Hugging Face's logo

furproxy
/
9b-60

Image-Text-to-Text
Transformers
Safetensors
qwen3_5
llama-factory
full
Generated from Trainer
conversational
Model card Files
xet
 Community
furproxy commited on
Commit
c22dd35
·
verified ·
1 Parent(s): 11a0adc

Upload folder using huggingface_hub

Browse files
Files changed (16) hide show
  1. .gitattributes +1 -0
  2. .ipynb_checkpoints/README-checkpoint.md +64 -0
  3. README.md +64 -0
  4. all_results.json +9 -0
  5. chat_template.jinja +154 -0
  6. config.json +113 -0
  7. generation_config.json +10 -0
  8. model.safetensors +3 -0
  9. processor_config.json +60 -0
  10. tokenizer.json +3 -0
  11. tokenizer_config.json +33 -0
  12. train_results.json +9 -0
  13. trainer_log.jsonl +465 -0
  14. trainer_state.json +3291 -0
  15. training_args.bin +3 -0
  16. training_loss.png +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
.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
+ back to smaller dataset more accurate
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", 2.5e-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
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
+ back to smaller dataset more accurate
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", 2.5e-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": 2201.716707715399,
3
+ "epoch": 4.0,
4
+ "total_flos": 3.61934117404672e+18,
5
+ "train_loss": 0.7849958032722874,
6
+ "train_runtime": 32566.8601,
7
+ "train_samples_per_second": 1.71,
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:bf7e29a376629e10c34cf5ab8027a92e1d40ee7725265431f2d3c9ef2f8c490f
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": 2201.716707715399,
3
+ "epoch": 4.0,
4
+ "total_flos": 3.61934117404672e+18,
5
+ "train_loss": 0.7849958032722874,
6
+ "train_runtime": 32566.8601,
7
+ "train_samples_per_second": 1.71,
8
+ "train_steps_per_second": 0.028
9
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,465 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 2, "total_steps": 928, "loss": 3.7070133686065674, "lr": 5.319148936170213e-07, "epoch": 0.008620689655172414, "percentage": 0.22, "elapsed_time": "0:01:18", "remaining_time": "10:09:18"}
2
+ {"current_steps": 4, "total_steps": 928, "loss": 2.565812349319458, "lr": 1.5957446808510639e-06, "epoch": 0.017241379310344827, "percentage": 0.43, "elapsed_time": "0:02:40", "remaining_time": "10:17:44"}
3
+ {"current_steps": 6, "total_steps": 928, "loss": 2.305459976196289, "lr": 2.6595744680851065e-06, "epoch": 0.02586206896551724, "percentage": 0.65, "elapsed_time": "0:03:40", "remaining_time": "9:25:44"}
4
+ {"current_steps": 8, "total_steps": 928, "loss": 1.9556151628494263, "lr": 3.723404255319149e-06, "epoch": 0.034482758620689655, "percentage": 0.86, "elapsed_time": "0:05:06", "remaining_time": "9:46:50"}
5
+ {"current_steps": 10, "total_steps": 928, "loss": 1.509597897529602, "lr": 4.787234042553191e-06, "epoch": 0.04310344827586207, "percentage": 1.08, "elapsed_time": "0:06:17", "remaining_time": "9:37:24"}
6
+ {"current_steps": 12, "total_steps": 928, "loss": 1.7625975608825684, "lr": 5.851063829787235e-06, "epoch": 0.05172413793103448, "percentage": 1.29, "elapsed_time": "0:07:43", "remaining_time": "9:49:55"}
7
+ {"current_steps": 14, "total_steps": 928, "loss": 1.8272696733474731, "lr": 6.914893617021277e-06, "epoch": 0.0603448275862069, "percentage": 1.51, "elapsed_time": "0:09:01", "remaining_time": "9:49:10"}
8
+ {"current_steps": 16, "total_steps": 928, "loss": 0.7375114560127258, "lr": 7.97872340425532e-06, "epoch": 0.06896551724137931, "percentage": 1.72, "elapsed_time": "0:09:44", "remaining_time": "9:15:02"}
9
+ {"current_steps": 18, "total_steps": 928, "loss": 0.8689496517181396, "lr": 9.042553191489362e-06, "epoch": 0.07758620689655173, "percentage": 1.94, "elapsed_time": "0:10:30", "remaining_time": "8:51:09"}
10
+ {"current_steps": 20, "total_steps": 928, "loss": 1.5243046283721924, "lr": 1.0106382978723404e-05, "epoch": 0.08620689655172414, "percentage": 2.16, "elapsed_time": "0:11:55", "remaining_time": "9:01:40"}
11
+ {"current_steps": 22, "total_steps": 928, "loss": 1.672706127166748, "lr": 1.1170212765957447e-05, "epoch": 0.09482758620689655, "percentage": 2.37, "elapsed_time": "0:13:07", "remaining_time": "9:00:23"}
12
+ {"current_steps": 24, "total_steps": 928, "loss": 1.5508843660354614, "lr": 1.223404255319149e-05, "epoch": 0.10344827586206896, "percentage": 2.59, "elapsed_time": "0:14:26", "remaining_time": "9:04:01"}
13
+ {"current_steps": 26, "total_steps": 928, "loss": 1.3349249362945557, "lr": 1.3297872340425532e-05, "epoch": 0.11206896551724138, "percentage": 2.8, "elapsed_time": "0:15:46", "remaining_time": "9:07:14"}
14
+ {"current_steps": 28, "total_steps": 928, "loss": 0.764009952545166, "lr": 1.4361702127659577e-05, "epoch": 0.1206896551724138, "percentage": 3.02, "elapsed_time": "0:16:29", "remaining_time": "8:50:09"}
15
+ {"current_steps": 30, "total_steps": 928, "loss": 1.4469249248504639, "lr": 1.5425531914893617e-05, "epoch": 0.12931034482758622, "percentage": 3.23, "elapsed_time": "0:17:54", "remaining_time": "8:56:11"}
16
+ {"current_steps": 32, "total_steps": 928, "loss": 1.205310583114624, "lr": 1.6489361702127658e-05, "epoch": 0.13793103448275862, "percentage": 3.45, "elapsed_time": "0:18:58", "remaining_time": "8:51:29"}
17
+ {"current_steps": 34, "total_steps": 928, "loss": 1.3685661554336548, "lr": 1.7553191489361703e-05, "epoch": 0.14655172413793102, "percentage": 3.66, "elapsed_time": "0:20:09", "remaining_time": "8:50:07"}
18
+ {"current_steps": 36, "total_steps": 928, "loss": 1.3990949392318726, "lr": 1.8617021276595745e-05, "epoch": 0.15517241379310345, "percentage": 3.88, "elapsed_time": "0:21:33", "remaining_time": "8:54:20"}
19
+ {"current_steps": 38, "total_steps": 928, "loss": 1.4243614673614502, "lr": 1.968085106382979e-05, "epoch": 0.16379310344827586, "percentage": 4.09, "elapsed_time": "0:22:25", "remaining_time": "8:45:10"}
20
+ {"current_steps": 40, "total_steps": 928, "loss": 1.4310336112976074, "lr": 2.074468085106383e-05, "epoch": 0.1724137931034483, "percentage": 4.31, "elapsed_time": "0:23:32", "remaining_time": "8:42:33"}
21
+ {"current_steps": 42, "total_steps": 928, "loss": 1.6425907611846924, "lr": 2.1808510638297873e-05, "epoch": 0.1810344827586207, "percentage": 4.53, "elapsed_time": "0:24:52", "remaining_time": "8:44:51"}
22
+ {"current_steps": 44, "total_steps": 928, "loss": 1.0640761852264404, "lr": 2.2872340425531915e-05, "epoch": 0.1896551724137931, "percentage": 4.74, "elapsed_time": "0:25:53", "remaining_time": "8:40:01"}
23
+ {"current_steps": 46, "total_steps": 928, "loss": 1.133437156677246, "lr": 2.393617021276596e-05, "epoch": 0.19827586206896552, "percentage": 4.96, "elapsed_time": "0:26:56", "remaining_time": "8:36:26"}
24
+ {"current_steps": 48, "total_steps": 928, "loss": 1.3954966068267822, "lr": 2.5e-05, "epoch": 0.20689655172413793, "percentage": 5.17, "elapsed_time": "0:28:16", "remaining_time": "8:38:16"}
25
+ {"current_steps": 50, "total_steps": 928, "loss": 1.5042736530303955, "lr": 2.4999713892919425e-05, "epoch": 0.21551724137931033, "percentage": 5.39, "elapsed_time": "0:29:28", "remaining_time": "8:37:32"}
26
+ {"current_steps": 52, "total_steps": 928, "loss": 0.8469613194465637, "lr": 2.4998855586230096e-05, "epoch": 0.22413793103448276, "percentage": 5.6, "elapsed_time": "0:30:08", "remaining_time": "8:27:49"}
27
+ {"current_steps": 54, "total_steps": 928, "loss": 1.439496636390686, "lr": 2.4997425123588484e-05, "epoch": 0.23275862068965517, "percentage": 5.82, "elapsed_time": "0:31:18", "remaining_time": "8:26:44"}
28
+ {"current_steps": 56, "total_steps": 928, "loss": 1.3860807418823242, "lr": 2.4995422577752894e-05, "epoch": 0.2413793103448276, "percentage": 6.03, "elapsed_time": "0:32:42", "remaining_time": "8:29:22"}
29
+ {"current_steps": 58, "total_steps": 928, "loss": 1.4035521745681763, "lr": 2.499284805057978e-05, "epoch": 0.25, "percentage": 6.25, "elapsed_time": "0:33:54", "remaining_time": "8:28:36"}
30
+ {"current_steps": 60, "total_steps": 928, "loss": 1.399681568145752, "lr": 2.4989701673018555e-05, "epoch": 0.25862068965517243, "percentage": 6.47, "elapsed_time": "0:35:19", "remaining_time": "8:31:02"}
31
+ {"current_steps": 62, "total_steps": 928, "loss": 1.099133849143982, "lr": 2.4985983605104933e-05, "epoch": 0.2672413793103448, "percentage": 6.68, "elapsed_time": "0:36:10", "remaining_time": "8:25:14"}
32
+ {"current_steps": 64, "total_steps": 928, "loss": 1.333014726638794, "lr": 2.4981694035952797e-05, "epoch": 0.27586206896551724, "percentage": 6.9, "elapsed_time": "0:37:35", "remaining_time": "8:27:22"}
33
+ {"current_steps": 66, "total_steps": 928, "loss": 1.316357135772705, "lr": 2.497683318374456e-05, "epoch": 0.28448275862068967, "percentage": 7.11, "elapsed_time": "0:38:50", "remaining_time": "8:27:17"}
34
+ {"current_steps": 68, "total_steps": 928, "loss": 1.3247264623641968, "lr": 2.497140129572009e-05, "epoch": 0.29310344827586204, "percentage": 7.33, "elapsed_time": "0:40:36", "remaining_time": "8:33:39"}
35
+ {"current_steps": 70, "total_steps": 928, "loss": 1.0423731803894043, "lr": 2.4965398648164114e-05, "epoch": 0.3017241379310345, "percentage": 7.54, "elapsed_time": "0:41:48", "remaining_time": "8:32:21"}
36
+ {"current_steps": 72, "total_steps": 928, "loss": 1.1186097860336304, "lr": 2.495882554639219e-05, "epoch": 0.3103448275862069, "percentage": 7.76, "elapsed_time": "0:42:46", "remaining_time": "8:28:37"}
37
+ {"current_steps": 74, "total_steps": 928, "loss": 1.0758298635482788, "lr": 2.4951682324735147e-05, "epoch": 0.31896551724137934, "percentage": 7.97, "elapsed_time": "0:43:55", "remaining_time": "8:26:50"}
38
+ {"current_steps": 76, "total_steps": 928, "loss": 1.177038311958313, "lr": 2.4943969346522113e-05, "epoch": 0.3275862068965517, "percentage": 8.19, "elapsed_time": "0:44:44", "remaining_time": "8:21:33"}
39
+ {"current_steps": 78, "total_steps": 928, "loss": 1.1172969341278076, "lr": 2.4935687004062e-05, "epoch": 0.33620689655172414, "percentage": 8.41, "elapsed_time": "0:45:30", "remaining_time": "8:15:59"}
40
+ {"current_steps": 80, "total_steps": 928, "loss": 0.925361156463623, "lr": 2.492683571862358e-05, "epoch": 0.3448275862068966, "percentage": 8.62, "elapsed_time": "0:46:40", "remaining_time": "8:14:45"}
41
+ {"current_steps": 82, "total_steps": 928, "loss": 1.3231241703033447, "lr": 2.491741594041404e-05, "epoch": 0.35344827586206895, "percentage": 8.84, "elapsed_time": "0:48:05", "remaining_time": "8:16:12"}
42
+ {"current_steps": 84, "total_steps": 928, "loss": 1.1561291217803955, "lr": 2.4907428148556096e-05, "epoch": 0.3620689655172414, "percentage": 9.05, "elapsed_time": "0:49:10", "remaining_time": "8:14:09"}
43
+ {"current_steps": 86, "total_steps": 928, "loss": 1.3458794355392456, "lr": 2.48968728510636e-05, "epoch": 0.3706896551724138, "percentage": 9.27, "elapsed_time": "0:50:32", "remaining_time": "8:14:48"}
44
+ {"current_steps": 88, "total_steps": 928, "loss": 1.2599420547485352, "lr": 2.4885750584815733e-05, "epoch": 0.3793103448275862, "percentage": 9.48, "elapsed_time": "0:51:57", "remaining_time": "8:15:54"}
45
+ {"current_steps": 90, "total_steps": 928, "loss": 1.0761513710021973, "lr": 2.487406191552967e-05, "epoch": 0.3879310344827586, "percentage": 9.7, "elapsed_time": "0:53:01", "remaining_time": "8:13:43"}
46
+ {"current_steps": 92, "total_steps": 928, "loss": 0.8632556200027466, "lr": 2.4861807437731816e-05, "epoch": 0.39655172413793105, "percentage": 9.91, "elapsed_time": "0:53:43", "remaining_time": "8:08:14"}
47
+ {"current_steps": 94, "total_steps": 928, "loss": 1.113997220993042, "lr": 2.4848987774727572e-05, "epoch": 0.4051724137931034, "percentage": 10.13, "elapsed_time": "0:54:53", "remaining_time": "8:07:04"}
48
+ {"current_steps": 96, "total_steps": 928, "loss": 1.225900650024414, "lr": 2.4835603578569627e-05, "epoch": 0.41379310344827586, "percentage": 10.34, "elapsed_time": "0:56:06", "remaining_time": "8:06:16"}
49
+ {"current_steps": 98, "total_steps": 928, "loss": 1.2307943105697632, "lr": 2.4821655530024785e-05, "epoch": 0.4224137931034483, "percentage": 10.56, "elapsed_time": "0:57:31", "remaining_time": "8:07:11"}
50
+ {"current_steps": 100, "total_steps": 928, "loss": 1.150252103805542, "lr": 2.4807144338539345e-05, "epoch": 0.43103448275862066, "percentage": 10.78, "elapsed_time": "0:58:40", "remaining_time": "8:05:46"}
51
+ {"current_steps": 102, "total_steps": 928, "loss": 1.1921312808990479, "lr": 2.479207074220303e-05, "epoch": 0.4396551724137931, "percentage": 10.99, "elapsed_time": "0:59:43", "remaining_time": "8:03:40"}
52
+ {"current_steps": 104, "total_steps": 928, "loss": 1.2748147249221802, "lr": 2.4776435507711404e-05, "epoch": 0.4482758620689655, "percentage": 11.21, "elapsed_time": "1:01:08", "remaining_time": "8:04:26"}
53
+ {"current_steps": 106, "total_steps": 928, "loss": 0.961470365524292, "lr": 2.476023943032693e-05, "epoch": 0.45689655172413796, "percentage": 11.42, "elapsed_time": "1:02:12", "remaining_time": "8:02:26"}
54
+ {"current_steps": 108, "total_steps": 928, "loss": 1.183750867843628, "lr": 2.474348333383848e-05, "epoch": 0.46551724137931033, "percentage": 11.64, "elapsed_time": "1:03:35", "remaining_time": "8:02:53"}
55
+ {"current_steps": 110, "total_steps": 928, "loss": 1.1943963766098022, "lr": 2.4726168070519457e-05, "epoch": 0.47413793103448276, "percentage": 11.85, "elapsed_time": "1:04:49", "remaining_time": "8:02:00"}
56
+ {"current_steps": 112, "total_steps": 928, "loss": 1.4336156845092773, "lr": 2.4708294521084412e-05, "epoch": 0.4827586206896552, "percentage": 12.07, "elapsed_time": "1:06:05", "remaining_time": "8:01:30"}
57
+ {"current_steps": 114, "total_steps": 928, "loss": 0.9574199318885803, "lr": 2.46898635946443e-05, "epoch": 0.49137931034482757, "percentage": 12.28, "elapsed_time": "1:07:08", "remaining_time": "7:59:21"}
58
+ {"current_steps": 116, "total_steps": 928, "loss": 0.8921310901641846, "lr": 2.4670876228660192e-05, "epoch": 0.5, "percentage": 12.5, "elapsed_time": "1:08:11", "remaining_time": "7:57:21"}
59
+ {"current_steps": 118, "total_steps": 928, "loss": 1.2967804670333862, "lr": 2.4651333388895617e-05, "epoch": 0.5086206896551724, "percentage": 12.72, "elapsed_time": "1:09:36", "remaining_time": "7:57:51"}
60
+ {"current_steps": 120, "total_steps": 928, "loss": 1.2260866165161133, "lr": 2.4631236069367435e-05, "epoch": 0.5172413793103449, "percentage": 12.93, "elapsed_time": "1:10:47", "remaining_time": "7:56:43"}
61
+ {"current_steps": 122, "total_steps": 928, "loss": 1.3115153312683105, "lr": 2.4610585292295278e-05, "epoch": 0.5258620689655172, "percentage": 13.15, "elapsed_time": "1:12:13", "remaining_time": "7:57:11"}
62
+ {"current_steps": 124, "total_steps": 928, "loss": 1.458106279373169, "lr": 2.458938210804955e-05, "epoch": 0.5344827586206896, "percentage": 13.36, "elapsed_time": "1:13:29", "remaining_time": "7:56:29"}
63
+ {"current_steps": 126, "total_steps": 928, "loss": 1.589352011680603, "lr": 2.456762759509801e-05, "epoch": 0.5431034482758621, "percentage": 13.58, "elapsed_time": "1:14:52", "remaining_time": "7:56:35"}
64
+ {"current_steps": 128, "total_steps": 928, "loss": 1.3105168342590332, "lr": 2.454532285995091e-05, "epoch": 0.5517241379310345, "percentage": 13.79, "elapsed_time": "1:15:46", "remaining_time": "7:53:33"}
65
+ {"current_steps": 130, "total_steps": 928, "loss": 1.2909717559814453, "lr": 2.4522469037104732e-05, "epoch": 0.5603448275862069, "percentage": 14.01, "elapsed_time": "1:17:00", "remaining_time": "7:52:43"}
66
+ {"current_steps": 132, "total_steps": 928, "loss": 1.2390798330307007, "lr": 2.449906728898445e-05, "epoch": 0.5689655172413793, "percentage": 14.22, "elapsed_time": "1:18:20", "remaining_time": "7:52:27"}
67
+ {"current_steps": 134, "total_steps": 928, "loss": 1.150702714920044, "lr": 2.447511880588444e-05, "epoch": 0.5775862068965517, "percentage": 14.44, "elapsed_time": "1:19:20", "remaining_time": "7:50:09"}
68
+ {"current_steps": 136, "total_steps": 928, "loss": 0.978553056716919, "lr": 2.445062480590792e-05, "epoch": 0.5862068965517241, "percentage": 14.66, "elapsed_time": "1:20:25", "remaining_time": "7:48:19"}
69
+ {"current_steps": 138, "total_steps": 928, "loss": 0.9309189915657043, "lr": 2.4425586534904986e-05, "epoch": 0.5948275862068966, "percentage": 14.87, "elapsed_time": "1:21:10", "remaining_time": "7:44:43"}
70
+ {"current_steps": 140, "total_steps": 928, "loss": 0.9206077456474304, "lr": 2.440000526640926e-05, "epoch": 0.603448275862069, "percentage": 15.09, "elapsed_time": "1:22:12", "remaining_time": "7:42:42"}
71
+ {"current_steps": 142, "total_steps": 928, "loss": 1.0149747133255005, "lr": 2.437388230157311e-05, "epoch": 0.6120689655172413, "percentage": 15.3, "elapsed_time": "1:23:18", "remaining_time": "7:41:05"}
72
+ {"current_steps": 144, "total_steps": 928, "loss": 1.004448413848877, "lr": 2.4347218969101456e-05, "epoch": 0.6206896551724138, "percentage": 15.52, "elapsed_time": "1:24:23", "remaining_time": "7:39:29"}
73
+ {"current_steps": 146, "total_steps": 928, "loss": 1.251673698425293, "lr": 2.4320016625184208e-05, "epoch": 0.6293103448275862, "percentage": 15.73, "elapsed_time": "1:25:29", "remaining_time": "7:37:53"}
74
+ {"current_steps": 148, "total_steps": 928, "loss": 1.2662962675094604, "lr": 2.4292276653427266e-05, "epoch": 0.6379310344827587, "percentage": 15.95, "elapsed_time": "1:26:29", "remaining_time": "7:35:51"}
75
+ {"current_steps": 150, "total_steps": 928, "loss": 1.05961275100708, "lr": 2.4264000464782154e-05, "epoch": 0.646551724137931, "percentage": 16.16, "elapsed_time": "1:27:29", "remaining_time": "7:33:45"}
76
+ {"current_steps": 152, "total_steps": 928, "loss": 0.8592603206634521, "lr": 2.423518949747426e-05, "epoch": 0.6551724137931034, "percentage": 16.38, "elapsed_time": "1:28:31", "remaining_time": "7:31:57"}
77
+ {"current_steps": 154, "total_steps": 928, "loss": 1.0430630445480347, "lr": 2.420584521692967e-05, "epoch": 0.6637931034482759, "percentage": 16.59, "elapsed_time": "1:29:43", "remaining_time": "7:30:56"}
78
+ {"current_steps": 156, "total_steps": 928, "loss": 1.1310160160064697, "lr": 2.4175969115700637e-05, "epoch": 0.6724137931034483, "percentage": 16.81, "elapsed_time": "1:30:57", "remaining_time": "7:30:07"}
79
+ {"current_steps": 158, "total_steps": 928, "loss": 1.237598180770874, "lr": 2.4145562713389665e-05, "epoch": 0.6810344827586207, "percentage": 17.03, "elapsed_time": "1:32:21", "remaining_time": "7:30:07"}
80
+ {"current_steps": 160, "total_steps": 928, "loss": 1.2904868125915527, "lr": 2.4114627556572236e-05, "epoch": 0.6896551724137931, "percentage": 17.24, "elapsed_time": "1:33:36", "remaining_time": "7:29:17"}
81
+ {"current_steps": 162, "total_steps": 928, "loss": 1.5490325689315796, "lr": 2.4083165218718094e-05, "epoch": 0.6982758620689655, "percentage": 17.46, "elapsed_time": "1:34:54", "remaining_time": "7:28:47"}
82
+ {"current_steps": 164, "total_steps": 928, "loss": 0.9844979643821716, "lr": 2.405117730011127e-05, "epoch": 0.7068965517241379, "percentage": 17.67, "elapsed_time": "1:35:59", "remaining_time": "7:27:08"}
83
+ {"current_steps": 166, "total_steps": 928, "loss": 1.264622449874878, "lr": 2.4018665427768652e-05, "epoch": 0.7155172413793104, "percentage": 17.89, "elapsed_time": "1:37:02", "remaining_time": "7:25:26"}
84
+ {"current_steps": 168, "total_steps": 928, "loss": 1.3305325508117676, "lr": 2.3985631255357243e-05, "epoch": 0.7241379310344828, "percentage": 18.1, "elapsed_time": "1:38:25", "remaining_time": "7:25:17"}
85
+ {"current_steps": 170, "total_steps": 928, "loss": 1.2382071018218994, "lr": 2.3952076463110038e-05, "epoch": 0.7327586206896551, "percentage": 18.32, "elapsed_time": "1:39:51", "remaining_time": "7:25:13"}
86
+ {"current_steps": 172, "total_steps": 928, "loss": 1.249595046043396, "lr": 2.3918002757740583e-05, "epoch": 0.7413793103448276, "percentage": 18.53, "elapsed_time": "1:41:13", "remaining_time": "7:24:53"}
87
+ {"current_steps": 174, "total_steps": 928, "loss": 0.9225230813026428, "lr": 2.3883411872356135e-05, "epoch": 0.75, "percentage": 18.75, "elapsed_time": "1:42:05", "remaining_time": "7:22:24"}
88
+ {"current_steps": 176, "total_steps": 928, "loss": 0.6734743714332581, "lr": 2.3848305566369543e-05, "epoch": 0.7586206896551724, "percentage": 18.97, "elapsed_time": "1:42:52", "remaining_time": "7:19:33"}
89
+ {"current_steps": 178, "total_steps": 928, "loss": 1.24442458152771, "lr": 2.3812685625409726e-05, "epoch": 0.7672413793103449, "percentage": 19.18, "elapsed_time": "1:44:07", "remaining_time": "7:18:42"}
90
+ {"current_steps": 180, "total_steps": 928, "loss": 1.170575499534607, "lr": 2.377655386123088e-05, "epoch": 0.7758620689655172, "percentage": 19.4, "elapsed_time": "1:45:13", "remaining_time": "7:17:15"}
91
+ {"current_steps": 182, "total_steps": 928, "loss": 0.8946050405502319, "lr": 2.3739912111620303e-05, "epoch": 0.7844827586206896, "percentage": 19.61, "elapsed_time": "1:46:15", "remaining_time": "7:15:34"}
92
+ {"current_steps": 184, "total_steps": 928, "loss": 0.7359169125556946, "lr": 2.3702762240304933e-05, "epoch": 0.7931034482758621, "percentage": 19.83, "elapsed_time": "1:47:18", "remaining_time": "7:13:54"}
93
+ {"current_steps": 186, "total_steps": 928, "loss": 1.377792477607727, "lr": 2.366510613685654e-05, "epoch": 0.8017241379310345, "percentage": 20.04, "elapsed_time": "1:48:31", "remaining_time": "7:12:56"}
94
+ {"current_steps": 188, "total_steps": 928, "loss": 1.5076112747192383, "lr": 2.362694571659563e-05, "epoch": 0.8103448275862069, "percentage": 20.26, "elapsed_time": "1:49:44", "remaining_time": "7:11:56"}
95
+ {"current_steps": 190, "total_steps": 928, "loss": 1.2605912685394287, "lr": 2.3588282920494007e-05, "epoch": 0.8189655172413793, "percentage": 20.47, "elapsed_time": "1:51:09", "remaining_time": "7:11:45"}
96
+ {"current_steps": 192, "total_steps": 928, "loss": 1.2289090156555176, "lr": 2.354911971507607e-05, "epoch": 0.8275862068965517, "percentage": 20.69, "elapsed_time": "1:52:35", "remaining_time": "7:11:36"}
97
+ {"current_steps": 194, "total_steps": 928, "loss": 1.2607473134994507, "lr": 2.3509458092318772e-05, "epoch": 0.8362068965517241, "percentage": 20.91, "elapsed_time": "1:53:46", "remaining_time": "7:10:27"}
98
+ {"current_steps": 196, "total_steps": 928, "loss": 1.2929327487945557, "lr": 2.3469300069550308e-05, "epoch": 0.8448275862068966, "percentage": 21.12, "elapsed_time": "1:55:09", "remaining_time": "7:10:04"}
99
+ {"current_steps": 198, "total_steps": 928, "loss": 0.8722022175788879, "lr": 2.342864768934751e-05, "epoch": 0.853448275862069, "percentage": 21.34, "elapsed_time": "1:56:13", "remaining_time": "7:08:30"}
100
+ {"current_steps": 200, "total_steps": 928, "loss": 1.184353232383728, "lr": 2.338750301943194e-05, "epoch": 0.8620689655172413, "percentage": 21.55, "elapsed_time": "1:57:20", "remaining_time": "7:07:08"}
101
+ {"current_steps": 202, "total_steps": 928, "loss": 1.3033404350280762, "lr": 2.3345868152564734e-05, "epoch": 0.8706896551724138, "percentage": 21.77, "elapsed_time": "1:58:40", "remaining_time": "7:06:32"}
102
+ {"current_steps": 204, "total_steps": 928, "loss": 0.8301224708557129, "lr": 2.3303745206440156e-05, "epoch": 0.8793103448275862, "percentage": 21.98, "elapsed_time": "1:59:45", "remaining_time": "7:04:59"}
103
+ {"current_steps": 206, "total_steps": 928, "loss": 1.2247326374053955, "lr": 2.3261136323577873e-05, "epoch": 0.8879310344827587, "percentage": 22.2, "elapsed_time": "2:01:00", "remaining_time": "7:04:05"}
104
+ {"current_steps": 208, "total_steps": 928, "loss": 1.275619626045227, "lr": 2.3218043671213988e-05, "epoch": 0.896551724137931, "percentage": 22.41, "elapsed_time": "2:02:25", "remaining_time": "7:03:46"}
105
