Hugging Face's logo

furproxy
/
9b-62

Image-Text-to-Text
Transformers
Safetensors
qwen3_5
llama-factory
full
Generated from Trainer
conversational
Model card Files
xet
 Community
furproxy commited on
Commit
f3ea8f9
·
verified ·
1 Parent(s): 714ffc1

Upload folder using huggingface_hub

Browse files
Files changed (16) hide show
  1. .gitattributes +1 -0
  2. .ipynb_checkpoints/README-checkpoint.md +63 -0
  3. README.md +63 -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 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - family_to_adamw_lr = {
39
+ "language": _fallback(getattr(training_args, "language_adamw_lr", 2e-5), language_lr),
40
+ "vision": _fallback(getattr(training_args, "vision_adamw_lr", 3e-6), vision_lr),
41
+ "merger": _fallback(getattr(training_args, "merger_adamw_lr", 8e-5), merger_lr),
42
+ }
43
+ - train_batch_size: 3
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - gradient_accumulation_steps: 20
48
+ - total_train_batch_size: 60
49
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
50
+ - lr_scheduler_type: cosine_with_min_lr
51
+ - lr_scheduler_warmup_steps: 0.05
52
+ - num_epochs: 4
53
+
54
+ ### Training results
55
+
56
+
57
+
58
+ ### Framework versions
59
+
60
+ - Transformers 5.5.3
61
+ - Pytorch 2.11.0+cu128
62
+ - Datasets 4.0.0
63
+ - Tokenizers 0.22.2
README.md ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - family_to_adamw_lr = {
39
+ "language": _fallback(getattr(training_args, "language_adamw_lr", 2e-5), language_lr),
40
+ "vision": _fallback(getattr(training_args, "vision_adamw_lr", 3e-6), vision_lr),
41
+ "merger": _fallback(getattr(training_args, "merger_adamw_lr", 8e-5), merger_lr),
42
+ }
43
+ - train_batch_size: 3
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - gradient_accumulation_steps: 20
48
+ - total_train_batch_size: 60
49
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
50
+ - lr_scheduler_type: cosine_with_min_lr
51
+ - lr_scheduler_warmup_steps: 0.05
52
+ - num_epochs: 4
53
+
54
+ ### Training results
55
+
56
+
57
+
58
+ ### Framework versions
59
+
60
+ - Transformers 5.5.3
61
+ - Pytorch 2.11.0+cu128
62
+ - Datasets 4.0.0
63
+ - Tokenizers 0.22.2
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "effective_tokens_per_sec": 7119.60252078194,
3
+ "epoch": 4.0,
4
+ "total_flos": 3.6942184710227886e+18,
5
+ "train_loss": 0.7407096255137103,
6
+ "train_runtime": 10071.2083,
7
+ "train_samples_per_second": 5.529,
8
+ "train_steps_per_second": 0.092
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 sha256:745da698dc31b514cf4782c4e6a53c4ac5940563ec457019fe7dc7cb594d4bd8
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": 7119.60252078194,
3
+ "epoch": 4.0,
4
+ "total_flos": 3.6942184710227886e+18,
5
+ "train_loss": 0.7407096255137103,
6
+ "train_runtime": 10071.2083,
7
+ "train_samples_per_second": 5.529,
8
+ "train_steps_per_second": 0.092
9
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,465 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 2, "total_steps": 928, "loss": 3.6935362815856934, "lr": 4.2553191489361704e-07, "epoch": 0.008620689655172414, "percentage": 0.22, "elapsed_time": "0:03:19", "remaining_time": "1 day, 1:36:40"}
2
+ {"current_steps": 4, "total_steps": 928, "loss": 2.4514498710632324, "lr": 1.276595744680851e-06, "epoch": 0.017241379310344827, "percentage": 0.43, "elapsed_time": "0:03:44", "remaining_time": "14:25:26"}
3
+ {"current_steps": 6, "total_steps": 928, "loss": 2.2222416400909424, "lr": 2.1276595744680853e-06, "epoch": 0.02586206896551724, "percentage": 0.65, "elapsed_time": "0:04:04", "remaining_time": "10:26:54"}
4
+ {"current_steps": 8, "total_steps": 928, "loss": 1.9781272411346436, "lr": 2.978723404255319e-06, "epoch": 0.034482758620689655, "percentage": 0.86, "elapsed_time": "0:04:30", "remaining_time": "8:38:11"}
5
+ {"current_steps": 10, "total_steps": 928, "loss": 1.768671989440918, "lr": 3.8297872340425535e-06, "epoch": 0.04310344827586207, "percentage": 1.08, "elapsed_time": "0:04:48", "remaining_time": "7:21:22"}
6
+ {"current_steps": 12, "total_steps": 928, "loss": 1.7612519264221191, "lr": 4.680851063829788e-06, "epoch": 0.05172413793103448, "percentage": 1.29, "elapsed_time": "0:05:14", "remaining_time": "6:39:55"}
7
+ {"current_steps": 14, "total_steps": 928, "loss": 1.8349356651306152, "lr": 5.531914893617022e-06, "epoch": 0.0603448275862069, "percentage": 1.51, "elapsed_time": "0:05:37", "remaining_time": "6:06:48"}
8
+ {"current_steps": 16, "total_steps": 928, "loss": 0.766176164150238, "lr": 6.382978723404256e-06, "epoch": 0.06896551724137931, "percentage": 1.72, "elapsed_time": "0:05:47", "remaining_time": "5:30:25"}
