
+ {
713
+ "epoch": 0.8706896551724138,
714
+ "grad_norm": 0.10266255587339401,
715
+ "learning_rate": 3.735338904410358e-05,
716
+ "loss": 1.333167552947998,
717
+ "step": 202
718
+ },
719
+ {
720
+ "epoch": 0.8793103448275862,
721
+ "grad_norm": 0.12212225794792175,
722
+ "learning_rate": 3.728599233030425e-05,
723
+ "loss": 0.894460916519165,
724
+ "step": 204
725
+ },
726
+ {
727
+ "epoch": 0.8879310344827587,
728
+ "grad_norm": 0.07256551086902618,
729
+ "learning_rate": 3.72178181177246e-05,
730
+ "loss": 1.2725472450256348,
731
+ "step": 206
732
+ },
733
+ {
734
+ "epoch": 0.896551724137931,
735
+ "grad_norm": 0.12780705094337463,
736
+ "learning_rate": 3.714886987394238e-05,
737
+ "loss": 1.3160998821258545,
738
+ "step": 208
739
+ },
740
+ {
741
+ "epoch": 0.9051724137931034,
742
+ "grad_norm": 0.9440551400184631,
743
+ "learning_rate": 3.70791511059053e-05,
744
+ "loss": 0.6522892117500305,
745
+ "step": 210
746
+ },
747
+ {
748
+ "epoch": 0.9137931034482759,
749
+ "grad_norm": 0.09754825383424759,
750
+ "learning_rate": 3.700866535975256e-05,
751
+ "loss": 0.9885504245758057,
752
+ "step": 212
753
+ },
754
+ {
755
+ "epoch": 0.9224137931034483,
756
+ "grad_norm": 0.4700559675693512,
757
+ "learning_rate": 3.69374162206346e-05,
758
+ "loss": 1.2210713624954224,
759
+ "step": 214
760
+ },
761
+ {
762
+ "epoch": 0.9310344827586207,
763
+ "grad_norm": 0.15046778321266174,
764
+ "learning_rate": 3.6865407312530635e-05,
765
+ "loss": 1.2765154838562012,
766
+ "step": 216
767
+ },
768
+ {
769
+ "epoch": 0.9396551724137931,
770
+ "grad_norm": 0.11940351128578186,
771
+ "learning_rate": 3.67926422980644e-05,
772
+ "loss": 1.0038096904754639,
773
+ "step": 218
774
+ },
775
+ {
776
+ "epoch": 0.9482758620689655,
777
+ "grad_norm": 0.11115628480911255,
778
+ "learning_rate": 3.671912487831783e-05,
779
+ "loss": 0.9949377179145813,
780
+ "step": 220
781
+ },
782
+ {
783
+ "epoch": 0.9568965517241379,
784
+ "grad_norm": 0.07952730357646942,
785
+ "learning_rate": 3.664485879264279e-05,
786
+ "loss": 1.3989291191101074,
787
+ "step": 222
788
+ },
789
+ {
790
+ "epoch": 0.9655172413793104,
791
+ "grad_norm": 0.09136626124382019,
792
+ "learning_rate": 3.656984781847094e-05,
793
+ "loss": 0.9785476922988892,
794
+ "step": 224
795
+ },
796
+ {
797
+ "epoch": 0.9741379310344828,
798
+ "grad_norm": 0.057783424854278564,
799
+ "learning_rate": 3.649409577112152e-05,
800
+ "loss": 0.9384239315986633,
801
+ "step": 226
802
+ },
803
+ {
804
+ "epoch": 0.9827586206896551,
805
+ "grad_norm": 0.07152694463729858,
806
+ "learning_rate": 3.641760650360736e-05,
807
+ "loss": 1.2639554738998413,
808
+ "step": 228
809
+ },
810
+ {
811
+ "epoch": 0.9913793103448276,
812
+ "grad_norm": 0.33899354934692383,
813
+ "learning_rate": 3.634038390643886e-05,
814
+ "loss": 0.9682251811027527,
815
+ "step": 230
816
+ },
817
+ {
818
+ "epoch": 1.0,
819
+ "grad_norm": 0.07940968871116638,
820
+ "learning_rate": 3.626243190742613e-05,
821
+ "loss": 0.870396614074707,
822
+ "step": 232
823
+ },
824
+ {
825
+ "epoch": 1.0086206896551724,
826
+ "grad_norm": 0.12955711781978607,
827
+ "learning_rate": 3.618375447147918e-05,
828
+ "loss": 0.9028921723365784,
829
+ "step": 234
830
+ },
831
+ {
832
+ "epoch": 1.0172413793103448,
833
+ "grad_norm": 0.18779002130031586,
834
+ "learning_rate": 3.6104355600406284e-05,
835
+ "loss": 0.7830209136009216,
836
+ "step": 236
837
+ },
838
+ {
839
+ "epoch": 1.0258620689655173,
840
+ "grad_norm": 0.9097674489021301,
841
+ "learning_rate": 3.6024239332710415e-05,
842
+ "loss": 0.8674835562705994,
843
+ "step": 238
844
+ },
845
+ {
846
+ "epoch": 1.0344827586206897,
847
+ "grad_norm": 0.27647626399993896,
848
+ "learning_rate": 3.5943409743383826e-05,
849
+ "loss": 0.6074855327606201,
850
+ "step": 240
851
+ },
852
+ {
853
+ "epoch": 1.043103448275862,
854
+ "grad_norm": 0.11918103694915771,
855
+ "learning_rate": 3.586187094370079e-05,
856
+ "loss": 1.2056649923324585,
857
+ "step": 242
858
+ },
859
+ {
860
+ "epoch": 1.0517241379310345,
861
+ "grad_norm": 0.11876146495342255,
862
+ "learning_rate": 3.577962708100851e-05,
863
+ "loss": 0.39286842942237854,
864
+ "step": 244
865
+ },
866
+ {
867
+ "epoch": 1.0603448275862069,
868
+ "grad_norm": 0.12174484878778458,
869
+ "learning_rate": 3.569668233851613e-05,
870
+ "loss": 0.9662060141563416,
871
+ "step": 246
872
+ },
873
+ {
874
+ "epoch": 1.0689655172413792,
875
+ "grad_norm": 0.06474713236093521,
876
+ "learning_rate": 3.561304093508198e-05,
877
+ "loss": 0.9487460851669312,
878
+ "step": 248
879
+ },
880
+ {
881
+ "epoch": 1.0775862068965518,
882
+ "grad_norm": 0.1427253633737564,
883
+ "learning_rate": 3.552870712499898e-05,
884
+ "loss": 0.8610017895698547,
885
+ "step": 250
886
+ },
887
+ {
888
+ "epoch": 1.0862068965517242,
889
+ "grad_norm": 0.1900222897529602,
890
+ "learning_rate": 3.54436851977783e-05,
891
+ "loss": 0.448039710521698,
892
+ "step": 252
893
+ },
894
+ {
895
+ "epoch": 1.0948275862068966,
896
+ "grad_norm": 0.22907602787017822,
897
+ "learning_rate": 3.535797947793111e-05,
898
+ "loss": 0.8761284351348877,
899
+ "step": 254
900
+ },
901
+ {
902
+ "epoch": 1.103448275862069,
903
+ "grad_norm": 0.15932975709438324,
904
+ "learning_rate": 3.527159432474865e-05,
905
+ "loss": 0.8125666975975037,
906
+ "step": 256
907
+ },
908
+ {
909
+ "epoch": 1.1120689655172413,
910
+ "grad_norm": 0.349263459444046,
911
+ "learning_rate": 3.518453413208053e-05,
912
+ "loss": 0.6569501757621765,
913
+ "step": 258
914
+ },
915
+ {
916
+ "epoch": 1.1206896551724137,
917
+ "grad_norm": 0.14833390712738037,
918
+ "learning_rate": 3.509680332811121e-05,
919
+ "loss": 0.7028253674507141,
920
+ "step": 260
921
+ },
922
+ {
923
+ "epoch": 1.1293103448275863,
924
+ "grad_norm": 0.05770875886082649,
925
+ "learning_rate": 3.5008406375134756e-05,
926
+ "loss": 1.1632713079452515,
927
+ "step": 262
928
+ },
929
+ {
930
+ "epoch": 1.1379310344827587,
931
+ "grad_norm": 0.055273283272981644,
932
+ "learning_rate": 3.491934776932791e-05,
933
+ "loss": 0.8371788263320923,
934
+ "step": 264
935
+ },
936
+ {
937
+ "epoch": 1.146551724137931,
938
+ "grad_norm": 0.05901844799518585,
939
+ "learning_rate": 3.482963204052139e-05,
940
+ "loss": 1.0003291368484497,
941
+ "step": 266
942
+ },
943
+ {
944
+ "epoch": 1.1551724137931034,
945
+ "grad_norm": 0.06095067784190178,
946
+ "learning_rate": 3.473926375196943e-05,
947
+ "loss": 0.9226457476615906,
948
+ "step": 268
949
+ },
950
+ {
951
+ "epoch": 1.1637931034482758,
952
+ "grad_norm": 0.07438337802886963,
953
+ "learning_rate": 3.464824750011779e-05,
954
+ "loss": 1.1149680614471436,
955
+ "step": 270
956
+ },
957
+ {
958
+ "epoch": 1.1724137931034484,
959
+ "grad_norm": 0.06853963434696198,
960
+ "learning_rate": 3.455658791436985e-05,
961
+ "loss": 1.0302170515060425,
962
+ "step": 272
963
+ },
964
+ {
965
+ "epoch": 1.1810344827586208,
966
+ "grad_norm": 0.0973813459277153,
967
+ "learning_rate": 3.446428965685121e-05,
968
+ "loss": 0.7569156289100647,
969
+ "step": 274
970
+ },
971
+ {
972
+ "epoch": 1.1896551724137931,
973
+ "grad_norm": 0.056327857077121735,
974
+ "learning_rate": 3.437135742217254e-05,
975
+ "loss": 0.3794441223144531,
976
+ "step": 276
977
+ },
978
+ {
979
+ "epoch": 1.1982758620689655,
980
+ "grad_norm": 0.07669582962989807,
981
+ "learning_rate": 3.427779593719079e-05,
982
+ "loss": 1.1280944347381592,
983
+ "step": 278
984
+ },
985
+ {
986
+ "epoch": 1.206896551724138,
987
+ "grad_norm": 0.058203186839818954,
988
+ "learning_rate": 3.4183609960768764e-05,
989
+ "loss": 0.9517163634300232,
990
+ "step": 280
991
+ },
992
+ {
993
+ "epoch": 1.2155172413793103,
994
+ "grad_norm": 0.06986816227436066,
995
+ "learning_rate": 3.4088804283533094e-05,
996
+ "loss": 0.6671708822250366,
997
+ "step": 282
998
+ },
999
+ {
1000
+ "epoch": 1.2241379310344827,
1001
+ "grad_norm": 0.08936108648777008,
1002
+ "learning_rate": 3.399338372763055e-05,
1003
+ "loss": 0.7694864869117737,
1004
+ "step": 284
1005
+ },
1006
+ {
1007
+ "epoch": 1.2327586206896552,
1008
+ "grad_norm": 0.12140902131795883,
1009
+ "learning_rate": 3.389735314648274e-05,
1010
+ "loss": 0.8068587183952332,
1011
+ "step": 286
1012
+ },
1013
+ {
1014
+ "epoch": 1.2413793103448276,
1015
+ "grad_norm": 0.04297681525349617,
1016
+ "learning_rate": 3.380071742453931e-05,
1017
+ "loss": 0.40287792682647705,
1018
+ "step": 288
1019
+ },
1020
+ {
1021
+ "epoch": 1.25,
1022
+ "grad_norm": 0.11908482015132904,
1023
+ "learning_rate": 3.370348147702949e-05,
1024
+ "loss": 1.0401684045791626,
1025
+ "step": 290
1026
+ },
1027
+ {
1028
+ "epoch": 1.2586206896551724,
1029
+ "grad_norm": 0.057489216327667236,
1030
+ "learning_rate": 3.360565024971202e-05,
1031
+ "loss": 0.8889655470848083,
1032
+ "step": 292
1033
+ },
1034
+ {
1035
+ "epoch": 1.2672413793103448,
1036
+ "grad_norm": 0.15609467029571533,
1037
+ "learning_rate": 3.350722871862368e-05,
1038
+ "loss": 0.9757481813430786,
1039
+ "step": 294
1040
+ },
1041
+ {
1042
+ "epoch": 1.2758620689655173,
1043
+ "grad_norm": 0.1248452365398407,
1044