+ {"current_steps": 210, "total_steps": 928, "loss": 0.763530969619751, "lr": 2.317446944119081e-05, "epoch": 0.9051724137931034, "percentage": 22.63, "elapsed_time": "2:03:27", "remaining_time": "7:02:05"}
106
+ {"current_steps": 212, "total_steps": 928, "loss": 0.8847395181655884, "lr": 2.313041584984535e-05, "epoch": 0.9137931034482759, "percentage": 22.84, "elapsed_time": "2:04:28", "remaining_time": "7:00:24"}
107
+ {"current_steps": 214, "total_steps": 928, "loss": 1.139022707939148, "lr": 2.3085885137896623e-05, "epoch": 0.9224137931034483, "percentage": 23.06, "elapsed_time": "2:05:40", "remaining_time": "6:59:17"}
108
+ {"current_steps": 216, "total_steps": 928, "loss": 1.2234928607940674, "lr": 2.3040879570331648e-05, "epoch": 0.9310344827586207, "percentage": 23.28, "elapsed_time": "2:06:52", "remaining_time": "6:58:13"}
109
+ {"current_steps": 218, "total_steps": 928, "loss": 0.7140777111053467, "lr": 2.2995401436290252e-05, "epoch": 0.9396551724137931, "percentage": 23.49, "elapsed_time": "2:07:57", "remaining_time": "6:56:45"}
110
+ {"current_steps": 220, "total_steps": 928, "loss": 0.8441833257675171, "lr": 2.294945304894864e-05, "epoch": 0.9482758620689655, "percentage": 23.71, "elapsed_time": "2:08:59", "remaining_time": "6:55:08"}
111
+ {"current_steps": 222, "total_steps": 928, "loss": 1.3300198316574097, "lr": 2.2903036745401742e-05, "epoch": 0.9568965517241379, "percentage": 23.92, "elapsed_time": "2:10:23", "remaining_time": "6:54:39"}
112
+ {"current_steps": 224, "total_steps": 928, "loss": 0.8017496466636658, "lr": 2.2856154886544338e-05, "epoch": 0.9655172413793104, "percentage": 24.14, "elapsed_time": "2:11:26", "remaining_time": "6:53:07"}
113
+ {"current_steps": 226, "total_steps": 928, "loss": 0.8339143991470337, "lr": 2.280880985695095e-05, "epoch": 0.9741379310344828, "percentage": 24.35, "elapsed_time": "2:12:49", "remaining_time": "6:52:35"}
114
+ {"current_steps": 228, "total_steps": 928, "loss": 1.2216969728469849, "lr": 2.27610040647546e-05, "epoch": 0.9827586206896551, "percentage": 24.57, "elapsed_time": "2:14:15", "remaining_time": "6:52:12"}
115
+ {"current_steps": 230, "total_steps": 928, "loss": 0.8863965272903442, "lr": 2.2712739941524287e-05, "epoch": 0.9913793103448276, "percentage": 24.78, "elapsed_time": "2:15:04", "remaining_time": "6:49:54"}
116
+ {"current_steps": 232, "total_steps": 928, "loss": 0.7733997702598572, "lr": 2.266401994214133e-05, "epoch": 1.0, "percentage": 25.0, "elapsed_time": "2:16:06", "remaining_time": "6:48:18"}
117
+ {"current_steps": 234, "total_steps": 928, "loss": 0.8990653157234192, "lr": 2.2614846544674486e-05, "epoch": 1.0086206896551724, "percentage": 25.22, "elapsed_time": "2:17:18", "remaining_time": "6:47:13"}
118
+ {"current_steps": 236, "total_steps": 928, "loss": 0.7269018888473511, "lr": 2.2565222250253925e-05, "epoch": 1.0172413793103448, "percentage": 25.43, "elapsed_time": "2:18:20", "remaining_time": "6:45:38"}
119
+ {"current_steps": 238, "total_steps": 928, "loss": 0.8784797787666321, "lr": 2.251514958294401e-05, "epoch": 1.0258620689655173, "percentage": 25.65, "elapsed_time": "2:19:31", "remaining_time": "6:44:31"}
120
+ {"current_steps": 240, "total_steps": 928, "loss": 0.5005398988723755, "lr": 2.246463108961489e-05, "epoch": 1.0344827586206897, "percentage": 25.86, "elapsed_time": "2:20:20", "remaining_time": "6:42:18"}
121
+ {"current_steps": 242, "total_steps": 928, "loss": 1.1350990533828735, "lr": 2.2413669339812997e-05, "epoch": 1.043103448275862, "percentage": 26.08, "elapsed_time": "2:21:36", "remaining_time": "6:41:23"}
122
+ {"current_steps": 244, "total_steps": 928, "loss": 0.36204367876052856, "lr": 2.2362266925630318e-05, "epoch": 1.0517241379310345, "percentage": 26.29, "elapsed_time": "2:22:20", "remaining_time": "6:39:00"}
123
+ {"current_steps": 246, "total_steps": 928, "loss": 1.1106600761413574, "lr": 2.231042646157258e-05, "epoch": 1.0603448275862069, "percentage": 26.51, "elapsed_time": "2:23:44", "remaining_time": "6:38:29"}
124
+ {"current_steps": 248, "total_steps": 928, "loss": 1.0469825267791748, "lr": 2.2258150584426235e-05, "epoch": 1.0689655172413792, "percentage": 26.72, "elapsed_time": "2:25:08", "remaining_time": "6:37:58"}
125
+ {"current_steps": 250, "total_steps": 928, "loss": 1.0015485286712646, "lr": 2.220544195312436e-05, "epoch": 1.0775862068965518, "percentage": 26.94, "elapsed_time": "2:26:16", "remaining_time": "6:36:43"}
126
+ {"current_steps": 252, "total_steps": 928, "loss": 0.36394670605659485, "lr": 2.2152303248611437e-05, "epoch": 1.0862068965517242, "percentage": 27.16, "elapsed_time": "2:26:56", "remaining_time": "6:34:10"}
127
+ {"current_steps": 254, "total_steps": 928, "loss": 1.0145013332366943, "lr": 2.2098737173706942e-05, "epoch": 1.0948275862068966, "percentage": 27.37, "elapsed_time": "2:28:09", "remaining_time": "6:33:07"}
128
+ {"current_steps": 256, "total_steps": 928, "loss": 0.7123383283615112, "lr": 2.2044746452967906e-05, "epoch": 1.103448275862069, "percentage": 27.59, "elapsed_time": "2:29:10", "remaining_time": "6:31:34"}
129
+ {"current_steps": 258, "total_steps": 928, "loss": 0.4990229606628418, "lr": 2.1990333832550333e-05, "epoch": 1.1120689655172413, "percentage": 27.8, "elapsed_time": "2:29:56", "remaining_time": "6:29:23"}
130
+ {"current_steps": 260, "total_steps": 928, "loss": 0.7109271287918091, "lr": 2.1935502080069505e-05, "epoch": 1.1206896551724137, "percentage": 28.02, "elapsed_time": "2:30:59", "remaining_time": "6:27:57"}
131
+ {"current_steps": 262, "total_steps": 928, "loss": 1.2138782739639282, "lr": 2.1880253984459224e-05, "epoch": 1.1293103448275863, "percentage": 28.23, "elapsed_time": "2:32:24", "remaining_time": "6:27:26"}
132
+ {"current_steps": 264, "total_steps": 928, "loss": 0.8597405552864075, "lr": 2.1824592355829945e-05, "epoch": 1.1379310344827587, "percentage": 28.45, "elapsed_time": "2:33:35", "remaining_time": "6:26:19"}
133
+ {"current_steps": 266, "total_steps": 928, "loss": 1.1350115537643433, "lr": 2.1768520025325866e-05, "epoch": 1.146551724137931, "percentage": 28.66, "elapsed_time": "2:34:50", "remaining_time": "6:25:21"}
134
+ {"current_steps": 268, "total_steps": 928, "loss": 1.1547629833221436, "lr": 2.1712039844980893e-05, "epoch": 1.1551724137931034, "percentage": 28.88, "elapsed_time": "2:36:03", "remaining_time": "6:24:20"}
135
+ {"current_steps": 270, "total_steps": 928, "loss": 1.1632627248764038, "lr": 2.165515468757362e-05, "epoch": 1.1637931034482758, "percentage": 29.09, "elapsed_time": "2:37:27", "remaining_time": "6:23:44"}
136
+ {"current_steps": 272, "total_steps": 928, "loss": 1.0813562870025635, "lr": 2.1597867446481155e-05, "epoch": 1.1724137931034484, "percentage": 29.31, "elapsed_time": "2:38:52", "remaining_time": "6:23:09"}
137
+ {"current_steps": 274, "total_steps": 928, "loss": 0.7284541130065918, "lr": 2.1540181035532004e-05, "epoch": 1.1810344827586208, "percentage": 29.53, "elapsed_time": "2:39:54", "remaining_time": "6:21:40"}
138
+ {"current_steps": 276, "total_steps": 928, "loss": 0.3468450903892517, "lr": 2.1482098388857834e-05, "epoch": 1.1896551724137931, "percentage": 29.74, "elapsed_time": "2:40:37", "remaining_time": "6:19:27"}
139
+ {"current_steps": 278, "total_steps": 928, "loss": 1.121955394744873, "lr": 2.1423622460744242e-05, "epoch": 1.1982758620689655, "percentage": 29.96, "elapsed_time": "2:42:03", "remaining_time": "6:18:55"}
140
+ {"current_steps": 280, "total_steps": 928, "loss": 1.0472161769866943, "lr": 2.1364756225480478e-05, "epoch": 1.206896551724138, "percentage": 30.17, "elapsed_time": "2:43:22", "remaining_time": "6:18:06"}
141
+ {"current_steps": 282, "total_steps": 928, "loss": 0.5945373773574829, "lr": 2.1305502677208184e-05, "epoch": 1.2155172413793103, "percentage": 30.39, "elapsed_time": "2:44:28", "remaining_time": "6:16:46"}
142
+ {"current_steps": 284, "total_steps": 928, "loss": 0.7433868646621704, "lr": 2.1245864829769092e-05, "epoch": 1.2241379310344827, "percentage": 30.6, "elapsed_time": "2:45:27", "remaining_time": "6:15:12"}
143
+ {"current_steps": 286, "total_steps": 928, "loss": 0.8308318853378296, "lr": 2.1185845716551712e-05, "epoch": 1.2327586206896552, "percentage": 30.82, "elapsed_time": "2:46:29", "remaining_time": "6:13:44"}
144
+ {"current_steps": 288, "total_steps": 928, "loss": 0.3982611894607544, "lr": 2.1125448390337072e-05, "epoch": 1.2413793103448276, "percentage": 31.03, "elapsed_time": "2:47:13", "remaining_time": "6:11:36"}
145
+ {"current_steps": 290, "total_steps": 928, "loss": 0.9474261403083801, "lr": 2.106467592314343e-05, "epoch": 1.25, "percentage": 31.25, "elapsed_time": "2:48:19", "remaining_time": "6:10:18"}
146
+ {"current_steps": 292, "total_steps": 928, "loss": 0.927026093006134, "lr": 2.1003531406070012e-05, "epoch": 1.2586206896551724, "percentage": 31.47, "elapsed_time": "2:49:31", "remaining_time": "6:09:13"}
147
+ {"current_steps": 294, "total_steps": 928, "loss": 1.0189390182495117, "lr": 2.09420179491398e-05, "epoch": 1.2672413793103448, "percentage": 31.68, "elapsed_time": "2:50:38", "remaining_time": "6:07:59"}
148
+ {"current_steps": 296, "total_steps": 928, "loss": 0.7990611791610718, "lr": 2.0880138681141345e-05, "epoch": 1.2758620689655173, "percentage": 31.9, "elapsed_time": "2:51:44", "remaining_time": "6:06:40"}
149
+ {"current_steps": 298, "total_steps": 928, "loss": 1.1012811660766602, "lr": 2.081789674946961e-05, "epoch": 1.2844827586206897, "percentage": 32.11, "elapsed_time": "2:53:11", "remaining_time": "6:06:08"}
150
+ {"current_steps": 300, "total_steps": 928, "loss": 1.0139261484146118, "lr": 2.075529531996591e-05, "epoch": 1.293103448275862, "percentage": 32.33, "elapsed_time": "2:54:33", "remaining_time": "6:05:23"}
151
+ {"current_steps": 302, "total_steps": 928, "loss": 0.6810011863708496, "lr": 2.0692337576756866e-05, "epoch": 1.3017241379310345, "percentage": 32.54, "elapsed_time": "2:55:37", "remaining_time": "6:04:01"}
152
+ {"current_steps": 304, "total_steps": 928, "loss": 0.41086244583129883, "lr": 2.0629026722092453e-05, "epoch": 1.3103448275862069, "percentage": 32.76, "elapsed_time": "2:56:20", "remaining_time": "6:01:58"}
153
+ {"current_steps": 306, "total_steps": 928, "loss": 1.3326225280761719, "lr": 2.056536597618312e-05, "epoch": 1.3189655172413794, "percentage": 32.97, "elapsed_time": "2:57:43", "remaining_time": "6:01:14"}
154
+ {"current_steps": 308, "total_steps": 928, "loss": 0.6605536341667175, "lr": 2.0501358577036003e-05, "epoch": 1.3275862068965516, "percentage": 33.19, "elapsed_time": "2:58:46", "remaining_time": "5:59:52"}
155
+ {"current_steps": 310, "total_steps": 928, "loss": 1.0086909532546997, "lr": 2.043700778029024e-05, "epoch": 1.3362068965517242, "percentage": 33.41, "elapsed_time": "2:59:58", "remaining_time": "5:58:46"}
156
+ {"current_steps": 312, "total_steps": 928, "loss": 0.9816857576370239, "lr": 2.0372316859051357e-05, "epoch": 1.3448275862068966, "percentage": 33.62, "elapsed_time": "3:01:05", "remaining_time": "5:57:33"}
157
+ {"current_steps": 314, "total_steps": 928, "loss": 1.206939458847046, "lr": 2.0307289103724793e-05, "epoch": 1.353448275862069, "percentage": 33.84, "elapsed_time": "3:02:11", "remaining_time": "5:56:16"}
158
+ {"current_steps": 316, "total_steps": 928, "loss": 1.3034409284591675, "lr": 2.0241927821848544e-05, "epoch": 1.3620689655172413, "percentage": 34.05, "elapsed_time": "3:03:31", "remaining_time": "5:55:25"}
159
+ {"current_steps": 318, "total_steps": 928, "loss": 1.0241037607192993, "lr": 2.0176236337924933e-05, "epoch": 1.3706896551724137, "percentage": 34.27, "elapsed_time": "3:04:47", "remaining_time": "5:54:28"}
160
+ {"current_steps": 320, "total_steps": 928, "loss": 0.6907064318656921, "lr": 2.01102179932515e-05, "epoch": 1.3793103448275863, "percentage": 34.48, "elapsed_time": "3:05:36", "remaining_time": "5:52:40"}
161
+ {"current_steps": 322, "total_steps": 928, "loss": 1.4752711057662964, "lr": 2.0043876145751067e-05, "epoch": 1.3879310344827587, "percentage": 34.7, "elapsed_time": "3:06:59", "remaining_time": "5:51:54"}
162
+ {"current_steps": 324, "total_steps": 928, "loss": 0.4461423456668854, "lr": 1.997721416980093e-05, "epoch": 1.396551724137931, "percentage": 34.91, "elapsed_time": "3:08:17", "remaining_time": "5:51:00"}
163
+ {"current_steps": 326, "total_steps": 928, "loss": 0.42419686913490295, "lr": 1.991023545606124e-05, "epoch": 1.4051724137931034, "percentage": 35.13, "elapsed_time": "3:08:58", "remaining_time": "5:48:58"}
164
+ {"current_steps": 328, "total_steps": 928, "loss": 0.9030317664146423, "lr": 1.9842943411302527e-05, "epoch": 1.4137931034482758, "percentage": 35.34, "elapsed_time": "3:10:02", "remaining_time": "5:47:38"}
165
+ {"current_steps": 330, "total_steps": 928, "loss": 0.768023669719696, "lr": 1.9775341458232436e-05, "epoch": 1.4224137931034484, "percentage": 35.56, "elapsed_time": "3:11:07", "remaining_time": "5:46:20"}
166
+ {"current_steps": 332, "total_steps": 928, "loss": 0.6390933394432068, "lr": 1.9707433035321613e-05, "epoch": 1.4310344827586206, "percentage": 35.78, "elapsed_time": "3:12:13", "remaining_time": "5:45:04"}
167
+ {"current_steps": 334, "total_steps": 928, "loss": 0.6768423914909363, "lr": 1.9639221596628836e-05, "epoch": 1.4396551724137931, "percentage": 35.99, "elapsed_time": "3:13:17", "remaining_time": "5:43:45"}
168
+ {"current_steps": 336, "total_steps": 928, "loss": 0.710463285446167, "lr": 1.957071061162533e-05, "epoch": 1.4482758620689655, "percentage": 36.21, "elapsed_time": "3:14:20", "remaining_time": "5:42:25"}
169
+ {"current_steps": 338, "total_steps": 928, "loss": 1.0271871089935303, "lr": 1.9501903565018273e-05, "epoch": 1.456896551724138, "percentage": 36.42, "elapsed_time": "3:15:42", "remaining_time": "5:41:37"}
170
+ {"current_steps": 340, "total_steps": 928, "loss": 1.2887871265411377, "lr": 1.943280395657357e-05, "epoch": 1.4655172413793103, "percentage": 36.64, "elapsed_time": "3:17:05", "remaining_time": "5:40:51"}
171
+ {"current_steps": 342, "total_steps": 928, "loss": 1.1640923023223877, "lr": 1.9363415300937846e-05, "epoch": 1.4741379310344827, "percentage": 36.85, "elapsed_time": "3:18:31", "remaining_time": "5:40:09"}
172
+ {"current_steps": 344, "total_steps": 928, "loss": 0.9469618201255798, "lr": 1.9293741127459665e-05, "epoch": 1.4827586206896552, "percentage": 37.07, "elapsed_time": "3:19:50", "remaining_time": "5:39:15"}
173
+ {"current_steps": 346, "total_steps": 928, "loss": 0.45884159207344055, "lr": 1.922378498001003e-05, "epoch": 1.4913793103448276, "percentage": 37.28, "elapsed_time": "3:20:31", "remaining_time": "5:37:17"}
174
+ {"current_steps": 348, "total_steps": 928, "loss": 0.8928775787353516, "lr": 1.9153550416802107e-05, "epoch": 1.5, "percentage": 37.5, "elapsed_time": "3:21:52", "remaining_time": "5:36:27"}
175
+ {"current_steps": 350, "total_steps": 928, "loss": 1.0797038078308105, "lr": 1.9083041010210273e-05, "epoch": 1.5086206896551724, "percentage": 37.72, "elapsed_time": "3:23:14", "remaining_time": "5:35:38"}
176
+ {"current_steps": 352, "total_steps": 928, "loss": 1.0333884954452515, "lr": 1.9012260346588394e-05, "epoch": 1.5172413793103448, "percentage": 37.93, "elapsed_time": "3:24:27", "remaining_time": "5:34:34"}
177
+ {"current_steps": 354, "total_steps": 928, "loss": 1.1652461290359497, "lr": 1.89412120260874e-05, "epoch": 1.5258620689655173, "percentage": 38.15, "elapsed_time": "3:25:54", "remaining_time": "5:33:51"}
178
+ {"current_steps": 356, "total_steps": 928, "loss": 0.8967676758766174, "lr": 1.8869899662472197e-05, "epoch": 1.5344827586206895, "percentage": 38.36, "elapsed_time": "3:27:16", "remaining_time": "5:33:01"}
179
+ {"current_steps": 358, "total_steps": 928, "loss": 0.7729952335357666, "lr": 1.8798326882937834e-05, "epoch": 1.543103448275862, "percentage": 38.58, "elapsed_time": "3:28:10", "remaining_time": "5:31:27"}
180
+ {"current_steps": 360, "total_steps": 928, "loss": 1.3566042184829712, "lr": 1.872649732792502e-05, "epoch": 1.5517241379310345, "percentage": 38.79, "elapsed_time": "3:29:23", "remaining_time": "5:30:22"}
181
+ {"current_steps": 362, "total_steps": 928, "loss": 1.0945885181427002, "lr": 1.8654414650934968e-05, "epoch": 1.5603448275862069, "percentage": 39.01, "elapsed_time": "3:30:38", "remaining_time": "5:29:21"}
182
+ {"current_steps": 364, "total_steps": 928, "loss": 1.1358948945999146, "lr": 1.858208251834354e-05, "epoch": 1.5689655172413794, "percentage": 39.22, "elapsed_time": "3:32:04", "remaining_time": "5:28:35"}
183
+ {"current_steps": 366, "total_steps": 928, "loss": 0.7162285447120667, "lr": 1.8509504609214794e-05, "epoch": 1.5775862068965516, "percentage": 39.44, "elapsed_time": "3:33:08", "remaining_time": "5:27:16"}
184
+ {"current_steps": 368, "total_steps": 928, "loss": 1.080881953239441, "lr": 1.8436684615113824e-05, "epoch": 1.5862068965517242, "percentage": 39.66, "elapsed_time": "3:34:34", "remaining_time": "5:26:32"}
185
+ {"current_steps": 370, "total_steps": 928, "loss": 1.054266333580017, "lr": 1.8363626239919022e-05, "epoch": 1.5948275862068966, "percentage": 39.87, "elapsed_time": "3:35:50", "remaining_time": "5:25:30"}
186
+ {"current_steps": 372, "total_steps": 928, "loss": 1.0251119136810303, "lr": 1.8290333199633665e-05, "epoch": 1.603448275862069, "percentage": 40.09, "elapsed_time": "3:37:12", "remaining_time": "5:24:38"}
187
+ {"current_steps": 374, "total_steps": 928, "loss": 0.7149249315261841, "lr": 1.821680922219691e-05, "epoch": 1.6120689655172413, "percentage": 40.3, "elapsed_time": "3:38:18", "remaining_time": "5:23:21"}
188
+ {"current_steps": 376, "total_steps": 928, "loss": 1.0818464756011963, "lr": 1.8143058047294197e-05, "epoch": 1.6206896551724137, "percentage": 40.52, "elapsed_time": "3:39:44", "remaining_time": "5:22:35"}
189
+ {"current_steps": 378, "total_steps": 928, "loss": 1.1972448825836182, "lr": 1.8069083426166995e-05, "epoch": 1.6293103448275863, "percentage": 40.73, "elapsed_time": "3:41:09", "remaining_time": "5:21:47"}
190
+ {"current_steps": 380, "total_steps": 928, "loss": 0.7313613891601562, "lr": 1.799488912142205e-05, "epoch": 1.6379310344827587, "percentage": 40.95, "elapsed_time": "3:42:00", "remaining_time": "5:20:10"}
191
+ {"current_steps": 382, "total_steps": 928, "loss": 1.3920011520385742, "lr": 1.7920478906839967e-05, "epoch": 1.646551724137931, "percentage": 41.16, "elapsed_time": "3:43:23", "remaining_time": "5:19:17"}
192
+ {"current_steps": 384, "total_steps": 928, "loss": 1.0703322887420654, "lr": 1.7845856567183278e-05, "epoch": 1.6551724137931034, "percentage": 41.38, "elapsed_time": "3:44:35", "remaining_time": "5:18:10"}
193
+ {"current_steps": 386, "total_steps": 928, "loss": 0.9152452349662781, "lr": 1.777102589800394e-05, "epoch": 1.6637931034482758, "percentage": 41.59, "elapsed_time": "3:45:48", "remaining_time": "5:17:03"}
194
+ {"current_steps": 388, "total_steps": 928, "loss": 1.360831618309021, "lr": 1.7695990705450276e-05, "epoch": 1.6724137931034484, "percentage": 41.81, "elapsed_time": "3:47:00", "remaining_time": "5:15:55"}
195
+ {"current_steps": 390, "total_steps": 928, "loss": 0.905778706073761, "lr": 1.7620754806073368e-05, "epoch": 1.6810344827586206, "percentage": 42.03, "elapsed_time": "3:48:12", "remaining_time": "5:14:48"}
196
+ {"current_steps": 392, "total_steps": 928, "loss": 0.43694019317626953, "lr": 1.754532202663297e-05, "epoch": 1.6896551724137931, "percentage": 42.24, "elapsed_time": "3:48:56", "remaining_time": "5:13:02"}
197
+ {"current_steps": 394, "total_steps": 928, "loss": 1.0095391273498535, "lr": 1.74696962039028e-05, "epoch": 1.6982758620689655, "percentage": 42.46, "elapsed_time": "3:50:18", "remaining_time": "5:12:09"}
198
+ {"current_steps": 396, "total_steps": 928, "loss": 1.0637139081954956, "lr": 1.7393881184475476e-05, "epoch": 1.706896551724138, "percentage": 42.67, "elapsed_time": "3:51:44", "remaining_time": "5:11:19"}
199
+ {"current_steps": 398, "total_steps": 928, "loss": 1.042551040649414, "lr": 1.7317880824566792e-05, "epoch": 1.7155172413793105, "percentage": 42.89, "elapsed_time": "3:52:55", "remaining_time": "5:10:10"}
200
+ {"current_steps": 400, "total_steps": 928, "loss": 1.02712082862854, "lr": 1.7241698989819604e-05, "epoch": 1.7241379310344827, "percentage": 43.1, "elapsed_time": "3:54:04", "remaining_time": "5:08:58"}
201
+ {"current_steps": 402, "total_steps": 928, "loss": 0.7308106422424316, "lr": 1.7165339555107223e-05, "epoch": 1.7327586206896552, "percentage": 43.32, "elapsed_time": "3:55:06", "remaining_time": "5:07:38"}
202
+ {"current_steps": 404, "total_steps": 928, "loss": 0.6191306710243225, "lr": 1.7088806404336298e-05, "epoch": 1.7413793103448276, "percentage": 43.53, "elapsed_time": "3:56:09", "remaining_time": "5:06:18"}
203
+ {"current_steps": 406, "total_steps": 928, "loss": 0.6533270478248596, "lr": 1.7012103430249292e-05, "epoch": 1.75, "percentage": 43.75, "elapsed_time": "3:57:14", "remaining_time": "5:05:01"}
204
+ {"current_steps": 408, "total_steps": 928, "loss": 0.7554537653923035, "lr": 1.6935234534226475e-05, "epoch": 1.7586206896551724, "percentage": 43.97, "elapsed_time": "3:58:08", "remaining_time": "5:03:30"}
205
+ {"current_steps": 410, "total_steps": 928, "loss": 0.8455532789230347, "lr": 1.6858203626087468e-05, "epoch": 1.7672413793103448, "percentage": 44.18, "elapsed_time": "3:59:13", "remaining_time": "5:02:14"}
206
+ {"current_steps": 412, "total_steps": 928, "loss": 0.7798787951469421, "lr": 1.6781014623892406e-05, "epoch": 1.7758620689655173, "percentage": 44.4, "elapsed_time": "4:00:36", "remaining_time": "5:01:20"}
207
+ {"current_steps": 414, "total_steps": 928, "loss": 0.6053640842437744, "lr": 1.6703671453742642e-05, "epoch": 1.7844827586206895, "percentage": 44.61, "elapsed_time": "4:01:40", "remaining_time": "5:00:02"}
208
+ {"current_steps": 416, "total_steps": 928, "loss": 0.8004411458969116, "lr": 1.6626178049581033e-05, "epoch": 1.793103448275862, "percentage": 44.83, "elapsed_time": "4:02:43", "remaining_time": "4:58:44"}
209
+ {"current_steps": 418, "total_steps": 928, "loss": 0.9836465716362, "lr": 1.654853835299187e-05, "epoch": 1.8017241379310345, "percentage": 45.04, "elapsed_time": "4:04:05", "remaining_time": "4:57:49"}
210
+ {"current_steps": 420, "total_steps": 928, "loss": 1.111708402633667, "lr": 1.6470756313000386e-05, "epoch": 1.8103448275862069, "percentage": 45.26, "elapsed_time": "4:05:27", "remaining_time": "4:56:53"}
211
+ {"current_steps": 422, "total_steps": 928, "loss": 1.1723716259002686, "lr": 1.6392835885871897e-05, "epoch": 1.8189655172413794, "percentage": 45.47, "elapsed_time": "4:06:51", "remaining_time": "4:55:59"}
212
+ {"current_steps": 424, "total_steps": 928, "loss": 1.006577491760254, "lr": 1.631478103491056e-05, "epoch": 1.8275862068965516, "percentage": 45.69, "elapsed_time": "4:07:55", "remaining_time": "4:54:42"}
213
+ {"current_steps": 426, "total_steps": 928, "loss": 0.6617158055305481, "lr": 1.623659573025781e-05, "epoch": 1.8362068965517242, "percentage": 45.91, "elapsed_time": "4:08:41", "remaining_time": "4:53:03"}
214
+ {"current_steps": 428, "total_steps": 928, "loss": 1.0630967617034912, "lr": 1.61582839486904e-05, "epoch": 1.8448275862068966, "percentage": 46.12, "elapsed_time": "4:10:06", "remaining_time": "4:52:10"}
215
+ {"current_steps": 430, "total_steps": 928, "loss": 1.0498086214065552, "lr": 1.6079849673418144e-05, "epoch": 1.853448275862069, "percentage": 46.34, "elapsed_time": "4:11:31", "remaining_time": "4:51:17"}
216
+ {"current_steps": 432, "total_steps": 928, "loss": 1.0054386854171753, "lr": 1.600129689388131e-05, "epoch": 1.8620689655172413, "percentage": 46.55, "elapsed_time": "4:12:41", "remaining_time": "4:50:07"}
217