9
+ {"current_steps": 18, "total_steps": 928, "loss": 0.9314699769020081, "lr": 7.234042553191491e-06, "epoch": 0.07758620689655173, "percentage": 1.94, "elapsed_time": "0:06:00", "remaining_time": "5:03:24"}
10
+ {"current_steps": 20, "total_steps": 928, "loss": 1.5076656341552734, "lr": 8.085106382978723e-06, "epoch": 0.08620689655172414, "percentage": 2.16, "elapsed_time": "0:06:26", "remaining_time": "4:52:39"}
11
+ {"current_steps": 22, "total_steps": 928, "loss": 1.6648786067962646, "lr": 8.936170212765958e-06, "epoch": 0.09482758620689655, "percentage": 2.37, "elapsed_time": "0:06:48", "remaining_time": "4:40:37"}
12
+ {"current_steps": 24, "total_steps": 928, "loss": 1.508145809173584, "lr": 9.787234042553192e-06, "epoch": 0.10344827586206896, "percentage": 2.59, "elapsed_time": "0:07:13", "remaining_time": "4:32:22"}
13
+ {"current_steps": 26, "total_steps": 928, "loss": 1.3233157396316528, "lr": 1.0638297872340426e-05, "epoch": 0.11206896551724138, "percentage": 2.8, "elapsed_time": "0:07:37", "remaining_time": "4:24:40"}
14
+ {"current_steps": 28, "total_steps": 928, "loss": 0.5060393214225769, "lr": 1.1489361702127662e-05, "epoch": 0.1206896551724138, "percentage": 3.02, "elapsed_time": "0:07:50", "remaining_time": "4:12:16"}
15
+ {"current_steps": 30, "total_steps": 928, "loss": 1.489959716796875, "lr": 1.2340425531914895e-05, "epoch": 0.12931034482758622, "percentage": 3.23, "elapsed_time": "0:08:17", "remaining_time": "4:08:02"}
16
+ {"current_steps": 32, "total_steps": 928, "loss": 1.14947509765625, "lr": 1.3191489361702127e-05, "epoch": 0.13793103448275862, "percentage": 3.45, "elapsed_time": "0:08:35", "remaining_time": "4:00:42"}
17
+ {"current_steps": 34, "total_steps": 928, "loss": 1.3647483587265015, "lr": 1.4042553191489363e-05, "epoch": 0.14655172413793102, "percentage": 3.66, "elapsed_time": "0:08:57", "remaining_time": "3:55:21"}
18
+ {"current_steps": 36, "total_steps": 928, "loss": 1.4109106063842773, "lr": 1.4893617021276596e-05, "epoch": 0.15517241379310345, "percentage": 3.88, "elapsed_time": "0:09:23", "remaining_time": "3:52:30"}
19
+ {"current_steps": 38, "total_steps": 928, "loss": 1.3988299369812012, "lr": 1.5744680851063832e-05, "epoch": 0.16379310344827586, "percentage": 4.09, "elapsed_time": "0:09:37", "remaining_time": "3:45:23"}
20
+ {"current_steps": 40, "total_steps": 928, "loss": 1.4429720640182495, "lr": 1.6595744680851064e-05, "epoch": 0.1724137931034483, "percentage": 4.31, "elapsed_time": "0:09:56", "remaining_time": "3:40:49"}
21
+ {"current_steps": 42, "total_steps": 928, "loss": 1.5474897623062134, "lr": 1.74468085106383e-05, "epoch": 0.1810344827586207, "percentage": 4.53, "elapsed_time": "0:10:21", "remaining_time": "3:38:35"}
22
+ {"current_steps": 44, "total_steps": 928, "loss": 0.9270882606506348, "lr": 1.8297872340425533e-05, "epoch": 0.1896551724137931, "percentage": 4.74, "elapsed_time": "0:10:39", "remaining_time": "3:34:01"}
23
+ {"current_steps": 46, "total_steps": 928, "loss": 1.1169898509979248, "lr": 1.914893617021277e-05, "epoch": 0.19827586206896552, "percentage": 4.96, "elapsed_time": "0:10:58", "remaining_time": "3:30:25"}
24
+ {"current_steps": 48, "total_steps": 928, "loss": 1.3745737075805664, "lr": 2e-05, "epoch": 0.20689655172413793, "percentage": 5.17, "elapsed_time": "0:11:23", "remaining_time": "3:28:43"}
25
+ {"current_steps": 50, "total_steps": 928, "loss": 1.4785292148590088, "lr": 1.999977111433554e-05, "epoch": 0.21551724137931033, "percentage": 5.39, "elapsed_time": "0:11:44", "remaining_time": "3:26:12"}
26
+ {"current_steps": 52, "total_steps": 928, "loss": 0.7019240260124207, "lr": 1.9999084468984077e-05, "epoch": 0.22413793103448276, "percentage": 5.6, "elapsed_time": "0:11:55", "remaining_time": "3:20:59"}
27
+ {"current_steps": 54, "total_steps": 928, "loss": 1.4191925525665283, "lr": 1.9997940098870788e-05, "epoch": 0.23275862068965517, "percentage": 5.82, "elapsed_time": "0:12:18", "remaining_time": "3:19:07"}
28
+ {"current_steps": 56, "total_steps": 928, "loss": 1.3712444305419922, "lr": 1.9996338062202316e-05, "epoch": 0.2413793103448276, "percentage": 6.03, "elapsed_time": "0:12:44", "remaining_time": "3:18:19"}
29
+ {"current_steps": 58, "total_steps": 928, "loss": 1.4150975942611694, "lr": 1.9994278440463823e-05, "epoch": 0.25, "percentage": 6.25, "elapsed_time": "0:13:06", "remaining_time": "3:16:32"}
30
+ {"current_steps": 60, "total_steps": 928, "loss": 1.3828074932098389, "lr": 1.9991761338414844e-05, "epoch": 0.25862068965517243, "percentage": 6.47, "elapsed_time": "0:13:32", "remaining_time": "3:15:56"}
31