+ "learning_rate": 3.340822188982616e-05,
1045
+ "loss": 0.7736673355102539,
1046
+ "step": 296
1047
+ },
1048
+ {
1049
+ "epoch": 1.2844827586206897,
1050
+ "grad_norm": 0.09071607887744904,
1051
+ "learning_rate": 3.330863479915138e-05,
1052
+ "loss": 1.0164954662322998,
1053
+ "step": 298
1054
+ },
1055
+ {
1056
+ "epoch": 1.293103448275862,
1057
+ "grad_norm": 0.06744378060102463,
1058
+ "learning_rate": 3.320847251194546e-05,
1059
+ "loss": 0.9475960731506348,
1060
+ "step": 300
1061
+ },
1062
+ {
1063
+ "epoch": 1.3017241379310345,
1064
+ "grad_norm": 0.07189597189426422,
1065
+ "learning_rate": 3.310774012281099e-05,
1066
+ "loss": 0.6825069785118103,
1067
+ "step": 302
1068
+ },
1069
+ {
1070
+ "epoch": 1.3103448275862069,
1071
+ "grad_norm": 0.07518645375967026,
1072
+ "learning_rate": 3.300644275534793e-05,
1073
+ "loss": 0.5717735290527344,
1074
+ "step": 304
1075
+ },
1076
+ {
1077
+ "epoch": 1.3189655172413794,
1078
+ "grad_norm": 0.09223438799381256,
1079
+ "learning_rate": 3.290458556189299e-05,
1080
+ "loss": 1.3711295127868652,
1081
+ "step": 306
1082
+ },
1083
+ {
1084
+ "epoch": 1.3275862068965516,
1085
+ "grad_norm": 0.14958783984184265,
1086
+ "learning_rate": 3.2802173723257604e-05,
1087
+ "loss": 0.6421374082565308,
1088
+ "step": 308
1089
+ },
1090
+ {
1091
+ "epoch": 1.3362068965517242,
1092
+ "grad_norm": 0.1238432452082634,
1093
+ "learning_rate": 3.2699212448464385e-05,
1094
+ "loss": 0.9758880734443665,
1095
+ "step": 310
1096
+ },
1097
+ {
1098
+ "epoch": 1.3448275862068966,
1099
+ "grad_norm": 0.06866496056318283,
1100
+ "learning_rate": 3.259570697448217e-05,
1101
+ "loss": 0.9329778552055359,
1102
+ "step": 312
1103
+ },
1104
+ {
1105
+ "epoch": 1.353448275862069,
1106
+ "grad_norm": 0.10072822868824005,
1107
+ "learning_rate": 3.249166256595967e-05,
1108
+ "loss": 1.2179062366485596,
1109
+ "step": 314
1110
+ },
1111
+ {
1112
+ "epoch": 1.3620689655172413,
1113
+ "grad_norm": 0.06878109276294708,
1114
+ "learning_rate": 3.2387084514957675e-05,
1115
+ "loss": 1.3471888303756714,
1116
+ "step": 316
1117
+ },
1118
+ {
1119
+ "epoch": 1.3706896551724137,
1120
+ "grad_norm": 0.06524922698736191,
1121
+ "learning_rate": 3.2281978140679894e-05,
1122
+ "loss": 0.9441757202148438,
1123
+ "step": 318
1124
+ },
1125
+ {
1126
+ "epoch": 1.3793103448275863,
1127
+ "grad_norm": 0.11429349333047867,
1128
+ "learning_rate": 3.21763487892024e-05,
1129
+ "loss": 0.7498874664306641,
1130
+ "step": 320
1131
+ },
1132
+ {
1133
+ "epoch": 1.3879310344827587,
1134
+ "grad_norm": 0.07464733719825745,
1135
+ "learning_rate": 3.207020183320171e-05,
1136
+ "loss": 1.4824918508529663,
1137
+ "step": 322
1138
+ },
1139
+ {
1140
+ "epoch": 1.396551724137931,
1141
+ "grad_norm": 0.11774388700723648,
1142
+ "learning_rate": 3.196354267168149e-05,
1143
+ "loss": 0.49022743105888367,
1144
+ "step": 324
1145
+ },
1146
+ {
1147
+ "epoch": 1.4051724137931034,
1148
+ "grad_norm": 0.16186490654945374,
1149
+ "learning_rate": 3.185637672969799e-05,
1150
+ "loss": 0.6543675661087036,
1151
+ "step": 326
1152
+ },
1153
+ {
1154
+ "epoch": 1.4137931034482758,
1155
+ "grad_norm": 0.10584386438131332,
1156
+ "learning_rate": 3.1748709458084045e-05,
1157
+ "loss": 0.8541685342788696,
1158
+ "step": 328
1159
+ },
1160
+ {
1161
+ "epoch": 1.4224137931034484,
1162
+ "grad_norm": 0.07407426834106445,
1163
+ "learning_rate": 3.1640546333171894e-05,
1164
+ "loss": 0.7656717300415039,
1165
+ "step": 330
1166
+ },
1167
+ {
1168
+ "epoch": 1.4310344827586206,
1169
+ "grad_norm": 0.16052280366420746,
1170
+ "learning_rate": 3.153189285651458e-05,
1171
+ "loss": 0.6957482695579529,
1172
+ "step": 332
1173
+ },
1174
+ {
1175
+ "epoch": 1.4396551724137931,
1176
+ "grad_norm": 0.13904230296611786,
1177
+ "learning_rate": 3.142275455460614e-05,
1178
+ "loss": 0.6638420224189758,
1179
+ "step": 334
1180
+ },
1181
+ {
1182
+ "epoch": 1.4482758620689655,
1183
+ "grad_norm": 0.11371087282896042,
1184
+ "learning_rate": 3.131313697860053e-05,
1185
+ "loss": 0.7661845088005066,
1186
+ "step": 336
1187
+ },
1188
+ {
1189
+ "epoch": 1.456896551724138,
1190
+ "grad_norm": 0.11575423926115036,
1191
+ "learning_rate": 3.120304570402924e-05,
1192
+ "loss": 1.1160173416137695,
1193
+ "step": 338
1194
+ },
1195
+ {
1196
+ "epoch": 1.4655172413793103,
1197
+ "grad_norm": 0.22061830759048462,
1198
+ "learning_rate": 3.1092486330517714e-05,
1199
+ "loss": 1.384441614151001,
1200
+ "step": 340
1201
+ },
1202
+ {
1203
+ "epoch": 1.4741379310344827,
1204
+ "grad_norm": 0.12608060240745544,
1205
+ "learning_rate": 3.098146448150055e-05,
1206
+ "loss": 1.145660638809204,
1207
+ "step": 342
1208
+ },
1209
+ {
1210
+ "epoch": 1.4827586206896552,
1211
+ "grad_norm": 0.06820492446422577,
1212
+ "learning_rate": 3.086998580393547e-05,
1213
+ "loss": 0.9891381859779358,
1214
+ "step": 344
1215
+ },
1216
+ {
1217
+ "epoch": 1.4913793103448276,
1218
+ "grad_norm": 0.11383876949548721,
1219
+ "learning_rate": 3.075805596801605e-05,
1220
+ "loss": 0.6093174815177917,
1221
+ "step": 346
1222
+ },
1223
+ {
1224
+ "epoch": 1.5,
1225
+ "grad_norm": 0.2013673037290573,
1226
+ "learning_rate": 3.0645680666883374e-05,
1227
+ "loss": 0.9298641681671143,
1228
+ "step": 348
1229
+ },
1230
+ {
1231
+ "epoch": 1.5086206896551724,
1232
+ "grad_norm": 0.08500847220420837,
1233
+ "learning_rate": 3.053286561633644e-05,
1234
+ "loss": 0.9974504113197327,
1235
+ "step": 350
1236
+ },
1237
+ {
1238
+ "epoch": 1.5172413793103448,
1239
+ "grad_norm": 0.14812250435352325,
1240
+ "learning_rate": 3.041961655454143e-05,
1241
+ "loss": 0.9739059209823608,
1242
+ "step": 352
1243
+ },
1244
+ {
1245
+ "epoch": 1.5258620689655173,
1246
+ "grad_norm": 0.11965472251176834,
1247
+ "learning_rate": 3.030593924173984e-05,
1248
+ "loss": 1.133984088897705,
1249
+ "step": 354
1250
+ },
1251
+ {
1252
+ "epoch": 1.5344827586206895,
1253
+ "grad_norm": 0.36424365639686584,
1254
+ "learning_rate": 3.0191839459955514e-05,
1255
+ "loss": 0.8807175755500793,
1256
+ "step": 356
1257
+ },
1258
+ {
1259
+ "epoch": 1.543103448275862,
1260
+ "grad_norm": 0.05107448622584343,
1261
+ "learning_rate": 3.0077323012700534e-05,
1262
+ "loss": 0.8361281156539917,
1263
+ "step": 358
1264
+ },
1265
+ {
1266
+ "epoch": 1.5517241379310345,
1267
+ "grad_norm": 0.09036049991846085,
1268
+ "learning_rate": 2.996239572468003e-05,
1269
+ "loss": 1.2387166023254395,
1270
+ "step": 360
1271
+ },
1272
+ {
1273
+ "epoch": 1.5603448275862069,
1274
+ "grad_norm": 0.06331617385149002,
1275
+ "learning_rate": 2.984706344149595e-05,
1276
+ "loss": 1.0467900037765503,
1277
+ "step": 362
1278
+ },
1279
+ {
1280
+ "epoch": 1.5689655172413794,
1281
+ "grad_norm": 0.06433523446321487,
1282
+ "learning_rate": 2.9731332029349667e-05,
1283
+ "loss": 1.0626113414764404,
1284
+ "step": 364
1285
+ },
1286
+ {
1287
+ "epoch": 1.5775862068965516,
1288
+ "grad_norm": 0.09752818942070007,
1289
+ "learning_rate": 2.961520737474367e-05,
1290
+ "loss": 1.0128107070922852,
1291
+ "step": 366
1292
+ },
1293
+ {
1294
+ "epoch": 1.5862068965517242,
1295
+ "grad_norm": 0.05285457894206047,
1296
+ "learning_rate": 2.9498695384182123e-05,
1297
+ "loss": 0.9877223968505859,
1298
+ "step": 368
1299
+ },
1300
+ {
1301
+ "epoch": 1.5948275862068966,
1302
+ "grad_norm": 0.05934653803706169,
1303
+ "learning_rate": 2.9381801983870435e-05,
1304
+ "loss": 0.9603118300437927,
1305
+ "step": 370
1306
+ },
1307
+ {
1308
+ "epoch": 1.603448275862069,
1309
+ "grad_norm": 0.22097375988960266,
1310
+ "learning_rate": 2.9264533119413866e-05,
1311
+ "loss": 1.081476092338562,
1312
+ "step": 372
1313
+ },
1314
+ {
1315
+ "epoch": 1.6120689655172413,
1316
+ "grad_norm": 0.10628407448530197,
1317
+ "learning_rate": 2.914689475551506e-05,
1318
+ "loss": 0.7714329957962036,
1319
+ "step": 374
1320
+ },
1321
+ {
1322
+ "epoch": 1.6206896551724137,
1323
+ "grad_norm": 0.10955756157636642,
1324
+ "learning_rate": 2.902889287567072e-05,
1325
+ "loss": 0.9913143515586853,
1326
+ "step": 376
1327
+ },
1328
+ {
1329
+ "epoch": 1.6293103448275863,
1330
+ "grad_norm": 0.07451241463422775,
1331
+ "learning_rate": 2.8910533481867195e-05,
1332
+ "loss": 1.1765313148498535,
1333
+ "step": 378
1334
+ },
1335
+ {
1336
+ "epoch": 1.6379310344827587,
1337
+ "grad_norm": 0.07359088957309723,
1338
+ "learning_rate": 2.879182259427528e-05,
1339
+ "loss": 0.7655573487281799,
1340
+ "step": 380
1341
+ },
1342
+ {
1343
+ "epoch": 1.646551724137931,
1344
+ "grad_norm": 0.13642138242721558,
1345
+ "learning_rate": 2.8672766250943947e-05,
1346
+ "loss": 1.3452657461166382,
1347
+ "step": 382
1348
+ },
1349
+ {
1350
+ "epoch": 1.6551724137931034,
1351
+ "grad_norm": 0.08765345811843872,
1352
+ "learning_rate": 2.8553370507493246e-05,
1353
+ "loss": 0.9972445964813232,
1354
+ "step": 384
1355
+ },
1356
+ {
1357
+ "epoch": 1.6637931034482758,
1358
+ "grad_norm": 0.0989682674407959,
1359
+ "learning_rate": 2.8433641436806306e-05,
1360
+ "loss": 0.8845785856246948,
1361
+ "step": 386
1362
+ },
1363
+ {
1364
+ "epoch": 1.6724137931034484,
1365
+ "grad_norm": 0.06875207275152206,
1366