+ {"current_steps": 434, "total_steps": 928, "loss": 1.107947826385498, "lr": 1.5922629605547708e-05, "epoch": 1.8706896551724137, "percentage": 46.77, "elapsed_time": "4:13:56", "remaining_time": "4:49:03"}
218
+ {"current_steps": 436, "total_steps": 928, "loss": 0.6936919093132019, "lr": 1.5843851809709462e-05, "epoch": 1.8793103448275863, "percentage": 46.98, "elapsed_time": "4:14:42", "remaining_time": "4:47:25"}
219
+ {"current_steps": 438, "total_steps": 928, "loss": 0.9637195467948914, "lr": 1.5764967513279498e-05, "epoch": 1.8879310344827587, "percentage": 47.2, "elapsed_time": "4:15:58", "remaining_time": "4:46:22"}
220
+ {"current_steps": 440, "total_steps": 928, "loss": 0.9367073774337769, "lr": 1.568598072858772e-05, "epoch": 1.896551724137931, "percentage": 47.41, "elapsed_time": "4:17:10", "remaining_time": "4:45:13"}
221
+ {"current_steps": 442, "total_steps": 928, "loss": 1.1365821361541748, "lr": 1.5606895473176952e-05, "epoch": 1.9051724137931034, "percentage": 47.63, "elapsed_time": "4:18:28", "remaining_time": "4:44:11"}
222
+ {"current_steps": 444, "total_steps": 928, "loss": 0.2478710114955902, "lr": 1.5527715769598582e-05, "epoch": 1.9137931034482758, "percentage": 47.84, "elapsed_time": "4:19:12", "remaining_time": "4:42:33"}
223
+ {"current_steps": 446, "total_steps": 928, "loss": 0.636488676071167, "lr": 1.544844564520796e-05, "epoch": 1.9224137931034484, "percentage": 48.06, "elapsed_time": "4:20:20", "remaining_time": "4:41:21"}
224
+ {"current_steps": 448, "total_steps": 928, "loss": 1.0994303226470947, "lr": 1.5369089131959546e-05, "epoch": 1.9310344827586206, "percentage": 48.28, "elapsed_time": "4:21:36", "remaining_time": "4:40:17"}
225
+ {"current_steps": 450, "total_steps": 928, "loss": 1.0804164409637451, "lr": 1.5289650266201842e-05, "epoch": 1.9396551724137931, "percentage": 48.49, "elapsed_time": "4:23:00", "remaining_time": "4:39:22"}
226
+ {"current_steps": 452, "total_steps": 928, "loss": 0.9485524296760559, "lr": 1.5210133088472096e-05, "epoch": 1.9482758620689655, "percentage": 48.71, "elapsed_time": "4:24:07", "remaining_time": "4:38:08"}
227
+ {"current_steps": 454, "total_steps": 928, "loss": 1.0982067584991455, "lr": 1.5130541643290755e-05, "epoch": 1.956896551724138, "percentage": 48.92, "elapsed_time": "4:25:22", "remaining_time": "4:37:03"}
228
+ {"current_steps": 456, "total_steps": 928, "loss": 1.069095492362976, "lr": 1.5050879978955784e-05, "epoch": 1.9655172413793105, "percentage": 49.14, "elapsed_time": "4:26:47", "remaining_time": "4:36:09"}
229
+ {"current_steps": 458, "total_steps": 928, "loss": 0.29495418071746826, "lr": 1.4971152147336735e-05, "epoch": 1.9741379310344827, "percentage": 49.35, "elapsed_time": "4:27:32", "remaining_time": "4:34:33"}
230
+ {"current_steps": 460, "total_steps": 928, "loss": 1.0610284805297852, "lr": 1.4891362203668646e-05, "epoch": 1.9827586206896552, "percentage": 49.57, "elapsed_time": "4:28:55", "remaining_time": "4:33:36"}
231
+ {"current_steps": 462, "total_steps": 928, "loss": 0.6210886240005493, "lr": 1.4811514206345812e-05, "epoch": 1.9913793103448276, "percentage": 49.78, "elapsed_time": "4:29:57", "remaining_time": "4:32:17"}
232
+ {"current_steps": 464, "total_steps": 928, "loss": 0.796215295791626, "lr": 1.4731612216715315e-05, "epoch": 2.0, "percentage": 50.0, "elapsed_time": "4:31:14", "remaining_time": "4:31:14"}
233
+ {"current_steps": 466, "total_steps": 928, "loss": 0.4970940947532654, "lr": 1.4651660298870496e-05, "epoch": 2.0086206896551726, "percentage": 50.22, "elapsed_time": "4:32:18", "remaining_time": "4:29:58"}
234
+ {"current_steps": 468, "total_steps": 928, "loss": 0.7648203372955322, "lr": 1.4571662519444208e-05, "epoch": 2.0172413793103448, "percentage": 50.43, "elapsed_time": "4:33:31", "remaining_time": "4:28:51"}
235
+ {"current_steps": 470, "total_steps": 928, "loss": 0.7012611627578735, "lr": 1.4491622947401976e-05, "epoch": 2.0258620689655173, "percentage": 50.65, "elapsed_time": "4:34:37", "remaining_time": "4:27:37"}
236
+ {"current_steps": 472, "total_steps": 928, "loss": 0.5726131796836853, "lr": 1.4411545653835056e-05, "epoch": 2.0344827586206895, "percentage": 50.86, "elapsed_time": "4:35:31", "remaining_time": "4:26:11"}
237
+ {"current_steps": 474, "total_steps": 928, "loss": 0.5506486296653748, "lr": 1.433143471175335e-05, "epoch": 2.043103448275862, "percentage": 51.08, "elapsed_time": "4:36:37", "remaining_time": "4:24:56"}
238
+ {"current_steps": 476, "total_steps": 928, "loss": 0.5927596688270569, "lr": 1.4251294195878231e-05, "epoch": 2.0517241379310347, "percentage": 51.29, "elapsed_time": "4:37:40", "remaining_time": "4:23:40"}
239
+ {"current_steps": 478, "total_steps": 928, "loss": 0.4717426598072052, "lr": 1.4171128182435312e-05, "epoch": 2.060344827586207, "percentage": 51.51, "elapsed_time": "4:38:45", "remaining_time": "4:22:25"}
240
+ {"current_steps": 480, "total_steps": 928, "loss": 0.8343894481658936, "lr": 1.409094074894709e-05, "epoch": 2.0689655172413794, "percentage": 51.72, "elapsed_time": "4:40:10", "remaining_time": "4:21:29"}
241
+ {"current_steps": 482, "total_steps": 928, "loss": 0.5780265927314758, "lr": 1.4010735974025565e-05, "epoch": 2.0775862068965516, "percentage": 51.94, "elapsed_time": "4:41:14", "remaining_time": "4:20:13"}
242
+ {"current_steps": 484, "total_steps": 928, "loss": 0.628856897354126, "lr": 1.3930517937164783e-05, "epoch": 2.086206896551724, "percentage": 52.16, "elapsed_time": "4:42:35", "remaining_time": "4:19:13"}
243
+ {"current_steps": 486, "total_steps": 928, "loss": 0.6124461889266968, "lr": 1.3850290718533334e-05, "epoch": 2.0948275862068964, "percentage": 52.37, "elapsed_time": "4:43:45", "remaining_time": "4:18:03"}
244
+ {"current_steps": 488, "total_steps": 928, "loss": 0.7476667761802673, "lr": 1.3770058398766825e-05, "epoch": 2.103448275862069, "percentage": 52.59, "elapsed_time": "4:45:06", "remaining_time": "4:17:03"}
245
+ {"current_steps": 490, "total_steps": 928, "loss": 0.8483129739761353, "lr": 1.3689825058760336e-05, "epoch": 2.1120689655172415, "percentage": 52.8, "elapsed_time": "4:46:31", "remaining_time": "4:16:07"}
246
+ {"current_steps": 492, "total_steps": 928, "loss": 0.8416441082954407, "lr": 1.3609594779460827e-05, "epoch": 2.1206896551724137, "percentage": 53.02, "elapsed_time": "4:47:48", "remaining_time": "4:15:03"}
247
+ {"current_steps": 494, "total_steps": 928, "loss": 0.8671878576278687, "lr": 1.3529371641659582e-05, "epoch": 2.1293103448275863, "percentage": 53.23, "elapsed_time": "4:49:13", "remaining_time": "4:14:05"}
248
+ {"current_steps": 496, "total_steps": 928, "loss": 0.919392466545105, "lr": 1.344915972578465e-05, "epoch": 2.1379310344827585, "percentage": 53.45, "elapsed_time": "4:50:22", "remaining_time": "4:12:54"}
249
+ {"current_steps": 498, "total_steps": 928, "loss": 0.3866020143032074, "lr": 1.3368963111693284e-05, "epoch": 2.146551724137931, "percentage": 53.66, "elapsed_time": "4:51:11", "remaining_time": "4:11:26"}
250
+ {"current_steps": 500, "total_steps": 928, "loss": 0.4366337060928345, "lr": 1.3288785878464447e-05, "epoch": 2.1551724137931036, "percentage": 53.88, "elapsed_time": "4:52:13", "remaining_time": "4:10:08"}
251
+ {"current_steps": 502, "total_steps": 928, "loss": 0.4676997661590576, "lr": 1.320863210419131e-05, "epoch": 2.163793103448276, "percentage": 54.09, "elapsed_time": "4:53:18", "remaining_time": "4:08:53"}
252
+ {"current_steps": 504, "total_steps": 928, "loss": 0.5133532285690308, "lr": 1.3128505865773855e-05, "epoch": 2.1724137931034484, "percentage": 54.31, "elapsed_time": "4:54:10", "remaining_time": "4:07:29"}
253
+ {"current_steps": 506, "total_steps": 928, "loss": 0.8324005603790283, "lr": 1.3048411238711478e-05, "epoch": 2.1810344827586206, "percentage": 54.53, "elapsed_time": "4:55:31", "remaining_time": "4:06:27"}
254
+ {"current_steps": 508, "total_steps": 928, "loss": 0.49498435854911804, "lr": 1.296835229689573e-05, "epoch": 2.189655172413793, "percentage": 54.74, "elapsed_time": "4:56:35", "remaining_time": "4:05:12"}
255
+ {"current_steps": 510, "total_steps": 928, "loss": 0.863714337348938, "lr": 1.2888333112403065e-05, "epoch": 2.1982758620689653, "percentage": 54.96, "elapsed_time": "4:57:49", "remaining_time": "4:04:05"}
256
+ {"current_steps": 512, "total_steps": 928, "loss": 0.4537322223186493, "lr": 1.2808357755287764e-05, "epoch": 2.206896551724138, "percentage": 55.17, "elapsed_time": "4:58:39", "remaining_time": "4:02:39"}
257
+ {"current_steps": 514, "total_steps": 928, "loss": 0.5666446685791016, "lr": 1.2728430293374871e-05, "epoch": 2.2155172413793105, "percentage": 55.39, "elapsed_time": "4:59:43", "remaining_time": "4:01:24"}
258
+ {"current_steps": 516, "total_steps": 928, "loss": 0.44022682309150696, "lr": 1.2648554792053342e-05, "epoch": 2.2241379310344827, "percentage": 55.6, "elapsed_time": "5:00:46", "remaining_time": "4:00:09"}
259
+ {"current_steps": 518, "total_steps": 928, "loss": 0.0981849655508995, "lr": 1.25687353140692e-05, "epoch": 2.2327586206896552, "percentage": 55.82, "elapsed_time": "5:01:30", "remaining_time": "3:58:38"}
260
+ {"current_steps": 520, "total_steps": 928, "loss": 0.15166987478733063, "lr": 1.2488975919318963e-05, "epoch": 2.2413793103448274, "percentage": 56.03, "elapsed_time": "5:02:29", "remaining_time": "3:57:20"}
261
+ {"current_steps": 522, "total_steps": 928, "loss": 0.8936961889266968, "lr": 1.2409280664643078e-05, "epoch": 2.25, "percentage": 56.25, "elapsed_time": "5:03:54", "remaining_time": "3:56:22"}
262
+ {"current_steps": 524, "total_steps": 928, "loss": 0.6037598848342896, "lr": 1.2329653603619629e-05, "epoch": 2.2586206896551726, "percentage": 56.47, "elapsed_time": "5:04:49", "remaining_time": "3:55:01"}
263
+ {"current_steps": 526, "total_steps": 928, "loss": 0.5711255669593811, "lr": 1.2250098786358108e-05, "epoch": 2.2672413793103448, "percentage": 56.68, "elapsed_time": "5:05:51", "remaining_time": "3:53:45"}
264
+ {"current_steps": 528, "total_steps": 928, "loss": 0.5425680875778198, "lr": 1.2170620259293458e-05, "epoch": 2.2758620689655173, "percentage": 56.9, "elapsed_time": "5:06:54", "remaining_time": "3:52:30"}
265
+ {"current_steps": 530, "total_steps": 928, "loss": 0.5258100628852844, "lr": 1.2091222064980228e-05, "epoch": 2.2844827586206895, "percentage": 57.11, "elapsed_time": "5:07:58", "remaining_time": "3:51:16"}
266
+ {"current_steps": 532, "total_steps": 928, "loss": 0.5525709390640259, "lr": 1.2011908241886966e-05, "epoch": 2.293103448275862, "percentage": 57.33, "elapsed_time": "5:09:02", "remaining_time": "3:50:02"}
267
+ {"current_steps": 534, "total_steps": 928, "loss": 0.6124492883682251, "lr": 1.1932682824190799e-05, "epoch": 2.3017241379310347, "percentage": 57.54, "elapsed_time": "5:10:08", "remaining_time": "3:48:49"}
268
+ {"current_steps": 536, "total_steps": 928, "loss": 0.5387366414070129, "lr": 1.1853549841572268e-05, "epoch": 2.310344827586207, "percentage": 57.76, "elapsed_time": "5:11:19", "remaining_time": "3:47:41"}
269
+ {"current_steps": 538, "total_steps": 928, "loss": 0.32567426562309265, "lr": 1.177451331901031e-05, "epoch": 2.3189655172413794, "percentage": 57.97, "elapsed_time": "5:12:10", "remaining_time": "3:46:17"}
270
+ {"current_steps": 540, "total_steps": 928, "loss": 0.46518173813819885, "lr": 1.169557727657761e-05, "epoch": 2.3275862068965516, "percentage": 58.19, "elapsed_time": "5:13:13", "remaining_time": "3:45:03"}
271
+ {"current_steps": 542, "total_steps": 928, "loss": 0.774487316608429, "lr": 1.1616745729236045e-05, "epoch": 2.336206896551724, "percentage": 58.41, "elapsed_time": "5:14:32", "remaining_time": "3:44:00"}
272
+ {"current_steps": 544, "total_steps": 928, "loss": 0.8165640830993652, "lr": 1.1538022686632536e-05, "epoch": 2.344827586206897, "percentage": 58.62, "elapsed_time": "5:15:51", "remaining_time": "3:42:57"}
273
+ {"current_steps": 546, "total_steps": 928, "loss": 0.19026756286621094, "lr": 1.145941215289505e-05, "epoch": 2.353448275862069, "percentage": 58.84, "elapsed_time": "5:16:33", "remaining_time": "3:41:28"}
274
+ {"current_steps": 548, "total_steps": 928, "loss": 0.8604028820991516, "lr": 1.1380918126428994e-05, "epoch": 2.3620689655172415, "percentage": 59.05, "elapsed_time": "5:17:48", "remaining_time": "3:40:22"}
275
+ {"current_steps": 550, "total_steps": 928, "loss": 0.4935520589351654, "lr": 1.1302544599713775e-05, "epoch": 2.3706896551724137, "percentage": 59.27, "elapsed_time": "5:18:48", "remaining_time": "3:39:06"}
276
+ {"current_steps": 552, "total_steps": 928, "loss": 0.7264618277549744, "lr": 1.122429555909979e-05, "epoch": 2.3793103448275863, "percentage": 59.48, "elapsed_time": "5:20:08", "remaining_time": "3:38:04"}
277
+ {"current_steps": 554, "total_steps": 928, "loss": 0.6168904900550842, "lr": 1.114617498460562e-05, "epoch": 2.3879310344827585, "percentage": 59.7, "elapsed_time": "5:21:20", "remaining_time": "3:36:56"}
278
+ {"current_steps": 556, "total_steps": 928, "loss": 0.6429441571235657, "lr": 1.1068186849715635e-05, "epoch": 2.396551724137931, "percentage": 59.91, "elapsed_time": "5:22:42", "remaining_time": "3:35:54"}
279
+ {"current_steps": 558, "total_steps": 928, "loss": 0.4293447434902191, "lr": 1.099033512117783e-05, "epoch": 2.405172413793103, "percentage": 60.13, "elapsed_time": "5:23:38", "remaining_time": "3:34:36"}
280
+ {"current_steps": 560, "total_steps": 928, "loss": 0.5241134762763977, "lr": 1.091262375880214e-05, "epoch": 2.413793103448276, "percentage": 60.34, "elapsed_time": "5:24:42", "remaining_time": "3:33:22"}
281
+ {"current_steps": 562, "total_steps": 928, "loss": 0.6372413039207458, "lr": 1.0835056715258956e-05, "epoch": 2.4224137931034484, "percentage": 60.56, "elapsed_time": "5:26:05", "remaining_time": "3:32:21"}
282
+ {"current_steps": 564, "total_steps": 928, "loss": 0.7959206104278564, "lr": 1.0757637935878126e-05, "epoch": 2.4310344827586206, "percentage": 60.78, "elapsed_time": "5:27:07", "remaining_time": "3:31:07"}
283
+ {"current_steps": 566, "total_steps": 928, "loss": 0.6432863473892212, "lr": 1.0680371358448266e-05, "epoch": 2.439655172413793, "percentage": 60.99, "elapsed_time": "5:28:29", "remaining_time": "3:30:05"}
284
+ {"current_steps": 568, "total_steps": 928, "loss": 0.31055551767349243, "lr": 1.0603260913016466e-05, "epoch": 2.4482758620689653, "percentage": 61.21, "elapsed_time": "5:29:15", "remaining_time": "3:28:40"}
285
+ {"current_steps": 570, "total_steps": 928, "loss": 0.9433165788650513, "lr": 1.05263105216884e-05, "epoch": 2.456896551724138, "percentage": 61.42, "elapsed_time": "5:30:37", "remaining_time": "3:27:39"}
286
+ {"current_steps": 572, "total_steps": 928, "loss": 0.5256112217903137, "lr": 1.044952409842884e-05, "epoch": 2.4655172413793105, "percentage": 61.64, "elapsed_time": "5:31:43", "remaining_time": "3:26:27"}
287
+ {"current_steps": 574, "total_steps": 928, "loss": 0.5775177478790283, "lr": 1.0372905548862554e-05, "epoch": 2.4741379310344827, "percentage": 61.85, "elapsed_time": "5:32:54", "remaining_time": "3:25:18"}
288
+ {"current_steps": 576, "total_steps": 928, "loss": 0.8348941802978516, "lr": 1.0296458770075702e-05, "epoch": 2.4827586206896552, "percentage": 62.07, "elapsed_time": "5:34:19", "remaining_time": "3:24:18"}
289
+ {"current_steps": 578, "total_steps": 928, "loss": 0.48024982213974, "lr": 1.0220187650417542e-05, "epoch": 2.4913793103448274, "percentage": 62.28, "elapsed_time": "5:35:23", "remaining_time": "3:23:05"}
290
+ {"current_steps": 580, "total_steps": 928, "loss": 0.8509948253631592, "lr": 1.0144096069302734e-05, "epoch": 2.5, "percentage": 62.5, "elapsed_time": "5:36:49", "remaining_time": "3:22:05"}
291
+ {"current_steps": 582, "total_steps": 928, "loss": 0.8317974209785461, "lr": 1.0068187897013957e-05, "epoch": 2.5086206896551726, "percentage": 62.72, "elapsed_time": "5:38:14", "remaining_time": "3:21:05"}
292
+ {"current_steps": 584, "total_steps": 928, "loss": 0.6833478808403015, "lr": 9.992466994505098e-06, "epoch": 2.5172413793103448, "percentage": 62.93, "elapsed_time": "5:39:14", "remaining_time": "3:19:49"}
293
+ {"current_steps": 586, "total_steps": 928, "loss": 0.6194900870323181, "lr": 9.916937213204835e-06, "epoch": 2.5258620689655173, "percentage": 63.15, "elapsed_time": "5:40:36", "remaining_time": "3:18:46"}
294
+ {"current_steps": 588, "total_steps": 928, "loss": 0.8086000680923462, "lr": 9.841602394820772e-06, "epoch": 2.5344827586206895, "percentage": 63.36, "elapsed_time": "5:41:59", "remaining_time": "3:17:45"}
295
+ {"current_steps": 590, "total_steps": 928, "loss": 0.8427749872207642, "lr": 9.766466371144e-06, "epoch": 2.543103448275862, "percentage": 63.58, "elapsed_time": "5:43:17", "remaining_time": "3:16:39"}
296
+ {"current_steps": 592, "total_steps": 928, "loss": 0.7539307475090027, "lr": 9.691532963854247e-06, "epoch": 2.5517241379310347, "percentage": 63.79, "elapsed_time": "5:44:29", "remaining_time": "3:15:31"}
297
+ {"current_steps": 594, "total_steps": 928, "loss": 0.4550768733024597, "lr": 9.616805984325434e-06, "epoch": 2.560344827586207, "percentage": 64.01, "elapsed_time": "5:45:34", "remaining_time": "3:14:18"}
298
+ {"current_steps": 596, "total_steps": 928, "loss": 0.4484248459339142, "lr": 9.542289233431875e-06, "epoch": 2.5689655172413794, "percentage": 64.22, "elapsed_time": "5:46:30", "remaining_time": "3:13:01"}
299
+ {"current_steps": 598, "total_steps": 928, "loss": 0.612811267375946, "lr": 9.467986501354905e-06, "epoch": 2.5775862068965516, "percentage": 64.44, "elapsed_time": "5:47:54", "remaining_time": "3:11:59"}
300
+ {"current_steps": 600, "total_steps": 928, "loss": 0.7580547332763672, "lr": 9.393901567390133e-06, "epoch": 2.586206896551724, "percentage": 64.66, "elapsed_time": "5:49:14", "remaining_time": "3:10:55"}
301
+ {"current_steps": 602, "total_steps": 928, "loss": 0.5622333288192749, "lr": 9.320038199755181e-06, "epoch": 2.594827586206897, "percentage": 64.87, "elapsed_time": "5:50:20", "remaining_time": "3:09:43"}
302
+ {"current_steps": 604, "total_steps": 928, "loss": 0.681129515171051, "lr": 9.246400155398052e-06, "epoch": 2.603448275862069, "percentage": 65.09, "elapsed_time": "5:51:42", "remaining_time": "3:08:39"}
303
+ {"current_steps": 606, "total_steps": 928, "loss": 0.5278960466384888, "lr": 9.17299117980601e-06, "epoch": 2.612068965517241, "percentage": 65.3, "elapsed_time": "5:52:47", "remaining_time": "3:07:27"}
304
+ {"current_steps": 608, "total_steps": 928, "loss": 0.8580256104469299, "lr": 9.09981500681509e-06, "epoch": 2.6206896551724137, "percentage": 65.52, "elapsed_time": "5:54:11", "remaining_time": "3:06:25"}
305
+ {"current_steps": 610, "total_steps": 928, "loss": 0.5199075937271118, "lr": 9.02687535842018e-06, "epoch": 2.6293103448275863, "percentage": 65.73, "elapsed_time": "5:55:12", "remaining_time": "3:05:10"}
306
+ {"current_steps": 612, "total_steps": 928, "loss": 0.8993173837661743, "lr": 8.9541759445857e-06, "epoch": 2.637931034482759, "percentage": 65.95, "elapsed_time": "5:56:25", "remaining_time": "3:04:02"}
307
+ {"current_steps": 614, "total_steps": 928, "loss": 0.7683141827583313, "lr": 8.881720463056895e-06, "epoch": 2.646551724137931, "percentage": 66.16, "elapsed_time": "5:57:48", "remaining_time": "3:02:58"}
308
+ {"current_steps": 616, "total_steps": 928, "loss": 0.5064551830291748, "lr": 8.809512599171781e-06, "epoch": 2.655172413793103, "percentage": 66.38, "elapsed_time": "5:58:53", "remaining_time": "3:01:46"}
309
+ {"current_steps": 618, "total_steps": 928, "loss": 0.8363734483718872, "lr": 8.737556025673664e-06, "epoch": 2.663793103448276, "percentage": 66.59, "elapsed_time": "6:00:18", "remaining_time": "3:00:44"}
310
+ {"current_steps": 620, "total_steps": 928, "loss": 0.4550926387310028, "lr": 8.665854402524356e-06, "epoch": 2.6724137931034484, "percentage": 66.81, "elapsed_time": "6:01:22", "remaining_time": "2:59:31"}
311
+ {"current_steps": 622, "total_steps": 928, "loss": 0.7523635029792786, "lr": 8.594411376718009e-06, "epoch": 2.6810344827586206, "percentage": 67.03, "elapsed_time": "6:02:46", "remaining_time": "2:58:28"}
312
+ {"current_steps": 624, "total_steps": 928, "loss": 0.4919755458831787, "lr": 8.523230582095595e-06, "epoch": 2.689655172413793, "percentage": 67.24, "elapsed_time": "6:03:48", "remaining_time": "2:57:14"}
313
+ {"current_steps": 626, "total_steps": 928, "loss": 0.7990663051605225, "lr": 8.452315639160121e-06, "epoch": 2.6982758620689653, "percentage": 67.46, "elapsed_time": "6:05:01", "remaining_time": "2:56:05"}
314
+ {"current_steps": 628, "total_steps": 928, "loss": 0.5231303572654724, "lr": 8.381670154892429e-06, "epoch": 2.706896551724138, "percentage": 67.67, "elapsed_time": "6:06:07", "remaining_time": "2:54:53"}
315
+ {"current_steps": 630, "total_steps": 928, "loss": 0.9687949419021606, "lr": 8.311297722567769e-06, "epoch": 2.7155172413793105, "percentage": 67.89, "elapsed_time": "6:07:33", "remaining_time": "2:53:51"}
316
+ {"current_steps": 632, "total_steps": 928, "loss": 0.42503124475479126, "lr": 8.241201921573018e-06, "epoch": 2.7241379310344827, "percentage": 68.1, "elapsed_time": "6:08:34", "remaining_time": "2:52:37"}
317
+ {"current_steps": 634, "total_steps": 928, "loss": 0.7973089814186096, "lr": 8.171386317224612e-06, "epoch": 2.7327586206896552, "percentage": 68.32, "elapsed_time": "6:09:47", "remaining_time": "2:51:28"}
318
+ {"current_steps": 636, "total_steps": 928, "loss": 0.5349655747413635, "lr": 8.101854460587225e-06, "epoch": 2.7413793103448274, "percentage": 68.53, "elapsed_time": "6:10:49", "remaining_time": "2:50:15"}
319
+ {"current_steps": 638, "total_steps": 928, "loss": 0.8247781991958618, "lr": 8.032609888293111e-06, "epoch": 2.75, "percentage": 68.75, "elapsed_time": "6:12:04", "remaining_time": "2:49:07"}
320
+ {"current_steps": 640, "total_steps": 928, "loss": 0.4262469708919525, "lr": 7.963656122362269e-06, "epoch": 2.7586206896551726, "percentage": 68.97, "elapsed_time": "6:13:11", "remaining_time": "2:47:56"}
321
+ {"current_steps": 642, "total_steps": 928, "loss": 0.6027237176895142, "lr": 7.894996670023252e-06, "epoch": 2.7672413793103448, "percentage": 69.18, "elapsed_time": "6:14:13", "remaining_time": "2:46:42"}
322
+ {"current_steps": 644, "total_steps": 928, "loss": 0.8435148000717163, "lr": 7.82663502353482e-06, "epoch": 2.7758620689655173, "percentage": 69.4, "elapsed_time": "6:15:39", "remaining_time": "2:45:39"}
323
+ {"current_steps": 646, "total_steps": 928, "loss": 0.7197084426879883, "lr": 7.758574660008268e-06, "epoch": 2.7844827586206895, "percentage": 69.61, "elapsed_time": "6:16:54", "remaining_time": "2:44:31"}
324
+ {"current_steps": 648, "total_steps": 928, "loss": 0.5047876834869385, "lr": 7.690819041230616e-06, "epoch": 2.793103448275862, "percentage": 69.83, "elapsed_time": "6:17:58", "remaining_time": "2:43:19"}
325
+ {"current_steps": 650, "total_steps": 928, "loss": 0.6992777585983276, "lr": 7.623371613488475e-06, "epoch": 2.8017241379310347, "percentage": 70.04, "elapsed_time": "6:19:10", "remaining_time": "2:42:10"}
326
+ {"current_steps": 652, "total_steps": 928, "loss": 0.7797609567642212, "lr": 7.55623580739282e-06, "epoch": 2.810344827586207, "percentage": 70.26, "elapsed_time": "6:20:31", "remaining_time": "2:41:04"}
327
+ {"current_steps": 654, "total_steps": 928, "loss": 0.7613651156425476, "lr": 7.489415037704437e-06, "epoch": 2.8189655172413794, "percentage": 70.47, "elapsed_time": "6:21:56", "remaining_time": "2:40:01"}
328
+ {"current_steps": 656, "total_steps": 928, "loss": 0.5962029695510864, "lr": 7.422912703160281e-06, "epoch": 2.8275862068965516, "percentage": 70.69, "elapsed_time": "6:23:02", "remaining_time": "2:38:49"}
329
+ {"current_steps": 658, "total_steps": 928, "loss": 0.8467994928359985, "lr": 7.3567321863005754e-06, "epoch": 2.836206896551724, "percentage": 70.91, "elapsed_time": "6:24:27", "remaining_time": "2:37:45"}
330