+ {"current_steps": 62, "total_steps": 928, "loss": 1.0158110857009888, "lr": 1.998878688408395e-05, "epoch": 0.2672413793103448, "percentage": 6.68, "elapsed_time": "0:13:46", "remaining_time": "3:12:27"}
32
+ {"current_steps": 64, "total_steps": 928, "loss": 1.3290989398956299, "lr": 1.9985355228762237e-05, "epoch": 0.27586206896551724, "percentage": 6.9, "elapsed_time": "0:14:12", "remaining_time": "3:11:54"}
33
+ {"current_steps": 66, "total_steps": 928, "loss": 1.303815484046936, "lr": 1.9981466546995648e-05, "epoch": 0.28448275862068967, "percentage": 7.11, "elapsed_time": "0:14:35", "remaining_time": "3:10:40"}
34
+ {"current_steps": 68, "total_steps": 928, "loss": 1.2936116456985474, "lr": 1.997712103657607e-05, "epoch": 0.29310344827586204, "percentage": 7.33, "elapsed_time": "0:15:27", "remaining_time": "3:15:28"}
35
+ {"current_steps": 70, "total_steps": 928, "loss": 1.0591557025909424, "lr": 1.9972318918531294e-05, "epoch": 0.3017241379310345, "percentage": 7.54, "elapsed_time": "0:15:46", "remaining_time": "3:13:26"}
36
+ {"current_steps": 72, "total_steps": 928, "loss": 0.9161477088928223, "lr": 1.9967060437113752e-05, "epoch": 0.3103448275862069, "percentage": 7.76, "elapsed_time": "0:15:59", "remaining_time": "3:10:06"}
37
+ {"current_steps": 74, "total_steps": 928, "loss": 1.130851149559021, "lr": 1.996134585978812e-05, "epoch": 0.31896551724137934, "percentage": 7.97, "elapsed_time": "0:16:18", "remaining_time": "3:08:17"}
38
+ {"current_steps": 76, "total_steps": 928, "loss": 1.167314052581787, "lr": 1.995517547721769e-05, "epoch": 0.3275862068965517, "percentage": 8.19, "elapsed_time": "0:16:34", "remaining_time": "3:05:43"}
39
+ {"current_steps": 78, "total_steps": 928, "loss": 1.1086735725402832, "lr": 1.9948549603249602e-05, "epoch": 0.33620689655172414, "percentage": 8.41, "elapsed_time": "0:16:47", "remaining_time": "3:02:55"}
40
+ {"current_steps": 80, "total_steps": 928, "loss": 1.0062172412872314, "lr": 1.9941468574898867e-05, "epoch": 0.3448275862068966, "percentage": 8.62, "elapsed_time": "0:17:07", "remaining_time": "3:01:29"}
41
+ {"current_steps": 82, "total_steps": 928, "loss": 1.3131077289581299, "lr": 1.993393275233123e-05, "epoch": 0.35344827586206895, "percentage": 8.84, "elapsed_time": "0:17:33", "remaining_time": "3:01:08"}
42
+ {"current_steps": 84, "total_steps": 928, "loss": 1.1570929288864136, "lr": 1.9925942518844875e-05, "epoch": 0.3620689655172414, "percentage": 9.05, "elapsed_time": "0:17:53", "remaining_time": "2:59:43"}
43
+ {"current_steps": 86, "total_steps": 928, "loss": 1.3317445516586304, "lr": 1.991749828085088e-05, "epoch": 0.3706896551724138, "percentage": 9.27, "elapsed_time": "0:18:18", "remaining_time": "2:59:12"}
44
+ {"current_steps": 88, "total_steps": 928, "loss": 1.2484678030014038, "lr": 1.9908600467852585e-05, "epoch": 0.3793103448275862, "percentage": 9.48, "elapsed_time": "0:18:44", "remaining_time": "2:58:54"}
45
+ {"current_steps": 90, "total_steps": 928, "loss": 1.0302324295043945, "lr": 1.9899249532423733e-05, "epoch": 0.3879310344827586, "percentage": 9.7, "elapsed_time": "0:19:03", "remaining_time": "2:57:29"}
46
+ {"current_steps": 92, "total_steps": 928, "loss": 0.8809990882873535, "lr": 1.9889445950185452e-05, "epoch": 0.39655172413793105, "percentage": 9.91, "elapsed_time": "0:19:14", "remaining_time": "2:54:49"}
47
+ {"current_steps": 94, "total_steps": 928, "loss": 1.0718659162521362, "lr": 1.9879190219782058e-05, "epoch": 0.4051724137931034, "percentage": 10.13, "elapsed_time": "0:19:35", "remaining_time": "2:53:50"}
48
+ {"current_steps": 96, "total_steps": 928, "loss": 1.2635523080825806, "lr": 1.98684828628557e-05, "epoch": 0.41379310344827586, "percentage": 10.34, "elapsed_time": "0:19:56", "remaining_time": "2:52:45"}
49
+ {"current_steps": 98, "total_steps": 928, "loss": 1.186186671257019, "lr": 1.9857324424019827e-05, "epoch": 0.4224137931034483, "percentage": 10.56, "elapsed_time": "0:20:20", "remaining_time": "2:52:19"}
50
+ {"current_steps": 100, "total_steps": 928, "loss": 1.1258275508880615, "lr": 1.9845715470831476e-05, "epoch": 0.43103448275862066, "percentage": 10.78, "elapsed_time": "0:20:38", "remaining_time": "2:50:55"}
51
+ {"current_steps": 102, "total_steps": 928, "loss": 1.2144218683242798, "lr": 1.983365659376242e-05, "epoch": 0.4396551724137931, "percentage": 10.99, "elapsed_time": "0:20:58", "remaining_time": "2:49:52"}
52
+ {"current_steps": 104, "total_steps": 928, "loss": 1.2573832273483276, "lr": 1.9821148406169124e-05, "epoch": 0.4482758620689655, "percentage": 11.21, "elapsed_time": "0:21:25", "remaining_time": "2:49:41"}
53