+ "learning_rate": 2.8313585128720444e-05,
1367
+ "loss": 1.3110713958740234,
1368
+ "step": 388
1369
+ },
1370
+ {
1371
+ "epoch": 1.6810344827586206,
1372
+ "grad_norm": 0.13957612216472626,
1373
+ "learning_rate": 2.8193207689717393e-05,
1374
+ "loss": 0.8128502368927002,
1375
+ "step": 390
1376
+ },
1377
+ {
1378
+ "epoch": 1.6896551724137931,
1379
+ "grad_norm": 0.6921377778053284,
1380
+ "learning_rate": 2.807251524261275e-05,
1381
+ "loss": 0.6244351863861084,
1382
+ "step": 392
1383
+ },
1384
+ {
1385
+ "epoch": 1.6982758620689655,
1386
+ "grad_norm": 0.30923035740852356,
1387
+ "learning_rate": 2.7951513926244484e-05,
1388
+ "loss": 1.127506136894226,
1389
+ "step": 394
1390
+ },
1391
+ {
1392
+ "epoch": 1.706896551724138,
1393
+ "grad_norm": 0.0620148703455925,
1394
+ "learning_rate": 2.7830209895160764e-05,
1395
+ "loss": 1.042289137840271,
1396
+ "step": 396
1397
+ },
1398
+ {
1399
+ "epoch": 1.7155172413793105,
1400
+ "grad_norm": 0.16145341098308563,
1401
+ "learning_rate": 2.770860931930687e-05,
1402
+ "loss": 1.0570330619812012,
1403
+ "step": 398
1404
+ },
1405
+ {
1406
+ "epoch": 1.7241379310344827,
1407
+ "grad_norm": 0.09267118573188782,
1408
+ "learning_rate": 2.7586718383711367e-05,
1409
+ "loss": 0.9959380626678467,
1410
+ "step": 400
1411
+ },
1412
+ {
1413
+ "epoch": 1.7327586206896552,
1414
+ "grad_norm": 0.07319535315036774,
1415
+ "learning_rate": 2.7464543288171558e-05,
1416
+ "loss": 1.0200254917144775,
1417
+ "step": 402
1418
+ },
1419
+ {
1420
+ "epoch": 1.7413793103448276,
1421
+ "grad_norm": 0.055158186703920364,
1422
+ "learning_rate": 2.7342090246938076e-05,
1423
+ "loss": 0.6205574870109558,
1424
+ "step": 404
1425
+ },
1426
+ {
1427
+ "epoch": 1.75,
1428
+ "grad_norm": 0.07343259453773499,
1429
+ "learning_rate": 2.721936548839887e-05,
1430
+ "loss": 0.8922735452651978,
1431
+ "step": 406
1432
+ },
1433
+ {
1434
+ "epoch": 1.7586206896551724,
1435
+ "grad_norm": 0.06107189506292343,
1436
+ "learning_rate": 2.709637525476236e-05,
1437
+ "loss": 0.6991145014762878,
1438
+ "step": 408
1439
+ },
1440
+ {
1441
+ "epoch": 1.7672413793103448,
1442
+ "grad_norm": 0.0519319549202919,
1443
+ "learning_rate": 2.697312580173995e-05,
1444
+ "loss": 0.8093492984771729,
1445
+ "step": 410
1446
+ },
1447
+ {
1448
+ "epoch": 1.7758620689655173,
1449
+ "grad_norm": 0.07292782515287399,
1450
+ "learning_rate": 2.684962339822785e-05,
1451
+ "loss": 0.7507970929145813,
1452
+ "step": 412
1453
+ },
1454
+ {
1455
+ "epoch": 1.7844827586206895,
1456
+ "grad_norm": 0.07456238567829132,
1457
+ "learning_rate": 2.672587432598823e-05,
1458
+ "loss": 0.5883830189704895,
1459
+ "step": 414
1460
+ },
1461
+ {
1462
+ "epoch": 1.793103448275862,
1463
+ "grad_norm": 0.11243204772472382,
1464
+ "learning_rate": 2.6601884879329653e-05,
1465
+ "loss": 0.7915773391723633,
1466
+ "step": 416
1467
+ },
1468
+ {
1469
+ "epoch": 1.8017241379310345,
1470
+ "grad_norm": 0.07653719186782837,
1471
+ "learning_rate": 2.6477661364786996e-05,
1472
+ "loss": 1.0269769430160522,
1473
+ "step": 418
1474
+ },
1475
+ {
1476
+ "epoch": 1.8103448275862069,
1477
+ "grad_norm": 0.14341171085834503,
1478
+ "learning_rate": 2.635321010080062e-05,
1479
+ "loss": 1.053789496421814,
1480
+ "step": 420
1481
+ },
1482
+ {
1483
+ "epoch": 1.8189655172413794,
1484
+ "grad_norm": 0.12033911049365997,
1485
+ "learning_rate": 2.6228537417395034e-05,
1486
+ "loss": 1.158492088317871,
1487
+ "step": 422
1488
+ },
1489
+ {
1490
+ "epoch": 1.8275862068965516,
1491
+ "grad_norm": 0.047955527901649475,
1492
+ "learning_rate": 2.61036496558569e-05,
1493
+ "loss": 0.9592758417129517,
1494
+ "step": 424
1495
+ },
1496
+ {
1497
+ "epoch": 1.8362068965517242,
1498
+ "grad_norm": 0.088678739964962,
1499
+ "learning_rate": 2.59785531684125e-05,
1500
+ "loss": 0.6086317300796509,
1501
+ "step": 426
1502
+ },
1503
+ {
1504
+ "epoch": 1.8448275862068966,
1505
+ "grad_norm": 0.07942725718021393,
1506
+ "learning_rate": 2.585325431790464e-05,
1507
+ "loss": 1.0528879165649414,
1508
+ "step": 428
1509
+ },
1510
+ {
1511
+ "epoch": 1.853448275862069,
1512
+ "grad_norm": 0.0694958046078682,
1513
+ "learning_rate": 2.572775947746903e-05,
1514
+ "loss": 1.0576783418655396,
1515
+ "step": 430
1516
+ },
1517
+ {
1518
+ "epoch": 1.8620689655172413,
1519
+ "grad_norm": 0.17858955264091492,
1520
+ "learning_rate": 2.5602075030210096e-05,
1521
+ "loss": 0.9204137325286865,
1522
+ "step": 432
1523
+ },
1524
+ {
1525
+ "epoch": 1.8706896551724137,
1526
+ "grad_norm": 0.296277791261673,
1527
+ "learning_rate": 2.5476207368876334e-05,
1528
+ "loss": 1.114011287689209,
1529
+ "step": 434
1530
+ },
1531
+ {
1532
+ "epoch": 1.8793103448275863,
1533
+ "grad_norm": 0.07735295593738556,
1534
+ "learning_rate": 2.535016289553514e-05,
1535
+ "loss": 0.7933326363563538,
1536
+ "step": 436
1537
+ },
1538
+ {
1539
+ "epoch": 1.8879310344827587,
1540
+ "grad_norm": 0.12477041035890579,
1541
+ "learning_rate": 2.5223948021247197e-05,
1542
+ "loss": 0.9807726144790649,
1543
+ "step": 438
1544
+ },
1545
+ {
1546
+ "epoch": 1.896551724137931,
1547
+ "grad_norm": 0.09196372330188751,
1548
+ "learning_rate": 2.509756916574035e-05,
1549
+ "loss": 1.0345503091812134,
1550
+ "step": 440
1551
+ },
1552
+ {
1553
+ "epoch": 1.9051724137931034,
1554
+ "grad_norm": 0.06840290129184723,
1555
+ "learning_rate": 2.4971032757083123e-05,
1556
+ "loss": 1.1201728582382202,
1557
+ "step": 442
1558
+ },
1559
+ {
1560
+ "epoch": 1.9137931034482758,
1561
+ "grad_norm": 0.11144451051950455,
1562
+ "learning_rate": 2.4844345231357734e-05,
1563
+ "loss": 0.28341731429100037,
1564
+ "step": 444
1565
+ },
1566
+ {
1567
+ "epoch": 1.9224137931034484,
1568
+ "grad_norm": 0.14570969343185425,
1569
+ "learning_rate": 2.4717513032332736e-05,
1570
+ "loss": 0.7789583206176758,
1571
+ "step": 446
1572
+ },
1573
+ {
1574
+ "epoch": 1.9310344827586206,
1575
+ "grad_norm": 0.05790058895945549,
1576
+ "learning_rate": 2.4590542611135274e-05,
1577
+ "loss": 1.012285590171814,
1578
+ "step": 448
1579
+ },
1580
+ {
1581
+ "epoch": 1.9396551724137931,
1582
+ "grad_norm": 0.05153496563434601,
1583
+ "learning_rate": 2.446344042592295e-05,
1584
+ "loss": 1.0196033716201782,
1585
+ "step": 450
1586
+ },
1587
+ {
1588
+ "epoch": 1.9482758620689655,
1589
+ "grad_norm": 0.057060956954956055,
1590
+ "learning_rate": 2.433621294155535e-05,
1591
+ "loss": 0.8052966594696045,
1592
+ "step": 452
1593
+ },
1594
+ {
1595
+ "epoch": 1.956896551724138,
1596
+ "grad_norm": 0.0602966733276844,
1597
+ "learning_rate": 2.420886662926521e-05,
1598
+ "loss": 0.9915321469306946,
1599
+ "step": 454
1600
+ },
1601
+ {
1602
+ "epoch": 1.9655172413793105,
1603
+ "grad_norm": 0.07094614952802658,
1604
+ "learning_rate": 2.4081407966329256e-05,
1605
+ "loss": 0.9689676761627197,
1606
+ "step": 456
1607
+ },
1608
+ {
1609
+ "epoch": 1.9741379310344827,
1610
+ "grad_norm": 0.08627466857433319,
1611
+ "learning_rate": 2.3953843435738775e-05,
1612
+ "loss": 0.41972166299819946,
1613
+ "step": 458
1614
+ },
1615
+ {
1616
+ "epoch": 1.9827586206896552,
1617
+ "grad_norm": 0.10626411437988281,
1618
+ "learning_rate": 2.3826179525869836e-05,
1619
+ "loss": 1.1633706092834473,
1620
+ "step": 460
1621
+ },
1622
+ {
1623
+ "epoch": 1.9913793103448276,
1624
+ "grad_norm": 0.15631678700447083,
1625
+ "learning_rate": 2.36984227301533e-05,
1626
+ "loss": 0.7487952709197998,
1627
+ "step": 462
1628
+ },
1629
+ {
1630
+ "epoch": 2.0,
1631
+ "grad_norm": 0.16628113389015198,
1632
+ "learning_rate": 2.3570579546744504e-05,
1633
+ "loss": 0.8847077488899231,
1634
+ "step": 464
1635
+ },
1636
+ {
1637
+ "epoch": 2.0086206896551726,
1638
+ "grad_norm": 0.06411660462617874,
1639
+ "learning_rate": 2.3442656478192794e-05,
1640
+ "loss": 0.484560489654541,
1641
+ "step": 466
1642
+ },
1643
+ {
1644
+ "epoch": 2.0172413793103448,
1645
+ "grad_norm": 0.3941573202610016,
1646
+ "learning_rate": 2.331466003111073e-05,
1647
+ "loss": 0.6984850764274597,
1648
+ "step": 468
1649
+ },
1650
+ {
1651
+ "epoch": 2.0258620689655173,
1652
+ "grad_norm": 0.044237978756427765,
1653
+ "learning_rate": 2.318659671584316e-05,
1654
+ "loss": 0.4863373935222626,
1655
+ "step": 470
1656
+ },
1657
+ {
1658
+ "epoch": 2.0344827586206895,
1659
+ "grad_norm": 0.0645633190870285,
1660
+ "learning_rate": 2.305847304613609e-05,
1661
+ "loss": 0.4588513970375061,
1662
+ "step": 472
1663
+ },
1664
+ {
1665
+ "epoch": 2.043103448275862,
1666
+ "grad_norm": 0.05587729066610336,
1667
+ "learning_rate": 2.293029553880536e-05,
1668
+ "loss": 0.4486234486103058,
1669
+ "step": 474
1670
+ },
1671
+ {
1672
+ "epoch": 2.0517241379310347,
1673
+ "grad_norm": 0.06679260730743408,
1674
+ "learning_rate": 2.280207071340517e-05,
1675
+ "loss": 0.5298870205879211,
1676
+ "step": 476
1677
+ },
1678
+ {
1679
+ "epoch": 2.060344827586207,
1680
+ "grad_norm": 0.08075322210788727,
1681
+ "learning_rate": 2.26738050918965e-05,
1682
+ "loss": 0.4382156729698181,
1683
+ "step": 478
1684
+ },
1685
+ {
1686
+ "epoch": 2.0689655172413794,
1687
+ "grad_norm": 0.06546280533075333,
1688