+ {"current_steps": 660, "total_steps": 928, "loss": 0.2079772651195526, "lr": 7.290876853296791e-06, "epoch": 2.844827586206897, "percentage": 71.12, "elapsed_time": "6:25:19", "remaining_time": "2:36:27"}
331
+ {"current_steps": 662, "total_steps": 928, "loss": 0.3967612087726593, "lr": 7.225350053780395e-06, "epoch": 2.853448275862069, "percentage": 71.34, "elapsed_time": "6:26:09", "remaining_time": "2:35:09"}
332
+ {"current_steps": 664, "total_steps": 928, "loss": 0.5651652216911316, "lr": 7.1601551206725165e-06, "epoch": 2.862068965517241, "percentage": 71.55, "elapsed_time": "6:27:20", "remaining_time": "2:34:00"}
333
+ {"current_steps": 666, "total_steps": 928, "loss": 0.4929698705673218, "lr": 7.0952953700143915e-06, "epoch": 2.8706896551724137, "percentage": 71.77, "elapsed_time": "6:28:25", "remaining_time": "2:32:48"}
334
+ {"current_steps": 668, "total_steps": 928, "loss": 0.3763951063156128, "lr": 7.0307741007987234e-06, "epoch": 2.8793103448275863, "percentage": 71.98, "elapsed_time": "6:29:24", "remaining_time": "2:31:33"}
335
+ {"current_steps": 670, "total_steps": 928, "loss": 0.8635995388031006, "lr": 6.9665945948018505e-06, "epoch": 2.887931034482759, "percentage": 72.2, "elapsed_time": "6:30:48", "remaining_time": "2:30:29"}
336
+ {"current_steps": 672, "total_steps": 928, "loss": 0.5167340636253357, "lr": 6.902760116416865e-06, "epoch": 2.896551724137931, "percentage": 72.41, "elapsed_time": "6:31:44", "remaining_time": "2:29:14"}
337
+ {"current_steps": 674, "total_steps": 928, "loss": 0.7079530358314514, "lr": 6.83927391248754e-06, "epoch": 2.905172413793103, "percentage": 72.63, "elapsed_time": "6:32:45", "remaining_time": "2:28:00"}
338
+ {"current_steps": 676, "total_steps": 928, "loss": 0.7067210674285889, "lr": 6.776139212143207e-06, "epoch": 2.913793103448276, "percentage": 72.84, "elapsed_time": "6:34:07", "remaining_time": "2:26:55"}
339
+ {"current_steps": 678, "total_steps": 928, "loss": 0.8161290287971497, "lr": 6.713359226634489e-06, "epoch": 2.9224137931034484, "percentage": 73.06, "elapsed_time": "6:35:22", "remaining_time": "2:25:47"}
340
+ {"current_steps": 680, "total_steps": 928, "loss": 0.6068652868270874, "lr": 6.650937149169996e-06, "epoch": 2.9310344827586206, "percentage": 73.28, "elapsed_time": "6:36:27", "remaining_time": "2:24:35"}
341
+ {"current_steps": 682, "total_steps": 928, "loss": 0.2772270143032074, "lr": 6.588876154753869e-06, "epoch": 2.939655172413793, "percentage": 73.49, "elapsed_time": "6:37:09", "remaining_time": "2:23:15"}
342
+ {"current_steps": 684, "total_steps": 928, "loss": 0.3592681288719177, "lr": 6.527179400024325e-06, "epoch": 2.9482758620689653, "percentage": 73.71, "elapsed_time": "6:38:11", "remaining_time": "2:22:02"}
343
+ {"current_steps": 686, "total_steps": 928, "loss": 0.7297719120979309, "lr": 6.46585002309307e-06, "epoch": 2.956896551724138, "percentage": 73.92, "elapsed_time": "6:39:25", "remaining_time": "2:20:54"}
344
+ {"current_steps": 688, "total_steps": 928, "loss": 0.7315416932106018, "lr": 6.4048911433857075e-06, "epoch": 2.9655172413793105, "percentage": 74.14, "elapsed_time": "6:40:39", "remaining_time": "2:19:45"}
345
+ {"current_steps": 690, "total_steps": 928, "loss": 0.9271728992462158, "lr": 6.344305861483049e-06, "epoch": 2.9741379310344827, "percentage": 74.35, "elapsed_time": "6:42:07", "remaining_time": "2:18:42"}
346
+ {"current_steps": 692, "total_steps": 928, "loss": 0.9229543209075928, "lr": 6.284097258963437e-06, "epoch": 2.9827586206896552, "percentage": 74.57, "elapsed_time": "6:43:32", "remaining_time": "2:17:37"}
347
+ {"current_steps": 694, "total_steps": 928, "loss": 0.503404974937439, "lr": 6.224268398245974e-06, "epoch": 2.9913793103448274, "percentage": 74.78, "elapsed_time": "6:44:30", "remaining_time": "2:16:23"}
348
+ {"current_steps": 696, "total_steps": 928, "loss": 0.9257153272628784, "lr": 6.164822322434786e-06, "epoch": 3.0, "percentage": 75.0, "elapsed_time": "6:45:54", "remaining_time": "2:15:18"}
349
+ {"current_steps": 698, "total_steps": 928, "loss": 0.33168870210647583, "lr": 6.105762055164216e-06, "epoch": 3.0086206896551726, "percentage": 75.22, "elapsed_time": "6:46:58", "remaining_time": "2:14:06"}
350
+ {"current_steps": 700, "total_steps": 928, "loss": 0.5617142915725708, "lr": 6.047090600445046e-06, "epoch": 3.0172413793103448, "percentage": 75.43, "elapsed_time": "6:48:22", "remaining_time": "2:13:00"}
351
+ {"current_steps": 702, "total_steps": 928, "loss": 0.3617658317089081, "lr": 5.988810942511703e-06, "epoch": 3.0258620689655173, "percentage": 75.65, "elapsed_time": "6:49:28", "remaining_time": "2:11:49"}
352
+ {"current_steps": 704, "total_steps": 928, "loss": 0.4505208134651184, "lr": 5.930926045670452e-06, "epoch": 3.0344827586206895, "percentage": 75.86, "elapsed_time": "6:50:47", "remaining_time": "2:10:42"}
353
+ {"current_steps": 706, "total_steps": 928, "loss": 0.07196473330259323, "lr": 5.873438854148648e-06, "epoch": 3.043103448275862, "percentage": 76.08, "elapsed_time": "6:51:31", "remaining_time": "2:09:24"}
354
+ {"current_steps": 708, "total_steps": 928, "loss": 0.5367481112480164, "lr": 5.816352291944968e-06, "epoch": 3.0517241379310347, "percentage": 76.29, "elapsed_time": "6:52:56", "remaining_time": "2:08:19"}
355
+ {"current_steps": 710, "total_steps": 928, "loss": 0.13180424273014069, "lr": 5.759669262680674e-06, "epoch": 3.060344827586207, "percentage": 76.51, "elapsed_time": "6:53:57", "remaining_time": "2:07:06"}
356
+ {"current_steps": 712, "total_steps": 928, "loss": 0.20174063742160797, "lr": 5.703392649451958e-06, "epoch": 3.0689655172413794, "percentage": 76.72, "elapsed_time": "6:55:05", "remaining_time": "2:05:55"}
357
+ {"current_steps": 714, "total_steps": 928, "loss": 0.7795681357383728, "lr": 5.647525314683258e-06, "epoch": 3.0775862068965516, "percentage": 76.94, "elapsed_time": "6:56:32", "remaining_time": "2:04:50"}
358
+ {"current_steps": 716, "total_steps": 928, "loss": 0.23963159322738647, "lr": 5.592070099981703e-06, "epoch": 3.086206896551724, "percentage": 77.16, "elapsed_time": "6:57:29", "remaining_time": "2:03:36"}
359
+ {"current_steps": 718, "total_steps": 928, "loss": 0.5431758761405945, "lr": 5.53702982599254e-06, "epoch": 3.0948275862068964, "percentage": 77.37, "elapsed_time": "6:58:51", "remaining_time": "2:02:30"}
360
+ {"current_steps": 720, "total_steps": 928, "loss": 0.14490705728530884, "lr": 5.4824072922557125e-06, "epoch": 3.103448275862069, "percentage": 77.59, "elapsed_time": "6:59:43", "remaining_time": "2:01:15"}
361
+ {"current_steps": 722, "total_steps": 928, "loss": 0.2814076840877533, "lr": 5.428205277063415e-06, "epoch": 3.1120689655172415, "percentage": 77.8, "elapsed_time": "7:00:51", "remaining_time": "2:00:04"}
362
+ {"current_steps": 724, "total_steps": 928, "loss": 0.4774016737937927, "lr": 5.374426537318827e-06, "epoch": 3.1206896551724137, "percentage": 78.02, "elapsed_time": "7:02:14", "remaining_time": "1:58:58"}
363
+ {"current_steps": 726, "total_steps": 928, "loss": 0.5318699479103088, "lr": 5.321073808395846e-06, "epoch": 3.1293103448275863, "percentage": 78.23, "elapsed_time": "7:03:28", "remaining_time": "1:57:49"}
364
+ {"current_steps": 728, "total_steps": 928, "loss": 0.5016626119613647, "lr": 5.268149803999993e-06, "epoch": 3.1379310344827585, "percentage": 78.45, "elapsed_time": "7:04:50", "remaining_time": "1:56:42"}
365
+ {"current_steps": 730, "total_steps": 928, "loss": 0.34347403049468994, "lr": 5.215657216030353e-06, "epoch": 3.146551724137931, "percentage": 78.66, "elapsed_time": "7:05:54", "remaining_time": "1:55:31"}
366
+ {"current_steps": 732, "total_steps": 928, "loss": 0.25022614002227783, "lr": 5.163598714442685e-06, "epoch": 3.1551724137931036, "percentage": 78.88, "elapsed_time": "7:06:55", "remaining_time": "1:54:18"}
367
+ {"current_steps": 734, "total_steps": 928, "loss": 0.5590460300445557, "lr": 5.111976947113585e-06, "epoch": 3.163793103448276, "percentage": 79.09, "elapsed_time": "7:08:21", "remaining_time": "1:53:13"}
368
+ {"current_steps": 736, "total_steps": 928, "loss": 0.27467963099479675, "lr": 5.0607945397058415e-06, "epoch": 3.1724137931034484, "percentage": 79.31, "elapsed_time": "7:09:27", "remaining_time": "1:52:01"}
369
+ {"current_steps": 738, "total_steps": 928, "loss": 0.5068318843841553, "lr": 5.010054095534857e-06, "epoch": 3.1810344827586206, "percentage": 79.53, "elapsed_time": "7:10:53", "remaining_time": "1:50:55"}
370
+ {"current_steps": 740, "total_steps": 928, "loss": 0.10915378481149673, "lr": 4.959758195436251e-06, "epoch": 3.189655172413793, "percentage": 79.74, "elapsed_time": "7:11:43", "remaining_time": "1:49:40"}
371
+ {"current_steps": 742, "total_steps": 928, "loss": 0.26452872157096863, "lr": 4.9099093976345804e-06, "epoch": 3.1982758620689653, "percentage": 79.96, "elapsed_time": "7:12:46", "remaining_time": "1:48:29"}
372
+ {"current_steps": 744, "total_steps": 928, "loss": 0.41886910796165466, "lr": 4.860510237613227e-06, "epoch": 3.206896551724138, "percentage": 80.17, "elapsed_time": "7:14:00", "remaining_time": "1:47:20"}
373
+ {"current_steps": 746, "total_steps": 928, "loss": 0.3879053592681885, "lr": 4.811563227985424e-06, "epoch": 3.2155172413793105, "percentage": 80.39, "elapsed_time": "7:15:13", "remaining_time": "1:46:10"}
374
+ {"current_steps": 748, "total_steps": 928, "loss": 0.5333488583564758, "lr": 4.763070858366472e-06, "epoch": 3.2241379310344827, "percentage": 80.6, "elapsed_time": "7:16:26", "remaining_time": "1:45:01"}
375
+ {"current_steps": 750, "total_steps": 928, "loss": 0.5044701099395752, "lr": 4.7150355952470814e-06, "epoch": 3.2327586206896552, "percentage": 80.82, "elapsed_time": "7:17:53", "remaining_time": "1:43:55"}
376
+ {"current_steps": 752, "total_steps": 928, "loss": 0.29571160674095154, "lr": 4.667459881867956e-06, "epoch": 3.2413793103448274, "percentage": 81.03, "elapsed_time": "7:18:57", "remaining_time": "1:42:44"}
377
+ {"current_steps": 754, "total_steps": 928, "loss": 0.567728579044342, "lr": 4.6203461380954775e-06, "epoch": 3.25, "percentage": 81.25, "elapsed_time": "7:20:24", "remaining_time": "1:41:37"}
378
+ {"current_steps": 756, "total_steps": 928, "loss": 0.3926270306110382, "lr": 4.5736967602986656e-06, "epoch": 3.2586206896551726, "percentage": 81.47, "elapsed_time": "7:21:35", "remaining_time": "1:40:28"}
379
+ {"current_steps": 758, "total_steps": 928, "loss": 0.46212252974510193, "lr": 4.527514121227256e-06, "epoch": 3.2672413793103448, "percentage": 81.68, "elapsed_time": "7:23:00", "remaining_time": "1:39:21"}
380
+ {"current_steps": 760, "total_steps": 928, "loss": 0.326346218585968, "lr": 4.481800569891035e-06, "epoch": 3.2758620689655173, "percentage": 81.9, "elapsed_time": "7:24:14", "remaining_time": "1:38:12"}
381
+ {"current_steps": 762, "total_steps": 928, "loss": 0.42745885252952576, "lr": 4.4365584314403535e-06, "epoch": 3.2844827586206895, "percentage": 82.11, "elapsed_time": "7:25:38", "remaining_time": "1:37:04"}
382
+ {"current_steps": 764, "total_steps": 928, "loss": 0.47123435139656067, "lr": 4.391790007047859e-06, "epoch": 3.293103448275862, "percentage": 82.33, "elapsed_time": "7:26:51", "remaining_time": "1:35:55"}
383
+ {"current_steps": 766, "total_steps": 928, "loss": 0.28395605087280273, "lr": 4.347497573791454e-06, "epoch": 3.3017241379310347, "percentage": 82.54, "elapsed_time": "7:27:56", "remaining_time": "1:34:44"}
384
+ {"current_steps": 768, "total_steps": 928, "loss": 0.43049338459968567, "lr": 4.303683384538477e-06, "epoch": 3.310344827586207, "percentage": 82.76, "elapsed_time": "7:29:12", "remaining_time": "1:33:35"}
385
+ {"current_steps": 770, "total_steps": 928, "loss": 0.14107981324195862, "lr": 4.260349667831108e-06, "epoch": 3.3189655172413794, "percentage": 82.97, "elapsed_time": "7:30:20", "remaining_time": "1:32:24"}
386
+ {"current_steps": 772, "total_steps": 928, "loss": 0.3092440068721771, "lr": 4.217498627773029e-06, "epoch": 3.3275862068965516, "percentage": 83.19, "elapsed_time": "7:31:21", "remaining_time": "1:31:12"}
387
+ {"current_steps": 774, "total_steps": 928, "loss": 0.5438483357429504, "lr": 4.175132443917291e-06, "epoch": 3.336206896551724, "percentage": 83.41, "elapsed_time": "7:32:35", "remaining_time": "1:30:03"}
388
+ {"current_steps": 776, "total_steps": 928, "loss": 0.5660426616668701, "lr": 4.133253271155489e-06, "epoch": 3.344827586206897, "percentage": 83.62, "elapsed_time": "7:34:06", "remaining_time": "1:28:56"}
389
+ {"current_steps": 778, "total_steps": 928, "loss": 0.522563636302948, "lr": 4.091863239608131e-06, "epoch": 3.353448275862069, "percentage": 83.84, "elapsed_time": "7:35:23", "remaining_time": "1:27:47"}
390
+ {"current_steps": 780, "total_steps": 928, "loss": 0.3490191698074341, "lr": 4.050964454516295e-06, "epoch": 3.3620689655172415, "percentage": 84.05, "elapsed_time": "7:36:46", "remaining_time": "1:26:40"}
391
+ {"current_steps": 782, "total_steps": 928, "loss": 0.3295826017856598, "lr": 4.010558996134562e-06, "epoch": 3.3706896551724137, "percentage": 84.27, "elapsed_time": "7:37:49", "remaining_time": "1:25:28"}
392
+ {"current_steps": 784, "total_steps": 928, "loss": 0.5116720199584961, "lr": 3.970648919625199e-06, "epoch": 3.3793103448275863, "percentage": 84.48, "elapsed_time": "7:39:08", "remaining_time": "1:24:19"}
393
+ {"current_steps": 786, "total_steps": 928, "loss": 0.3242761194705963, "lr": 3.9312362549536195e-06, "epoch": 3.3879310344827585, "percentage": 84.7, "elapsed_time": "7:40:12", "remaining_time": "1:23:08"}
394
+ {"current_steps": 788, "total_steps": 928, "loss": 0.49483513832092285, "lr": 3.892323006785148e-06, "epoch": 3.396551724137931, "percentage": 84.91, "elapsed_time": "7:41:37", "remaining_time": "1:22:00"}
395
+ {"current_steps": 790, "total_steps": 928, "loss": 0.4984126091003418, "lr": 3.853911154383037e-06, "epoch": 3.405172413793103, "percentage": 85.13, "elapsed_time": "7:42:52", "remaining_time": "1:20:51"}
396
+ {"current_steps": 792, "total_steps": 928, "loss": 0.2216302752494812, "lr": 3.8160026515078165e-06, "epoch": 3.413793103448276, "percentage": 85.34, "elapsed_time": "7:43:43", "remaining_time": "1:19:37"}
397
+ {"current_steps": 794, "total_steps": 928, "loss": 0.24825817346572876, "lr": 3.7785994263178978e-06, "epoch": 3.4224137931034484, "percentage": 85.56, "elapsed_time": "7:44:43", "remaining_time": "1:18:25"}
398
+ {"current_steps": 796, "total_steps": 928, "loss": 0.6700829863548279, "lr": 3.741703381271519e-06, "epoch": 3.4310344827586206, "percentage": 85.78, "elapsed_time": "7:46:09", "remaining_time": "1:17:18"}
399
+ {"current_steps": 798, "total_steps": 928, "loss": 0.3420178294181824, "lr": 3.7053163930299625e-06, "epoch": 3.439655172413793, "percentage": 85.99, "elapsed_time": "7:47:32", "remaining_time": "1:16:10"}
400
+ {"current_steps": 800, "total_steps": 928, "loss": 0.5975306034088135, "lr": 3.6694403123621143e-06, "epoch": 3.4482758620689653, "percentage": 86.21, "elapsed_time": "7:48:58", "remaining_time": "1:15:02"}
401
+ {"current_steps": 802, "total_steps": 928, "loss": 0.4603136479854584, "lr": 3.634076964050321e-06, "epoch": 3.456896551724138, "percentage": 86.42, "elapsed_time": "7:50:22", "remaining_time": "1:13:53"}
402
+ {"current_steps": 804, "total_steps": 928, "loss": 0.4385143518447876, "lr": 3.5992281467975773e-06, "epoch": 3.4655172413793105, "percentage": 86.64, "elapsed_time": "7:51:33", "remaining_time": "1:12:43"}
403
+ {"current_steps": 806, "total_steps": 928, "loss": 0.4603370428085327, "lr": 3.5648956331360366e-06, "epoch": 3.4741379310344827, "percentage": 86.85, "elapsed_time": "7:52:50", "remaining_time": "1:11:34"}
404
+ {"current_steps": 808, "total_steps": 928, "loss": 0.5407328605651855, "lr": 3.5310811693368584e-06, "epoch": 3.4827586206896552, "percentage": 87.07, "elapsed_time": "7:53:58", "remaining_time": "1:10:23"}
405
+ {"current_steps": 810, "total_steps": 928, "loss": 0.14333094656467438, "lr": 3.4977864753213727e-06, "epoch": 3.4913793103448274, "percentage": 87.28, "elapsed_time": "7:54:49", "remaining_time": "1:09:10"}
406
+ {"current_steps": 812, "total_steps": 928, "loss": 0.3744501769542694, "lr": 3.4650132445736226e-06, "epoch": 3.5, "percentage": 87.5, "elapsed_time": "7:56:12", "remaining_time": "1:08:01"}
407
+ {"current_steps": 814, "total_steps": 928, "loss": 0.49726125597953796, "lr": 3.4327631440542018e-06, "epoch": 3.5086206896551726, "percentage": 87.72, "elapsed_time": "7:57:25", "remaining_time": "1:06:51"}
408
+ {"current_steps": 816, "total_steps": 928, "loss": 0.441265344619751, "lr": 3.401037814115489e-06, "epoch": 3.5172413793103448, "percentage": 87.93, "elapsed_time": "7:58:38", "remaining_time": "1:05:41"}
409
+ {"current_steps": 818, "total_steps": 928, "loss": 0.5781593322753906, "lr": 3.3698388684181987e-06, "epoch": 3.5258620689655173, "percentage": 88.15, "elapsed_time": "8:00:04", "remaining_time": "1:04:33"}
410
+ {"current_steps": 820, "total_steps": 928, "loss": 0.5013993382453918, "lr": 3.339167893849313e-06, "epoch": 3.5344827586206895, "percentage": 88.36, "elapsed_time": "8:01:18", "remaining_time": "1:03:23"}
411
+ {"current_steps": 822, "total_steps": 928, "loss": 0.2739977240562439, "lr": 3.3090264504413618e-06, "epoch": 3.543103448275862, "percentage": 88.58, "elapsed_time": "8:02:40", "remaining_time": "1:02:14"}
412
+ {"current_steps": 824, "total_steps": 928, "loss": 0.33216506242752075, "lr": 3.279416071293083e-06, "epoch": 3.5517241379310347, "percentage": 88.79, "elapsed_time": "8:03:33", "remaining_time": "1:01:01"}
413
+ {"current_steps": 826, "total_steps": 928, "loss": 0.08031827956438065, "lr": 3.2503382624914287e-06, "epoch": 3.560344827586207, "percentage": 89.01, "elapsed_time": "8:04:20", "remaining_time": "0:59:48"}
414
+ {"current_steps": 828, "total_steps": 928, "loss": 0.13542333245277405, "lr": 3.2217945030349785e-06, "epoch": 3.5689655172413794, "percentage": 89.22, "elapsed_time": "8:05:03", "remaining_time": "0:58:34"}
415
+ {"current_steps": 830, "total_steps": 928, "loss": 0.3155873119831085, "lr": 3.1937862447586945e-06, "epoch": 3.5775862068965516, "percentage": 89.44, "elapsed_time": "8:06:16", "remaining_time": "0:57:24"}
416
+ {"current_steps": 832, "total_steps": 928, "loss": 0.32241955399513245, "lr": 3.1663149122600946e-06, "epoch": 3.586206896551724, "percentage": 89.66, "elapsed_time": "8:07:18", "remaining_time": "0:56:13"}
417
+ {"current_steps": 834, "total_steps": 928, "loss": 0.5564541220664978, "lr": 3.1393819028267712e-06, "epoch": 3.594827586206897, "percentage": 89.87, "elapsed_time": "8:08:34", "remaining_time": "0:55:04"}
418
+ {"current_steps": 836, "total_steps": 928, "loss": 0.3382505476474762, "lr": 3.112988586365343e-06, "epoch": 3.603448275862069, "percentage": 90.09, "elapsed_time": "8:09:37", "remaining_time": "0:53:52"}
419
+ {"current_steps": 838, "total_steps": 928, "loss": 0.4953022599220276, "lr": 3.0871363053317565e-06, "epoch": 3.612068965517241, "percentage": 90.3, "elapsed_time": "8:11:04", "remaining_time": "0:52:44"}
420
+ {"current_steps": 840, "total_steps": 928, "loss": 0.12510839104652405, "lr": 3.0618263746630195e-06, "epoch": 3.6206896551724137, "percentage": 90.52, "elapsed_time": "8:12:07", "remaining_time": "0:51:33"}
421
+ {"current_steps": 842, "total_steps": 928, "loss": 0.12819121778011322, "lr": 3.0370600817103054e-06, "epoch": 3.6293103448275863, "percentage": 90.73, "elapsed_time": "8:12:48", "remaining_time": "0:50:20"}
422
+ {"current_steps": 844, "total_steps": 928, "loss": 0.34562334418296814, "lr": 3.0128386861734907e-06, "epoch": 3.637931034482759, "percentage": 90.95, "elapsed_time": "8:14:11", "remaining_time": "0:49:11"}
423
+ {"current_steps": 846, "total_steps": 928, "loss": 0.2538186311721802, "lr": 2.9891634200370615e-06, "epoch": 3.646551724137931, "percentage": 91.16, "elapsed_time": "8:15:29", "remaining_time": "0:48:01"}
424
+ {"current_steps": 848, "total_steps": 928, "loss": 0.09849962592124939, "lr": 2.9660354875074694e-06, "epoch": 3.655172413793103, "percentage": 91.38, "elapsed_time": "8:16:11", "remaining_time": "0:46:48"}
425
+ {"current_steps": 850, "total_steps": 928, "loss": 0.31217214465141296, "lr": 2.9434560649518708e-06, "epoch": 3.663793103448276, "percentage": 91.59, "elapsed_time": "8:17:16", "remaining_time": "0:45:37"}
426
+ {"current_steps": 852, "total_steps": 928, "loss": 0.5859465003013611, "lr": 2.9214263008382943e-06, "epoch": 3.6724137931034484, "percentage": 91.81, "elapsed_time": "8:18:41", "remaining_time": "0:44:29"}
427
+ {"current_steps": 854, "total_steps": 928, "loss": 0.32843440771102905, "lr": 2.8999473156772322e-06, "epoch": 3.6810344827586206, "percentage": 92.03, "elapsed_time": "8:19:45", "remaining_time": "0:43:18"}
428
+ {"current_steps": 856, "total_steps": 928, "loss": 0.525081992149353, "lr": 2.8790202019646327e-06, "epoch": 3.689655172413793, "percentage": 92.24, "elapsed_time": "8:21:10", "remaining_time": "0:42:09"}
429
+ {"current_steps": 858, "total_steps": 928, "loss": 0.5662536025047302, "lr": 2.858646024126349e-06, "epoch": 3.6982758620689653, "percentage": 92.46, "elapsed_time": "8:22:39", "remaining_time": "0:41:00"}
430
+ {"current_steps": 860, "total_steps": 928, "loss": 0.26983019709587097, "lr": 2.8388258184639895e-06, "epoch": 3.706896551724138, "percentage": 92.67, "elapsed_time": "8:23:45", "remaining_time": "0:39:49"}
431
+ {"current_steps": 862, "total_steps": 928, "loss": 0.325879842042923, "lr": 2.8195605931022044e-06, "epoch": 3.7155172413793105, "percentage": 92.89, "elapsed_time": "8:25:04", "remaining_time": "0:38:40"}
432
+ {"current_steps": 864, "total_steps": 928, "loss": 0.26106154918670654, "lr": 2.8008513279374227e-06, "epoch": 3.7241379310344827, "percentage": 93.1, "elapsed_time": "8:26:10", "remaining_time": "0:37:29"}
433
+ {"current_steps": 866, "total_steps": 928, "loss": 0.0877346470952034, "lr": 2.7826989745879937e-06, "epoch": 3.7327586206896552, "percentage": 93.32, "elapsed_time": "8:26:52", "remaining_time": "0:36:17"}
434
+ {"current_steps": 868, "total_steps": 928, "loss": 0.32902559638023376, "lr": 2.7651044563457983e-06, "epoch": 3.7413793103448274, "percentage": 93.53, "elapsed_time": "8:27:55", "remaining_time": "0:35:06"}
435
+ {"current_steps": 870, "total_steps": 928, "loss": 0.40092846751213074, "lr": 2.748068668129279e-06, "epoch": 3.75, "percentage": 93.75, "elapsed_time": "8:29:06", "remaining_time": "0:33:56"}
436
+ {"current_steps": 872, "total_steps": 928, "loss": 0.12595315277576447, "lr": 2.7315924764379292e-06, "epoch": 3.7586206896551726, "percentage": 93.97, "elapsed_time": "8:30:12", "remaining_time": "0:32:45"}
437
+ {"current_steps": 874, "total_steps": 928, "loss": 0.30979275703430176, "lr": 2.715676719308208e-06, "epoch": 3.7672413793103448, "percentage": 94.18, "elapsed_time": "8:31:16", "remaining_time": "0:31:35"}
438
+ {"current_steps": 876, "total_steps": 928, "loss": 0.5150949358940125, "lr": 2.7003222062709295e-06, "epoch": 3.7758620689655173, "percentage": 94.4, "elapsed_time": "8:32:37", "remaining_time": "0:30:25"}
439
+ {"current_steps": 878, "total_steps": 928, "loss": 0.41893500089645386, "lr": 2.6855297183100716e-06, "epoch": 3.7844827586206895, "percentage": 94.61, "elapsed_time": "8:33:59", "remaining_time": "0:29:16"}
440
+ {"current_steps": 880, "total_steps": 928, "loss": 0.28298234939575195, "lr": 2.671300007823068e-06, "epoch": 3.793103448275862, "percentage": 94.83, "elapsed_time": "8:34:52", "remaining_time": "0:28:05"}
441
+ {"current_steps": 882, "total_steps": 928, "loss": 0.29594770073890686, "lr": 2.657633798582525e-06, "epoch": 3.8017241379310347, "percentage": 95.04, "elapsed_time": "8:35:56", "remaining_time": "0:26:54"}
442
+ {"current_steps": 884, "total_steps": 928, "loss": 0.5765542984008789, "lr": 2.6445317856994187e-06, "epoch": 3.810344827586207, "percentage": 95.26, "elapsed_time": "8:37:20", "remaining_time": "0:25:45"}