+ {"current_steps": 106, "total_steps": 928, "loss": 0.9863552451133728, "lr": 1.9808191544261545e-05, "epoch": 0.45689655172413796, "percentage": 11.42, "elapsed_time": "0:21:43", "remaining_time": "2:48:24"}
54
+ {"current_steps": 108, "total_steps": 928, "loss": 1.1030296087265015, "lr": 1.9794786667070786e-05, "epoch": 0.46551724137931033, "percentage": 11.64, "elapsed_time": "0:22:08", "remaining_time": "2:48:07"}
55
+ {"current_steps": 110, "total_steps": 928, "loss": 1.1802244186401367, "lr": 1.9780934456415567e-05, "epoch": 0.47413793103448276, "percentage": 11.85, "elapsed_time": "0:22:30", "remaining_time": "2:47:22"}
56
+ {"current_steps": 112, "total_steps": 928, "loss": 1.410926103591919, "lr": 1.976663561686753e-05, "epoch": 0.4827586206896552, "percentage": 12.07, "elapsed_time": "0:22:53", "remaining_time": "2:46:44"}
57
+ {"current_steps": 114, "total_steps": 928, "loss": 0.8382052183151245, "lr": 1.975189087571544e-05, "epoch": 0.49137931034482757, "percentage": 12.28, "elapsed_time": "0:23:11", "remaining_time": "2:45:38"}
58
+ {"current_steps": 116, "total_steps": 928, "loss": 0.8902297616004944, "lr": 1.9736700982928156e-05, "epoch": 0.5, "percentage": 12.5, "elapsed_time": "0:23:31", "remaining_time": "2:44:42"}
59
+ {"current_steps": 118, "total_steps": 928, "loss": 1.2678132057189941, "lr": 1.9721066711116496e-05, "epoch": 0.5086206896551724, "percentage": 12.72, "elapsed_time": "0:23:58", "remaining_time": "2:44:33"}
60
+ {"current_steps": 120, "total_steps": 928, "loss": 1.1739540100097656, "lr": 1.9704988855493948e-05, "epoch": 0.5172413793103449, "percentage": 12.93, "elapsed_time": "0:24:19", "remaining_time": "2:43:50"}
61
+ {"current_steps": 122, "total_steps": 928, "loss": 1.2742866277694702, "lr": 1.9688468233836224e-05, "epoch": 0.5258620689655172, "percentage": 13.15, "elapsed_time": "0:24:46", "remaining_time": "2:43:40"}
62
+ {"current_steps": 124, "total_steps": 928, "loss": 1.3943136930465698, "lr": 1.9671505686439637e-05, "epoch": 0.5344827586206896, "percentage": 13.36, "elapsed_time": "0:25:09", "remaining_time": "2:43:05"}
63
+ {"current_steps": 126, "total_steps": 928, "loss": 1.564052939414978, "lr": 1.9654102076078406e-05, "epoch": 0.5431034482758621, "percentage": 13.58, "elapsed_time": "0:25:34", "remaining_time": "2:42:50"}
64
+ {"current_steps": 128, "total_steps": 928, "loss": 1.28196382522583, "lr": 1.963625828796073e-05, "epoch": 0.5517241379310345, "percentage": 13.79, "elapsed_time": "0:25:50", "remaining_time": "2:41:32"}
65
+ {"current_steps": 130, "total_steps": 928, "loss": 1.2830570936203003, "lr": 1.9617975229683784e-05, "epoch": 0.5603448275862069, "percentage": 14.01, "elapsed_time": "0:26:12", "remaining_time": "2:40:53"}
66
+ {"current_steps": 132, "total_steps": 928, "loss": 1.1657885313034058, "lr": 1.959925383118756e-05, "epoch": 0.5689655172413793, "percentage": 14.22, "elapsed_time": "0:26:37", "remaining_time": "2:40:33"}
67
+ {"current_steps": 134, "total_steps": 928, "loss": 0.7719605565071106, "lr": 1.9580095044707553e-05, "epoch": 0.5775862068965517, "percentage": 14.44, "elapsed_time": "0:26:56", "remaining_time": "2:39:37"}
68
+ {"current_steps": 136, "total_steps": 928, "loss": 0.9887097477912903, "lr": 1.956049984472634e-05, "epoch": 0.5862068965517241, "percentage": 14.66, "elapsed_time": "0:27:14", "remaining_time": "2:38:41"}
69
+ {"current_steps": 138, "total_steps": 928, "loss": 0.9446215033531189, "lr": 1.954046922792399e-05, "epoch": 0.5948275862068966, "percentage": 14.87, "elapsed_time": "0:27:25", "remaining_time": "2:37:01"}
70
+ {"current_steps": 140, "total_steps": 928, "loss": 0.9638822078704834, "lr": 1.9520004213127412e-05, "epoch": 0.603448275862069, "percentage": 15.09, "elapsed_time": "0:27:45", "remaining_time": "2:36:15"}
71
+ {"current_steps": 142, "total_steps": 928, "loss": 1.028863549232483, "lr": 1.9499105841258488e-05, "epoch": 0.6120689655172413, "percentage": 15.3, "elapsed_time": "0:28:05", "remaining_time": "2:35:31"}
72
+ {"current_steps": 144, "total_steps": 928, "loss": 1.0197412967681885, "lr": 1.9477775175281166e-05, "epoch": 0.6206896551724138, "percentage": 15.52, "elapsed_time": "0:28:25", "remaining_time": "2:34:43"}
73
+ {"current_steps": 146, "total_steps": 928, "loss": 1.2361186742782593, "lr": 1.945601330014737e-05, "epoch": 0.6293103448275862, "percentage": 15.73, "elapsed_time": "0:28:43", "remaining_time": "2:33:53"}
74
+ {"current_steps": 148, "total_steps": 928, "loss": 1.2686266899108887, "lr": 1.9433821322741814e-05, "epoch": 0.6379310344827587, "percentage": 15.95, "elapsed_time": "0:29:01", "remaining_time": "2:32:59"}
75
+ {"current_steps": 150, "total_steps": 928, "loss": 1.1161521673202515, "lr": 1.9411200371825724e-05, "epoch": 0.646551724137931, "percentage": 16.16, "elapsed_time": "0:29:18", "remaining_time": "2:32:01"}