+ "learning_rate": 2.2545505198315346e-05,
1689
+ "loss": 0.5762298107147217,
1690
+ "step": 480
1691
+ },
1692
+ {
1693
+ "epoch": 2.0775862068965516,
1694
+ "grad_norm": 0.11915218830108643,
1695
+ "learning_rate": 2.2417177558440907e-05,
1696
+ "loss": 0.36859992146492004,
1697
+ "step": 482
1698
+ },
1699
+ {
1700
+ "epoch": 2.086206896551724,
1701
+ "grad_norm": 0.22198820114135742,
1702
+ "learning_rate": 2.2288828699463652e-05,
1703
+ "loss": 0.5293700098991394,
1704
+ "step": 484
1705
+ },
1706
+ {
1707
+ "epoch": 2.0948275862068964,
1708
+ "grad_norm": 0.0842965617775917,
1709
+ "learning_rate": 2.2160465149653337e-05,
1710
+ "loss": 0.49147215485572815,
1711
+ "step": 486
1712
+ },
1713
+ {
1714
+ "epoch": 2.103448275862069,
1715
+ "grad_norm": 0.11753598600625992,
1716
+ "learning_rate": 2.203209343802692e-05,
1717
+ "loss": 0.5180780291557312,
1718
+ "step": 488
1719
+ },
1720
+ {
1721
+ "epoch": 2.1120689655172415,
1722
+ "grad_norm": 0.37540075182914734,
1723
+ "learning_rate": 2.1903720094016537e-05,
1724
+ "loss": 0.581203818321228,
1725
+ "step": 490
1726
+ },
1727
+ {
1728
+ "epoch": 2.1206896551724137,
1729
+ "grad_norm": 0.062044426798820496,
1730
+ "learning_rate": 2.1775351647137323e-05,
1731
+ "loss": 0.4889185130596161,
1732
+ "step": 492
1733
+ },
1734
+ {
1735
+ "epoch": 2.1293103448275863,
1736
+ "grad_norm": 0.07434380799531937,
1737
+ "learning_rate": 2.1646994626655332e-05,
1738
+ "loss": 0.6391059756278992,
1739
+ "step": 494
1740
+ },
1741
+ {
1742
+ "epoch": 2.1379310344827585,
1743
+ "grad_norm": 0.10223301500082016,
1744
+ "learning_rate": 2.151865556125544e-05,
1745
+ "loss": 0.6237853169441223,
1746
+ "step": 496
1747
+ },
1748
+ {
1749
+ "epoch": 2.146551724137931,
1750
+ "grad_norm": 0.14267216622829437,
1751
+ "learning_rate": 2.1390340978709254e-05,
1752
+ "loss": 0.36577755212783813,
1753
+ "step": 498
1754
+ },
1755
+ {
1756
+ "epoch": 2.1551724137931036,
1757
+ "grad_norm": 0.13929963111877441,
1758
+ "learning_rate": 2.1262057405543115e-05,
1759
+ "loss": 0.49633127450942993,
1760
+ "step": 500
1761
+ },
1762
+ {
1763
+ "epoch": 2.163793103448276,
1764
+ "grad_norm": 0.05517968162894249,
1765
+ "learning_rate": 2.1133811366706097e-05,
1766
+ "loss": 0.38259175419807434,
1767
+ "step": 502
1768
+ },
1769
+ {
1770
+ "epoch": 2.1724137931034484,
1771
+ "grad_norm": 0.058835044503211975,
1772
+ "learning_rate": 2.100560938523817e-05,
1773
+ "loss": 0.4427034258842468,
1774
+ "step": 504
1775
+ },
1776
+ {
1777
+ "epoch": 2.1810344827586206,
1778
+ "grad_norm": 0.15045633912086487,
1779
+ "learning_rate": 2.0877457981938364e-05,
1780
+ "loss": 0.6942803263664246,
1781
+ "step": 506
1782
+ },
1783
+ {
1784
+ "epoch": 2.189655172413793,
1785
+ "grad_norm": 2.292686700820923,
1786
+ "learning_rate": 2.074936367503317e-05,
1787
+ "loss": 0.5671365261077881,
1788
+ "step": 508
1789
+ },
1790
+ {
1791
+ "epoch": 2.1982758620689653,
1792
+ "grad_norm": 0.046695832163095474,
1793
+ "learning_rate": 2.0621332979844904e-05,
1794
+ "loss": 0.6063480377197266,
1795
+ "step": 510
1796
+ },
1797
+ {
1798
+ "epoch": 2.206896551724138,
1799
+ "grad_norm": 0.16905461251735687,
1800
+ "learning_rate": 2.0493372408460425e-05,
1801
+ "loss": 0.6027957201004028,
1802
+ "step": 512
1803
+ },
1804
+ {
1805
+ "epoch": 2.2155172413793105,
1806
+ "grad_norm": 0.06160572171211243,
1807
+ "learning_rate": 2.0365488469399795e-05,
1808
+ "loss": 0.6078309416770935,
1809
+ "step": 514
1810
+ },
1811
+ {
1812
+ "epoch": 2.2241379310344827,
1813
+ "grad_norm": 0.07821284979581833,
1814
+ "learning_rate": 2.0237687667285345e-05,
1815
+ "loss": 0.3304949402809143,
1816
+ "step": 516
1817
+ },
1818
+ {
1819
+ "epoch": 2.2327586206896552,
1820
+ "grad_norm": 0.34748536348342896,
1821
+ "learning_rate": 2.010997650251072e-05,
1822
+ "loss": 0.12825970351696014,
1823
+ "step": 518
1824
+ },
1825
+ {
1826
+ "epoch": 2.2413793103448274,
1827
+ "grad_norm": 0.11893010139465332,
1828
+ "learning_rate": 1.9982361470910342e-05,
1829
+ "loss": 0.1828547865152359,
1830
+ "step": 520
1831
+ },
1832
+ {
1833
+ "epoch": 2.25,
1834
+ "grad_norm": 0.12491466104984283,
1835
+ "learning_rate": 1.9854849063428926e-05,
1836
+ "loss": 0.6522985696792603,
1837
+ "step": 522
1838
+ },
1839
+ {
1840
+ "epoch": 2.2586206896551726,
1841
+ "grad_norm": 0.15903355181217194,
1842
+ "learning_rate": 1.9727445765791405e-05,
1843
+ "loss": 0.47932472825050354,
1844
+ "step": 524
1845
+ },
1846
+ {
1847
+ "epoch": 2.2672413793103448,
1848
+ "grad_norm": 0.09779471158981323,
1849
+ "learning_rate": 1.9600158058172974e-05,
1850
+ "loss": 0.4181676208972931,
1851
+ "step": 526
1852
+ },
1853
+ {
1854
+ "epoch": 2.2758620689655173,
1855
+ "grad_norm": 0.07378951460123062,
1856
+ "learning_rate": 1.9472992414869534e-05,
1857
+ "loss": 0.46739447116851807,
1858
+ "step": 528
1859
+ },
1860
+ {
1861
+ "epoch": 2.2844827586206895,
1862
+ "grad_norm": 0.04063527286052704,
1863
+ "learning_rate": 1.9345955303968365e-05,
1864
+ "loss": 0.38251054286956787,
1865
+ "step": 530
1866
+ },
1867
+ {
1868
+ "epoch": 2.293103448275862,
1869
+ "grad_norm": 0.08258794993162155,
1870
+ "learning_rate": 1.9219053187019144e-05,
1871
+ "loss": 0.4366922080516815,
1872
+ "step": 532
1873
+ },
1874
+ {
1875
+ "epoch": 2.3017241379310347,
1876
+ "grad_norm": 0.09015543758869171,
1877
+ "learning_rate": 1.909229251870528e-05,
1878
+ "loss": 0.4965798556804657,
1879
+ "step": 534
1880
+ },
1881
+ {
1882
+ "epoch": 2.310344827586207,
1883
+ "grad_norm": 0.08743222802877426,
1884
+ "learning_rate": 1.8965679746515628e-05,
1885
+ "loss": 0.43146276473999023,
1886
+ "step": 536
1887
+ },
1888
+ {
1889
+ "epoch": 2.3189655172413794,
1890
+ "grad_norm": 0.084476038813591,
1891
+ "learning_rate": 1.88392213104165e-05,
1892
+ "loss": 0.2771337330341339,
1893
+ "step": 538
1894
+ },
1895
+ {
1896
+ "epoch": 2.3275862068965516,
1897
+ "grad_norm": 0.07576002180576324,
1898
+ "learning_rate": 1.8712923642524175e-05,
1899
+ "loss": 0.36878013610839844,
1900
+ "step": 540
1901
+ },
1902
+ {
1903
+ "epoch": 2.336206896551724,
1904
+ "grad_norm": 0.10497633367776871,
1905
+ "learning_rate": 1.858679316677767e-05,
1906
+ "loss": 0.6058629751205444,
1907
+ "step": 542
1908
+ },
1909
+ {
1910
+ "epoch": 2.344827586206897,
1911
+ "grad_norm": 0.13856923580169678,
1912
+ "learning_rate": 1.8460836298612056e-05,
1913
+ "loss": 0.6428977251052856,
1914
+ "step": 544
1915
+ },
1916
+ {
1917
+ "epoch": 2.353448275862069,
1918
+ "grad_norm": 0.1172226220369339,
1919
+ "learning_rate": 1.8335059444632078e-05,
1920
+ "loss": 0.2821408212184906,
1921
+ "step": 546
1922
+ },
1923
+ {
1924
+ "epoch": 2.3620689655172415,
1925
+ "grad_norm": 0.1798970252275467,
1926
+ "learning_rate": 1.820946900228639e-05,
1927
+ "loss": 0.8290093541145325,
1928
+ "step": 548
1929
+ },
1930
+ {
1931
+ "epoch": 2.3706896551724137,
1932
+ "grad_norm": 0.2738807499408722,
1933
+ "learning_rate": 1.808407135954204e-05,
1934
+ "loss": 0.5475698709487915,
1935
+ "step": 550
1936
+ },
1937
+ {
1938
+ "epoch": 2.3793103448275863,
1939
+ "grad_norm": 0.20505401492118835,
1940
+ "learning_rate": 1.7958872894559666e-05,
1941
+ "loss": 0.6245191693305969,
1942
+ "step": 552
1943
+ },
1944
+ {
1945
+ "epoch": 2.3879310344827585,
1946
+ "grad_norm": 0.05477019026875496,
1947
+ "learning_rate": 1.7833879975368994e-05,
1948
+ "loss": 0.5108689665794373,
1949
+ "step": 554
1950
+ },
1951
+ {
1952
+ "epoch": 2.396551724137931,
1953
+ "grad_norm": 0.09034960716962814,
1954
+ "learning_rate": 1.7709098959545015e-05,
1955
+ "loss": 0.5519805550575256,
1956
+ "step": 556
1957
+ },
1958
+ {
1959
+ "epoch": 2.405172413793103,
1960
+ "grad_norm": 0.1560261845588684,
1961
+ "learning_rate": 1.758453619388453e-05,
1962
+ "loss": 0.4397192597389221,
1963
+ "step": 558
1964
+ },
1965
+ {
1966
+ "epoch": 2.413793103448276,
1967
+ "grad_norm": 0.118907131254673,
1968
+ "learning_rate": 1.7460198014083424e-05,
1969
+ "loss": 0.38739266991615295,
1970
+ "step": 560
1971
+ },
1972
+ {
1973
+ "epoch": 2.4224137931034484,
1974
+ "grad_norm": 0.23784895241260529,
1975
+ "learning_rate": 1.733609074441433e-05,
1976
+ "loss": 0.5064358711242676,
1977
+ "step": 562
1978
+ },
1979
+ {
1980
+ "epoch": 2.4310344827586206,
1981
+ "grad_norm": 0.09993483871221542,
1982
+ "learning_rate": 1.7212220697405003e-05,
1983
+ "loss": 0.540324330329895,
1984
+ "step": 564
1985
+ },
1986
+ {
1987
+ "epoch": 2.439655172413793,
1988
+ "grad_norm": 0.7780280113220215,
1989
+ "learning_rate": 1.7088594173517225e-05,
1990
+ "loss": 0.5431786179542542,
1991
+ "step": 566
1992
+ },
1993
+ {
1994
+ "epoch": 2.4482758620689653,
1995
+ "grad_norm": 0.14646178483963013,
1996
+ "learning_rate": 1.6965217460826345e-05,
1997
+ "loss": 0.3365917205810547,
1998
+ "step": 568
1999
+ },
2000
+ {
2001
+ "epoch": 2.456896551724138,
2002
+ "grad_norm": 0.07466763257980347,
2003
+ "learning_rate": 1.6842096834701443e-05,
2004
+ "loss": 0.6636412739753723,
2005
+ "step": 570
2006
+ },
2007
+ {
2008
+ "epoch": 2.4655172413793105,
2009