443
+ {"current_steps": 886, "total_steps": 928, "loss": 0.18173623085021973, "lr": 2.6319946355877313e-06, "epoch": 3.8189655172413794, "percentage": 95.47, "elapsed_time": "8:38:25", "remaining_time": "0:24:34"}
444
+ {"current_steps": 888, "total_steps": 928, "loss": 0.3384171724319458, "lr": 2.62002298593056e-06, "epoch": 3.8275862068965516, "percentage": 95.69, "elapsed_time": "8:39:48", "remaining_time": "0:23:24"}
445
+ {"current_steps": 890, "total_steps": 928, "loss": 0.10825614631175995, "lr": 2.6086174456476774e-06, "epoch": 3.836206896551724, "percentage": 95.91, "elapsed_time": "8:40:37", "remaining_time": "0:22:13"}
446
+ {"current_steps": 892, "total_steps": 928, "loss": 0.36479508876800537, "lr": 2.5977785948645674e-06, "epoch": 3.844827586206897, "percentage": 96.12, "elapsed_time": "8:41:59", "remaining_time": "0:21:03"}
447
+ {"current_steps": 894, "total_steps": 928, "loss": 0.4843851923942566, "lr": 2.58750698488291e-06, "epoch": 3.853448275862069, "percentage": 96.34, "elapsed_time": "8:43:05", "remaining_time": "0:19:53"}
448
+ {"current_steps": 896, "total_steps": 928, "loss": 0.34008628129959106, "lr": 2.577803138152543e-06, "epoch": 3.862068965517241, "percentage": 96.55, "elapsed_time": "8:44:16", "remaining_time": "0:18:43"}
449
+ {"current_steps": 898, "total_steps": 928, "loss": 0.2734866142272949, "lr": 2.56866754824489e-06, "epoch": 3.8706896551724137, "percentage": 96.77, "elapsed_time": "8:45:18", "remaining_time": "0:17:32"}
450
+ {"current_steps": 900, "total_steps": 928, "loss": 0.5301221013069153, "lr": 2.560100679827858e-06, "epoch": 3.8793103448275863, "percentage": 96.98, "elapsed_time": "8:46:43", "remaining_time": "0:16:23"}
451
+ {"current_steps": 902, "total_steps": 928, "loss": 0.40625, "lr": 2.5521029686421927e-06, "epoch": 3.887931034482759, "percentage": 97.2, "elapsed_time": "8:47:36", "remaining_time": "0:15:12"}
452
+ {"current_steps": 904, "total_steps": 928, "loss": 0.412536084651947, "lr": 2.544674821479325e-06, "epoch": 3.896551724137931, "percentage": 97.41, "elapsed_time": "8:48:39", "remaining_time": "0:14:02"}
453
+ {"current_steps": 906, "total_steps": 928, "loss": 0.5061193108558655, "lr": 2.5378166161606777e-06, "epoch": 3.905172413793103, "percentage": 97.63, "elapsed_time": "8:50:03", "remaining_time": "0:12:52"}
454
+ {"current_steps": 908, "total_steps": 928, "loss": 0.2978528141975403, "lr": 2.5315287015184486e-06, "epoch": 3.913793103448276, "percentage": 97.84, "elapsed_time": "8:51:06", "remaining_time": "0:11:41"}
455
+ {"current_steps": 910, "total_steps": 928, "loss": 0.21743205189704895, "lr": 2.525811397377861e-06, "epoch": 3.9224137931034484, "percentage": 98.06, "elapsed_time": "8:51:52", "remaining_time": "0:10:31"}
456
+ {"current_steps": 912, "total_steps": 928, "loss": 0.2807198464870453, "lr": 2.520664994540907e-06, "epoch": 3.9310344827586206, "percentage": 98.28, "elapsed_time": "8:52:59", "remaining_time": "0:09:21"}
457
+ {"current_steps": 914, "total_steps": 928, "loss": 0.537182629108429, "lr": 2.5160897547715503e-06, "epoch": 3.939655172413793, "percentage": 98.49, "elapsed_time": "8:54:30", "remaining_time": "0:08:11"}
458
+ {"current_steps": 916, "total_steps": 928, "loss": 0.4790320098400116, "lr": 2.5120859107824107e-06, "epoch": 3.9482758620689653, "percentage": 98.71, "elapsed_time": "8:55:51", "remaining_time": "0:07:01"}
459
+ {"current_steps": 918, "total_steps": 928, "loss": 0.3746393024921417, "lr": 2.508653666222932e-06, "epoch": 3.956896551724138, "percentage": 98.92, "elapsed_time": "8:56:43", "remaining_time": "0:05:50"}
460
+ {"current_steps": 920, "total_steps": 928, "loss": 0.3896881341934204, "lr": 2.5057931956690186e-06, "epoch": 3.9655172413793105, "percentage": 99.14, "elapsed_time": "8:57:47", "remaining_time": "0:04:40"}
461
+ {"current_steps": 922, "total_steps": 928, "loss": 0.40220990777015686, "lr": 2.5035046446141617e-06, "epoch": 3.9741379310344827, "percentage": 99.35, "elapsed_time": "8:58:57", "remaining_time": "0:03:30"}
462
+ {"current_steps": 924, "total_steps": 928, "loss": 0.5496131777763367, "lr": 2.501788129462033e-06, "epoch": 3.9827586206896552, "percentage": 99.57, "elapsed_time": "9:00:12", "remaining_time": "0:02:20"}
463
+ {"current_steps": 926, "total_steps": 928, "loss": 0.5806067585945129, "lr": 2.5006437375205682e-06, "epoch": 3.9913793103448274, "percentage": 99.78, "elapsed_time": "9:01:25", "remaining_time": "0:01:10"}
464
+ {"current_steps": 928, "total_steps": 928, "loss": 0.39202290773391724, "lr": 2.5000715269975277e-06, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "9:02:46", "remaining_time": "0:00:00"}
465
+ {"current_steps": 928, "total_steps": 928, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "9:02:46", "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": 0.26599958539009094,
15
+ "learning_rate": 5.319148936170213e-07,
16
+ "loss": 3.7070133686065674,
17
+ "step": 2
18
+ },
19
+ {
20
+ "epoch": 0.017241379310344827,
21
+ "grad_norm": 0.8651937246322632,
22
+ "learning_rate": 1.5957446808510639e-06,
23
+ "loss": 2.565812349319458,
24
+ "step": 4
25
+ },
26
+ {
27
+ "epoch": 0.02586206896551724,
28
+ "grad_norm": 1.8721028566360474,
29
+ "learning_rate": 2.6595744680851065e-06,
30
+ "loss": 2.305459976196289,
31
+ "step": 6
32
+ },
33
+ {
34
+ "epoch": 0.034482758620689655,
35
+ "grad_norm": 0.3591924011707306,
36
+ "learning_rate": 3.723404255319149e-06,
37
+ "loss": 1.9556151628494263,
38
+ "step": 8
39
+ },
40
+ {
41
+ "epoch": 0.04310344827586207,
42
+ "grad_norm": 0.9290058016777039,
43
+ "learning_rate": 4.787234042553191e-06,
44
+ "loss": 1.509597897529602,
45
+ "step": 10
46
+ },
47
+ {
48
+ "epoch": 0.05172413793103448,
49
+ "grad_norm": 0.21377742290496826,
50
+ "learning_rate": 5.851063829787235e-06,
51
+ "loss": 1.7625975608825684,
52
+ "step": 12
53
+ },
54
+ {
55
+ "epoch": 0.0603448275862069,
56
+ "grad_norm": 0.32525238394737244,
57
+ "learning_rate": 6.914893617021277e-06,
58
+ "loss": 1.8272696733474731,
59
+ "step": 14
60
+ },
61
+ {
62
+ "epoch": 0.06896551724137931,
63
+ "grad_norm": 0.1417679488658905,
64
+ "learning_rate": 7.97872340425532e-06,
65
+ "loss": 0.7375114560127258,
66
+ "step": 16
67
+ },
68
+ {
69
+ "epoch": 0.07758620689655173,
70
+ "grad_norm": 0.31613945960998535,
71
+ "learning_rate": 9.042553191489362e-06,
72
+ "loss": 0.8689496517181396,
73
+ "step": 18
74
+ },
75
+ {
76
+ "epoch": 0.08620689655172414,
77
+ "grad_norm": 0.13069696724414825,
78
+ "learning_rate": 1.0106382978723404e-05,
79
+ "loss": 1.5243046283721924,
80
+ "step": 20
81
+ },
82
+ {
83
+ "epoch": 0.09482758620689655,
84
+ "grad_norm": 0.12042803317308426,
85
+ "learning_rate": 1.1170212765957447e-05,
86
+ "loss": 1.672706127166748,
87
+ "step": 22
88
+ },
89
+ {
90
+ "epoch": 0.10344827586206896,
91
+ "grad_norm": 0.24695225059986115,
92
+ "learning_rate": 1.223404255319149e-05,
93
+ "loss": 1.5508843660354614,
94
+ "step": 24
95
+ },
96
+ {
97
+ "epoch": 0.11206896551724138,
98
+ "grad_norm": 0.2681316137313843,
99
+ "learning_rate": 1.3297872340425532e-05,
100
+ "loss": 1.3349249362945557,
101
+ "step": 26
102
+ },
103
+ {
104
+ "epoch": 0.1206896551724138,
105
+ "grad_norm": 2.3104257583618164,
106
+ "learning_rate": 1.4361702127659577e-05,
107
+ "loss": 0.764009952545166,
108
+ "step": 28
109
+ },
110
+ {
111
+ "epoch": 0.12931034482758622,
112
+ "grad_norm": 0.16677935421466827,
113
+ "learning_rate": 1.5425531914893617e-05,
114
+ "loss": 1.4469249248504639,
115
+ "step": 30
116
+ },
117
+ {
118
+ "epoch": 0.13793103448275862,
119
+ "grad_norm": 0.2230503112077713,
120
+ "learning_rate": 1.6489361702127658e-05,
121
+ "loss": 1.205310583114624,
122
+ "step": 32
123
+ },
124
+ {
125
+ "epoch": 0.14655172413793102,
126
+ "grad_norm": 0.1180746778845787,
127
+ "learning_rate": 1.7553191489361703e-05,
128
+ "loss": 1.3685661554336548,
129
+ "step": 34
130
+ },
131
+ {
132
+ "epoch": 0.15517241379310345,
133
+ "grad_norm": 0.12433399260044098,
134
+ "learning_rate": 1.8617021276595745e-05,
135
+ "loss": 1.3990949392318726,
136
+ "step": 36
137
+ },
138
+ {
139
+ "epoch": 0.16379310344827586,
140
+ "grad_norm": 0.20617301762104034,
141
+ "learning_rate": 1.968085106382979e-05,
142
+ "loss": 1.4243614673614502,
143
+ "step": 38
144
+ },
145
+ {
146
+ "epoch": 0.1724137931034483,
147
+ "grad_norm": 0.12327367067337036,
148
+ "learning_rate": 2.074468085106383e-05,
149
+ "loss": 1.4310336112976074,
150
+ "step": 40
151
+ },
152
+ {
153
+ "epoch": 0.1810344827586207,
154
+ "grad_norm": 0.23117953538894653,
155
+ "learning_rate": 2.1808510638297873e-05,
156
+ "loss": 1.6425907611846924,
157
+ "step": 42
158
+ },
159
+ {
160
+ "epoch": 0.1896551724137931,
161
+ "grad_norm": 0.2595376968383789,
162
+ "learning_rate": 2.2872340425531915e-05,
163
+ "loss": 1.0640761852264404,
164
+ "step": 44
165
+ },
166
+ {
167
+ "epoch": 0.19827586206896552,
168
+ "grad_norm": 0.21752837300300598,
169
+ "learning_rate": 2.393617021276596e-05,
170
+ "loss": 1.133437156677246,
171
+ "step": 46
172
+ },
173
+ {
174
+ "epoch": 0.20689655172413793,
175
+ "grad_norm": 0.27208733558654785,
176
+ "learning_rate": 2.5e-05,
177
+ "loss": 1.3954966068267822,
178
+ "step": 48
179
+ },
180
+ {
181
+ "epoch": 0.21551724137931033,
182
+ "grad_norm": 0.1533847600221634,
183
+ "learning_rate": 2.4999713892919425e-05,
184
+ "loss": 1.5042736530303955,
185
+ "step": 50
186
+ },
187
+ {
188
+ "epoch": 0.22413793103448276,
189
+ "grad_norm": 0.8918558955192566,
190
+ "learning_rate": 2.4998855586230096e-05,
191
+ "loss": 0.8469613194465637,
192
+ "step": 52
193
+ },
194
+ {
195
+ "epoch": 0.23275862068965517,
196
+ "grad_norm": 0.2088381052017212,
197
+ "learning_rate": 2.4997425123588484e-05,
198
+ "loss": 1.439496636390686,
199
+ "step": 54
200
+ },
201
+ {
202
+ "epoch": 0.2413793103448276,
203
+ "grad_norm": 0.08935986459255219,
204
+ "learning_rate": 2.4995422577752894e-05,
205
+ "loss": 1.3860807418823242,
206
+ "step": 56
207
+ },
208
+ {
209
+ "epoch": 0.25,
210
+ "grad_norm": 0.27968838810920715,
211
+ "learning_rate": 2.499284805057978e-05,
212
+ "loss": 1.4035521745681763,
213
+ "step": 58
214
+ },
215
+ {
216
+ "epoch": 0.25862068965517243,
217
+ "grad_norm": 0.1754721850156784,
218
+ "learning_rate": 2.4989701673018555e-05,
219
+ "loss": 1.399681568145752,
220
+ "step": 60
221
+ },
222
+ {
223
+ "epoch": 0.2672413793103448,
224
+ "grad_norm": 0.4435209035873413,
225
+ "learning_rate": 2.4985983605104933e-05,
226
+ "loss": 1.099133849143982,
227
+ "step": 62
228
+ },
229
+ {
230
+ "epoch": 0.27586206896551724,
231
+ "grad_norm": 0.09654241800308228,
232
+ "learning_rate": 2.4981694035952797e-05,
233
+ "loss": 1.333014726638794,
234
+ "step": 64
235
+ },
236
+ {
237
+ "epoch": 0.28448275862068967,
238
+ "grad_norm": 0.15928423404693604,
239
+ "learning_rate": 2.497683318374456e-05,
240
+ "loss": 1.316357135772705,
241
+ "step": 66
242
+ },
243
+ {
244
+ "epoch": 0.29310344827586204,
245
+ "grad_norm": 0.09688454866409302,
246
+ "learning_rate": 2.497140129572009e-05,
247
+ "loss": 1.3247264623641968,
248
+ "step": 68
249
+ },
250
+ {
251
+ "epoch": 0.3017241379310345,
252
+ "grad_norm": 0.09453880041837692,
253
+ "learning_rate": 2.4965398648164114e-05,
254
+ "loss": 1.0423731803894043,
255
+ "step": 70
256
+ },
257
+ {
258
+ "epoch": 0.3103448275862069,
259
+ "grad_norm": 0.2843893766403198,
260
+ "learning_rate": 2.495882554639219e-05,
261
+ "loss": 1.1186097860336304,
262
+ "step": 72
263
+ },
264
+ {
265
+ "epoch": 0.31896551724137934,
266
+ "grad_norm": 0.21019835770130157,
267
+ "learning_rate": 2.4951682324735147e-05,
268
+ "loss": 1.0758298635482788,
269
+ "step": 74
270
+ },
271
+ {
272
+ "epoch": 0.3275862068965517,
273
+ "grad_norm": 0.12580540776252747,
274
+ "learning_rate": 2.4943969346522113e-05,
275
+ "loss": 1.177038311958313,
276
+ "step": 76
277
+ },
278
+ {
279
+ "epoch": 0.33620689655172414,
280
+ "grad_norm": 0.06072475388646126,
281
+ "learning_rate": 2.4935687004062e-05,
282
+ "loss": 1.1172969341278076,
283
+ "step": 78
284
+ },
285
+ {
286
+ "epoch": 0.3448275862068966,
287
+ "grad_norm": 0.45244401693344116,
288
+ "learning_rate": 2.492683571862358e-05,
289
+ "loss": 0.925361156463623,
290
+ "step": 80
291
+ },
292
+ {
293
+ "epoch": 0.35344827586206895,
294
+ "grad_norm": 0.12420380115509033,
295
+ "learning_rate": 2.491741594041404e-05,
296
+ "loss": 1.3231241703033447,
297
+ "step": 82
298
+ },
299
+ {
300
+ "epoch": 0.3620689655172414,
301
+ "grad_norm": 0.27926552295684814,
302
+ "learning_rate": 2.4907428148556096e-05,
303
+ "loss": 1.1561291217803955,
304
+ "step": 84
305
+ },
306
+ {
307
+ "epoch": 0.3706896551724138,
308
+ "grad_norm": 0.13415296375751495,
309
+ "learning_rate": 2.48968728510636e-05,
310
+ "loss": 1.3458794355392456,
311
+ "step": 86
312
+ },
313
+ {
314
+ "epoch": 0.3793103448275862,
315
+ "grad_norm": 0.09626008570194244,
316
+ "learning_rate": 2.4885750584815733e-05,
317
+ "loss": 1.2599420547485352,
318
+ "step": 88
319
+ },
320
+ {
321
+ "epoch": 0.3879310344827586,
322
+ "grad_norm": 0.4228016138076782,
323
+ "learning_rate": 2.487406191552967e-05,
324
+ "loss": 1.0761513710021973,
325
+ "step": 90
326
+ },
327
+ {
328
+ "epoch": 0.39655172413793105,
329
+ "grad_norm": 0.09193041175603867,
330
+ "learning_rate": 2.4861807437731816e-05,
331
+ "loss": 0.8632556200027466,
332
+ "step": 92
333
+ },
334
+ {
335
+ "epoch": 0.4051724137931034,
336
+ "grad_norm": 0.4261094629764557,
337
+ "learning_rate": 2.4848987774727572e-05,
338
+ "loss": 1.113997220993042,
339
+ "step": 94
340
+ },
341
+ {
342
+ "epoch": 0.41379310344827586,
343
+ "grad_norm": 0.33825597167015076,
344
+ "learning_rate": 2.4835603578569627e-05,
345
+ "loss": 1.225900650024414,
346
+ "step": 96
347
+ },
348
+ {
349
+ "epoch": 0.4224137931034483,
350
+ "grad_norm": 0.32547953724861145,
351
+ "learning_rate": 2.4821655530024785e-05,
352
+ "loss": 1.2307943105697632,
353
+ "step": 98
354
+ },
355
+ {
356
+ "epoch": 0.43103448275862066,
357
+ "grad_norm": 0.20468811690807343,
358
+ "learning_rate": 2.4807144338539345e-05,
359
+ "loss": 1.150252103805542,
360
+ "step": 100
361
+ },
362
+ {
363
+ "epoch": 0.4396551724137931,
364
+ "grad_norm": 0.23082198202610016,
365
+ "learning_rate": 2.479207074220303e-05,
366
+ "loss": 1.1921312808990479,
367
+ "step": 102
368
+ },
369
+ {
370
+ "epoch": 0.4482758620689655,
371
+ "grad_norm": 0.12201318144798279,
372
+ "learning_rate": 2.4776435507711404e-05,
373
+ "loss": 1.2748147249221802,
374
+ "step": 104
375
+ },
376
+ {
377
+ "epoch": 0.45689655172413796,
378
+ "grad_norm": 0.11089494824409485,
379
+ "learning_rate": 2.476023943032693e-05,
380
+ "loss": 0.961470365524292,
381
+ "step": 106
382
+ },
383
+ {
384
+ "epoch": 0.46551724137931033,
385
+ "grad_norm": 0.21629932522773743,
386
+ "learning_rate": 2.474348333383848e-05,
387
+ "loss": 1.183750867843628,
388
+ "step": 108
389
+ },
390
+ {
391
+ "epoch": 0.47413793103448276,
392
+ "grad_norm": 0.18608631193637848,
393
+ "learning_rate": 2.4726168070519457e-05,
394
+ "loss": 1.1943963766098022,
395
+ "step": 110
396
+ },
397
+ {
398
+ "epoch": 0.4827586206896552,
399
+ "grad_norm": 0.10688410699367523,
400
+ "learning_rate": 2.4708294521084412e-05,
401
+ "loss": 1.4336156845092773,
402
+ "step": 112
403
+ },
404
+ {
405
+ "epoch": 0.49137931034482757,
406
+ "grad_norm": 0.48181021213531494,
407
+ "learning_rate": 2.46898635946443e-05,
408
+ "loss": 0.9574199318885803,
409
+ "step": 114
410
+ },
411
+ {
412
+ "epoch": 0.5,
413
+ "grad_norm": 0.15253476798534393,
414
+ "learning_rate": 2.4670876228660192e-05,
415
+ "loss": 0.8921310901641846,
416
+ "step": 116
417
+ },
418
+ {
419
+ "epoch": 0.5086206896551724,
420
+ "grad_norm": 0.12271574139595032,
421
+ "learning_rate": 2.4651333388895617e-05,
422
+ "loss": 1.2967804670333862,
423
+ "step": 118
424
+ },
425
+ {
426
+ "epoch": 0.5172413793103449,
427
+ "grad_norm": 0.0918755903840065,
428
+ "learning_rate": 2.4631236069367435e-05,
429
+ "loss": 1.2260866165161133,
430
+ "step": 120
431
+ },
432
+ {
433
+ "epoch": 0.5258620689655172,
434
+ "grad_norm": 0.09145987033843994,
435
+ "learning_rate": 2.4610585292295278e-05,
436
+ "loss": 1.3115153312683105,
437
+ "step": 122
438
+ },
439
+ {
440
+ "epoch": 0.5344827586206896,
441
+ "grad_norm": 0.22018814086914062,
442
+ "learning_rate": 2.458938210804955e-05,
443
+ "loss": 1.458106279373169,
444
+ "step": 124
445
+ },
446
+ {
447
+ "epoch": 0.5431034482758621,
448
+ "grad_norm": 0.10475312173366547,
449
+ "learning_rate": 2.456762759509801e-05,
450
+ "loss": 1.589352011680603,
451
+ "step": 126
452
+ },
453
+ {
454
+ "epoch": 0.5517241379310345,
455
+ "grad_norm": 0.074061319231987,
456
+ "learning_rate": 2.454532285995091e-05,
457
+ "loss": 1.3105168342590332,
458
+ "step": 128
459
+ },
460
+ {
461
+ "epoch": 0.5603448275862069,
462
+ "grad_norm": 0.07942219823598862,
463
+ "learning_rate": 2.4522469037104732e-05,
464
+ "loss": 1.2909717559814453,
465
+ "step": 130
466
+ },
467
+ {
468
+ "epoch": 0.5689655172413793,
469
+ "grad_norm": 0.1517823338508606,
470
+ "learning_rate": 2.449906728898445e-05,
471
+ "loss": 1.2390798330307007,
472
+ "step": 132
473
+ },
474
+ {
475
+ "epoch": 0.5775862068965517,
476
+ "grad_norm": 0.3011615574359894,
477
+ "learning_rate": 2.447511880588444e-05,
478
+ "loss": 1.150702714920044,
479
+ "step": 134
480
+ },
481
+ {
482
+ "epoch": 0.5862068965517241,
483
+ "grad_norm": 0.12555329501628876,
484
+ "learning_rate": 2.445062480590792e-05,
485
+ "loss": 0.978553056716919,
486
+ "step": 136
487
+ },
488
+ {
489
+ "epoch": 0.5948275862068966,
490
+ "grad_norm": 0.049786102026700974,
491
+ "learning_rate": 2.4425586534904986e-05,
492
+ "loss": 0.9309189915657043,
493
+ "step": 138
494
+ },
495
+ {
496
+ "epoch": 0.603448275862069,
497
+ "grad_norm": 0.13792221248149872,
498
+ "learning_rate": 2.440000526640926e-05,
499
+ "loss": 0.9206077456474304,
500
+ "step": 140
501
+ },
502
+ {
503
+ "epoch": 0.6120689655172413,
504
+ "grad_norm": 0.08182723075151443,
505
+ "learning_rate": 2.437388230157311e-05,
506
+ "loss": 1.0149747133255005,
507
+ "step": 142
508
+ },
509
+ {
510
+ "epoch": 0.6206896551724138,
511
+ "grad_norm": 0.09423603862524033,
512
+ "learning_rate": 2.4347218969101456e-05,
513
+ "loss": 1.004448413848877,
514
+ "step": 144
515
+ },
516
+ {
517
+ "epoch": 0.6293103448275862,
518
+ "grad_norm": 0.0868963673710823,
519
+ "learning_rate": 2.4320016625184208e-05,
520
+ "loss": 1.251673698425293,
521
+ "step": 146
522
+ },
523
+ {
524
+ "epoch": 0.6379310344827587,
525
+ "grad_norm": 0.24749471247196198,
526
+ "learning_rate": 2.4292276653427266e-05,
527
+ "loss": 1.2662962675094604,
528
+ "step": 148
529
+ },
530
+ {
531
+ "epoch": 0.646551724137931,
532
+ "grad_norm": 0.1099071204662323,
533
+ "learning_rate": 2.4264000464782154e-05,
534
+ "loss": 1.05961275100708,
535
+ "step": 150
536
+ },
537
+ {
538
+ "epoch": 0.6551724137931034,
539
+ "grad_norm": 0.13175362348556519,
540
+ "learning_rate": 2.423518949747426e-05,
541
+ "loss": 0.8592603206634521,
542
+ "step": 152
543
+ },
544
+ {
545
+ "epoch": 0.6637931034482759,
546
+ "grad_norm": 0.25698038935661316,
547
+ "learning_rate": 2.420584521692967e-05,
548
+ "loss": 1.0430630445480347,
549
+ "step": 154
550
+ },
551
+ {
552
+ "epoch": 0.6724137931034483,
553
+ "grad_norm": 1.2381486892700195,
554
+ "learning_rate": 2.4175969115700637e-05,
555
+ "loss": 1.1310160160064697,
556
+ "step": 156
557
+ },
558
+ {
559
+ "epoch": 0.6810344827586207,
560
+ "grad_norm": 0.27287977933883667,
561
+ "learning_rate": 2.4145562713389665e-05,
562
+ "loss": 1.237598180770874,
563
+ "step": 158
564
+ },
565
+ {
566
+ "epoch": 0.6896551724137931,
567
+ "grad_norm": 0.18096014857292175,
568
+ "learning_rate": 2.4114627556572236e-05,
569
+ "loss": 1.2904868125915527,
570
+ "step": 160
571
+ },
572
+ {
573
+ "epoch": 0.6982758620689655,
574
+ "grad_norm": 0.12753747403621674,
575
+ "learning_rate": 2.4083165218718094e-05,
576
+ "loss": 1.5490325689315796,
577
+ "step": 162
578
+ },
579
+ {
580
+ "epoch": 0.7068965517241379,
581
+ "grad_norm": 0.1123153418302536,
582
+ "learning_rate": 2.405117730011127e-05,
583
+ "loss": 0.9844979643821716,
584
+ "step": 164
585
+ },
586
+ {
587
+ "epoch": 0.7155172413793104,
588
+ "grad_norm": 0.14886538684368134,
589
+ "learning_rate": 2.4018665427768652e-05,
590
+ "loss": 1.264622449874878,
591
+ "step": 166
592
+ },
593
+ {
594
+ "epoch": 0.7241379310344828,
595
+ "grad_norm": 0.09712914377450943,
596
+ "learning_rate": 2.3985631255357243e-05,
597
+ "loss": 1.3305325508117676,
598
+ "step": 168
599
+ },
600
+ {
601
+ "epoch": 0.7327586206896551,
602
+ "grad_norm": 0.11504893749952316,
603
+ "learning_rate": 2.3952076463110038e-05,
604
+ "loss": 1.2382071018218994,
605
+ "step": 170
606
+ },
607
+ {
608
+ "epoch": 0.7413793103448276,
609
+ "grad_norm": 0.23046067357063293,
610
+ "learning_rate": 2.3918002757740583e-05,
611
+ "loss": 1.249595046043396,
612
+ "step": 172
613
+ },
614
+ {
615
+ "epoch": 0.75,
616
+ "grad_norm": 0.1944531351327896,
617
+ "learning_rate": 2.3883411872356135e-05,
618
+ "loss": 0.9225230813026428,
619
+ "step": 174
620
+ },
621
+ {
622
+ "epoch": 0.7586206896551724,
623
+ "grad_norm": 0.13637058436870575,
624
+ "learning_rate": 2.3848305566369543e-05,
625
+ "loss": 0.6734743714332581,
626
+ "step": 176
627
+ },
628
+ {
629
+ "epoch": 0.7672413793103449,
630
+ "grad_norm": 0.1864972859621048,
631
+ "learning_rate": 2.3812685625409726e-05,
632
+ "loss": 1.24442458152771,
633
+ "step": 178
634
+ },
635
+ {
636
+ "epoch": 0.7758620689655172,
637
+ "grad_norm": 0.0796503946185112,
638
+ "learning_rate": 2.377655386123088e-05,
639
+ "loss": 1.170575499534607,
640
+ "step": 180
641
+ },
642
+ {
643
+ "epoch": 0.7844827586206896,
644
+ "grad_norm": 0.1396905928850174,
645
+ "learning_rate": 2.3739912111620303e-05,
646
+ "loss": 0.8946050405502319,
647
+ "step": 182
648
+ },
649
+ {
650
+ "epoch": 0.7931034482758621,
651
+ "grad_norm": 0.07272544503211975,
652
+ "learning_rate": 2.3702762240304933e-05,
653
+ "loss": 0.7359169125556946,
654
+ "step": 184
655
+ },
656
+ {
657
+ "epoch": 0.8017241379310345,
658
+ "grad_norm": 0.21276403963565826,
659
+ "learning_rate": 2.366510613685654e-05,
660
+ "loss": 1.377792477607727,
661
+ "step": 186
662
+ },
663
+ {
664
+ "epoch": 0.8103448275862069,
665
+ "grad_norm": 0.6310688853263855,
666
+ "learning_rate": 2.362694571659563e-05,
667
+ "loss": 1.5076112747192383,
668
+ "step": 188
669
+ },
670
+ {
671
+ "epoch": 0.8189655172413793,
672
+ "grad_norm": 0.09929846227169037,
673
+ "learning_rate": 2.3588282920494007e-05,
674
+ "loss": 1.2605912685394287,
675
+ "step": 190
676
+ },
677
+ {
678
+ "epoch": 0.8275862068965517,
679
+ "grad_norm": 0.12678399682044983,
680
+ "learning_rate": 2.354911971507607e-05,
681
+ "loss": 1.2289090156555176,
682
+ "step": 192
683
+ },
684
+ {
685
+ "epoch": 0.8362068965517241,
686
+ "grad_norm": 0.09680190682411194,
687
+ "learning_rate": 2.3509458092318772e-05,
688
+ "loss": 1.2607473134994507,
689
+ "step": 194
690
+ },
691
+ {
692
+ "epoch": 0.8448275862068966,
693
+ "grad_norm": 0.11455738544464111,
694
+ "learning_rate": 2.3469300069550308e-05,
695
+ "loss": 1.2929327487945557,
696
+ "step": 196
697
+ },
698
+ {
699
+ "epoch": 0.853448275862069,
700
+ "grad_norm": 0.22343184053897858,
701
+ "learning_rate": 2.342864768934751e-05,
702
+ "loss": 0.8722022175788879,
703
+ "step": 198
704
+ },
705
+ {
706
+ "epoch": 0.8620689655172413,
707
+ "grad_norm": 0.1278611570596695,
708
+ "learning_rate": 2.338750301943194e-05,
709
+ "loss": 1.184353232383728,
710
+ "step": 200
711
+ },
712
+ {
713
+ "epoch": 0.8706896551724138,
714
+ "grad_norm": 0.1743975579738617,
715
+ "learning_rate": 2.3345868152564734e-05,
716
+ "loss": 1.3033404350280762,