76
+ {"current_steps": 152, "total_steps": 928, "loss": 0.8658434152603149, "lr": 1.9388151597979407e-05, "epoch": 0.6551724137931034, "percentage": 16.38, "elapsed_time": "0:29:36", "remaining_time": "2:31:10"}
77
+ {"current_steps": 154, "total_steps": 928, "loss": 1.0216702222824097, "lr": 1.9364676173543734e-05, "epoch": 0.6637931034482759, "percentage": 16.59, "elapsed_time": "0:29:58", "remaining_time": "2:30:40"}
78
+ {"current_steps": 156, "total_steps": 928, "loss": 1.0728117227554321, "lr": 1.934077529256051e-05, "epoch": 0.6724137931034483, "percentage": 16.81, "elapsed_time": "0:30:21", "remaining_time": "2:30:15"}
79
+ {"current_steps": 158, "total_steps": 928, "loss": 1.2310281991958618, "lr": 1.9316450170711732e-05, "epoch": 0.6810344827586207, "percentage": 17.03, "elapsed_time": "0:30:48", "remaining_time": "2:30:07"}
80
+ {"current_steps": 160, "total_steps": 928, "loss": 1.2321813106536865, "lr": 1.929170204525779e-05, "epoch": 0.6896551724137931, "percentage": 17.24, "elapsed_time": "0:31:10", "remaining_time": "2:29:37"}
81
+ {"current_steps": 162, "total_steps": 928, "loss": 1.5227911472320557, "lr": 1.9266532174974476e-05, "epoch": 0.6982758620689655, "percentage": 17.46, "elapsed_time": "0:31:34", "remaining_time": "2:29:18"}
82
+ {"current_steps": 164, "total_steps": 928, "loss": 0.9460239410400391, "lr": 1.9240941840089015e-05, "epoch": 0.7068965517241379, "percentage": 17.67, "elapsed_time": "0:31:55", "remaining_time": "2:28:41"}
83
+ {"current_steps": 166, "total_steps": 928, "loss": 1.24778151512146, "lr": 1.9214932342214923e-05, "epoch": 0.7155172413793104, "percentage": 17.89, "elapsed_time": "0:32:13", "remaining_time": "2:27:56"}
84
+ {"current_steps": 168, "total_steps": 928, "loss": 1.3147329092025757, "lr": 1.9188505004285795e-05, "epoch": 0.7241379310344828, "percentage": 18.1, "elapsed_time": "0:32:39", "remaining_time": "2:27:46"}
85
+ {"current_steps": 170, "total_steps": 928, "loss": 1.223319411277771, "lr": 1.916166117048803e-05, "epoch": 0.7327586206896551, "percentage": 18.32, "elapsed_time": "0:33:06", "remaining_time": "2:27:37"}
86
+ {"current_steps": 172, "total_steps": 928, "loss": 1.2373883724212646, "lr": 1.9134402206192468e-05, "epoch": 0.7413793103448276, "percentage": 18.53, "elapsed_time": "0:33:32", "remaining_time": "2:27:25"}
87
+ {"current_steps": 174, "total_steps": 928, "loss": 0.937210202217102, "lr": 1.910672949788491e-05, "epoch": 0.75, "percentage": 18.75, "elapsed_time": "0:33:47", "remaining_time": "2:26:24"}
88
+ {"current_steps": 176, "total_steps": 928, "loss": 0.6929088234901428, "lr": 1.9078644453095634e-05, "epoch": 0.7586206896551724, "percentage": 18.97, "elapsed_time": "0:33:58", "remaining_time": "2:25:08"}
89
+ {"current_steps": 178, "total_steps": 928, "loss": 1.1937777996063232, "lr": 1.9050148500327783e-05, "epoch": 0.7672413793103449, "percentage": 19.18, "elapsed_time": "0:34:20", "remaining_time": "2:24:43"}
90
+ {"current_steps": 180, "total_steps": 928, "loss": 0.9337822198867798, "lr": 1.9021243088984705e-05, "epoch": 0.7758620689655172, "percentage": 19.4, "elapsed_time": "0:34:41", "remaining_time": "2:24:10"}
91
+ {"current_steps": 182, "total_steps": 928, "loss": 0.9228455424308777, "lr": 1.8991929689296245e-05, "epoch": 0.7844827586206896, "percentage": 19.61, "elapsed_time": "0:35:00", "remaining_time": "2:23:31"}
92
+ {"current_steps": 184, "total_steps": 928, "loss": 0.77471923828125, "lr": 1.896220979224395e-05, "epoch": 0.7931034482758621, "percentage": 19.83, "elapsed_time": "0:35:19", "remaining_time": "2:22:50"}
93
+ {"current_steps": 186, "total_steps": 928, "loss": 1.335769772529602, "lr": 1.8932084909485233e-05, "epoch": 0.8017241379310345, "percentage": 20.04, "elapsed_time": "0:35:41", "remaining_time": "2:22:23"}
94
+ {"current_steps": 188, "total_steps": 928, "loss": 1.4579006433486938, "lr": 1.89015565732765e-05, "epoch": 0.8103448275862069, "percentage": 20.26, "elapsed_time": "0:36:04", "remaining_time": "2:21:58"}
95
+ {"current_steps": 190, "total_steps": 928, "loss": 1.2512203454971313, "lr": 1.8870626336395204e-05, "epoch": 0.8189655172413793, "percentage": 20.47, "elapsed_time": "0:36:31", "remaining_time": "2:21:50"}
96
+ {"current_steps": 192, "total_steps": 928, "loss": 1.217123031616211, "lr": 1.8839295772060856e-05, "epoch": 0.8275862068965517, "percentage": 20.69, "elapsed_time": "0:36:57", "remaining_time": "2:21:40"}
97
+ {"current_steps": 194, "total_steps": 928, "loss": 1.2150022983551025, "lr": 1.8807566473855018e-05, "epoch": 0.8362068965517241, "percentage": 20.91, "elapsed_time": "0:37:19", "remaining_time": "2:21:13"}