+ "grad_norm": 0.3850714862346649,
2010
+ "learning_rate": 1.6719238557486143e-05,
2011
+ "loss": 0.3930183947086334,
2012
+ "step": 572
2013
+ },
2014
+ {
2015
+ "epoch": 2.4741379310344827,
2016
+ "grad_norm": 0.12653613090515137,
2017
+ "learning_rate": 1.6596648878180088e-05,
2018
+ "loss": 0.4772527813911438,
2019
+ "step": 574
2020
+ },
2021
+ {
2022
+ "epoch": 2.4827586206896552,
2023
+ "grad_norm": 0.10766978561878204,
2024
+ "learning_rate": 1.647433403212112e-05,
2025
+ "loss": 0.6689369082450867,
2026
+ "step": 576
2027
+ },
2028
+ {
2029
+ "epoch": 2.4913793103448274,
2030
+ "grad_norm": 0.1643172800540924,
2031
+ "learning_rate": 1.635230024066807e-05,
2032
+ "loss": 0.5050515532493591,
2033
+ "step": 578
2034
+ },
2035
+ {
2036
+ "epoch": 2.5,
2037
+ "grad_norm": 0.06176433712244034,
2038
+ "learning_rate": 1.6230553710884373e-05,
2039
+ "loss": 0.6936325430870056,
2040
+ "step": 580
2041
+ },
2042
+ {
2043
+ "epoch": 2.5086206896551726,
2044
+ "grad_norm": 0.17540457844734192,
2045
+ "learning_rate": 1.610910063522233e-05,
2046
+ "loss": 0.5566367506980896,
2047
+ "step": 582
2048
+ },
2049
+ {
2050
+ "epoch": 2.5172413793103448,
2051
+ "grad_norm": 0.09146937727928162,
2052
+ "learning_rate": 1.598794719120816e-05,
2053
+ "loss": 0.5264196991920471,
2054
+ "step": 584
2055
+ },
2056
+ {
2057
+ "epoch": 2.5258620689655173,
2058
+ "grad_norm": 0.08665334433317184,
2059
+ "learning_rate": 1.5867099541127737e-05,
2060
+ "loss": 0.4999127686023712,
2061
+ "step": 586
2062
+ },
2063
+ {
2064
+ "epoch": 2.5344827586206895,
2065
+ "grad_norm": 0.05140522122383118,
2066
+ "learning_rate": 1.5746563831713236e-05,
2067
+ "loss": 0.5660111308097839,
2068
+ "step": 588
2069
+ },
2070
+ {
2071
+ "epoch": 2.543103448275862,
2072
+ "grad_norm": 0.08618345856666565,
2073
+ "learning_rate": 1.56263461938304e-05,
2074
+ "loss": 0.7160353064537048,
2075
+ "step": 590
2076
+ },
2077
+ {
2078
+ "epoch": 2.5517241379310347,
2079
+ "grad_norm": 0.05319703742861748,
2080
+ "learning_rate": 1.5506452742166796e-05,
2081
+ "loss": 0.575738251209259,
2082
+ "step": 592
2083
+ },
2084
+ {
2085
+ "epoch": 2.560344827586207,
2086
+ "grad_norm": 0.29011279344558716,
2087
+ "learning_rate": 1.5386889574920692e-05,
2088
+ "loss": 0.35511380434036255,
2089
+ "step": 594
2090
+ },
2091
+ {
2092
+ "epoch": 2.5689655172413794,
2093
+ "grad_norm": 0.07296542078256607,
2094
+ "learning_rate": 1.5267662773491e-05,
2095
+ "loss": 0.40391749143600464,
2096
+ "step": 596
2097
+ },
2098
+ {
2099
+ "epoch": 2.5775862068965516,
2100
+ "grad_norm": 0.09713292866945267,
2101
+ "learning_rate": 1.514877840216785e-05,
2102
+ "loss": 0.5037810802459717,
2103
+ "step": 598
2104
+ },
2105
+ {
2106
+ "epoch": 2.586206896551724,
2107
+ "grad_norm": 0.1726667881011963,
2108
+ "learning_rate": 1.5030242507824215e-05,
2109
+ "loss": 0.6312216520309448,
2110
+ "step": 600
2111
+ },
2112
+ {
2113
+ "epoch": 2.594827586206897,
2114
+ "grad_norm": 0.0342765673995018,
2115
+ "learning_rate": 1.4912061119608292e-05,
2116
+ "loss": 0.39456382393836975,
2117
+ "step": 602
2118
+ },
2119
+ {
2120
+ "epoch": 2.603448275862069,
2121
+ "grad_norm": 0.45015275478363037,
2122
+ "learning_rate": 1.4794240248636885e-05,
2123
+ "loss": 0.5595788359642029,
2124
+ "step": 604
2125
+ },
2126
+ {
2127
+ "epoch": 2.612068965517241,
2128
+ "grad_norm": 0.10634768009185791,
2129
+ "learning_rate": 1.4676785887689614e-05,
2130
+ "loss": 0.41876575350761414,
2131
+ "step": 606
2132
+ },
2133
+ {
2134
+ "epoch": 2.6206896551724137,
2135
+ "grad_norm": 0.06522602587938309,
2136
+ "learning_rate": 1.4559704010904145e-05,
2137
+ "loss": 0.6346225142478943,
2138
+ "step": 608
2139
+ },
2140
+ {
2141
+ "epoch": 2.6293103448275863,
2142
+ "grad_norm": 0.24831700325012207,
2143
+ "learning_rate": 1.444300057347229e-05,
2144
+ "loss": 0.5777739882469177,
2145
+ "step": 610
2146
+ },
2147
+ {
2148
+ "epoch": 2.637931034482759,
2149
+ "grad_norm": 0.06677041202783585,
2150
+ "learning_rate": 1.432668151133712e-05,
2151
+ "loss": 0.5916672945022583,
2152
+ "step": 612
2153
+ },
2154
+ {
2155
+ "epoch": 2.646551724137931,
2156
+ "grad_norm": 0.09093949943780899,
2157
+ "learning_rate": 1.4210752740891032e-05,
2158
+ "loss": 0.5175487995147705,
2159
+ "step": 614
2160
+ },
2161
+ {
2162
+ "epoch": 2.655172413793103,
2163
+ "grad_norm": 0.1291448175907135,
2164
+ "learning_rate": 1.4095220158674851e-05,
2165
+ "loss": 0.37486380338668823,
2166
+ "step": 616
2167
+ },
2168
+ {
2169
+ "epoch": 2.663793103448276,
2170
+ "grad_norm": 0.10089799761772156,
2171
+ "learning_rate": 1.3980089641077864e-05,
2172
+ "loss": 0.5902385115623474,
2173
+ "step": 618
2174
+ },
2175
+ {
2176
+ "epoch": 2.6724137931034484,
2177
+ "grad_norm": 0.3151969611644745,
2178
+ "learning_rate": 1.3865367044038972e-05,
2179
+ "loss": 0.3626130223274231,
2180
+ "step": 620
2181
+ },
2182
+ {
2183
+ "epoch": 2.6810344827586206,
2184
+ "grad_norm": 0.10858116298913956,
2185
+ "learning_rate": 1.3751058202748815e-05,
2186
+ "loss": 0.6260622143745422,
2187
+ "step": 622
2188
+ },
2189
+ {
2190
+ "epoch": 2.689655172413793,
2191
+ "grad_norm": 0.09145694226026535,
2192
+ "learning_rate": 1.3637168931352952e-05,
2193
+ "loss": 0.3847617506980896,
2194
+ "step": 624
2195
+ },
2196
+ {
2197
+ "epoch": 2.6982758620689653,
2198
+ "grad_norm": 0.10181720554828644,
2199
+ "learning_rate": 1.3523705022656194e-05,
2200
+ "loss": 0.5213911533355713,
2201
+ "step": 626
2202
+ },
2203
+ {
2204
+ "epoch": 2.706896551724138,
2205
+ "grad_norm": 0.07265552878379822,
2206
+ "learning_rate": 1.3410672247827887e-05,
2207
+ "loss": 0.3843521475791931,
2208
+ "step": 628
2209
+ },
2210
+ {
2211
+ "epoch": 2.7155172413793105,
2212
+ "grad_norm": 0.06394084542989731,
2213
+ "learning_rate": 1.3298076356108431e-05,
2214
+ "loss": 0.7390468716621399,
2215
+ "step": 630
2216
+ },
2217
+ {
2218
+ "epoch": 2.7241379310344827,
2219
+ "grad_norm": 0.08277060091495514,
2220
+ "learning_rate": 1.318592307451683e-05,
2221
+ "loss": 0.3152429461479187,
2222
+ "step": 632
2223
+ },
2224
+ {
2225
+ "epoch": 2.7327586206896552,
2226
+ "grad_norm": 0.06954030692577362,
2227
+ "learning_rate": 1.307421810755938e-05,
2228
+ "loss": 0.5903550982475281,
2229
+ "step": 634
2230
+ },
2231
+ {
2232
+ "epoch": 2.7413793103448274,
2233
+ "grad_norm": 0.14430810511112213,
2234
+ "learning_rate": 1.296296713693956e-05,
2235
+ "loss": 0.4196533262729645,
2236
+ "step": 636
2237
+ },
2238
+ {
2239
+ "epoch": 2.75,
2240
+ "grad_norm": 0.049837417900562286,
2241
+ "learning_rate": 1.2852175821268977e-05,
2242
+ "loss": 0.5849826335906982,
2243
+ "step": 638
2244
+ },
2245
+ {
2246
+ "epoch": 2.7586206896551726,
2247
+ "grad_norm": 0.16439993679523468,
2248
+ "learning_rate": 1.274184979577963e-05,
2249
+ "loss": 0.40721848607063293,
2250
+ "step": 640
2251
+ },
2252
+ {
2253
+ "epoch": 2.7672413793103448,
2254
+ "grad_norm": 0.15708234906196594,
2255
+ "learning_rate": 1.2631994672037205e-05,
2256
+ "loss": 0.5138668417930603,
2257
+ "step": 642
2258
+ },
2259
+ {
2260
+ "epoch": 2.7758620689655173,
2261
+ "grad_norm": 0.0595339760184288,
2262
+ "learning_rate": 1.2522616037655713e-05,
2263
+ "loss": 0.6097421646118164,
2264
+ "step": 644
2265
+ },
2266
+ {
2267
+ "epoch": 2.7844827586206895,
2268
+ "grad_norm": 0.14719434082508087,
2269
+ "learning_rate": 1.2413719456013231e-05,
2270
+ "loss": 0.5522211194038391,
2271
+ "step": 646
2272
+ },
2273
+ {
2274
+ "epoch": 2.793103448275862,
2275
+ "grad_norm": 0.06864980608224869,
2276
+ "learning_rate": 1.2305310465968985e-05,
2277
+ "loss": 0.3453619182109833,
2278
+ "step": 648
2279
+ },
2280
+ {
2281
+ "epoch": 2.8017241379310347,
2282
+ "grad_norm": 0.05219966545701027,
2283
+ "learning_rate": 1.2197394581581561e-05,
2284
+ "loss": 0.7121859788894653,
2285
+ "step": 650
2286
+ },
2287
+ {
2288
+ "epoch": 2.810344827586207,
2289
+ "grad_norm": 0.24679023027420044,
2290
+ "learning_rate": 1.2089977291828512e-05,
2291
+ "loss": 0.7990239262580872,
2292
+ "step": 652
2293
+ },
2294
+ {
2295
+ "epoch": 2.8189655172413794,
2296
+ "grad_norm": 0.15024927258491516,
2297
+ "learning_rate": 1.1983064060327098e-05,
2298
+ "loss": 0.6081220507621765,
2299
+ "step": 654
2300
+ },
2301
+ {
2302
+ "epoch": 2.8275862068965516,
2303
+ "grad_norm": 0.05443995073437691,
2304
+ "learning_rate": 1.187666032505645e-05,
2305
+ "loss": 0.43975335359573364,
2306
+ "step": 656
2307
+ },
2308
+ {
2309
+ "epoch": 2.836206896551724,
2310
+ "grad_norm": 0.05697048828005791,
2311
+ "learning_rate": 1.1770771498080921e-05,
2312
+ "loss": 0.6137202978134155,
2313
+ "step": 658
2314
+ },
2315
+ {
2316
+ "epoch": 2.844827586206897,
2317
+ "grad_norm": 0.11451619118452072,
2318
+ "learning_rate": 1.1665402965274866e-05,
2319
+ "loss": 0.20562584698200226,
2320
+ "step": 660
2321
+ },
2322
+ {
2323
+ "epoch": 2.853448275862069,
2324
+ "grad_norm": 0.22301547229290009,
2325
+ "learning_rate": 1.1560560086048632e-05,
2326
+ "loss": 0.42035165429115295,
2327
+ "step": 662
2328
+ },
2329
+ {
2330
+ "epoch": 2.862068965517241,
2331