717
+ "step": 202
718
+ },
719
+ {
720
+ "epoch": 0.8793103448275862,
721
+ "grad_norm": 0.09879707545042038,
722
+ "learning_rate": 2.3303745206440156e-05,
723
+ "loss": 0.8301224708557129,
724
+ "step": 204
725
+ },
726
+ {
727
+ "epoch": 0.8879310344827587,
728
+ "grad_norm": 0.07442842423915863,
729
+ "learning_rate": 2.3261136323577873e-05,
730
+ "loss": 1.2247326374053955,
731
+ "step": 206
732
+ },
733
+ {
734
+ "epoch": 0.896551724137931,
735
+ "grad_norm": 0.0767381563782692,
736
+ "learning_rate": 2.3218043671213988e-05,
737
+ "loss": 1.275619626045227,
738
+ "step": 208
739
+ },
740
+ {
741
+ "epoch": 0.9051724137931034,
742
+ "grad_norm": 0.21888962388038635,
743
+ "learning_rate": 2.317446944119081e-05,
744
+ "loss": 0.763530969619751,
745
+ "step": 210
746
+ },
747
+ {
748
+ "epoch": 0.9137931034482759,
749
+ "grad_norm": 0.2872910797595978,
750
+ "learning_rate": 2.313041584984535e-05,
751
+ "loss": 0.8847395181655884,
752
+ "step": 212
753
+ },
754
+ {
755
+ "epoch": 0.9224137931034483,
756
+ "grad_norm": 0.22158221900463104,
757
+ "learning_rate": 2.3085885137896623e-05,
758
+ "loss": 1.139022707939148,
759
+ "step": 214
760
+ },
761
+ {
762
+ "epoch": 0.9310344827586207,
763
+ "grad_norm": 0.11198285222053528,
764
+ "learning_rate": 2.3040879570331648e-05,
765
+ "loss": 1.2234928607940674,
766
+ "step": 216
767
+ },
768
+ {
769
+ "epoch": 0.9396551724137931,
770
+ "grad_norm": 0.09761999547481537,
771
+ "learning_rate": 2.2995401436290252e-05,
772
+ "loss": 0.7140777111053467,
773
+ "step": 218
774
+ },
775
+ {
776
+ "epoch": 0.9482758620689655,
777
+ "grad_norm": 0.22312681376934052,
778
+ "learning_rate": 2.294945304894864e-05,
779
+ "loss": 0.8441833257675171,
780
+ "step": 220
781
+ },
782
+ {
783
+ "epoch": 0.9568965517241379,
784
+ "grad_norm": 0.2743966579437256,
785
+ "learning_rate": 2.2903036745401742e-05,
786
+ "loss": 1.3300198316574097,
787
+ "step": 222
788
+ },
789
+ {
790
+ "epoch": 0.9655172413793104,
791
+ "grad_norm": 0.6734777688980103,
792
+ "learning_rate": 2.2856154886544338e-05,
793
+ "loss": 0.8017496466636658,
794
+ "step": 224
795
+ },
796
+ {
797
+ "epoch": 0.9741379310344828,
798
+ "grad_norm": 0.1572013795375824,
799
+ "learning_rate": 2.280880985695095e-05,
800
+ "loss": 0.8339143991470337,
801
+ "step": 226
802
+ },
803
+ {
804
+ "epoch": 0.9827586206896551,
805
+ "grad_norm": 0.12929460406303406,
806
+ "learning_rate": 2.27610040647546e-05,
807
+ "loss": 1.2216969728469849,
808
+ "step": 228
809
+ },
810
+ {
811
+ "epoch": 0.9913793103448276,
812
+ "grad_norm": 0.24367918074131012,
813
+ "learning_rate": 2.2712739941524287e-05,
814
+ "loss": 0.8863965272903442,
815
+ "step": 230
816
+ },
817
+ {
818
+ "epoch": 1.0,
819
+ "grad_norm": 0.11734236776828766,
820
+ "learning_rate": 2.266401994214133e-05,
821
+ "loss": 0.7733997702598572,
822
+ "step": 232
823
+ },
824
+ {
825
+ "epoch": 1.0086206896551724,
826
+ "grad_norm": 0.3386397063732147,
827
+ "learning_rate": 2.2614846544674486e-05,
828
+ "loss": 0.8990653157234192,
829
+ "step": 234
830
+ },
831
+ {
832
+ "epoch": 1.0172413793103448,
833
+ "grad_norm": 0.20136962831020355,
834
+ "learning_rate": 2.2565222250253925e-05,
835
+ "loss": 0.7269018888473511,
836
+ "step": 236
837
+ },
838
+ {
839
+ "epoch": 1.0258620689655173,
840
+ "grad_norm": 0.27934548258781433,
841
+ "learning_rate": 2.251514958294401e-05,
842
+ "loss": 0.8784797787666321,
843
+ "step": 238
844
+ },
845
+ {
846
+ "epoch": 1.0344827586206897,
847
+ "grad_norm": 0.45081236958503723,
848
+ "learning_rate": 2.246463108961489e-05,
849
+ "loss": 0.5005398988723755,
850
+ "step": 240
851
+ },
852
+ {
853
+ "epoch": 1.043103448275862,
854
+ "grad_norm": 0.10048855096101761,
855
+ "learning_rate": 2.2413669339812997e-05,
856
+ "loss": 1.1350990533828735,
857
+ "step": 242
858
+ },
859
+ {
860
+ "epoch": 1.0517241379310345,
861
+ "grad_norm": 0.21551190316677094,
862
+ "learning_rate": 2.2362266925630318e-05,
863
+ "loss": 0.36204367876052856,
864
+ "step": 244
865
+ },
866
+ {
867
+ "epoch": 1.0603448275862069,
868
+ "grad_norm": 0.19904279708862305,
869
+ "learning_rate": 2.231042646157258e-05,
870
+ "loss": 1.1106600761413574,
871
+ "step": 246
872
+ },
873
+ {
874
+ "epoch": 1.0689655172413792,
875
+ "grad_norm": 0.15923143923282623,
876
+ "learning_rate": 2.2258150584426235e-05,
877
+ "loss": 1.0469825267791748,
878
+ "step": 248
879
+ },
880
+ {
881
+ "epoch": 1.0775862068965518,
882
+ "grad_norm": 0.1504082828760147,
883
+ "learning_rate": 2.220544195312436e-05,
884
+ "loss": 1.0015485286712646,
885
+ "step": 250
886
+ },
887
+ {
888
+ "epoch": 1.0862068965517242,
889
+ "grad_norm": 0.13122083246707916,
890
+ "learning_rate": 2.2152303248611437e-05,
891
+ "loss": 0.36394670605659485,
892
+ "step": 252
893
+ },
894
+ {
895
+ "epoch": 1.0948275862068966,
896
+ "grad_norm": 0.3954187333583832,
897
+ "learning_rate": 2.2098737173706942e-05,
898
+ "loss": 1.0145013332366943,
899
+ "step": 254
900
+ },
901
+ {
902
+ "epoch": 1.103448275862069,
903
+ "grad_norm": 0.24796895682811737,
904
+ "learning_rate": 2.2044746452967906e-05,
905
+ "loss": 0.7123383283615112,
906
+ "step": 256
907
+ },
908
+ {
909
+ "epoch": 1.1120689655172413,
910
+ "grad_norm": 0.653165340423584,
911
+ "learning_rate": 2.1990333832550333e-05,
912
+ "loss": 0.4990229606628418,
913
+ "step": 258
914
+ },
915
+ {
916
+ "epoch": 1.1206896551724137,
917
+ "grad_norm": 0.08383104205131531,
918
+ "learning_rate": 2.1935502080069505e-05,
919
+ "loss": 0.7109271287918091,
920
+ "step": 260
921
+ },
922
+ {
923
+ "epoch": 1.1293103448275863,
924
+ "grad_norm": 0.1300533413887024,
925
+ "learning_rate": 2.1880253984459224e-05,
926
+ "loss": 1.2138782739639282,
927
+ "step": 262
928
+ },
929
+ {
930
+ "epoch": 1.1379310344827587,
931
+ "grad_norm": 0.1580507606267929,
932
+ "learning_rate": 2.1824592355829945e-05,
933
+ "loss": 0.8597405552864075,
934
+ "step": 264
935
+ },
936
+ {
937
+ "epoch": 1.146551724137931,
938
+ "grad_norm": 0.3914850354194641,
939
+ "learning_rate": 2.1768520025325866e-05,
940
+ "loss": 1.1350115537643433,
941
+ "step": 266
942
+ },
943
+ {
944
+ "epoch": 1.1551724137931034,
945
+ "grad_norm": 0.10050661861896515,
946
+ "learning_rate": 2.1712039844980893e-05,
947
+ "loss": 1.1547629833221436,
948
+ "step": 268
949
+ },
950
+ {
951
+ "epoch": 1.1637931034482758,
952
+ "grad_norm": 0.08129733055830002,
953
+ "learning_rate": 2.165515468757362e-05,
954
+ "loss": 1.1632627248764038,
955
+ "step": 270
956
+ },
957
+ {
958
+ "epoch": 1.1724137931034484,
959
+ "grad_norm": 0.10255050659179688,
960
+ "learning_rate": 2.1597867446481155e-05,
961
+ "loss": 1.0813562870025635,
962
+ "step": 272
963
+ },
964
+ {
965
+ "epoch": 1.1810344827586208,
966
+ "grad_norm": 0.30835890769958496,
967
+ "learning_rate": 2.1540181035532004e-05,
968
+ "loss": 0.7284541130065918,
969
+ "step": 274
970
+ },
971
+ {
972
+ "epoch": 1.1896551724137931,
973
+ "grad_norm": 0.08633577823638916,
974
+ "learning_rate": 2.1482098388857834e-05,
975
+ "loss": 0.3468450903892517,
976
+ "step": 276
977
+ },
978
+ {
979
+ "epoch": 1.1982758620689655,
980
+ "grad_norm": 0.09107495844364166,
981
+ "learning_rate": 2.1423622460744242e-05,
982
+ "loss": 1.121955394744873,
983
+ "step": 278
984
+ },
985
+ {
986
+ "epoch": 1.206896551724138,
987
+ "grad_norm": 0.0790572538971901,
988
+ "learning_rate": 2.1364756225480478e-05,
989
+ "loss": 1.0472161769866943,
990
+ "step": 280
991
+ },
992
+ {
993
+ "epoch": 1.2155172413793103,
994
+ "grad_norm": 0.18506139516830444,
995
+ "learning_rate": 2.1305502677208184e-05,
996
+ "loss": 0.5945373773574829,
997
+ "step": 282
998
+ },
999
+ {
1000
+ "epoch": 1.2241379310344827,
1001
+ "grad_norm": 0.11592728644609451,
1002
+ "learning_rate": 2.1245864829769092e-05,
1003
+ "loss": 0.7433868646621704,
1004
+ "step": 284
1005
+ },
1006
+ {
1007
+ "epoch": 1.2327586206896552,
1008
+ "grad_norm": 0.2357645183801651,
1009
+ "learning_rate": 2.1185845716551712e-05,
1010
+ "loss": 0.8308318853378296,
1011
+ "step": 286
1012
+ },
1013
+ {
1014
+ "epoch": 1.2413793103448276,
1015
+ "grad_norm": 0.2785407304763794,
1016
+ "learning_rate": 2.1125448390337072e-05,
1017
+ "loss": 0.3982611894607544,
1018
+ "step": 288
1019
+ },
1020
+ {
1021
+ "epoch": 1.25,
1022
+ "grad_norm": 0.17765875160694122,
1023
+ "learning_rate": 2.106467592314343e-05,
1024
+ "loss": 0.9474261403083801,
1025
+ "step": 290
1026
+ },
1027
+ {
1028
+ "epoch": 1.2586206896551724,
1029
+ "grad_norm": 0.10548825562000275,
1030
+ "learning_rate": 2.1003531406070012e-05,
1031
+ "loss": 0.927026093006134,
1032
+ "step": 292
1033
+ },
1034
+ {
1035
+ "epoch": 1.2672413793103448,
1036
+ "grad_norm": 0.32261741161346436,
1037
+ "learning_rate": 2.09420179491398e-05,
1038
+ "loss": 1.0189390182495117,
1039
+ "step": 294
1040
+ },
1041
+ {
1042
+ "epoch": 1.2758620689655173,
1043
+ "grad_norm": 0.08806228637695312,
1044
+ "learning_rate": 2.0880138681141345e-05,
1045
+ "loss": 0.7990611791610718,
1046
+ "step": 296
1047
+ },
1048
+ {
1049
+ "epoch": 1.2844827586206897,
1050
+ "grad_norm": 0.07184045016765594,
1051
+ "learning_rate": 2.081789674946961e-05,
1052
+ "loss": 1.1012811660766602,
1053
+ "step": 298
1054
+ },
1055
+ {
1056
+ "epoch": 1.293103448275862,
1057
+ "grad_norm": 0.09509000182151794,
1058
+ "learning_rate": 2.075529531996591e-05,
1059
+ "loss": 1.0139261484146118,
1060
+ "step": 300
1061
+ },
1062
+ {
1063
+ "epoch": 1.3017241379310345,
1064
+ "grad_norm": 0.10540059953927994,
1065
+ "learning_rate": 2.0692337576756866e-05,
1066
+ "loss": 0.6810011863708496,
1067
+ "step": 302
1068
+ },
1069
+ {
1070
+ "epoch": 1.3103448275862069,
1071
+ "grad_norm": 0.0546269416809082,
1072
+ "learning_rate": 2.0629026722092453e-05,
1073
+ "loss": 0.41086244583129883,
1074
+ "step": 304
1075
+ },
1076
+ {
1077
+ "epoch": 1.3189655172413794,
1078
+ "grad_norm": 0.08956003934144974,
1079
+ "learning_rate": 2.056536597618312e-05,
1080
+ "loss": 1.3326225280761719,
1081
+ "step": 306
1082
+ },
1083
+ {
1084
+ "epoch": 1.3275862068965516,
1085
+ "grad_norm": 0.06451157480478287,
1086
+ "learning_rate": 2.0501358577036003e-05,
1087
+ "loss": 0.6605536341667175,
1088
+ "step": 308
1089
+ },
1090
+ {
1091
+ "epoch": 1.3362068965517242,
1092
+ "grad_norm": 0.3918032944202423,
1093
+ "learning_rate": 2.043700778029024e-05,
1094
+ "loss": 1.0086909532546997,
1095
+ "step": 310
1096
+ },
1097
+ {
1098
+ "epoch": 1.3448275862068966,
1099
+ "grad_norm": 0.09106232970952988,
1100
+ "learning_rate": 2.0372316859051357e-05,
1101
+ "loss": 0.9816857576370239,
1102
+ "step": 312
1103
+ },
1104
+ {
1105
+ "epoch": 1.353448275862069,
1106
+ "grad_norm": 0.3334841728210449,
1107
+ "learning_rate": 2.0307289103724793e-05,
1108
+ "loss": 1.206939458847046,
1109
+ "step": 314
1110
+ },
1111
+ {
1112
+ "epoch": 1.3620689655172413,
1113
+ "grad_norm": 0.1430651843547821,
1114
+ "learning_rate": 2.0241927821848544e-05,
1115
+ "loss": 1.3034409284591675,
1116
+ "step": 316
1117
+ },
1118
+ {
1119
+ "epoch": 1.3706896551724137,
1120
+ "grad_norm": 0.12443620711565018,
1121
+ "learning_rate": 2.0176236337924933e-05,
1122
+ "loss": 1.0241037607192993,
1123
+ "step": 318
1124
+ },
1125
+ {
1126
+ "epoch": 1.3793103448275863,
1127
+ "grad_norm": 0.15698978304862976,
1128
+ "learning_rate": 2.01102179932515e-05,
1129
+ "loss": 0.6907064318656921,
1130
+ "step": 320
1131
+ },
1132
+ {
1133
+ "epoch": 1.3879310344827587,
1134
+ "grad_norm": 0.10246633738279343,
1135
+ "learning_rate": 2.0043876145751067e-05,
1136
+ "loss": 1.4752711057662964,
1137
+ "step": 322
1138
+ },
1139
+ {
1140
+ "epoch": 1.396551724137931,
1141
+ "grad_norm": 0.11384667456150055,
1142
+ "learning_rate": 1.997721416980093e-05,
1143
+ "loss": 0.4461423456668854,
1144
+ "step": 324
1145
+ },
1146
+ {
1147
+ "epoch": 1.4051724137931034,
1148
+ "grad_norm": 0.9319630265235901,
1149
+ "learning_rate": 1.991023545606124e-05,
1150
+ "loss": 0.42419686913490295,
1151
+ "step": 326
1152
+ },
1153
+ {
1154
+ "epoch": 1.4137931034482758,
1155
+ "grad_norm": 0.1355047971010208,
1156
+ "learning_rate": 1.9842943411302527e-05,
1157
+ "loss": 0.9030317664146423,
1158
+ "step": 328
1159
+ },
1160
+ {
1161
+ "epoch": 1.4224137931034484,
1162
+ "grad_norm": 0.12845957279205322,
1163
+ "learning_rate": 1.9775341458232436e-05,
1164
+ "loss": 0.768023669719696,
1165
+ "step": 330
1166
+ },
1167
+ {
1168
+ "epoch": 1.4310344827586206,
1169
+ "grad_norm": 0.058132827281951904,
1170
+ "learning_rate": 1.9707433035321613e-05,
1171
+ "loss": 0.6390933394432068,
1172
+ "step": 332
1173
+ },
1174
+ {
1175
+ "epoch": 1.4396551724137931,
1176
+ "grad_norm": 0.14224818348884583,
1177
+ "learning_rate": 1.9639221596628836e-05,
1178
+ "loss": 0.6768423914909363,
1179
+ "step": 334
1180
+ },
1181
+ {
1182
+ "epoch": 1.4482758620689655,
1183
+ "grad_norm": 0.30597057938575745,
1184
+ "learning_rate": 1.957071061162533e-05,
1185
+ "loss": 0.710463285446167,
1186
+ "step": 336
1187
+ },
1188
+ {
1189
+ "epoch": 1.456896551724138,
1190
+ "grad_norm": 0.08679758757352829,
1191
+ "learning_rate": 1.9501903565018273e-05,
1192
+ "loss": 1.0271871089935303,
1193
+ "step": 338
1194
+ },
1195
+ {
1196
+ "epoch": 1.4655172413793103,
1197
+ "grad_norm": 0.17182646691799164,
1198
+ "learning_rate": 1.943280395657357e-05,
1199
+ "loss": 1.2887871265411377,
1200
+ "step": 340
1201
+ },
1202
+ {
1203
+ "epoch": 1.4741379310344827,
1204
+ "grad_norm": 0.09934200346469879,
1205
+ "learning_rate": 1.9363415300937846e-05,
1206
+ "loss": 1.1640923023223877,
1207
+ "step": 342
1208
+ },
1209
+ {
1210
+ "epoch": 1.4827586206896552,
1211
+ "grad_norm": 0.114693783223629,
1212
+ "learning_rate": 1.9293741127459665e-05,
1213
+ "loss": 0.9469618201255798,
1214
+ "step": 344
1215
+ },
1216
+ {
1217
+ "epoch": 1.4913793103448276,
1218
+ "grad_norm": 0.1328352689743042,
1219
+ "learning_rate": 1.922378498001003e-05,
1220
+ "loss": 0.45884159207344055,
1221
+ "step": 346
1222
+ },
1223
+ {
1224
+ "epoch": 1.5,
1225
+ "grad_norm": 0.3843913972377777,
1226
+ "learning_rate": 1.9153550416802107e-05,
1227
+ "loss": 0.8928775787353516,
1228
+ "step": 348
1229
+ },
1230
+ {
1231
+ "epoch": 1.5086206896551724,
1232
+ "grad_norm": 0.2543632686138153,
1233
+ "learning_rate": 1.9083041010210273e-05,
1234
+ "loss": 1.0797038078308105,
1235
+ "step": 350
1236
+ },
1237
+ {
1238
+ "epoch": 1.5172413793103448,
1239
+ "grad_norm": 0.1487339287996292,
1240
+ "learning_rate": 1.9012260346588394e-05,
1241
+ "loss": 1.0333884954452515,
1242
+ "step": 352
1243
+ },
1244
+ {
1245
+ "epoch": 1.5258620689655173,
1246
+ "grad_norm": 0.14135690033435822,
1247
+ "learning_rate": 1.89412120260874e-05,
1248
+ "loss": 1.1652461290359497,
1249
+ "step": 354
1250
+ },
1251
+ {
1252
+ "epoch": 1.5344827586206895,
1253
+ "grad_norm": 0.3962077796459198,
1254
+ "learning_rate": 1.8869899662472197e-05,
1255
+ "loss": 0.8967676758766174,
1256
+ "step": 356
1257
+ },
1258
+ {
1259
+ "epoch": 1.543103448275862,
1260
+ "grad_norm": 0.05238835886120796,
1261
+ "learning_rate": 1.8798326882937834e-05,
1262
+ "loss": 0.7729952335357666,
1263
+ "step": 358
1264
+ },
1265
+ {
1266
+ "epoch": 1.5517241379310345,
1267
+ "grad_norm": 0.09740617871284485,
1268
+ "learning_rate": 1.872649732792502e-05,
1269
+ "loss": 1.3566042184829712,
1270
+ "step": 360
1271
+ },
1272
+ {
1273
+ "epoch": 1.5603448275862069,
1274
+ "grad_norm": 0.06796287000179291,
1275
+ "learning_rate": 1.8654414650934968e-05,
1276
+ "loss": 1.0945885181427002,
1277
+ "step": 362
1278
+ },
1279
+ {
1280
+ "epoch": 1.5689655172413794,
1281
+ "grad_norm": 0.08173417299985886,
1282
+ "learning_rate": 1.858208251834354e-05,
1283
+ "loss": 1.1358948945999146,
1284
+ "step": 364
1285
+ },
1286
+ {
1287
+ "epoch": 1.5775862068965516,
1288
+ "grad_norm": 0.13112913072109222,
1289
+ "learning_rate": 1.8509504609214794e-05,
1290
+ "loss": 0.7162285447120667,
1291
+ "step": 366
1292
+ },
1293
+ {
1294
+ "epoch": 1.5862068965517242,
1295
+ "grad_norm": 0.08164367079734802,
1296
+ "learning_rate": 1.8436684615113824e-05,
1297
+ "loss": 1.080881953239441,
1298
+ "step": 368
1299
+ },
1300
+ {
1301
+ "epoch": 1.5948275862068966,
1302
+ "grad_norm": 0.07548147439956665,
1303
+ "learning_rate": 1.8363626239919022e-05,
1304
+ "loss": 1.054266333580017,
1305
+ "step": 370
1306
+ },
1307
+ {
1308
+ "epoch": 1.603448275862069,
1309
+ "grad_norm": 0.3800515830516815,
1310
+ "learning_rate": 1.8290333199633665e-05,
1311
+ "loss": 1.0251119136810303,
1312
+ "step": 372
1313
+ },
1314
+ {
1315
+ "epoch": 1.6120689655172413,
1316
+ "grad_norm": 0.058936282992362976,
1317
+ "learning_rate": 1.821680922219691e-05,
1318
+ "loss": 0.7149249315261841,
1319
+ "step": 374
1320
+ },
1321
+ {
1322
+ "epoch": 1.6206896551724137,
1323
+ "grad_norm": 0.08577758073806763,
1324
+ "learning_rate": 1.8143058047294197e-05,
1325
+ "loss": 1.0818464756011963,
1326
+ "step": 376
1327
+ },
1328
+ {
1329
+ "epoch": 1.6293103448275863,
1330
+ "grad_norm": 15.305747985839844,
1331
+ "learning_rate": 1.8069083426166995e-05,
1332
+ "loss": 1.1972448825836182,
1333
+ "step": 378
1334
+ },
1335
+ {
1336
+ "epoch": 1.6379310344827587,
1337
+ "grad_norm": 0.08515272289514542,
1338
+ "learning_rate": 1.799488912142205e-05,
1339
+ "loss": 0.7313613891601562,
1340
+ "step": 380
1341
+ },
1342
+ {
1343
+ "epoch": 1.646551724137931,
1344
+ "grad_norm": 0.14937828481197357,
1345
+ "learning_rate": 1.7920478906839967e-05,
1346
+ "loss": 1.3920011520385742,
1347
+ "step": 382
1348
+ },
1349
+ {
1350
+ "epoch": 1.6551724137931034,
1351
+ "grad_norm": 0.11451909691095352,
1352
+ "learning_rate": 1.7845856567183278e-05,
1353
+ "loss": 1.0703322887420654,
1354
+ "step": 384
1355
+ },
1356
+ {
1357
+ "epoch": 1.6637931034482758,
1358
+ "grad_norm": 0.11318498849868774,
1359
+ "learning_rate": 1.777102589800394e-05,
1360
+ "loss": 0.9152452349662781,
1361
+ "step": 386
1362
+ },
1363
+ {
1364
+ "epoch": 1.6724137931034484,
1365
+ "grad_norm": 0.11082017421722412,
1366
+ "learning_rate": 1.7695990705450276e-05,
1367
+ "loss": 1.360831618309021,
1368
+ "step": 388
1369
+ },
1370
+ {
1371
+ "epoch": 1.6810344827586206,
1372
+ "grad_norm": 0.6852989792823792,
1373
+ "learning_rate": 1.7620754806073368e-05,
1374
+ "loss": 0.905778706073761,
1375
+ "step": 390
1376
+ },
1377
+ {
1378
+ "epoch": 1.6896551724137931,
1379
+ "grad_norm": 0.14728988707065582,
1380
+ "learning_rate": 1.754532202663297e-05,
1381
+ "loss": 0.43694019317626953,
1382
+ "step": 392
1383
+ },
1384
+ {
1385
+ "epoch": 1.6982758620689655,
1386
+ "grad_norm": 0.48538804054260254,
1387
+ "learning_rate": 1.74696962039028e-05,
1388
+ "loss": 1.0095391273498535,
1389
+ "step": 394
1390
+ },
1391
+ {
1392
+ "epoch": 1.706896551724138,
1393
+ "grad_norm": 0.3051482141017914,
1394
+ "learning_rate": 1.7393881184475476e-05,
1395
+ "loss": 1.0637139081954956,
1396
+ "step": 396
1397
+ },
1398
+ {
1399
+ "epoch": 1.7155172413793105,
1400
+ "grad_norm": 0.1540980041027069,
1401
+ "learning_rate": 1.7317880824566792e-05,
1402
+ "loss": 1.042551040649414,
1403
+ "step": 398
1404
+ },
1405
+ {
1406
+ "epoch": 1.7241379310344827,
1407
+ "grad_norm": 0.16515707969665527,
1408
+ "learning_rate": 1.7241698989819604e-05,
1409
+ "loss": 1.02712082862854,
1410
+ "step": 400
1411
+ },
1412
+ {
1413
+ "epoch": 1.7327586206896552,
1414
+ "grad_norm": 0.13348740339279175,
1415
+ "learning_rate": 1.7165339555107223e-05,
1416
+ "loss": 0.7308106422424316,
1417
+ "step": 402
1418
+ },
1419
+ {
1420
+ "epoch": 1.7413793103448276,
1421
+ "grad_norm": 0.13249894976615906,
1422
+ "learning_rate": 1.7088806404336298e-05,
1423
+ "loss": 0.6191306710243225,
1424
+ "step": 404
1425
+ },
1426
+ {
1427
+ "epoch": 1.75,
1428
+ "grad_norm": 0.26619619131088257,
1429
+ "learning_rate": 1.7012103430249292e-05,
1430
+ "loss": 0.6533270478248596,
1431
+ "step": 406
1432
+ },
1433
+ {
1434
+ "epoch": 1.7586206896551724,
1435
+ "grad_norm": 0.07099182903766632,
1436
+ "learning_rate": 1.6935234534226475e-05,
1437
+ "loss": 0.7554537653923035,
1438
+ "step": 408
1439
+ },
1440
+ {
1441
+ "epoch": 1.7672413793103448,
1442
+ "grad_norm": 0.12172161787748337,
1443
+ "learning_rate": 1.6858203626087468e-05,
1444
+ "loss": 0.8455532789230347,
1445
+ "step": 410
1446
+ },
1447
+ {
1448
+ "epoch": 1.7758620689655173,
1449
+ "grad_norm": 0.11975358426570892,
1450
+ "learning_rate": 1.6781014623892406e-05,
1451
+ "loss": 0.7798787951469421,
1452
+ "step": 412
1453
+ },
1454
+ {
1455
+ "epoch": 1.7844827586206895,
1456
+ "grad_norm": 0.11753620207309723,
1457
+ "learning_rate": 1.6703671453742642e-05,
1458
+ "loss": 0.6053640842437744,
1459
+ "step": 414
1460
+ },
1461
+ {
1462
+ "epoch": 1.793103448275862,
1463
+ "grad_norm": 0.14249694347381592,
1464
+ "learning_rate": 1.6626178049581033e-05,
1465
+ "loss": 0.8004411458969116,
1466
+ "step": 416
1467
+ },
1468
+ {
1469
+ "epoch": 1.8017241379310345,
1470
+ "grad_norm": 0.12212086468935013,
1471
+ "learning_rate": 1.654853835299187e-05,
1472
+ "loss": 0.9836465716362,
1473
+ "step": 418
1474
+ },
1475
+ {
1476
+ "epoch": 1.8103448275862069,
1477
+ "grad_norm": 0.4444316625595093,
1478
+ "learning_rate": 1.6470756313000386e-05,
1479
+ "loss": 1.111708402633667,
1480
+ "step": 420
1481
+ },
1482
+ {
1483
+ "epoch": 1.8189655172413794,
1484
+ "grad_norm": 0.12780721485614777,
1485
+ "learning_rate": 1.6392835885871897e-05,
1486
+ "loss": 1.1723716259002686,
1487
+ "step": 422
1488
+ },
1489
+ {
1490
+ "epoch": 1.8275862068965516,
1491
+ "grad_norm": 0.11766626685857773,
1492
+ "learning_rate": 1.631478103491056e-05,
1493
+ "loss": 1.006577491760254,
1494
+ "step": 424
1495
+ },
1496
+ {
1497
+ "epoch": 1.8362068965517242,
1498
+ "grad_norm": 0.10621563345193863,
1499
+ "learning_rate": 1.623659573025781e-05,
1500
+ "loss": 0.6617158055305481,
1501
+ "step": 426
1502
+ },
1503
+ {
1504
+ "epoch": 1.8448275862068966,
1505
+ "grad_norm": 0.19948941469192505,
1506
+ "learning_rate": 1.61582839486904e-05,
1507
+ "loss": 1.0630967617034912,
1508
+ "step": 428
1509
+ },
1510
+ {
1511
+ "epoch": 1.853448275862069,
1512
+ "grad_norm": 0.08539006859064102,
1513
+ "learning_rate": 1.6079849673418144e-05,
1514
+ "loss": 1.0498086214065552,
1515
+ "step": 430
1516
+ },
1517
+ {
1518
+ "epoch": 1.8620689655172413,
1519
+ "grad_norm": 0.15522508323192596,
1520
+ "learning_rate": 1.600129689388131e-05,
1521
+ "loss": 1.0054386854171753,
1522
+ "step": 432
1523
+ },
1524
+ {
1525
+ "epoch": 1.8706896551724137,
1526
+ "grad_norm": 0.23854467272758484,
1527
+ "learning_rate": 1.5922629605547708e-05,
1528
+ "loss": 1.107947826385498,
1529
+ "step": 434
1530
+ },
1531
+ {
1532
+ "epoch": 1.8793103448275863,
1533
+ "grad_norm": 0.14075548946857452,
1534
+ "learning_rate": 1.5843851809709462e-05,
1535
+ "loss": 0.6936919093132019,
1536
+ "step": 436
1537
+ },
1538
+ {
1539
+ "epoch": 1.8879310344827587,
1540
+ "grad_norm": 0.20101788640022278,
1541
+ "learning_rate": 1.5764967513279498e-05,
1542
+ "loss": 0.9637195467948914,
1543
+ "step": 438
1544
+ },
1545
+ {
1546
+ "epoch": 1.896551724137931,
1547
+ "grad_norm": 0.09223437309265137,
1548
+ "learning_rate": 1.568598072858772e-05,
1549
+ "loss": 0.9367073774337769,
1550
+ "step": 440
1551
+ },
1552
+ {
1553
+ "epoch": 1.9051724137931034,
1554
+ "grad_norm": 0.1555858999490738,
1555
+ "learning_rate": 1.5606895473176952e-05,
1556
+ "loss": 1.1365821361541748,
1557
+ "step": 442
1558
+ },
1559
+ {
1560
+ "epoch": 1.9137931034482758,
1561
+ "grad_norm": 0.20968762040138245,
1562
+ "learning_rate": 1.5527715769598582e-05,
1563
+ "loss": 0.2478710114955902,
1564
+ "step": 444
1565
+ },
1566
+ {
1567
+ "epoch": 1.9224137931034484,
1568
+ "grad_norm": 0.25190311670303345,
1569
+ "learning_rate": 1.544844564520796e-05,
1570
+ "loss": 0.636488676071167,
1571
+ "step": 446
1572
+ },
1573
+ {
1574
+ "epoch": 1.9310344827586206,
1575
+ "grad_norm": 0.09660618007183075,