98
+ {"current_steps": 196, "total_steps": 928, "loss": 1.2615362405776978, "lr": 1.8775440055640245e-05, "epoch": 0.8448275862068966, "percentage": 21.12, "elapsed_time": "0:37:45", "remaining_time": "2:21:00"}
99
+ {"current_steps": 198, "total_steps": 928, "loss": 0.9542128443717957, "lr": 1.8742918151478008e-05, "epoch": 0.853448275862069, "percentage": 21.34, "elapsed_time": "0:38:05", "remaining_time": "2:20:26"}
100
+ {"current_steps": 200, "total_steps": 928, "loss": 1.2213218212127686, "lr": 1.8710002415545552e-05, "epoch": 0.8620689655172413, "percentage": 21.55, "elapsed_time": "0:38:25", "remaining_time": "2:19:53"}
101
+ {"current_steps": 202, "total_steps": 928, "loss": 1.2830454111099243, "lr": 1.867669452205179e-05, "epoch": 0.8706896551724138, "percentage": 21.77, "elapsed_time": "0:38:49", "remaining_time": "2:19:32"}
102
+ {"current_steps": 204, "total_steps": 928, "loss": 0.8405295610427856, "lr": 1.8642996165152126e-05, "epoch": 0.8793103448275862, "percentage": 21.98, "elapsed_time": "0:39:09", "remaining_time": "2:18:59"}
103
+ {"current_steps": 206, "total_steps": 928, "loss": 1.2193654775619507, "lr": 1.86089090588623e-05, "epoch": 0.8879310344827587, "percentage": 22.2, "elapsed_time": "0:39:32", "remaining_time": "2:18:35"}
104
+ {"current_steps": 208, "total_steps": 928, "loss": 1.2507728338241577, "lr": 1.857443493697119e-05, "epoch": 0.896551724137931, "percentage": 22.41, "elapsed_time": "0:39:59", "remaining_time": "2:18:25"}
105
+ {"current_steps": 210, "total_steps": 928, "loss": 0.5853071212768555, "lr": 1.853957555295265e-05, "epoch": 0.9051724137931034, "percentage": 22.63, "elapsed_time": "0:40:18", "remaining_time": "2:17:49"}
106
+ {"current_steps": 212, "total_steps": 928, "loss": 0.9337871074676514, "lr": 1.850433267987628e-05, "epoch": 0.9137931034482759, "percentage": 22.84, "elapsed_time": "0:40:37", "remaining_time": "2:17:10"}
107
+ {"current_steps": 214, "total_steps": 928, "loss": 1.1255706548690796, "lr": 1.84687081103173e-05, "epoch": 0.9224137931034483, "percentage": 23.06, "elapsed_time": "0:40:58", "remaining_time": "2:16:43"}
108
+ {"current_steps": 216, "total_steps": 928, "loss": 1.2182952165603638, "lr": 1.8432703656265317e-05, "epoch": 0.9310344827586207, "percentage": 23.28, "elapsed_time": "0:41:22", "remaining_time": "2:16:21"}
109
+ {"current_steps": 218, "total_steps": 928, "loss": 0.755200207233429, "lr": 1.83963211490322e-05, "epoch": 0.9396551724137931, "percentage": 23.49, "elapsed_time": "0:41:40", "remaining_time": "2:15:45"}
110
+ {"current_steps": 220, "total_steps": 928, "loss": 0.9116122722625732, "lr": 1.8359562439158915e-05, "epoch": 0.9482758620689655, "percentage": 23.71, "elapsed_time": "0:41:59", "remaining_time": "2:15:06"}
111
+ {"current_steps": 222, "total_steps": 928, "loss": 1.314684271812439, "lr": 1.8322429396321395e-05, "epoch": 0.9568965517241379, "percentage": 23.92, "elapsed_time": "0:42:24", "remaining_time": "2:14:51"}
112
+ {"current_steps": 224, "total_steps": 928, "loss": 0.8659108877182007, "lr": 1.828492390923547e-05, "epoch": 0.9655172413793104, "percentage": 24.14, "elapsed_time": "0:42:44", "remaining_time": "2:14:20"}
113
+ {"current_steps": 226, "total_steps": 928, "loss": 0.8610473275184631, "lr": 1.824704788556076e-05, "epoch": 0.9741379310344828, "percentage": 24.35, "elapsed_time": "0:43:10", "remaining_time": "2:14:05"}
114
+ {"current_steps": 228, "total_steps": 928, "loss": 1.2119630575180054, "lr": 1.820880325180368e-05, "epoch": 0.9827586206896551, "percentage": 24.57, "elapsed_time": "0:43:36", "remaining_time": "2:13:53"}
115
+ {"current_steps": 230, "total_steps": 928, "loss": 0.886069118976593, "lr": 1.817019195321943e-05, "epoch": 0.9913793103448276, "percentage": 24.78, "elapsed_time": "0:43:52", "remaining_time": "2:13:08"}
116
+ {"current_steps": 232, "total_steps": 928, "loss": 0.7485522031784058, "lr": 1.8131215953713065e-05, "epoch": 1.0, "percentage": 25.0, "elapsed_time": "0:44:10", "remaining_time": "2:12:30"}
117
+ {"current_steps": 234, "total_steps": 928, "loss": 0.8965163826942444, "lr": 1.809187723573959e-05, "epoch": 1.0086206896551724, "percentage": 25.22, "elapsed_time": "0:44:35", "remaining_time": "2:12:16"}
118
+ {"current_steps": 236, "total_steps": 928, "loss": 0.745244026184082, "lr": 1.8052177800203142e-05, "epoch": 1.0172413793103448, "percentage": 25.43, "elapsed_time": "0:44:56", "remaining_time": "2:11:48"}
119
+ {"current_steps": 238, "total_steps": 928, "loss": 0.8928873538970947, "lr": 1.8012119666355207e-05, "epoch": 1.0258620689655173, "percentage": 25.65, "elapsed_time": "0:45:18", "remaining_time": "2:11:22"}