+ "grad_norm": 0.15491816401481628,
2332
+ "learning_rate": 1.1456248193076027e-05,
2333
+ "loss": 0.6786882877349854,
2334
+ "step": 664
2335
+ },
2336
+ {
2337
+ "epoch": 2.8706896551724137,
2338
+ "grad_norm": 0.06417909264564514,
2339
+ "learning_rate": 1.1352472592023026e-05,
2340
+ "loss": 0.34481775760650635,
2341
+ "step": 666
2342
+ },
2343
+ {
2344
+ "epoch": 2.8793103448275863,
2345
+ "grad_norm": 0.2559848129749298,
2346
+ "learning_rate": 1.1249238561277957e-05,
2347
+ "loss": 0.37077146768569946,
2348
+ "step": 668
2349
+ },
2350
+ {
2351
+ "epoch": 2.887931034482759,
2352
+ "grad_norm": 0.07367434352636337,
2353
+ "learning_rate": 1.1146551351682962e-05,
2354
+ "loss": 0.6234573125839233,
2355
+ "step": 670
2356
+ },
2357
+ {
2358
+ "epoch": 2.896551724137931,
2359
+ "grad_norm": 0.13318119943141937,
2360
+ "learning_rate": 1.1044416186266985e-05,
2361
+ "loss": 0.43646591901779175,
2362
+ "step": 672
2363
+ },
2364
+ {
2365
+ "epoch": 2.905172413793103,
2366
+ "grad_norm": 0.04189766198396683,
2367
+ "learning_rate": 1.0942838259980065e-05,
2368
+ "loss": 0.6099374890327454,
2369
+ "step": 674
2370
+ },
2371
+ {
2372
+ "epoch": 2.913793103448276,
2373
+ "grad_norm": 0.16093385219573975,
2374
+ "learning_rate": 1.0841822739429131e-05,
2375
+ "loss": 0.5961918830871582,
2376
+ "step": 676
2377
+ },
2378
+ {
2379
+ "epoch": 2.9224137931034484,
2380
+ "grad_norm": 0.05338941141963005,
2381
+ "learning_rate": 1.0741374762615181e-05,
2382
+ "loss": 0.5247670412063599,
2383
+ "step": 678
2384
+ },
2385
+ {
2386
+ "epoch": 2.9310344827586206,
2387
+ "grad_norm": 0.06662659347057343,
2388
+ "learning_rate": 1.0641499438671994e-05,
2389
+ "loss": 0.4245750606060028,
2390
+ "step": 680
2391
+ },
2392
+ {
2393
+ "epoch": 2.939655172413793,
2394
+ "grad_norm": 0.03824161738157272,
2395
+ "learning_rate": 1.054220184760619e-05,
2396
+ "loss": 0.21983936429023743,
2397
+ "step": 682
2398
+ },
2399
+ {
2400
+ "epoch": 2.9482758620689653,
2401
+ "grad_norm": 0.061386823654174805,
2402
+ "learning_rate": 1.0443487040038919e-05,
2403
+ "loss": 0.3854738771915436,
2404
+ "step": 684
2405
+ },
2406
+ {
2407
+ "epoch": 2.956896551724138,
2408
+ "grad_norm": 0.06032966449856758,
2409
+ "learning_rate": 1.0345360036948912e-05,
2410
+ "loss": 0.6782163381576538,
2411
+ "step": 686
2412
+ },
2413
+ {
2414
+ "epoch": 2.9655172413793105,
2415
+ "grad_norm": 0.06708291918039322,
2416
+ "learning_rate": 1.0247825829417132e-05,
2417
+ "loss": 0.5401458740234375,
2418
+ "step": 688
2419
+ },
2420
+ {
2421
+ "epoch": 2.9741379310344827,
2422
+ "grad_norm": 0.0782044380903244,
2423
+ "learning_rate": 1.0150889378372878e-05,
2424
+ "loss": 0.7114209532737732,
2425
+ "step": 690
2426
+ },
2427
+ {
2428
+ "epoch": 2.9827586206896552,
2429
+ "grad_norm": 0.06770720332860947,
2430
+ "learning_rate": 1.00545556143415e-05,
2431
+ "loss": 0.660466730594635,
2432
+ "step": 692
2433
+ },
2434
+ {
2435
+ "epoch": 2.9913793103448274,
2436
+ "grad_norm": 0.07091684639453888,
2437
+ "learning_rate": 9.958829437193558e-06,
2438
+ "loss": 0.4320341944694519,
2439
+ "step": 694
2440
+ },
2441
+ {
2442
+ "epoch": 3.0,
2443
+ "grad_norm": 0.06834368407726288,
2444
+ "learning_rate": 9.863715715895658e-06,
2445
+ "loss": 0.6856396198272705,
2446
+ "step": 696
2447
+ },
2448
+ {
2449
+ "epoch": 3.0086206896551726,
2450
+ "grad_norm": 0.03995652124285698,
2451
+ "learning_rate": 9.769219288262745e-06,
2452
+ "loss": 0.16509434580802917,
2453
+ "step": 698
2454
+ },
2455
+ {
2456
+ "epoch": 3.0172413793103448,
2457
+ "grad_norm": 0.043883178383111954,
2458
+ "learning_rate": 9.675344960712074e-06,
2459
+ "loss": 0.29928964376449585,
2460
+ "step": 700
2461
+ },
2462
+ {
2463
+ "epoch": 3.0258620689655173,
2464
+ "grad_norm": 0.0733269527554512,
2465
+ "learning_rate": 9.582097508018724e-06,
2466
+ "loss": 0.25162428617477417,
2467
+ "step": 702
2468
+ },
2469
+ {
2470
+ "epoch": 3.0344827586206895,
2471
+ "grad_norm": 0.12920475006103516,
2472
+ "learning_rate": 9.489481673072723e-06,
2473
+ "loss": 0.3514169454574585,
2474
+ "step": 704
2475
+ },
2476
+ {
2477
+ "epoch": 3.043103448275862,
2478
+ "grad_norm": 0.017986657097935677,
2479
+ "learning_rate": 9.397502166637837e-06,
2480
+ "loss": 0.07074951380491257,
2481
+ "step": 706
2482
+ },
2483
+ {
2484
+ "epoch": 3.0517241379310347,
2485
+ "grad_norm": 0.07337481528520584,
2486
+ "learning_rate": 9.30616366711195e-06,
2487
+ "loss": 0.20599356293678284,
2488
+ "step": 708
2489
+ },
2490
+ {
2491
+ "epoch": 3.060344827586207,
2492
+ "grad_norm": 0.03576648607850075,
2493
+ "learning_rate": 9.21547082028908e-06,
2494
+ "loss": 0.11480194330215454,
2495
+ "step": 710
2496
+ },
2497
+ {
2498
+ "epoch": 3.0689655172413794,
2499
+ "grad_norm": 0.38087305426597595,
2500
+ "learning_rate": 9.125428239123133e-06,
2501
+ "loss": 0.26979854702949524,
2502
+ "step": 712
2503
+ },
2504
+ {
2505
+ "epoch": 3.0775862068965516,
2506
+ "grad_norm": 0.0725908949971199,
2507
+ "learning_rate": 9.036040503493213e-06,
2508
+ "loss": 0.42210009694099426,
2509
+ "step": 714
2510
+ },
2511
+ {
2512
+ "epoch": 3.086206896551724,
2513
+ "grad_norm": 0.14822497963905334,
2514
+ "learning_rate": 8.947312159970725e-06,
2515
+ "loss": 0.1675470620393753,
2516
+ "step": 716
2517
+ },
2518
+ {
2519
+ "epoch": 3.0948275862068964,
2520
+ "grad_norm": 0.08073808997869492,
2521
+ "learning_rate": 8.859247721588064e-06,
2522
+ "loss": 0.20833522081375122,
2523
+ "step": 718
2524
+ },
2525
+ {
2526
+ "epoch": 3.103448275862069,
2527
+ "grad_norm": 0.057046178728342056,
2528
+ "learning_rate": 8.77185166760914e-06,
2529
+ "loss": 0.16950953006744385,
2530
+ "step": 720
2531
+ },
2532
+ {
2533
+ "epoch": 3.1120689655172415,
2534
+ "grad_norm": 0.10354648530483246,
2535
+ "learning_rate": 8.685128443301465e-06,
2536
+ "loss": 0.12641456723213196,
2537
+ "step": 722
2538
+ },
2539
+ {
2540
+ "epoch": 3.1206896551724137,
2541
+ "grad_norm": 0.05845208466053009,
2542
+ "learning_rate": 8.599082459710125e-06,
2543
+ "loss": 0.13568329811096191,
2544
+ "step": 724
2545
+ },
2546
+ {
2547
+ "epoch": 3.1293103448275863,
2548
+ "grad_norm": 0.04908813536167145,
2549
+ "learning_rate": 8.513718093433354e-06,
2550
+ "loss": 0.21239104866981506,
2551
+ "step": 726
2552
+ },
2553
+ {
2554
+ "epoch": 3.1379310344827585,
2555
+ "grad_norm": 0.13193517923355103,
2556
+ "learning_rate": 8.42903968639999e-06,
2557
+ "loss": 0.2763456702232361,
2558
+ "step": 728
2559
+ },
2560
+ {
2561
+ "epoch": 3.146551724137931,
2562
+ "grad_norm": 0.03571261465549469,
2563
+ "learning_rate": 8.345051545648565e-06,
2564
+ "loss": 0.12836386263370514,
2565
+ "step": 730
2566
+ },
2567
+ {
2568
+ "epoch": 3.1551724137931036,
2569
+ "grad_norm": 0.06112167611718178,
2570
+ "learning_rate": 8.261757943108296e-06,
2571
+ "loss": 0.16560682654380798,
2572
+ "step": 732
2573
+ },
2574
+ {
2575
+ "epoch": 3.163793103448276,
2576
+ "grad_norm": 0.0860171988606453,
2577
+ "learning_rate": 8.179163115381737e-06,
2578
+ "loss": 0.2081730216741562,
2579
+ "step": 734
2580
+ },
2581
+ {
2582
+ "epoch": 3.1724137931034484,
2583
+ "grad_norm": 0.03247256577014923,
2584
+ "learning_rate": 8.097271263529346e-06,
2585
+ "loss": 0.14392191171646118,
2586
+ "step": 736
2587
+ },
2588
+ {
2589
+ "epoch": 3.1810344827586206,
2590
+ "grad_norm": 0.0918356403708458,
2591
+ "learning_rate": 8.016086552855771e-06,
2592
+ "loss": 0.15577132999897003,
2593
+ "step": 738
2594
+ },
2595
+ {
2596
+ "epoch": 3.189655172413793,
2597
+ "grad_norm": 0.06287133693695068,
2598
+ "learning_rate": 7.935613112698003e-06,
2599
+ "loss": 0.0789552852511406,
2600
+ "step": 740
2601
+ },
2602
+ {
2603
+ "epoch": 3.1982758620689653,
2604
+ "grad_norm": 0.03986852988600731,
2605
+ "learning_rate": 7.855855036215328e-06,
2606
+ "loss": 0.10101716220378876,
2607
+ "step": 742
2608
+ },
2609
+ {
2610
+ "epoch": 3.206896551724138,
2611
+ "grad_norm": 0.13693907856941223,
2612
+ "learning_rate": 7.776816380181165e-06,
2613
+ "loss": 0.1658182144165039,
2614
+ "step": 744
2615
+ },
2616
+ {
2617
+ "epoch": 3.2155172413793105,
2618
+ "grad_norm": 0.14548790454864502,
2619
+ "learning_rate": 7.698501164776679e-06,
2620
+ "loss": 0.19248032569885254,
2621
+ "step": 746
2622
+ },
2623
+ {
2624
+ "epoch": 3.2241379310344827,
2625
+ "grad_norm": 0.05582420900464058,
2626
+ "learning_rate": 7.620913373386356e-06,
2627
+ "loss": 0.21470694243907928,
2628
+ "step": 748
2629
+ },
2630
+ {
2631
+ "epoch": 3.2327586206896552,
2632
+ "grad_norm": 0.04277574643492699,
2633
+ "learning_rate": 7.5440569523953315e-06,
2634
+ "loss": 0.15740104019641876,
2635
+ "step": 750
2636
+ },
2637
+ {
2638
+ "epoch": 3.2413793103448274,
2639
+ "grad_norm": 0.14733938872814178,
2640
+ "learning_rate": 7.467935810988729e-06,
2641
+ "loss": 0.18646365404129028,
2642
+ "step": 752
2643
+ },
2644
+ {
2645
+ "epoch": 3.25,
2646
+ "grad_norm": 0.06095249578356743,
2647
+ "learning_rate": 7.392553820952764e-06,
2648
+ "loss": 0.22709967195987701,
2649
+ "step": 754
2650
+ },
2651
+ {
2652
+ "epoch": 3.2586206896551726,