1576
+ "learning_rate": 1.5369089131959546e-05,
1577
+ "loss": 1.0994303226470947,
1578
+ "step": 448
1579
+ },
1580
+ {
1581
+ "epoch": 1.9396551724137931,
1582
+ "grad_norm": 0.09424697607755661,
1583
+ "learning_rate": 1.5289650266201842e-05,
1584
+ "loss": 1.0804164409637451,
1585
+ "step": 450
1586
+ },
1587
+ {
1588
+ "epoch": 1.9482758620689655,
1589
+ "grad_norm": 0.12286921590566635,
1590
+ "learning_rate": 1.5210133088472096e-05,
1591
+ "loss": 0.9485524296760559,
1592
+ "step": 452
1593
+ },
1594
+ {
1595
+ "epoch": 1.956896551724138,
1596
+ "grad_norm": 0.23227761685848236,
1597
+ "learning_rate": 1.5130541643290755e-05,
1598
+ "loss": 1.0982067584991455,
1599
+ "step": 454
1600
+ },
1601
+ {
1602
+ "epoch": 1.9655172413793105,
1603
+ "grad_norm": 0.1315997689962387,
1604
+ "learning_rate": 1.5050879978955784e-05,
1605
+ "loss": 1.069095492362976,
1606
+ "step": 456
1607
+ },
1608
+ {
1609
+ "epoch": 1.9741379310344827,
1610
+ "grad_norm": 0.36187416315078735,
1611
+ "learning_rate": 1.4971152147336735e-05,
1612
+ "loss": 0.29495418071746826,
1613
+ "step": 458
1614
+ },
1615
+ {
1616
+ "epoch": 1.9827586206896552,
1617
+ "grad_norm": 0.28215736150741577,
1618
+ "learning_rate": 1.4891362203668646e-05,
1619
+ "loss": 1.0610284805297852,
1620
+ "step": 460
1621
+ },
1622
+ {
1623
+ "epoch": 1.9913793103448276,
1624
+ "grad_norm": 0.2604519724845886,
1625
+ "learning_rate": 1.4811514206345812e-05,
1626
+ "loss": 0.6210886240005493,
1627
+ "step": 462
1628
+ },
1629
+ {
1630
+ "epoch": 2.0,
1631
+ "grad_norm": 0.8820472359657288,
1632
+ "learning_rate": 1.4731612216715315e-05,
1633
+ "loss": 0.796215295791626,
1634
+ "step": 464
1635
+ },
1636
+ {
1637
+ "epoch": 2.0086206896551726,
1638
+ "grad_norm": 0.090216726064682,
1639
+ "learning_rate": 1.4651660298870496e-05,
1640
+ "loss": 0.4970940947532654,
1641
+ "step": 466
1642
+ },
1643
+ {
1644
+ "epoch": 2.0172413793103448,
1645
+ "grad_norm": 0.17966338992118835,
1646
+ "learning_rate": 1.4571662519444208e-05,
1647
+ "loss": 0.7648203372955322,
1648
+ "step": 468
1649
+ },
1650
+ {
1651
+ "epoch": 2.0258620689655173,
1652
+ "grad_norm": 0.8345552682876587,
1653
+ "learning_rate": 1.4491622947401976e-05,
1654
+ "loss": 0.7012611627578735,
1655
+ "step": 470
1656
+ },
1657
+ {
1658
+ "epoch": 2.0344827586206895,
1659
+ "grad_norm": 0.14064906537532806,
1660
+ "learning_rate": 1.4411545653835056e-05,
1661
+ "loss": 0.5726131796836853,
1662
+ "step": 472
1663
+ },
1664
+ {
1665
+ "epoch": 2.043103448275862,
1666
+ "grad_norm": 0.13400284945964813,
1667
+ "learning_rate": 1.433143471175335e-05,
1668
+ "loss": 0.5506486296653748,
1669
+ "step": 474
1670
+ },
1671
+ {
1672
+ "epoch": 2.0517241379310347,
1673
+ "grad_norm": 0.10555726289749146,
1674
+ "learning_rate": 1.4251294195878231e-05,
1675
+ "loss": 0.5927596688270569,
1676
+ "step": 476
1677
+ },
1678
+ {
1679
+ "epoch": 2.060344827586207,
1680
+ "grad_norm": 0.10835747420787811,
1681
+ "learning_rate": 1.4171128182435312e-05,
1682
+ "loss": 0.4717426598072052,
1683
+ "step": 478
1684
+ },
1685
+ {
1686
+ "epoch": 2.0689655172413794,
1687
+ "grad_norm": 0.14888739585876465,
1688
+ "learning_rate": 1.409094074894709e-05,
1689
+ "loss": 0.8343894481658936,
1690
+ "step": 480
1691
+ },
1692
+ {
1693
+ "epoch": 2.0775862068965516,
1694
+ "grad_norm": 0.13630028069019318,
1695
+ "learning_rate": 1.4010735974025565e-05,
1696
+ "loss": 0.5780265927314758,
1697
+ "step": 482
1698
+ },
1699
+ {
1700
+ "epoch": 2.086206896551724,
1701
+ "grad_norm": 0.1016228199005127,
1702
+ "learning_rate": 1.3930517937164783e-05,
1703
+ "loss": 0.628856897354126,
1704
+ "step": 484
1705
+ },
1706
+ {
1707
+ "epoch": 2.0948275862068964,
1708
+ "grad_norm": 0.09641385078430176,
1709
+ "learning_rate": 1.3850290718533334e-05,
1710
+ "loss": 0.6124461889266968,
1711
+ "step": 486
1712
+ },
1713
+ {
1714
+ "epoch": 2.103448275862069,
1715
+ "grad_norm": 0.15994173288345337,
1716
+ "learning_rate": 1.3770058398766825e-05,
1717
+ "loss": 0.7476667761802673,
1718
+ "step": 488
1719
+ },
1720
+ {
1721
+ "epoch": 2.1120689655172415,
1722
+ "grad_norm": 0.2043793946504593,
1723
+ "learning_rate": 1.3689825058760336e-05,
1724
+ "loss": 0.8483129739761353,
1725
+ "step": 490
1726
+ },
1727
+ {
1728
+ "epoch": 2.1206896551724137,
1729
+ "grad_norm": 0.26761284470558167,
1730
+ "learning_rate": 1.3609594779460827e-05,
1731
+ "loss": 0.8416441082954407,
1732
+ "step": 492
1733
+ },
1734
+ {
1735
+ "epoch": 2.1293103448275863,
1736
+ "grad_norm": 0.1123756691813469,
1737
+ "learning_rate": 1.3529371641659582e-05,
1738
+ "loss": 0.8671878576278687,
1739
+ "step": 494
1740
+ },
1741
+ {
1742
+ "epoch": 2.1379310344827585,
1743
+ "grad_norm": 0.3551185429096222,
1744
+ "learning_rate": 1.344915972578465e-05,
1745
+ "loss": 0.919392466545105,
1746
+ "step": 496
1747
+ },
1748
+ {
1749
+ "epoch": 2.146551724137931,
1750
+ "grad_norm": 0.27395322918891907,
1751
+ "learning_rate": 1.3368963111693284e-05,
1752
+ "loss": 0.3866020143032074,
1753
+ "step": 498
1754
+ },
1755
+ {
1756
+ "epoch": 2.1551724137931036,
1757
+ "grad_norm": 0.16738073527812958,
1758
+ "learning_rate": 1.3288785878464447e-05,
1759
+ "loss": 0.4366337060928345,
1760
+ "step": 500
1761
+ },
1762
+ {
1763
+ "epoch": 2.163793103448276,
1764
+ "grad_norm": 0.12588895857334137,
1765
+ "learning_rate": 1.320863210419131e-05,
1766
+ "loss": 0.4676997661590576,
1767
+ "step": 502
1768
+ },
1769
+ {
1770
+ "epoch": 2.1724137931034484,
1771
+ "grad_norm": 0.08290023356676102,
1772
+ "learning_rate": 1.3128505865773855e-05,
1773
+ "loss": 0.5133532285690308,
1774
+ "step": 504
1775
+ },
1776
+ {
1777
+ "epoch": 2.1810344827586206,
1778
+ "grad_norm": 0.30187880992889404,
1779
+ "learning_rate": 1.3048411238711478e-05,
1780
+ "loss": 0.8324005603790283,
1781
+ "step": 506
1782
+ },
1783
+ {
1784
+ "epoch": 2.189655172413793,
1785
+ "grad_norm": 0.12202301621437073,
1786
+ "learning_rate": 1.296835229689573e-05,
1787
+ "loss": 0.49498435854911804,
1788
+ "step": 508
1789
+ },
1790
+ {
1791
+ "epoch": 2.1982758620689653,
1792
+ "grad_norm": 0.11802731454372406,
1793
+ "learning_rate": 1.2888333112403065e-05,
1794
+ "loss": 0.863714337348938,
1795
+ "step": 510
1796
+ },
1797
+ {
1798
+ "epoch": 2.206896551724138,
1799
+ "grad_norm": 0.3642047941684723,
1800
+ "learning_rate": 1.2808357755287764e-05,
1801
+ "loss": 0.4537322223186493,
1802
+ "step": 512
1803
+ },
1804
+ {
1805
+ "epoch": 2.2155172413793105,
1806
+ "grad_norm": 0.14233991503715515,
1807
+ "learning_rate": 1.2728430293374871e-05,
1808
+ "loss": 0.5666446685791016,
1809
+ "step": 514
1810
+ },
1811
+ {
1812
+ "epoch": 2.2241379310344827,
1813
+ "grad_norm": 0.11399707943201065,
1814
+ "learning_rate": 1.2648554792053342e-05,
1815
+ "loss": 0.44022682309150696,
1816
+ "step": 516
1817
+ },
1818
+ {
1819
+ "epoch": 2.2327586206896552,
1820
+ "grad_norm": 0.083638034760952,
1821
+ "learning_rate": 1.25687353140692e-05,
1822
+ "loss": 0.0981849655508995,
1823
+ "step": 518
1824
+ },
1825
+ {
1826
+ "epoch": 2.2413793103448274,
1827
+ "grad_norm": 0.7049289345741272,
1828
+ "learning_rate": 1.2488975919318963e-05,
1829
+ "loss": 0.15166987478733063,
1830
+ "step": 520
1831
+ },
1832
+ {
1833
+ "epoch": 2.25,
1834
+ "grad_norm": 0.09605469554662704,
1835
+ "learning_rate": 1.2409280664643078e-05,
1836
+ "loss": 0.8936961889266968,
1837
+ "step": 522
1838
+ },
1839
+ {
1840
+ "epoch": 2.2586206896551726,
1841
+ "grad_norm": 0.09633728861808777,
1842
+ "learning_rate": 1.2329653603619629e-05,
1843
+ "loss": 0.6037598848342896,
1844
+ "step": 524
1845
+ },
1846
+ {
1847
+ "epoch": 2.2672413793103448,
1848
+ "grad_norm": 0.2239246815443039,
1849
+ "learning_rate": 1.2250098786358108e-05,
1850
+ "loss": 0.5711255669593811,
1851
+ "step": 526
1852
+ },
1853
+ {
1854
+ "epoch": 2.2758620689655173,
1855
+ "grad_norm": 0.0958482176065445,
1856
+ "learning_rate": 1.2170620259293458e-05,
1857
+ "loss": 0.5425680875778198,
1858
+ "step": 528
1859
+ },
1860
+ {
1861
+ "epoch": 2.2844827586206895,
1862
+ "grad_norm": 0.21729297935962677,
1863
+ "learning_rate": 1.2091222064980228e-05,
1864
+ "loss": 0.5258100628852844,
1865
+ "step": 530
1866
+ },
1867
+ {
1868
+ "epoch": 2.293103448275862,
1869
+ "grad_norm": 0.10278700292110443,
1870
+ "learning_rate": 1.2011908241886966e-05,
1871
+ "loss": 0.5525709390640259,
1872
+ "step": 532
1873
+ },
1874
+ {
1875
+ "epoch": 2.3017241379310347,
1876
+ "grad_norm": 0.13626424968242645,
1877
+ "learning_rate": 1.1932682824190799e-05,
1878
+ "loss": 0.6124492883682251,
1879
+ "step": 534
1880
+ },
1881
+ {
1882
+ "epoch": 2.310344827586207,
1883
+ "grad_norm": 0.08790895342826843,
1884
+ "learning_rate": 1.1853549841572268e-05,
1885
+ "loss": 0.5387366414070129,
1886
+ "step": 536
1887
+ },
1888
+ {
1889
+ "epoch": 2.3189655172413794,
1890
+ "grad_norm": 0.07563458383083344,
1891
+ "learning_rate": 1.177451331901031e-05,
1892
+ "loss": 0.32567426562309265,
1893
+ "step": 538
1894
+ },
1895
+ {
1896
+ "epoch": 2.3275862068965516,
1897
+ "grad_norm": 0.14534932374954224,
1898
+ "learning_rate": 1.169557727657761e-05,
1899
+ "loss": 0.46518173813819885,
1900
+ "step": 540
1901
+ },
1902
+ {
1903
+ "epoch": 2.336206896551724,
1904
+ "grad_norm": 0.16097325086593628,
1905
+ "learning_rate": 1.1616745729236045e-05,
1906
+ "loss": 0.774487316608429,
1907
+ "step": 542
1908
+ },
1909
+ {
1910
+ "epoch": 2.344827586206897,
1911
+ "grad_norm": 0.22597487270832062,
1912
+ "learning_rate": 1.1538022686632536e-05,
1913
+ "loss": 0.8165640830993652,
1914
+ "step": 544
1915
+ },
1916
+ {
1917
+ "epoch": 2.353448275862069,
1918
+ "grad_norm": 0.0724526047706604,
1919
+ "learning_rate": 1.145941215289505e-05,
1920
+ "loss": 0.19026756286621094,
1921
+ "step": 546
1922
+ },
1923
+ {
1924
+ "epoch": 2.3620689655172415,
1925
+ "grad_norm": 0.21904027462005615,
1926
+ "learning_rate": 1.1380918126428994e-05,
1927
+ "loss": 0.8604028820991516,
1928
+ "step": 548
1929
+ },
1930
+ {
1931
+ "epoch": 2.3706896551724137,
1932
+ "grad_norm": 0.2525452971458435,
1933
+ "learning_rate": 1.1302544599713775e-05,
1934
+ "loss": 0.4935520589351654,
1935
+ "step": 550
1936
+ },
1937
+ {
1938
+ "epoch": 2.3793103448275863,
1939
+ "grad_norm": 0.14142468571662903,
1940
+ "learning_rate": 1.122429555909979e-05,
1941
+ "loss": 0.7264618277549744,
1942
+ "step": 552
1943
+ },
1944
+ {
1945
+ "epoch": 2.3879310344827585,
1946
+ "grad_norm": 0.09017884731292725,
1947
+ "learning_rate": 1.114617498460562e-05,
1948
+ "loss": 0.6168904900550842,
1949
+ "step": 554
1950
+ },
1951
+ {
1952
+ "epoch": 2.396551724137931,
1953
+ "grad_norm": 0.10239855945110321,
1954
+ "learning_rate": 1.1068186849715635e-05,
1955
+ "loss": 0.6429441571235657,
1956
+ "step": 556
1957
+ },
1958
+ {
1959
+ "epoch": 2.405172413793103,
1960
+ "grad_norm": 0.14052097499370575,
1961
+ "learning_rate": 1.099033512117783e-05,
1962
+ "loss": 0.4293447434902191,
1963
+ "step": 558
1964
+ },
1965
+ {
1966
+ "epoch": 2.413793103448276,
1967
+ "grad_norm": 0.18142321705818176,
1968
+ "learning_rate": 1.091262375880214e-05,
1969
+ "loss": 0.5241134762763977,
1970
+ "step": 560
1971
+ },
1972
+ {
1973
+ "epoch": 2.4224137931034484,
1974
+ "grad_norm": 0.24674156308174133,
1975
+ "learning_rate": 1.0835056715258956e-05,
1976
+ "loss": 0.6372413039207458,
1977
+ "step": 562
1978
+ },
1979
+ {
1980
+ "epoch": 2.4310344827586206,
1981
+ "grad_norm": 0.16510127484798431,
1982
+ "learning_rate": 1.0757637935878126e-05,
1983
+ "loss": 0.7959206104278564,
1984
+ "step": 564
1985
+ },
1986
+ {
1987
+ "epoch": 2.439655172413793,
1988
+ "grad_norm": 0.4282341003417969,
1989
+ "learning_rate": 1.0680371358448266e-05,
1990
+ "loss": 0.6432863473892212,
1991
+ "step": 566
1992
+ },
1993
+ {
1994
+ "epoch": 2.4482758620689653,
1995
+ "grad_norm": 0.25148487091064453,
1996
+ "learning_rate": 1.0603260913016466e-05,
1997
+ "loss": 0.31055551767349243,
1998
+ "step": 568
1999
+ },
2000
+ {
2001
+ "epoch": 2.456896551724138,
2002
+ "grad_norm": 0.15222759544849396,
2003
+ "learning_rate": 1.05263105216884e-05,
2004
+ "loss": 0.9433165788650513,
2005
+ "step": 570
2006
+ },
2007
+ {
2008
+ "epoch": 2.4655172413793105,
2009
+ "grad_norm": 0.46396973729133606,
2010
+ "learning_rate": 1.044952409842884e-05,
2011
+ "loss": 0.5256112217903137,
2012
+ "step": 572
2013
+ },
2014
+ {
2015
+ "epoch": 2.4741379310344827,
2016
+ "grad_norm": 0.11589831113815308,
2017
+ "learning_rate": 1.0372905548862554e-05,
2018
+ "loss": 0.5775177478790283,
2019
+ "step": 574
2020
+ },
2021
+ {
2022
+ "epoch": 2.4827586206896552,
2023
+ "grad_norm": 0.12475285679101944,
2024
+ "learning_rate": 1.0296458770075702e-05,
2025
+ "loss": 0.8348941802978516,
2026
+ "step": 576
2027
+ },
2028
+ {
2029
+ "epoch": 2.4913793103448274,
2030
+ "grad_norm": 0.08721303939819336,
2031
+ "learning_rate": 1.0220187650417542e-05,
2032
+ "loss": 0.48024982213974,
2033
+ "step": 578
2034
+ },
2035
+ {
2036
+ "epoch": 2.5,
2037
+ "grad_norm": 0.09861183911561966,
2038
+ "learning_rate": 1.0144096069302734e-05,
2039
+ "loss": 0.8509948253631592,
2040
+ "step": 580
2041
+ },
2042
+ {
2043
+ "epoch": 2.5086206896551726,
2044
+ "grad_norm": 0.11032187938690186,
2045
+ "learning_rate": 1.0068187897013957e-05,
2046
+ "loss": 0.8317974209785461,
2047
+ "step": 582
2048
+ },
2049
+ {
2050
+ "epoch": 2.5172413793103448,
2051
+ "grad_norm": 0.12203815579414368,
2052
+ "learning_rate": 9.992466994505098e-06,
2053
+ "loss": 0.6833478808403015,
2054
+ "step": 584
2055
+ },
2056
+ {
2057
+ "epoch": 2.5258620689655173,
2058
+ "grad_norm": 0.1454346776008606,
2059
+ "learning_rate": 9.916937213204835e-06,
2060
+ "loss": 0.6194900870323181,
2061
+ "step": 586
2062
+ },
2063
+ {
2064
+ "epoch": 2.5344827586206895,
2065
+ "grad_norm": 0.10960187017917633,
2066
+ "learning_rate": 9.841602394820772e-06,
2067
+ "loss": 0.8086000680923462,
2068
+ "step": 588
2069
+ },
2070
+ {
2071
+ "epoch": 2.543103448275862,
2072
+ "grad_norm": 0.43562886118888855,
2073
+ "learning_rate": 9.766466371144e-06,
2074
+ "loss": 0.8427749872207642,
2075
+ "step": 590
2076
+ },
2077
+ {
2078
+ "epoch": 2.5517241379310347,
2079
+ "grad_norm": 0.4390433430671692,
2080
+ "learning_rate": 9.691532963854247e-06,
2081
+ "loss": 0.7539307475090027,
2082
+ "step": 592
2083
+ },
2084
+ {
2085
+ "epoch": 2.560344827586207,
2086
+ "grad_norm": 0.08623161911964417,
2087
+ "learning_rate": 9.616805984325434e-06,
2088
+ "loss": 0.4550768733024597,
2089
+ "step": 594
2090
+ },
2091
+ {
2092
+ "epoch": 2.5689655172413794,
2093
+ "grad_norm": 0.07664068043231964,
2094
+ "learning_rate": 9.542289233431875e-06,
2095
+ "loss": 0.4484248459339142,
2096
+ "step": 596
2097
+ },
2098
+ {
2099
+ "epoch": 2.5775862068965516,
2100
+ "grad_norm": 0.12998774647712708,
2101
+ "learning_rate": 9.467986501354905e-06,
2102
+ "loss": 0.612811267375946,
2103
+ "step": 598
2104
+ },
2105
+ {
2106
+ "epoch": 2.586206896551724,
2107
+ "grad_norm": 0.20723579823970795,
2108
+ "learning_rate": 9.393901567390133e-06,
2109
+ "loss": 0.7580547332763672,
2110
+ "step": 600
2111
+ },
2112
+ {
2113
+ "epoch": 2.594827586206897,
2114
+ "grad_norm": 0.050058234483003616,
2115
+ "learning_rate": 9.320038199755181e-06,
2116
+ "loss": 0.5622333288192749,
2117
+ "step": 602
2118
+ },
2119
+ {
2120
+ "epoch": 2.603448275862069,
2121
+ "grad_norm": 0.2059715837240219,
2122
+ "learning_rate": 9.246400155398052e-06,
2123
+ "loss": 0.681129515171051,
2124
+ "step": 604
2125
+ },
2126
+ {
2127
+ "epoch": 2.612068965517241,
2128
+ "grad_norm": 0.1687193214893341,
2129
+ "learning_rate": 9.17299117980601e-06,
2130
+ "loss": 0.5278960466384888,
2131
+ "step": 606
2132
+ },
2133
+ {
2134
+ "epoch": 2.6206896551724137,
2135
+ "grad_norm": 0.0910777598619461,
2136
+ "learning_rate": 9.09981500681509e-06,
2137
+ "loss": 0.8580256104469299,
2138
+ "step": 608
2139
+ },
2140
+ {
2141
+ "epoch": 2.6293103448275863,
2142
+ "grad_norm": 0.17509189248085022,
2143
+ "learning_rate": 9.02687535842018e-06,
2144
+ "loss": 0.5199075937271118,
2145
+ "step": 610
2146
+ },
2147
+ {
2148
+ "epoch": 2.637931034482759,
2149
+ "grad_norm": 0.1662158966064453,
2150
+ "learning_rate": 8.9541759445857e-06,
2151
+ "loss": 0.8993173837661743,
2152
+ "step": 612
2153
+ },
2154
+ {
2155
+ "epoch": 2.646551724137931,
2156
+ "grad_norm": 0.12328854948282242,
2157
+ "learning_rate": 8.881720463056895e-06,
2158
+ "loss": 0.7683141827583313,
2159
+ "step": 614
2160
+ },
2161
+ {
2162
+ "epoch": 2.655172413793103,
2163
+ "grad_norm": 0.1306183785200119,
2164
+ "learning_rate": 8.809512599171781e-06,
2165
+ "loss": 0.5064551830291748,
2166
+ "step": 616
2167
+ },
2168
+ {
2169
+ "epoch": 2.663793103448276,
2170
+ "grad_norm": 0.09901191294193268,
2171
+ "learning_rate": 8.737556025673664e-06,
2172
+ "loss": 0.8363734483718872,
2173
+ "step": 618
2174
+ },
2175
+ {
2176
+ "epoch": 2.6724137931034484,
2177
+ "grad_norm": 0.02962668426334858,
2178
+ "learning_rate": 8.665854402524356e-06,
2179
+ "loss": 0.4550926387310028,
2180
+ "step": 620
2181
+ },
2182
+ {
2183
+ "epoch": 2.6810344827586206,
2184
+ "grad_norm": 0.14856787025928497,
2185
+ "learning_rate": 8.594411376718009e-06,
2186
+ "loss": 0.7523635029792786,
2187
+ "step": 622
2188
+ },
2189
+ {
2190
+ "epoch": 2.689655172413793,
2191
+ "grad_norm": 0.09993023425340652,
2192
+ "learning_rate": 8.523230582095595e-06,
2193
+ "loss": 0.4919755458831787,
2194
+ "step": 624
2195
+ },
2196
+ {
2197
+ "epoch": 2.6982758620689653,
2198
+ "grad_norm": 0.09833712875843048,
2199
+ "learning_rate": 8.452315639160121e-06,
2200
+ "loss": 0.7990663051605225,
2201
+ "step": 626
2202
+ },
2203
+ {
2204
+ "epoch": 2.706896551724138,
2205
+ "grad_norm": 0.1576949805021286,
2206
+ "learning_rate": 8.381670154892429e-06,
2207
+ "loss": 0.5231303572654724,
2208
+ "step": 628
2209
+ },
2210
+ {
2211
+ "epoch": 2.7155172413793105,
2212
+ "grad_norm": 0.07004235684871674,
2213
+ "learning_rate": 8.311297722567769e-06,
2214
+ "loss": 0.9687949419021606,
2215
+ "step": 630
2216
+ },
2217
+ {
2218
+ "epoch": 2.7241379310344827,
2219
+ "grad_norm": 0.7200844883918762,
2220
+ "learning_rate": 8.241201921573018e-06,
2221
+ "loss": 0.42503124475479126,
2222
+ "step": 632
2223
+ },
2224
+ {
2225
+ "epoch": 2.7327586206896552,
2226
+ "grad_norm": 0.1017894297838211,
2227
+ "learning_rate": 8.171386317224612e-06,
2228
+ "loss": 0.7973089814186096,
2229
+ "step": 634
2230
+ },
2231
+ {
2232
+ "epoch": 2.7413793103448274,
2233
+ "grad_norm": 0.11164805293083191,
2234
+ "learning_rate": 8.101854460587225e-06,
2235
+ "loss": 0.5349655747413635,
2236
+ "step": 636
2237
+ },
2238
+ {
2239
+ "epoch": 2.75,
2240
+ "grad_norm": 0.15979333221912384,
2241
+ "learning_rate": 8.032609888293111e-06,
2242
+ "loss": 0.8247781991958618,
2243
+ "step": 638
2244
+ },
2245
+ {
2246
+ "epoch": 2.7586206896551726,
2247
+ "grad_norm": 0.12597469985485077,
2248
+ "learning_rate": 7.963656122362269e-06,
2249
+ "loss": 0.4262469708919525,
2250
+ "step": 640
2251
+ },
2252
+ {
2253
+ "epoch": 2.7672413793103448,
2254
+ "grad_norm": 0.33888915181159973,
2255
+ "learning_rate": 7.894996670023252e-06,
2256
+ "loss": 0.6027237176895142,
2257
+ "step": 642
2258
+ },
2259
+ {
2260
+ "epoch": 2.7758620689655173,
2261
+ "grad_norm": 0.10856196284294128,
2262
+ "learning_rate": 7.82663502353482e-06,
2263
+ "loss": 0.8435148000717163,
2264
+ "step": 644
2265
+ },
2266
+ {
2267
+ "epoch": 2.7844827586206895,
2268
+ "grad_norm": 0.11784857511520386,
2269
+ "learning_rate": 7.758574660008268e-06,
2270
+ "loss": 0.7197084426879883,
2271
+ "step": 646
2272
+ },
2273
+ {
2274
+ "epoch": 2.793103448275862,
2275
+ "grad_norm": 0.09773028641939163,
2276
+ "learning_rate": 7.690819041230616e-06,
2277
+ "loss": 0.5047876834869385,
2278
+ "step": 648
2279
+ },
2280
+ {
2281
+ "epoch": 2.8017241379310347,
2282
+ "grad_norm": 0.07827575504779816,
2283
+ "learning_rate": 7.623371613488475e-06,
2284
+ "loss": 0.6992777585983276,
2285
+ "step": 650
2286
+ },
2287
+ {
2288
+ "epoch": 2.810344827586207,
2289
+ "grad_norm": 0.28669506311416626,
2290
+ "learning_rate": 7.55623580739282e-06,
2291
+ "loss": 0.7797609567642212,
2292
+ "step": 652
2293
+ },
2294
+ {
2295
+ "epoch": 2.8189655172413794,
2296
+ "grad_norm": 0.30535227060317993,
2297
+ "learning_rate": 7.489415037704437e-06,
2298
+ "loss": 0.7613651156425476,
2299
+ "step": 654
2300
+ },
2301
+ {
2302
+ "epoch": 2.8275862068965516,
2303
+ "grad_norm": 0.09482012689113617,
2304
+ "learning_rate": 7.422912703160281e-06,
2305
+ "loss": 0.5962029695510864,
2306
+ "step": 656
2307
+ },
2308
+ {
2309
+ "epoch": 2.836206896551724,
2310
+ "grad_norm": 0.08960616588592529,
2311
+ "learning_rate": 7.3567321863005754e-06,
2312
+ "loss": 0.8467994928359985,
2313
+ "step": 658
2314
+ },
2315
+ {
2316
+ "epoch": 2.844827586206897,
2317
+ "grad_norm": 0.07423507422208786,
2318
+ "learning_rate": 7.290876853296791e-06,
2319
+ "loss": 0.2079772651195526,
2320
+ "step": 660
2321
+ },
2322
+ {
2323
+ "epoch": 2.853448275862069,
2324
+ "grad_norm": 0.8393888473510742,
2325
+ "learning_rate": 7.225350053780395e-06,
2326
+ "loss": 0.3967612087726593,
2327
+ "step": 662
2328
+ },
2329
+ {
2330
+ "epoch": 2.862068965517241,
2331
+ "grad_norm": 0.17007187008857727,
2332
+ "learning_rate": 7.1601551206725165e-06,
2333
+ "loss": 0.5651652216911316,
2334
+ "step": 664
2335
+ },
2336
+ {
2337
+ "epoch": 2.8706896551724137,
2338
+ "grad_norm": 0.09160315245389938,
2339
+ "learning_rate": 7.0952953700143915e-06,
2340
+ "loss": 0.4929698705673218,
2341
+ "step": 666
2342
+ },
2343
+ {
2344
+ "epoch": 2.8793103448275863,
2345
+ "grad_norm": 0.5120726823806763,
2346
+ "learning_rate": 7.0307741007987234e-06,
2347
+ "loss": 0.3763951063156128,
2348
+ "step": 668
2349
+ },
2350
+ {
2351
+ "epoch": 2.887931034482759,
2352
+ "grad_norm": 0.14259645342826843,
2353
+ "learning_rate": 6.9665945948018505e-06,
2354
+ "loss": 0.8635995388031006,
2355
+ "step": 670
2356
+ },
2357
+ {
2358
+ "epoch": 2.896551724137931,
2359
+ "grad_norm": 0.14974215626716614,
2360
+ "learning_rate": 6.902760116416865e-06,
2361
+ "loss": 0.5167340636253357,
2362
+ "step": 672
2363
+ },
2364
+ {
2365
+ "epoch": 2.905172413793103,
2366
+ "grad_norm": 0.1967867761850357,
2367
+ "learning_rate": 6.83927391248754e-06,
2368
+ "loss": 0.7079530358314514,
2369
+ "step": 674
2370
+ },
2371
+ {
2372
+ "epoch": 2.913793103448276,
2373
+ "grad_norm": 0.7743811011314392,
2374
+ "learning_rate": 6.776139212143207e-06,
2375
+ "loss": 0.7067210674285889,
2376
+ "step": 676
2377
+ },
2378
+ {
2379
+ "epoch": 2.9224137931034484,
2380
+ "grad_norm": 0.06761736422777176,
2381
+ "learning_rate": 6.713359226634489e-06,
2382
+ "loss": 0.8161290287971497,
2383
+ "step": 678
2384
+ },
2385
+ {
2386
+ "epoch": 2.9310344827586206,
2387
+ "grad_norm": 0.08846219629049301,
2388
+ "learning_rate": 6.650937149169996e-06,
2389
+ "loss": 0.6068652868270874,
2390
+ "step": 680
2391
+ },
2392
+ {
2393
+ "epoch": 2.939655172413793,
2394
+ "grad_norm": 0.04591520130634308,
2395
+ "learning_rate": 6.588876154753869e-06,
2396
+ "loss": 0.2772270143032074,
2397
+ "step": 682
2398
+ },
2399
+ {
2400
+ "epoch": 2.9482758620689653,
2401
+ "grad_norm": 0.07999559491872787,
2402
+ "learning_rate": 6.527179400024325e-06,
2403
+ "loss": 0.3592681288719177,
2404
+ "step": 684
2405
+ },
2406
+ {
2407
+ "epoch": 2.956896551724138,
2408
+ "grad_norm": 0.10797300934791565,
2409
+ "learning_rate": 6.46585002309307e-06,
2410
+ "loss": 0.7297719120979309,
2411
+ "step": 686
2412
+ },
2413
+ {
2414
+ "epoch": 2.9655172413793105,
2415
+ "grad_norm": 0.06590516865253448,
2416
+ "learning_rate": 6.4048911433857075e-06,
2417
+ "loss": 0.7315416932106018,
2418
+ "step": 688
2419
+ },
2420
+ {
2421
+ "epoch": 2.9741379310344827,
2422
+ "grad_norm": 0.11769416928291321,
2423
+ "learning_rate": 6.344305861483049e-06,
2424