120
+ {"current_steps": 240, "total_steps": 928, "loss": 0.5402289628982544, "lr": 1.7971704871691913e-05, "epoch": 1.0344827586206897, "percentage": 25.86, "elapsed_time": "0:45:33", "remaining_time": "2:10:37"}
121
+ {"current_steps": 242, "total_steps": 928, "loss": 1.0471925735473633, "lr": 1.7930935471850397e-05, "epoch": 1.043103448275862, "percentage": 26.08, "elapsed_time": "0:45:56", "remaining_time": "2:10:13"}
122
+ {"current_steps": 244, "total_steps": 928, "loss": 0.3569323420524597, "lr": 1.7889813540504255e-05, "epoch": 1.0517241379310345, "percentage": 26.29, "elapsed_time": "0:46:07", "remaining_time": "2:09:16"}
123
+ {"current_steps": 246, "total_steps": 928, "loss": 1.0155720710754395, "lr": 1.7848341169258066e-05, "epoch": 1.0603448275862069, "percentage": 26.51, "elapsed_time": "0:46:33", "remaining_time": "2:09:03"}
124
+ {"current_steps": 248, "total_steps": 928, "loss": 0.9613431096076965, "lr": 1.780652046754099e-05, "epoch": 1.0689655172413792, "percentage": 26.72, "elapsed_time": "0:46:59", "remaining_time": "2:08:50"}
125
+ {"current_steps": 250, "total_steps": 928, "loss": 0.9193274974822998, "lr": 1.776435356249949e-05, "epoch": 1.0775862068965518, "percentage": 26.94, "elapsed_time": "0:47:18", "remaining_time": "2:08:18"}
126
+ {"current_steps": 252, "total_steps": 928, "loss": 0.42583054304122925, "lr": 1.772184259888915e-05, "epoch": 1.0862068965517242, "percentage": 27.16, "elapsed_time": "0:47:32", "remaining_time": "2:07:32"}
127
+ {"current_steps": 254, "total_steps": 928, "loss": 0.9037254452705383, "lr": 1.7678989738965554e-05, "epoch": 1.0948275862068966, "percentage": 27.37, "elapsed_time": "0:47:55", "remaining_time": "2:07:09"}
128
+ {"current_steps": 256, "total_steps": 928, "loss": 0.6561298370361328, "lr": 1.7635797162374325e-05, "epoch": 1.103448275862069, "percentage": 27.59, "elapsed_time": "0:48:12", "remaining_time": "2:06:32"}
129
+ {"current_steps": 258, "total_steps": 928, "loss": 0.5268462896347046, "lr": 1.7592267066040266e-05, "epoch": 1.1120689655172413, "percentage": 27.8, "elapsed_time": "0:48:23", "remaining_time": "2:05:40"}
130
+ {"current_steps": 260, "total_steps": 928, "loss": 0.6683530211448669, "lr": 1.7548401664055605e-05, "epoch": 1.1206896551724137, "percentage": 28.02, "elapsed_time": "0:48:44", "remaining_time": "2:05:13"}
131
+ {"current_steps": 262, "total_steps": 928, "loss": 1.1267895698547363, "lr": 1.7504203187567378e-05, "epoch": 1.1293103448275863, "percentage": 28.23, "elapsed_time": "0:49:10", "remaining_time": "2:05:01"}
132
+ {"current_steps": 264, "total_steps": 928, "loss": 0.864532470703125, "lr": 1.7459673884663956e-05, "epoch": 1.1379310344827587, "percentage": 28.45, "elapsed_time": "0:49:33", "remaining_time": "2:04:38"}
133
+ {"current_steps": 266, "total_steps": 928, "loss": 1.043933629989624, "lr": 1.7414816020260694e-05, "epoch": 1.146551724137931, "percentage": 28.66, "elapsed_time": "0:49:56", "remaining_time": "2:04:16"}
134
+ {"current_steps": 268, "total_steps": 928, "loss": 1.0288755893707275, "lr": 1.7369631875984715e-05, "epoch": 1.1551724137931034, "percentage": 28.88, "elapsed_time": "0:50:18", "remaining_time": "2:03:53"}
135
+ {"current_steps": 270, "total_steps": 928, "loss": 1.0703890323638916, "lr": 1.7324123750058895e-05, "epoch": 1.1637931034482758, "percentage": 29.09, "elapsed_time": "0:50:44", "remaining_time": "2:03:39"}
136
+ {"current_steps": 272, "total_steps": 928, "loss": 1.0030405521392822, "lr": 1.7278293957184924e-05, "epoch": 1.1724137931034484, "percentage": 29.31, "elapsed_time": "0:51:10", "remaining_time": "2:03:25"}
137
+ {"current_steps": 274, "total_steps": 928, "loss": 0.8162426352500916, "lr": 1.7232144828425606e-05, "epoch": 1.1810344827586208, "percentage": 29.53, "elapsed_time": "0:51:30", "remaining_time": "2:02:55"}
138
+ {"current_steps": 276, "total_steps": 928, "loss": 0.5113853216171265, "lr": 1.718567871108627e-05, "epoch": 1.1896551724137931, "percentage": 29.74, "elapsed_time": "0:51:43", "remaining_time": "2:02:10"}
139
+ {"current_steps": 278, "total_steps": 928, "loss": 1.0502547025680542, "lr": 1.7138897968595394e-05, "epoch": 1.1982758620689655, "percentage": 29.96, "elapsed_time": "0:52:09", "remaining_time": "2:01:56"}
140
+ {"current_steps": 280, "total_steps": 928, "loss": 0.9660123586654663, "lr": 1.7091804980384382e-05, "epoch": 1.206896551724138, "percentage": 30.17, "elapsed_time": "0:52:33", "remaining_time": "2:01:38"}
141
+ {"current_steps": 282, "total_steps": 928, "loss": 0.5680896043777466, "lr": 1.7044402141766547e-05, "epoch": 1.2155172413793103, "percentage": 30.39, "elapsed_time": "0:52:54", "remaining_time": "2:01:11"}