2653
+ "grad_norm": 0.04888584464788437,
2654
+ "learning_rate": 7.317914816477865e-06,
2655
+ "loss": 0.1782107949256897,
2656
+ "step": 756
2657
+ },
2658
+ {
2659
+ "epoch": 3.2672413793103448,
2660
+ "grad_norm": 0.2761983275413513,
2661
+ "learning_rate": 7.244022593963609e-06,
2662
+ "loss": 0.19192323088645935,
2663
+ "step": 758
2664
+ },
2665
+ {
2666
+ "epoch": 3.2758620689655173,
2667
+ "grad_norm": 0.041269440203905106,
2668
+ "learning_rate": 7.170880911825657e-06,
2669
+ "loss": 0.13779321312904358,
2670
+ "step": 760
2671
+ },
2672
+ {
2673
+ "epoch": 3.2844827586206895,
2674
+ "grad_norm": 0.2219523787498474,
2675
+ "learning_rate": 7.098493490304566e-06,
2676
+ "loss": 0.24427469074726105,
2677
+ "step": 762
2678
+ },
2679
+ {
2680
+ "epoch": 3.293103448275862,
2681
+ "grad_norm": 0.7461491227149963,
2682
+ "learning_rate": 7.026864011276575e-06,
2683
+ "loss": 0.32002437114715576,
2684
+ "step": 764
2685
+ },
2686
+ {
2687
+ "epoch": 3.3017241379310347,
2688
+ "grad_norm": 0.100465789437294,
2689
+ "learning_rate": 6.955996118066326e-06,
2690
+ "loss": 0.11214806139469147,
2691
+ "step": 766
2692
+ },
2693
+ {
2694
+ "epoch": 3.310344827586207,
2695
+ "grad_norm": 0.06019704416394234,
2696
+ "learning_rate": 6.8858934152615646e-06,
2697
+ "loss": 0.1987936794757843,
2698
+ "step": 768
2699
+ },
2700
+ {
2701
+ "epoch": 3.3189655172413794,
2702
+ "grad_norm": 0.12379293888807297,
2703
+ "learning_rate": 6.816559468529773e-06,
2704
+ "loss": 0.058321211487054825,
2705
+ "step": 770
2706
+ },
2707
+ {
2708
+ "epoch": 3.3275862068965516,
2709
+ "grad_norm": 0.3285755515098572,
2710
+ "learning_rate": 6.747997804436846e-06,
2711
+ "loss": 0.08903615176677704,
2712
+ "step": 772
2713
+ },
2714
+ {
2715
+ "epoch": 3.336206896551724,
2716
+ "grad_norm": 0.11563495546579361,
2717
+ "learning_rate": 6.680211910267665e-06,
2718
+ "loss": 0.35364535450935364,
2719
+ "step": 774
2720
+ },
2721
+ {
2722
+ "epoch": 3.344827586206897,
2723
+ "grad_norm": 0.07364711910486221,
2724
+ "learning_rate": 6.613205233848783e-06,
2725
+ "loss": 0.20209553837776184,
2726
+ "step": 776
2727
+ },
2728
+ {
2729
+ "epoch": 3.353448275862069,
2730
+ "grad_norm": 0.0495804026722908,
2731
+ "learning_rate": 6.546981183373009e-06,
2732
+ "loss": 0.19359779357910156,
2733
+ "step": 778
2734
+ },
2735
+ {
2736
+ "epoch": 3.3620689655172415,
2737
+ "grad_norm": 0.13539589941501617,
2738
+ "learning_rate": 6.481543127226073e-06,
2739
+ "loss": 0.28171947598457336,
2740
+ "step": 780
2741
+ },
2742
+ {
2743
+ "epoch": 3.3706896551724137,
2744
+ "grad_norm": 0.07525072246789932,
2745
+ "learning_rate": 6.4168943938153e-06,
2746
+ "loss": 0.1644493192434311,
2747
+ "step": 782
2748
+ },
2749
+ {
2750
+ "epoch": 3.3793103448275863,
2751
+ "grad_norm": 0.04455971717834473,
2752
+ "learning_rate": 6.353038271400319e-06,
2753
+ "loss": 0.17818251252174377,
2754
+ "step": 784
2755
+ },
2756
+ {
2757
+ "epoch": 3.3879310344827585,
2758
+ "grad_norm": 0.04888049513101578,
2759
+ "learning_rate": 6.289978007925791e-06,
2760
+ "loss": 0.08893375098705292,
2761
+ "step": 786
2762
+ },
2763
+ {
2764
+ "epoch": 3.396551724137931,
2765
+ "grad_norm": 0.0592099204659462,
2766
+ "learning_rate": 6.227716810856235e-06,
2767
+ "loss": 0.16159863770008087,
2768
+ "step": 788
2769
+ },
2770
+ {
2771
+ "epoch": 3.405172413793103,
2772
+ "grad_norm": 0.05931266024708748,
2773
+ "learning_rate": 6.1662578470128595e-06,
2774
+ "loss": 0.19914919137954712,
2775
+ "step": 790
2776
+ },
2777
+ {
2778
+ "epoch": 3.413793103448276,
2779
+ "grad_norm": 0.05199761316180229,
2780
+ "learning_rate": 6.105604242412507e-06,
2781
+ "loss": 0.1833517998456955,
2782
+ "step": 792
2783
+ },
2784
+ {
2785
+ "epoch": 3.4224137931034484,
2786
+ "grad_norm": 0.07053744047880173,
2787
+ "learning_rate": 6.0457590821086364e-06,
2788
+ "loss": 0.1568892002105713,
2789
+ "step": 794
2790
+ },
2791
+ {
2792
+ "epoch": 3.4310344827586206,
2793
+ "grad_norm": 0.11103974282741547,
2794
+ "learning_rate": 5.9867254100344305e-06,
2795
+ "loss": 0.5605343580245972,
2796
+ "step": 796
2797
+ },
2798
+ {
2799
+ "epoch": 3.439655172413793,
2800
+ "grad_norm": 0.1462671458721161,
2801
+ "learning_rate": 5.92850622884794e-06,
2802
+ "loss": 0.2542985677719116,
2803
+ "step": 798
2804
+ },
2805
+ {
2806
+ "epoch": 3.4482758620689653,
2807
+ "grad_norm": 0.07662937045097351,
2808
+ "learning_rate": 5.871104499779383e-06,
2809
+ "loss": 0.3042844533920288,
2810
+ "step": 800
2811
+ },
2812
+ {
2813
+ "epoch": 3.456896551724138,
2814
+ "grad_norm": 0.40208032727241516,
2815
+ "learning_rate": 5.814523142480514e-06,
2816
+ "loss": 0.23688863217830658,
2817
+ "step": 802
2818
+ },
2819
+ {
2820
+ "epoch": 3.4655172413793105,
2821
+ "grad_norm": 0.0428071990609169,
2822
+ "learning_rate": 5.758765034876124e-06,
2823
+ "loss": 0.1413598358631134,
2824
+ "step": 804
2825
+ },
2826
+ {
2827
+ "epoch": 3.4741379310344827,
2828
+ "grad_norm": 0.18820738792419434,
2829
+ "learning_rate": 5.703833013017659e-06,
2830
+ "loss": 0.26621344685554504,
2831
+ "step": 806
2832
+ },
2833
+ {
2834
+ "epoch": 3.4827586206896552,
2835
+ "grad_norm": 0.0474395789206028,
2836
+ "learning_rate": 5.649729870938974e-06,
2837
+ "loss": 0.1856929361820221,
2838
+ "step": 808
2839
+ },
2840
+ {
2841
+ "epoch": 3.4913793103448274,
2842
+ "grad_norm": 0.041845474392175674,
2843
+ "learning_rate": 5.596458360514197e-06,
2844
+ "loss": 0.11116787791252136,
2845
+ "step": 810
2846
+ },
2847
+ {
2848
+ "epoch": 3.5,
2849
+ "grad_norm": 0.06532273441553116,
2850
+ "learning_rate": 5.544021191317797e-06,
2851
+ "loss": 0.2585083842277527,
2852
+ "step": 812
2853
+ },
2854
+ {
2855
+ "epoch": 3.5086206896551726,
2856
+ "grad_norm": 0.17792125046253204,
2857
+ "learning_rate": 5.492421030486723e-06,
2858
+ "loss": 0.24390508234500885,
2859
+ "step": 814
2860
+ },
2861
+ {
2862
+ "epoch": 3.5172413793103448,
2863
+ "grad_norm": 0.06345438957214355,
2864
+ "learning_rate": 5.441660502584782e-06,
2865
+ "loss": 0.19690856337547302,
2866
+ "step": 816
2867
+ },
2868
+ {
2869
+ "epoch": 3.5258620689655173,
2870
+ "grad_norm": 0.06546366214752197,
2871
+ "learning_rate": 5.391742189469118e-06,
2872
+ "loss": 0.18222372233867645,
2873
+ "step": 818
2874
+ },
2875
+ {
2876
+ "epoch": 3.5344827586206895,
2877
+ "grad_norm": 0.06039542332291603,
2878
+ "learning_rate": 5.342668630158901e-06,
2879
+ "loss": 0.14991328120231628,
2880
+ "step": 820
2881
+ },
2882
+ {
2883
+ "epoch": 3.543103448275862,
2884
+ "grad_norm": 0.08110994100570679,
2885
+ "learning_rate": 5.294442320706179e-06,
2886
+ "loss": 0.12025367468595505,
2887
+ "step": 822
2888
+ },
2889
+ {
2890
+ "epoch": 3.5517241379310347,
2891
+ "grad_norm": 0.04069434478878975,
2892
+ "learning_rate": 5.247065714068933e-06,
2893
+ "loss": 0.0922561064362526,
2894
+ "step": 824
2895
+ },
2896
+ {
2897
+ "epoch": 3.560344827586207,
2898
+ "grad_norm": 0.012679479084908962,
2899
+ "learning_rate": 5.200541219986286e-06,
2900
+ "loss": 0.03818206116557121,
2901
+ "step": 826
2902
+ },
2903
+ {
2904
+ "epoch": 3.5689655172413794,
2905
+ "grad_norm": 0.22782403230667114,
2906
+ "learning_rate": 5.1548712048559655e-06,
2907
+ "loss": 0.2238304615020752,
2908
+ "step": 828
2909
+ },
2910
+ {
2911
+ "epoch": 3.5775862068965516,
2912
+ "grad_norm": 0.05799900367856026,
2913
+ "learning_rate": 5.110057991613912e-06,
2914
+ "loss": 0.1549633890390396,
2915
+ "step": 830
2916
+ },
2917
+ {
2918
+ "epoch": 3.586206896551724,
2919
+ "grad_norm": 0.05750008672475815,
2920
+ "learning_rate": 5.0661038596161515e-06,
2921
+ "loss": 0.14927032589912415,
2922
+ "step": 832
2923
+ },
2924
+ {
2925
+ "epoch": 3.594827586206897,
2926
+ "grad_norm": 0.24478773772716522,
2927
+ "learning_rate": 5.023011044522834e-06,
2928
+ "loss": 0.2999204397201538,
2929
+ "step": 834
2930
+ },
2931
+ {
2932
+ "epoch": 3.603448275862069,
2933
+ "grad_norm": 0.07759716361761093,
2934
+ "learning_rate": 4.980781738184549e-06,
2935
+ "loss": 0.20024727284908295,
2936
+ "step": 836
2937
+ },
2938
+ {
2939
+ "epoch": 3.612068965517241,
2940
+ "grad_norm": 0.0670485645532608,
2941
+ "learning_rate": 4.939418088530811e-06,
2942
+ "loss": 0.13863810896873474,
2943
+ "step": 838
2944
+ },
2945
+ {
2946
+ "epoch": 3.6206896551724137,
2947
+ "grad_norm": 0.12198883295059204,
2948
+ "learning_rate": 4.898922199460831e-06,
2949
+ "loss": 0.18534965813159943,
2950
+ "step": 840
2951
+ },
2952
+ {
2953
+ "epoch": 3.6293103448275863,
2954
+ "grad_norm": 0.1402168571949005,
2955
+ "learning_rate": 4.859296130736489e-06,
2956
+ "loss": 0.15518294274806976,
2957
+ "step": 842
2958
+ },
2959
+ {
2960
+ "epoch": 3.637931034482759,
2961
+ "grad_norm": 0.06257359683513641,
2962
+ "learning_rate": 4.820541897877585e-06,
2963
+ "loss": 0.23298737406730652,
2964
+ "step": 844
2965
+ },
2966
+ {
2967
+ "epoch": 3.646551724137931,
2968
+ "grad_norm": 0.11765491217374802,
2969
+ "learning_rate": 4.782661472059298e-06,
2970
+ "loss": 0.264419823884964,
2971
+ "step": 846
2972
+ },
2973
+ {
2974