+ "loss": 0.9271728992462158,
2425
+ "step": 690
2426
+ },
2427
+ {
2428
+ "epoch": 2.9827586206896552,
2429
+ "grad_norm": 0.08777745068073273,
2430
+ "learning_rate": 6.284097258963437e-06,
2431
+ "loss": 0.9229543209075928,
2432
+ "step": 692
2433
+ },
2434
+ {
2435
+ "epoch": 2.9913793103448274,
2436
+ "grad_norm": 0.21901434659957886,
2437
+ "learning_rate": 6.224268398245974e-06,
2438
+ "loss": 0.503404974937439,
2439
+ "step": 694
2440
+ },
2441
+ {
2442
+ "epoch": 3.0,
2443
+ "grad_norm": 0.10449380427598953,
2444
+ "learning_rate": 6.164822322434786e-06,
2445
+ "loss": 0.9257153272628784,
2446
+ "step": 696
2447
+ },
2448
+ {
2449
+ "epoch": 3.0086206896551726,
2450
+ "grad_norm": 0.09770919382572174,
2451
+ "learning_rate": 6.105762055164216e-06,
2452
+ "loss": 0.33168870210647583,
2453
+ "step": 698
2454
+ },
2455
+ {
2456
+ "epoch": 3.0172413793103448,
2457
+ "grad_norm": 0.08789953589439392,
2458
+ "learning_rate": 6.047090600445046e-06,
2459
+ "loss": 0.5617142915725708,
2460
+ "step": 700
2461
+ },
2462
+ {
2463
+ "epoch": 3.0258620689655173,
2464
+ "grad_norm": 0.20638461410999298,
2465
+ "learning_rate": 5.988810942511703e-06,
2466
+ "loss": 0.3617658317089081,
2467
+ "step": 702
2468
+ },
2469
+ {
2470
+ "epoch": 3.0344827586206895,
2471
+ "grad_norm": 0.1723642200231552,
2472
+ "learning_rate": 5.930926045670452e-06,
2473
+ "loss": 0.4505208134651184,
2474
+ "step": 704
2475
+ },
2476
+ {
2477
+ "epoch": 3.043103448275862,
2478
+ "grad_norm": 0.015494012273848057,
2479
+ "learning_rate": 5.873438854148648e-06,
2480
+ "loss": 0.07196473330259323,
2481
+ "step": 706
2482
+ },
2483
+ {
2484
+ "epoch": 3.0517241379310347,
2485
+ "grad_norm": 0.3174591660499573,
2486
+ "learning_rate": 5.816352291944968e-06,
2487
+ "loss": 0.5367481112480164,
2488
+ "step": 708
2489
+ },
2490
+ {
2491
+ "epoch": 3.060344827586207,
2492
+ "grad_norm": 0.04589797183871269,
2493
+ "learning_rate": 5.759669262680674e-06,
2494
+ "loss": 0.13180424273014069,
2495
+ "step": 710
2496
+ },
2497
+ {
2498
+ "epoch": 3.0689655172413794,
2499
+ "grad_norm": 0.1757347136735916,
2500
+ "learning_rate": 5.703392649451958e-06,
2501
+ "loss": 0.20174063742160797,
2502
+ "step": 712
2503
+ },
2504
+ {
2505
+ "epoch": 3.0775862068965516,
2506
+ "grad_norm": 0.2314152717590332,
2507
+ "learning_rate": 5.647525314683258e-06,
2508
+ "loss": 0.7795681357383728,
2509
+ "step": 714
2510
+ },
2511
+ {
2512
+ "epoch": 3.086206896551724,
2513
+ "grad_norm": 0.12419137358665466,
2514
+ "learning_rate": 5.592070099981703e-06,
2515
+ "loss": 0.23963159322738647,
2516
+ "step": 716
2517
+ },
2518
+ {
2519
+ "epoch": 3.0948275862068964,
2520
+ "grad_norm": 0.08404622972011566,
2521
+ "learning_rate": 5.53702982599254e-06,
2522
+ "loss": 0.5431758761405945,
2523
+ "step": 718
2524
+ },
2525
+ {
2526
+ "epoch": 3.103448275862069,
2527
+ "grad_norm": 0.055214498192071915,
2528
+ "learning_rate": 5.4824072922557125e-06,
2529
+ "loss": 0.14490705728530884,
2530
+ "step": 720
2531
+ },
2532
+ {
2533
+ "epoch": 3.1120689655172415,
2534
+ "grad_norm": 0.09599906206130981,
2535
+ "learning_rate": 5.428205277063415e-06,
2536
+ "loss": 0.2814076840877533,
2537
+ "step": 722
2538
+ },
2539
+ {
2540
+ "epoch": 3.1206896551724137,
2541
+ "grad_norm": 0.1080830916762352,
2542
+ "learning_rate": 5.374426537318827e-06,
2543
+ "loss": 0.4774016737937927,
2544
+ "step": 724
2545
+ },
2546
+ {
2547
+ "epoch": 3.1293103448275863,
2548
+ "grad_norm": 0.06993446499109268,
2549
+ "learning_rate": 5.321073808395846e-06,
2550
+ "loss": 0.5318699479103088,
2551
+ "step": 726
2552
+ },
2553
+ {
2554
+ "epoch": 3.1379310344827585,
2555
+ "grad_norm": 0.2722335755825043,
2556
+ "learning_rate": 5.268149803999993e-06,
2557
+ "loss": 0.5016626119613647,
2558
+ "step": 728
2559
+ },
2560
+ {
2561
+ "epoch": 3.146551724137931,
2562
+ "grad_norm": 0.08205164968967438,
2563
+ "learning_rate": 5.215657216030353e-06,
2564
+ "loss": 0.34347403049468994,
2565
+ "step": 730
2566
+ },
2567
+ {
2568
+ "epoch": 3.1551724137931036,
2569
+ "grad_norm": 0.08843845874071121,
2570
+ "learning_rate": 5.163598714442685e-06,
2571
+ "loss": 0.25022614002227783,
2572
+ "step": 732
2573
+ },
2574
+ {
2575
+ "epoch": 3.163793103448276,
2576
+ "grad_norm": 0.12089242041110992,
2577
+ "learning_rate": 5.111976947113585e-06,
2578
+ "loss": 0.5590460300445557,
2579
+ "step": 734
2580
+ },
2581
+ {
2582
+ "epoch": 3.1724137931034484,
2583
+ "grad_norm": 0.02552367001771927,
2584
+ "learning_rate": 5.0607945397058415e-06,
2585
+ "loss": 0.27467963099479675,
2586
+ "step": 736
2587
+ },
2588
+ {
2589
+ "epoch": 3.1810344827586206,
2590
+ "grad_norm": 0.6171451807022095,
2591
+ "learning_rate": 5.010054095534857e-06,
2592
+ "loss": 0.5068318843841553,
2593
+ "step": 738
2594
+ },
2595
+ {
2596
+ "epoch": 3.189655172413793,
2597
+ "grad_norm": 0.09212607145309448,
2598
+ "learning_rate": 4.959758195436251e-06,
2599
+ "loss": 0.10915378481149673,
2600
+ "step": 740
2601
+ },
2602
+ {
2603
+ "epoch": 3.1982758620689653,
2604
+ "grad_norm": 0.03245019167661667,
2605
+ "learning_rate": 4.9099093976345804e-06,
2606
+ "loss": 0.26452872157096863,
2607
+ "step": 742
2608
+ },
2609
+ {
2610
+ "epoch": 3.206896551724138,
2611
+ "grad_norm": 0.09578347951173782,
2612
+ "learning_rate": 4.860510237613227e-06,
2613
+ "loss": 0.41886910796165466,
2614
+ "step": 744
2615
+ },
2616
+ {
2617
+ "epoch": 3.2155172413793105,
2618
+ "grad_norm": 0.1211688295006752,
2619
+ "learning_rate": 4.811563227985424e-06,
2620
+ "loss": 0.3879053592681885,
2621
+ "step": 746
2622
+ },
2623
+ {
2624
+ "epoch": 3.2241379310344827,
2625
+ "grad_norm": 0.09857846051454544,
2626
+ "learning_rate": 4.763070858366472e-06,
2627
+ "loss": 0.5333488583564758,
2628
+ "step": 748
2629
+ },
2630
+ {
2631
+ "epoch": 3.2327586206896552,
2632
+ "grad_norm": 0.10673126578330994,
2633
+ "learning_rate": 4.7150355952470814e-06,
2634
+ "loss": 0.5044701099395752,
2635
+ "step": 750
2636
+ },
2637
+ {
2638
+ "epoch": 3.2413793103448274,
2639
+ "grad_norm": 0.10853103548288345,
2640
+ "learning_rate": 4.667459881867956e-06,
2641
+ "loss": 0.29571160674095154,
2642
+ "step": 752
2643
+ },
2644
+ {
2645
+ "epoch": 3.25,
2646
+ "grad_norm": 0.13593336939811707,
2647
+ "learning_rate": 4.6203461380954775e-06,
2648
+ "loss": 0.567728579044342,
2649
+ "step": 754
2650
+ },
2651
+ {
2652
+ "epoch": 3.2586206896551726,
2653
+ "grad_norm": 0.2100203037261963,
2654
+ "learning_rate": 4.5736967602986656e-06,
2655
+ "loss": 0.3926270306110382,
2656
+ "step": 756
2657
+ },
2658
+ {
2659
+ "epoch": 3.2672413793103448,
2660
+ "grad_norm": 0.10009382665157318,
2661
+ "learning_rate": 4.527514121227256e-06,
2662
+ "loss": 0.46212252974510193,
2663
+ "step": 758
2664
+ },
2665
+ {
2666
+ "epoch": 3.2758620689655173,
2667
+ "grad_norm": 0.03222054988145828,
2668
+ "learning_rate": 4.481800569891035e-06,
2669
+ "loss": 0.326346218585968,
2670
+ "step": 760
2671
+ },
2672
+ {
2673
+ "epoch": 3.2844827586206895,
2674
+ "grad_norm": 0.18550357222557068,
2675
+ "learning_rate": 4.4365584314403535e-06,
2676
+ "loss": 0.42745885252952576,
2677
+ "step": 762
2678
+ },
2679
+ {
2680
+ "epoch": 3.293103448275862,
2681
+ "grad_norm": 0.44906678795814514,
2682
+ "learning_rate": 4.391790007047859e-06,
2683
+ "loss": 0.47123435139656067,
2684
+ "step": 764
2685
+ },
2686
+ {
2687
+ "epoch": 3.3017241379310347,
2688
+ "grad_norm": 0.04999677091836929,
2689
+ "learning_rate": 4.347497573791454e-06,
2690
+ "loss": 0.28395605087280273,
2691
+ "step": 766
2692
+ },
2693
+ {
2694
+ "epoch": 3.310344827586207,
2695
+ "grad_norm": 0.15627415478229523,
2696
+ "learning_rate": 4.303683384538477e-06,
2697
+ "loss": 0.43049338459968567,
2698
+ "step": 768
2699
+ },
2700
+ {
2701
+ "epoch": 3.3189655172413794,
2702
+ "grad_norm": 0.18941667675971985,
2703
+ "learning_rate": 4.260349667831108e-06,
2704
+ "loss": 0.14107981324195862,
2705
+ "step": 770
2706
+ },
2707
+ {
2708
+ "epoch": 3.3275862068965516,
2709
+ "grad_norm": 0.11692267656326294,
2710
+ "learning_rate": 4.217498627773029e-06,
2711
+ "loss": 0.3092440068721771,
2712
+ "step": 772
2713
+ },
2714
+ {
2715
+ "epoch": 3.336206896551724,
2716
+ "grad_norm": 0.2257245033979416,
2717
+ "learning_rate": 4.175132443917291e-06,
2718
+ "loss": 0.5438483357429504,
2719
+ "step": 774
2720
+ },
2721
+ {
2722
+ "epoch": 3.344827586206897,
2723
+ "grad_norm": 0.10574281215667725,
2724
+ "learning_rate": 4.133253271155489e-06,
2725
+ "loss": 0.5660426616668701,
2726
+ "step": 776
2727
+ },
2728
+ {
2729
+ "epoch": 3.353448275862069,
2730
+ "grad_norm": 0.09007226675748825,
2731
+ "learning_rate": 4.091863239608131e-06,
2732
+ "loss": 0.522563636302948,
2733
+ "step": 778
2734
+ },
2735
+ {
2736
+ "epoch": 3.3620689655172415,
2737
+ "grad_norm": 0.2889229357242584,
2738
+ "learning_rate": 4.050964454516295e-06,
2739
+ "loss": 0.3490191698074341,
2740
+ "step": 780
2741
+ },
2742
+ {
2743
+ "epoch": 3.3706896551724137,
2744
+ "grad_norm": 0.11218395084142685,
2745
+ "learning_rate": 4.010558996134562e-06,
2746
+ "loss": 0.3295826017856598,
2747
+ "step": 782
2748
+ },
2749
+ {
2750
+ "epoch": 3.3793103448275863,
2751
+ "grad_norm": 0.08473268151283264,
2752
+ "learning_rate": 3.970648919625199e-06,
2753
+ "loss": 0.5116720199584961,
2754
+ "step": 784
2755
+ },
2756
+ {
2757
+ "epoch": 3.3879310344827585,
2758
+ "grad_norm": 0.10021858662366867,
2759
+ "learning_rate": 3.9312362549536195e-06,
2760
+ "loss": 0.3242761194705963,
2761
+ "step": 786
2762
+ },
2763
+ {
2764
+ "epoch": 3.396551724137931,
2765
+ "grad_norm": 0.09062958508729935,
2766
+ "learning_rate": 3.892323006785148e-06,
2767
+ "loss": 0.49483513832092285,
2768
+ "step": 788
2769
+ },
2770
+ {
2771
+ "epoch": 3.405172413793103,
2772
+ "grad_norm": 0.09515772014856339,
2773
+ "learning_rate": 3.853911154383037e-06,
2774
+ "loss": 0.4984126091003418,
2775
+ "step": 790
2776
+ },
2777
+ {
2778
+ "epoch": 3.413793103448276,
2779
+ "grad_norm": 0.0715399831533432,
2780
+ "learning_rate": 3.8160026515078165e-06,
2781
+ "loss": 0.2216302752494812,
2782
+ "step": 792
2783
+ },
2784
+ {
2785
+ "epoch": 3.4224137931034484,
2786
+ "grad_norm": 0.125926673412323,
2787
+ "learning_rate": 3.7785994263178978e-06,
2788
+ "loss": 0.24825817346572876,
2789
+ "step": 794
2790
+ },
2791
+ {
2792
+ "epoch": 3.4310344827586206,
2793
+ "grad_norm": 0.10221999138593674,
2794
+ "learning_rate": 3.741703381271519e-06,
2795
+ "loss": 0.6700829863548279,
2796
+ "step": 796
2797
+ },
2798
+ {
2799
+ "epoch": 3.439655172413793,
2800
+ "grad_norm": 0.18892526626586914,
2801
+ "learning_rate": 3.7053163930299625e-06,
2802
+ "loss": 0.3420178294181824,
2803
+ "step": 798
2804
+ },
2805
+ {
2806
+ "epoch": 3.4482758620689653,
2807
+ "grad_norm": 0.13281142711639404,
2808
+ "learning_rate": 3.6694403123621143e-06,
2809
+ "loss": 0.5975306034088135,
2810
+ "step": 800
2811
+ },
2812
+ {
2813
+ "epoch": 3.456896551724138,
2814
+ "grad_norm": 0.0972677543759346,
2815
+ "learning_rate": 3.634076964050321e-06,
2816
+ "loss": 0.4603136479854584,
2817
+ "step": 802
2818
+ },
2819
+ {
2820
+ "epoch": 3.4655172413793105,
2821
+ "grad_norm": 0.11561470478773117,
2822
+ "learning_rate": 3.5992281467975773e-06,
2823
+ "loss": 0.4385143518447876,
2824
+ "step": 804
2825
+ },
2826
+ {
2827
+ "epoch": 3.4741379310344827,
2828
+ "grad_norm": 0.13390004634857178,
2829
+ "learning_rate": 3.5648956331360366e-06,
2830
+ "loss": 0.4603370428085327,
2831
+ "step": 806
2832
+ },
2833
+ {
2834
+ "epoch": 3.4827586206896552,
2835
+ "grad_norm": 0.09613792598247528,
2836
+ "learning_rate": 3.5310811693368584e-06,
2837
+ "loss": 0.5407328605651855,
2838
+ "step": 808
2839
+ },
2840
+ {
2841
+ "epoch": 3.4913793103448274,
2842
+ "grad_norm": 0.02905660681426525,
2843
+ "learning_rate": 3.4977864753213727e-06,
2844
+ "loss": 0.14333094656467438,
2845
+ "step": 810
2846
+ },
2847
+ {
2848
+ "epoch": 3.5,
2849
+ "grad_norm": 0.40168488025665283,
2850
+ "learning_rate": 3.4650132445736226e-06,
2851
+ "loss": 0.3744501769542694,
2852
+ "step": 812
2853
+ },
2854
+ {
2855
+ "epoch": 3.5086206896551726,
2856
+ "grad_norm": 0.18611806631088257,
2857
+ "learning_rate": 3.4327631440542018e-06,
2858
+ "loss": 0.49726125597953796,
2859
+ "step": 814
2860
+ },
2861
+ {
2862
+ "epoch": 3.5172413793103448,
2863
+ "grad_norm": 0.1080632358789444,
2864
+ "learning_rate": 3.401037814115489e-06,
2865
+ "loss": 0.441265344619751,
2866
+ "step": 816
2867
+ },
2868
+ {
2869
+ "epoch": 3.5258620689655173,
2870
+ "grad_norm": 0.17538908123970032,
2871
+ "learning_rate": 3.3698388684181987e-06,
2872
+ "loss": 0.5781593322753906,
2873
+ "step": 818
2874
+ },
2875
+ {
2876
+ "epoch": 3.5344827586206895,
2877
+ "grad_norm": 0.08419691771268845,
2878
+ "learning_rate": 3.339167893849313e-06,
2879
+ "loss": 0.5013993382453918,
2880
+ "step": 820
2881
+ },
2882
+ {
2883
+ "epoch": 3.543103448275862,
2884
+ "grad_norm": 0.23287096619606018,
2885
+ "learning_rate": 3.3090264504413618e-06,
2886
+ "loss": 0.2739977240562439,
2887
+ "step": 822
2888
+ },
2889
+ {
2890
+ "epoch": 3.5517241379310347,
2891
+ "grad_norm": 0.08283835649490356,
2892
+ "learning_rate": 3.279416071293083e-06,
2893
+ "loss": 0.33216506242752075,
2894
+ "step": 824
2895
+ },
2896
+ {
2897
+ "epoch": 3.560344827586207,
2898
+ "grad_norm": 0.03011569194495678,
2899
+ "learning_rate": 3.2503382624914287e-06,
2900
+ "loss": 0.08031827956438065,
2901
+ "step": 826
2902
+ },
2903
+ {
2904
+ "epoch": 3.5689655172413794,
2905
+ "grad_norm": 0.15765611827373505,
2906
+ "learning_rate": 3.2217945030349785e-06,
2907
+ "loss": 0.13542333245277405,
2908
+ "step": 828
2909
+ },
2910
+ {
2911
+ "epoch": 3.5775862068965516,
2912
+ "grad_norm": 0.1197817251086235,
2913
+ "learning_rate": 3.1937862447586945e-06,
2914
+ "loss": 0.3155873119831085,
2915
+ "step": 830
2916
+ },
2917
+ {
2918
+ "epoch": 3.586206896551724,
2919
+ "grad_norm": 0.12214954197406769,
2920
+ "learning_rate": 3.1663149122600946e-06,
2921
+ "loss": 0.32241955399513245,
2922
+ "step": 832
2923
+ },
2924
+ {
2925
+ "epoch": 3.594827586206897,
2926
+ "grad_norm": 0.11514101922512054,
2927
+ "learning_rate": 3.1393819028267712e-06,
2928
+ "loss": 0.5564541220664978,
2929
+ "step": 834
2930
+ },
2931
+ {
2932
+ "epoch": 3.603448275862069,
2933
+ "grad_norm": 0.12636107206344604,
2934
+ "learning_rate": 3.112988586365343e-06,
2935
+ "loss": 0.3382505476474762,
2936
+ "step": 836
2937
+ },
2938
+ {
2939
+ "epoch": 3.612068965517241,
2940
+ "grad_norm": 0.13530541956424713,
2941
+ "learning_rate": 3.0871363053317565e-06,
2942
+ "loss": 0.4953022599220276,
2943
+ "step": 838
2944
+ },
2945
+ {
2946
+ "epoch": 3.6206896551724137,
2947
+ "grad_norm": 0.21053758263587952,
2948
+ "learning_rate": 3.0618263746630195e-06,
2949
+ "loss": 0.12510839104652405,
2950
+ "step": 840
2951
+ },
2952
+ {
2953
+ "epoch": 3.6293103448275863,
2954
+ "grad_norm": 0.20605450868606567,
2955
+ "learning_rate": 3.0370600817103054e-06,
2956
+ "loss": 0.12819121778011322,
2957
+ "step": 842
2958
+ },
2959
+ {
2960
+ "epoch": 3.637931034482759,
2961
+ "grad_norm": 0.08630102127790451,
2962
+ "learning_rate": 3.0128386861734907e-06,
2963
+ "loss": 0.34562334418296814,
2964
+ "step": 844
2965
+ },
2966
+ {
2967
+ "epoch": 3.646551724137931,
2968
+ "grad_norm": 0.21599891781806946,
2969
+ "learning_rate": 2.9891634200370615e-06,
2970
+ "loss": 0.2538186311721802,
2971
+ "step": 846
2972
+ },
2973
+ {
2974
+ "epoch": 3.655172413793103,
2975
+ "grad_norm": 0.08544554561376572,
2976
+ "learning_rate": 2.9660354875074694e-06,
2977
+ "loss": 0.09849962592124939,
2978
+ "step": 848
2979
+ },
2980
+ {
2981
+ "epoch": 3.663793103448276,
2982
+ "grad_norm": 0.058089643716812134,
2983
+ "learning_rate": 2.9434560649518708e-06,
2984
+ "loss": 0.31217214465141296,
2985
+ "step": 850
2986
+ },
2987
+ {
2988
+ "epoch": 3.6724137931034484,
2989
+ "grad_norm": 0.144460067152977,
2990
+ "learning_rate": 2.9214263008382943e-06,
2991
+ "loss": 0.5859465003013611,
2992
+ "step": 852
2993
+ },
2994
+ {
2995
+ "epoch": 3.6810344827586206,
2996
+ "grad_norm": 0.11717334389686584,
2997
+ "learning_rate": 2.8999473156772322e-06,
2998
+ "loss": 0.32843440771102905,
2999
+ "step": 854
3000
+ },
3001
+ {
3002
+ "epoch": 3.689655172413793,
3003
+ "grad_norm": 0.42045170068740845,
3004
+ "learning_rate": 2.8790202019646327e-06,
3005
+ "loss": 0.525081992149353,
3006
+ "step": 856
3007
+ },
3008
+ {
3009
+ "epoch": 3.6982758620689653,
3010
+ "grad_norm": 0.10836631059646606,
3011
+ "learning_rate": 2.858646024126349e-06,
3012
+ "loss": 0.5662536025047302,
3013
+ "step": 858
3014
+ },
3015
+ {
3016
+ "epoch": 3.706896551724138,
3017
+ "grad_norm": 0.044400930404663086,
3018
+ "learning_rate": 2.8388258184639895e-06,
3019
+ "loss": 0.26983019709587097,
3020
+ "step": 860
3021
+ },
3022
+ {
3023
+ "epoch": 3.7155172413793105,
3024
+ "grad_norm": 0.20253317058086395,
3025
+ "learning_rate": 2.8195605931022044e-06,
3026
+ "loss": 0.325879842042923,
3027
+ "step": 862
3028
+ },
3029
+ {
3030
+ "epoch": 3.7241379310344827,
3031
+ "grad_norm": 0.021246735006570816,
3032
+ "learning_rate": 2.8008513279374227e-06,
3033
+ "loss": 0.26106154918670654,
3034
+ "step": 864
3035
+ },
3036
+ {
3037
+ "epoch": 3.7327586206896552,
3038
+ "grad_norm": 0.2902374267578125,
3039
+ "learning_rate": 2.7826989745879937e-06,
3040
+ "loss": 0.0877346470952034,
3041
+ "step": 866
3042
+ },
3043
+ {
3044
+ "epoch": 3.7413793103448274,
3045
+ "grad_norm": 0.16706323623657227,
3046
+ "learning_rate": 2.7651044563457983e-06,
3047
+ "loss": 0.32902559638023376,
3048
+ "step": 868
3049
+ },
3050
+ {
3051
+ "epoch": 3.75,
3052
+ "grad_norm": 0.06072617322206497,
3053
+ "learning_rate": 2.748068668129279e-06,
3054
+ "loss": 0.40092846751213074,
3055
+ "step": 870
3056
+ },
3057
+ {
3058
+ "epoch": 3.7586206896551726,
3059
+ "grad_norm": 0.02342204377055168,
3060
+ "learning_rate": 2.7315924764379292e-06,
3061
+ "loss": 0.12595315277576447,
3062
+ "step": 872
3063
+ },
3064
+ {
3065
+ "epoch": 3.7672413793103448,
3066
+ "grad_norm": 0.019321495667099953,
3067
+ "learning_rate": 2.715676719308208e-06,
3068
+ "loss": 0.30979275703430176,
3069
+ "step": 874
3070
+ },
3071
+ {
3072
+ "epoch": 3.7758620689655173,
3073
+ "grad_norm": 0.10496604442596436,
3074
+ "learning_rate": 2.7003222062709295e-06,
3075
+ "loss": 0.5150949358940125,
3076
+ "step": 876
3077
+ },
3078
+ {
3079
+ "epoch": 3.7844827586206895,
3080
+ "grad_norm": 0.2970009446144104,
3081
+ "learning_rate": 2.6855297183100716e-06,
3082
+ "loss": 0.41893500089645386,
3083
+ "step": 878
3084
+ },
3085
+ {
3086
+ "epoch": 3.793103448275862,
3087
+ "grad_norm": 0.1417495757341385,
3088
+ "learning_rate": 2.671300007823068e-06,
3089
+ "loss": 0.28298234939575195,
3090
+ "step": 880
3091
+ },
3092
+ {
3093
+ "epoch": 3.8017241379310347,
3094
+ "grad_norm": 0.13228200376033783,
3095
+ "learning_rate": 2.657633798582525e-06,
3096
+ "loss": 0.29594770073890686,
3097
+ "step": 882
3098
+ },
3099
+ {
3100
+ "epoch": 3.810344827586207,
3101
+ "grad_norm": 0.23043398559093475,
3102
+ "learning_rate": 2.6445317856994187e-06,
3103
+ "loss": 0.5765542984008789,
3104
+ "step": 884
3105
+ },
3106
+ {
3107
+ "epoch": 3.8189655172413794,
3108
+ "grad_norm": 0.19636346399784088,
3109
+ "learning_rate": 2.6319946355877313e-06,
3110
+ "loss": 0.18173623085021973,
3111
+ "step": 886
3112
+ },
3113
+ {
3114
+ "epoch": 3.8275862068965516,
3115
+ "grad_norm": 0.09358557313680649,
3116
+ "learning_rate": 2.62002298593056e-06,
3117
+ "loss": 0.3384171724319458,
3118
+ "step": 888
3119
+ },
3120
+ {
3121
+ "epoch": 3.836206896551724,
3122
+ "grad_norm": 0.0909893587231636,
3123
+ "learning_rate": 2.6086174456476774e-06,
3124
+ "loss": 0.10825614631175995,
3125
+ "step": 890
3126
+ },
3127
+ {
3128
+ "epoch": 3.844827586206897,
3129
+ "grad_norm": 0.15050019323825836,
3130
+ "learning_rate": 2.5977785948645674e-06,
3131
+ "loss": 0.36479508876800537,
3132
+ "step": 892
3133
+ },
3134
+ {
3135
+ "epoch": 3.853448275862069,
3136
+ "grad_norm": 0.07940158993005753,
3137
+ "learning_rate": 2.58750698488291e-06,
3138
+ "loss": 0.4843851923942566,
3139
+ "step": 894
3140
+ },
3141
+ {
3142
+ "epoch": 3.862068965517241,
3143
+ "grad_norm": 0.08099526911973953,
3144
+ "learning_rate": 2.577803138152543e-06,
3145
+ "loss": 0.34008628129959106,
3146
+ "step": 896
3147
+ },
3148
+ {
3149
+ "epoch": 3.8706896551724137,
3150
+ "grad_norm": 0.14477792382240295,
3151
+ "learning_rate": 2.56866754824489e-06,
3152
+ "loss": 0.2734866142272949,
3153
+ "step": 898
3154
+ },
3155
+ {
3156
+ "epoch": 3.8793103448275863,
3157
+ "grad_norm": 0.08499010652303696,
3158
+ "learning_rate": 2.560100679827858e-06,
3159
+ "loss": 0.5301221013069153,
3160
+ "step": 900
3161
+ },
3162
+ {
3163
+ "epoch": 3.887931034482759,
3164
+ "grad_norm": 0.1628270447254181,
3165
+ "learning_rate": 2.5521029686421927e-06,
3166
+ "loss": 0.40625,
3167
+ "step": 902
3168
+ },
3169
+ {
3170
+ "epoch": 3.896551724137931,
3171
+ "grad_norm": 0.1032799556851387,
3172
+ "learning_rate": 2.544674821479325e-06,
3173
+ "loss": 0.412536084651947,
3174
+ "step": 904
3175
+ },
3176
+ {
3177
+ "epoch": 3.905172413793103,
3178
+ "grad_norm": 0.19824664294719696,
3179
+ "learning_rate": 2.5378166161606777e-06,
3180
+ "loss": 0.5061193108558655,
3181
+ "step": 906
3182
+ },
3183
+ {
3184
+ "epoch": 3.913793103448276,
3185
+ "grad_norm": 0.2031039297580719,
3186
+ "learning_rate": 2.5315287015184486e-06,
3187
+ "loss": 0.2978528141975403,
3188
+ "step": 908
3189
+ },
3190
+ {
3191
+ "epoch": 3.9224137931034484,
3192
+ "grad_norm": 0.1463300883769989,
3193
+ "learning_rate": 2.525811397377861e-06,
3194
+ "loss": 0.21743205189704895,
3195
+ "step": 910
3196
+ },
3197
+ {
3198
+ "epoch": 3.9310344827586206,
3199
+ "grad_norm": 0.10597418993711472,
3200
+ "learning_rate": 2.520664994540907e-06,
3201
+ "loss": 0.2807198464870453,
3202
+ "step": 912
3203
+ },
3204
+ {
3205
+ "epoch": 3.939655172413793,
3206
+ "grad_norm": 0.11543630063533783,
3207
+ "learning_rate": 2.5160897547715503e-06,
3208
+ "loss": 0.537182629108429,
3209
+ "step": 914
3210
+ },
3211
+ {
3212
+ "epoch": 3.9482758620689653,
3213
+ "grad_norm": 0.12250090390443802,
3214
+ "learning_rate": 2.5120859107824107e-06,
3215
+ "loss": 0.4790320098400116,
3216
+ "step": 916
3217
+ },
3218
+ {
3219
+ "epoch": 3.956896551724138,
3220
+ "grad_norm": 0.1139577180147171,
3221
+ "learning_rate": 2.508653666222932e-06,
3222
+ "loss": 0.3746393024921417,
3223
+ "step": 918
3224
+ },
3225
+ {
3226
+ "epoch": 3.9655172413793105,
3227
+ "grad_norm": 0.34079810976982117,
3228
+ "learning_rate": 2.5057931956690186e-06,
3229
+ "loss": 0.3896881341934204,
3230
+ "step": 920
3231
+ },
3232
+ {
3233
+ "epoch": 3.9741379310344827,
3234
+ "grad_norm": 0.1102144867181778,
3235
+ "learning_rate": 2.5035046446141617e-06,
3236
+ "loss": 0.40220990777015686,
3237
+ "step": 922
3238
+ },
3239
+ {
3240
+ "epoch": 3.9827586206896552,
3241
+ "grad_norm": 0.11032397300004959,
3242
+ "learning_rate": 2.501788129462033e-06,
3243
+ "loss": 0.5496131777763367,
3244
+ "step": 924
3245
+ },
3246
+ {
3247
+ "epoch": 3.9913793103448274,
3248
+ "grad_norm": 0.12315075099468231,
3249
+ "learning_rate": 2.5006437375205682e-06,
3250
+ "loss": 0.5806067585945129,
3251
+ "step": 926
3252
+ },
3253
+ {
3254
+ "epoch": 4.0,
3255
+ "grad_norm": 0.18088358640670776,
3256
+ "learning_rate": 2.5000715269975277e-06,
3257
+ "loss": 0.39202290773391724,
3258
+ "step": 928
3259
+ },
3260
+ {
3261
+ "epoch": 4.0,
3262
+ "step": 928,
3263
+ "total_flos": 3.61934117404672e+18,
3264
+ "train_loss": 0.7849958032722874,
3265
+ "train_runtime": 32566.8601,
3266
+ "train_samples_per_second": 1.71,
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 sha256:0a82a97d70c47618b48dfe1aecc6edc824c0eca26128a76c0f1bf4d2e15c7d62
3
+ size 5649
training_loss.png ADDED