+ "epoch": 3.655172413793103,
2975
+ "grad_norm": 0.03430064767599106,
2976
+ "learning_rate": 4.745656780011951e-06,
2977
+ "loss": 0.15973711013793945,
2978
+ "step": 848
2979
+ },
2980
+ {
2981
+ "epoch": 3.663793103448276,
2982
+ "grad_norm": 0.03395868092775345,
2983
+ "learning_rate": 4.709529703922993e-06,
2984
+ "loss": 0.17208503186702728,
2985
+ "step": 850
2986
+ },
2987
+ {
2988
+ "epoch": 3.6724137931034484,
2989
+ "grad_norm": 0.08469868451356888,
2990
+ "learning_rate": 4.674282081341271e-06,
2991
+ "loss": 0.19475609064102173,
2992
+ "step": 852
2993
+ },
2994
+ {
2995
+ "epoch": 3.6810344827586206,
2996
+ "grad_norm": 0.06020957604050636,
2997
+ "learning_rate": 4.639915705083572e-06,
2998
+ "loss": 0.1562570333480835,
2999
+ "step": 854
3000
+ },
3001
+ {
3002
+ "epoch": 3.689655172413793,
3003
+ "grad_norm": 0.07793577015399933,
3004
+ "learning_rate": 4.606432323143412e-06,
3005
+ "loss": 0.15900962054729462,
3006
+ "step": 856
3007
+ },
3008
+ {
3009
+ "epoch": 3.6982758620689653,
3010
+ "grad_norm": 0.13332881033420563,
3011
+ "learning_rate": 4.573833638602159e-06,
3012
+ "loss": 0.22381483018398285,
3013
+ "step": 858
3014
+ },
3015
+ {
3016
+ "epoch": 3.706896551724138,
3017
+ "grad_norm": 0.02578054927289486,
3018
+ "learning_rate": 4.542121309542383e-06,
3019
+ "loss": 0.09598782658576965,
3020
+ "step": 860
3021
+ },
3022
+ {
3023
+ "epoch": 3.7155172413793105,
3024
+ "grad_norm": 0.15609142184257507,
3025
+ "learning_rate": 4.511296948963527e-06,
3026
+ "loss": 0.19943147897720337,
3027
+ "step": 862
3028
+ },
3029
+ {
3030
+ "epoch": 3.7241379310344827,
3031
+ "grad_norm": 0.01152154989540577,
3032
+ "learning_rate": 4.4813621246998765e-06,
3033
+ "loss": 0.07272744178771973,
3034
+ "step": 864
3035
+ },
3036
+ {
3037
+ "epoch": 3.7327586206896552,
3038
+ "grad_norm": 0.06737919896841049,
3039
+ "learning_rate": 4.45231835934079e-06,
3040
+ "loss": 0.1303609311580658,
3041
+ "step": 866
3042
+ },
3043
+ {
3044
+ "epoch": 3.7413793103448274,
3045
+ "grad_norm": 0.2695164084434509,
3046
+ "learning_rate": 4.424167130153277e-06,
3047
+ "loss": 0.18073761463165283,
3048
+ "step": 868
3049
+ },
3050
+ {
3051
+ "epoch": 3.75,
3052
+ "grad_norm": 0.044187769293785095,
3053
+ "learning_rate": 4.396909869006847e-06,
3054
+ "loss": 0.12275875359773636,
3055
+ "step": 870
3056
+ },
3057
+ {
3058
+ "epoch": 3.7586206896551726,
3059
+ "grad_norm": 0.008070076815783978,
3060
+ "learning_rate": 4.3705479623006866e-06,
3061
+ "loss": 0.06019383668899536,
3062
+ "step": 872
3063
+ },
3064
+ {
3065
+ "epoch": 3.7672413793103448,
3066
+ "grad_norm": 0.06783478707075119,
3067
+ "learning_rate": 4.345082750893132e-06,
3068
+ "loss": 0.10059908032417297,
3069
+ "step": 874
3070
+ },
3071
+ {
3072
+ "epoch": 3.7758620689655173,
3073
+ "grad_norm": 0.10238350927829742,
3074
+ "learning_rate": 4.320515530033487e-06,
3075
+ "loss": 0.30081015825271606,
3076
+ "step": 876
3077
+ },
3078
+ {
3079
+ "epoch": 3.7844827586206895,
3080
+ "grad_norm": 0.14017876982688904,
3081
+ "learning_rate": 4.296847549296115e-06,
3082
+ "loss": 0.30149415135383606,
3083
+ "step": 878
3084
+ },
3085
+ {
3086
+ "epoch": 3.793103448275862,
3087
+ "grad_norm": 0.06336027383804321,
3088
+ "learning_rate": 4.274080012516909e-06,
3089
+ "loss": 0.13996456563472748,
3090
+ "step": 880
3091
+ },
3092
+ {
3093
+ "epoch": 3.8017241379310347,
3094
+ "grad_norm": 0.04175444692373276,
3095
+ "learning_rate": 4.2522140777320404e-06,
3096
+ "loss": 0.09511252492666245,
3097
+ "step": 882
3098
+ },
3099
+ {
3100
+ "epoch": 3.810344827586207,
3101
+ "grad_norm": 0.05849481746554375,
3102
+ "learning_rate": 4.23125085711907e-06,
3103
+ "loss": 0.23412549495697021,
3104
+ "step": 884
3105
+ },
3106
+ {
3107
+ "epoch": 3.8189655172413794,
3108
+ "grad_norm": 0.06667976826429367,
3109
+ "learning_rate": 4.21119141694037e-06,
3110
+ "loss": 0.160780131816864,
3111
+ "step": 886
3112
+ },
3113
+ {
3114
+ "epoch": 3.8275862068965516,
3115
+ "grad_norm": 0.056150369346141815,
3116
+ "learning_rate": 4.192036777488896e-06,
3117
+ "loss": 0.11835036426782608,
3118
+ "step": 888
3119
+ },
3120
+ {
3121
+ "epoch": 3.836206896551724,
3122
+ "grad_norm": 0.12212974578142166,
3123
+ "learning_rate": 4.173787913036284e-06,
3124
+ "loss": 0.11370360106229782,
3125
+ "step": 890
3126
+ },
3127
+ {
3128
+ "epoch": 3.844827586206897,
3129
+ "grad_norm": 0.08489777147769928,
3130
+ "learning_rate": 4.156445751783308e-06,
3131
+ "loss": 0.17437399923801422,
3132
+ "step": 892
3133
+ },
3134
+ {
3135
+ "epoch": 3.853448275862069,
3136
+ "grad_norm": 0.03076266683638096,
3137
+ "learning_rate": 4.140011175812656e-06,
3138
+ "loss": 0.15946733951568604,
3139
+ "step": 894
3140
+ },
3141
+ {
3142
+ "epoch": 3.862068965517241,
3143
+ "grad_norm": 0.044967204332351685,
3144
+ "learning_rate": 4.124485021044069e-06,
3145
+ "loss": 0.16160649061203003,
3146
+ "step": 896
3147
+ },
3148
+ {
3149
+ "epoch": 3.8706896551724137,
3150
+ "grad_norm": 0.06540035456418991,
3151
+ "learning_rate": 4.1098680771918245e-06,
3152
+ "loss": 0.13039463758468628,
3153
+ "step": 898
3154
+ },
3155
+ {
3156
+ "epoch": 3.8793103448275863,
3157
+ "grad_norm": 0.0530594103038311,
3158
+ "learning_rate": 4.096161087724573e-06,
3159
+ "loss": 0.16959071159362793,
3160
+ "step": 900
3161
+ },
3162
+ {
3163
+ "epoch": 3.887931034482759,
3164
+ "grad_norm": 0.15922929346561432,
3165
+ "learning_rate": 4.0833647498275085e-06,
3166
+ "loss": 0.20945216715335846,
3167
+ "step": 902
3168
+ },
3169
+ {
3170
+ "epoch": 3.896551724137931,
3171
+ "grad_norm": 0.06301749497652054,
3172
+ "learning_rate": 4.07147971436692e-06,
3173
+ "loss": 0.22212789952754974,
3174
+ "step": 904
3175
+ },
3176
+ {
3177
+ "epoch": 3.905172413793103,
3178
+ "grad_norm": 0.15187880396842957,
3179
+ "learning_rate": 4.060506585857085e-06,
3180
+ "loss": 0.21481694281101227,
3181
+ "step": 906
3182
+ },
3183
+ {
3184
+ "epoch": 3.913793103448276,
3185
+ "grad_norm": 0.32235345244407654,
3186
+ "learning_rate": 4.0504459224295174e-06,
3187
+ "loss": 0.16184020042419434,
3188
+ "step": 908
3189
+ },
3190
+ {
3191
+ "epoch": 3.9224137931034484,
3192
+ "grad_norm": 0.07125352323055267,
3193
+ "learning_rate": 4.041298235804577e-06,
3194
+ "loss": 0.1316578984260559,
3195
+ "step": 910
3196
+ },
3197
+ {
3198
+ "epoch": 3.9310344827586206,
3199
+ "grad_norm": 0.04981033504009247,
3200
+ "learning_rate": 4.0330639912654516e-06,
3201
+ "loss": 0.11852114647626877,
3202
+ "step": 912
3203
+ },
3204
+ {
3205
+ "epoch": 3.939655172413793,
3206
+ "grad_norm": 0.11605612933635712,
3207
+ "learning_rate": 4.02574360763448e-06,
3208
+ "loss": 0.16175302863121033,
3209
+ "step": 914
3210
+ },
3211
+ {
3212
+ "epoch": 3.9482758620689653,
3213
+ "grad_norm": 0.05728490650653839,
3214
+ "learning_rate": 4.019337457251857e-06,
3215
+ "loss": 0.16411718726158142,
3216
+ "step": 916
3217
+ },
3218
+ {
3219
+ "epoch": 3.956896551724138,
3220
+ "grad_norm": 0.08225003629922867,
3221
+ "learning_rate": 4.013845865956692e-06,
3222
+ "loss": 0.22733992338180542,
3223
+ "step": 918
3224
+ },
3225
+ {
3226
+ "epoch": 3.9655172413793105,
3227
+ "grad_norm": 0.2269326150417328,
3228
+ "learning_rate": 4.00926911307043e-06,
3229
+ "loss": 0.1820860654115677,
3230
+ "step": 920
3231
+ },
3232
+ {
3233
+ "epoch": 3.9741379310344827,
3234
+ "grad_norm": 0.06667070835828781,
3235
+ "learning_rate": 4.005607431382659e-06,
3236
+ "loss": 0.15438126027584076,
3237
+ "step": 922
3238
+ },
3239
+ {
3240
+ "epoch": 3.9827586206896552,
3241
+ "grad_norm": 0.05220466107130051,
3242
+ "learning_rate": 4.002861007139253e-06,
3243
+ "loss": 0.17508190870285034,
3244
+ "step": 924
3245
+ },
3246
+ {
3247
+ "epoch": 3.9913793103448274,
3248
+ "grad_norm": 0.07188162952661514,
3249
+ "learning_rate": 4.001029980032909e-06,
3250
+ "loss": 0.1996261328458786,
3251
+ "step": 926
3252
+ },
3253
+ {
3254
+ "epoch": 4.0,
3255
+ "grad_norm": 0.3331678509712219,
3256
+ "learning_rate": 4.000114443196044e-06,
3257
+ "loss": 0.2806675136089325,
3258
+ "step": 928
3259
+ },
3260
+ {
3261
+ "epoch": 4.0,
3262
+ "step": 928,
3263
+ "total_flos": 3.61934117404672e+18,
3264
+ "train_loss": 0.7229323635552207,
3265
+ "train_runtime": 32622.9814,
3266
+ "train_samples_per_second": 1.707,
3267
+ "train_steps_per_second": 0.028
3268
+ }
3269
+ ],
3270
+ "logging_steps": 2,
3271
+ "max_steps": 928,
3272
+ "num_input_tokens_seen": 0,
3273
+ "num_train_epochs": 4,
3274
+ "save_steps": 99999,
3275
+ "stateful_callbacks": {
3276
+ "TrainerControl": {
3277
+ "args": {
3278
+ "should_epoch_stop": false,
3279
+ "should_evaluate": false,
3280
+ "should_log": false,
3281
+ "should_save": false,
3282
+ "should_training_stop": false
3283
+ },
3284
+ "attributes": {}
3285
+ }
3286
+ },
3287
+ "total_flos": 3.61934117404672e+18,
3288
+ "train_batch_size": 1,
3289
+ "trial_name": null,
3290
+ "trial_params": null
3291
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:706321121b2516fbeb580854aa96535003a143b2398602adaeb2ae06857a47dd
3
+ size 5649
training_loss.png ADDED