142
+ {"current_steps": 284, "total_steps": 928, "loss": 0.6948413848876953, "lr": 1.6996691863815274e-05, "epoch": 1.2241379310344827, "percentage": 30.6, "elapsed_time": "0:53:11", "remaining_time": "2:00:36"}
143
+ {"current_steps": 286, "total_steps": 928, "loss": 0.8091163635253906, "lr": 1.694867657324137e-05, "epoch": 1.2327586206896552, "percentage": 30.82, "elapsed_time": "0:53:30", "remaining_time": "2:00:06"}
144
+ {"current_steps": 288, "total_steps": 928, "loss": 0.4585968852043152, "lr": 1.6900358712269656e-05, "epoch": 1.2413793103448276, "percentage": 31.03, "elapsed_time": "0:53:44", "remaining_time": "1:59:24"}
145
+ {"current_steps": 290, "total_steps": 928, "loss": 0.9265656471252441, "lr": 1.6851740738514745e-05, "epoch": 1.25, "percentage": 31.25, "elapsed_time": "0:54:03", "remaining_time": "1:58:56"}
146
+ {"current_steps": 292, "total_steps": 928, "loss": 0.8860665559768677, "lr": 1.680282512485601e-05, "epoch": 1.2586206896551724, "percentage": 31.47, "elapsed_time": "0:54:26", "remaining_time": "1:58:33"}
147
+ {"current_steps": 294, "total_steps": 928, "loss": 1.003743052482605, "lr": 1.675361435931184e-05, "epoch": 1.2672413793103448, "percentage": 31.68, "elapsed_time": "0:54:46", "remaining_time": "1:58:07"}
148
+ {"current_steps": 296, "total_steps": 928, "loss": 0.7721095085144043, "lr": 1.670411094491308e-05, "epoch": 1.2758620689655173, "percentage": 31.9, "elapsed_time": "0:55:06", "remaining_time": "1:57:40"}
149
+ {"current_steps": 298, "total_steps": 928, "loss": 1.0421152114868164, "lr": 1.665431739957569e-05, "epoch": 1.2844827586206897, "percentage": 32.11, "elapsed_time": "0:55:32", "remaining_time": "1:57:25"}
150
+ {"current_steps": 300, "total_steps": 928, "loss": 0.9568123817443848, "lr": 1.660423625597273e-05, "epoch": 1.293103448275862, "percentage": 32.33, "elapsed_time": "0:55:57", "remaining_time": "1:57:09"}
151
+ {"current_steps": 302, "total_steps": 928, "loss": 0.610411524772644, "lr": 1.6553870061405496e-05, "epoch": 1.3017241379310345, "percentage": 32.54, "elapsed_time": "0:56:18", "remaining_time": "1:56:42"}
152
+ {"current_steps": 304, "total_steps": 928, "loss": 0.5043181777000427, "lr": 1.6503221377673965e-05, "epoch": 1.3103448275862069, "percentage": 32.76, "elapsed_time": "0:56:32", "remaining_time": "1:56:02"}
153
+ {"current_steps": 306, "total_steps": 928, "loss": 1.1974719762802124, "lr": 1.6452292780946495e-05, "epoch": 1.3189655172413794, "percentage": 32.97, "elapsed_time": "0:56:57", "remaining_time": "1:55:47"}
154
+ {"current_steps": 308, "total_steps": 928, "loss": 0.6518297791481018, "lr": 1.6401086861628802e-05, "epoch": 1.3275862068965516, "percentage": 33.19, "elapsed_time": "0:57:16", "remaining_time": "1:55:17"}
155
+ {"current_steps": 310, "total_steps": 928, "loss": 0.9824197292327881, "lr": 1.6349606224232193e-05, "epoch": 1.3362068965517242, "percentage": 33.41, "elapsed_time": "0:57:38", "remaining_time": "1:54:54"}
156
+ {"current_steps": 312, "total_steps": 928, "loss": 0.9464894533157349, "lr": 1.6297853487241087e-05, "epoch": 1.3448275862068966, "percentage": 33.62, "elapsed_time": "0:57:56", "remaining_time": "1:54:23"}
157
+ {"current_steps": 314, "total_steps": 928, "loss": 1.1367565393447876, "lr": 1.6245831282979834e-05, "epoch": 1.353448275862069, "percentage": 33.84, "elapsed_time": "0:58:15", "remaining_time": "1:53:54"}
158
+ {"current_steps": 316, "total_steps": 928, "loss": 1.3186949491500854, "lr": 1.6193542257478837e-05, "epoch": 1.3620689655172413, "percentage": 34.05, "elapsed_time": "0:58:39", "remaining_time": "1:53:35"}
159
+ {"current_steps": 318, "total_steps": 928, "loss": 0.9798977375030518, "lr": 1.6140989070339947e-05, "epoch": 1.3706896551724137, "percentage": 34.27, "elapsed_time": "0:59:02", "remaining_time": "1:53:15"}
160
+ {"current_steps": 320, "total_steps": 928, "loss": 0.7535231709480286, "lr": 1.60881743946012e-05, "epoch": 1.3793103448275863, "percentage": 34.48, "elapsed_time": "0:59:16", "remaining_time": "1:52:36"}
161
+ {"current_steps": 322, "total_steps": 928, "loss": 1.412162184715271, "lr": 1.6035100916600856e-05, "epoch": 1.3879310344827587, "percentage": 34.7, "elapsed_time": "0:59:41", "remaining_time": "1:52:20"}
162
+ {"current_steps": 324, "total_steps": 928, "loss": 0.5099858641624451, "lr": 1.5981771335840744e-05, "epoch": 1.396551724137931, "percentage": 34.91, "elapsed_time": "1:00:06", "remaining_time": "1:52:02"}
163
+ {"current_steps": 326, "total_steps": 928, "loss": 0.5017666220664978, "lr": 1.5928188364848995e-05, "epoch": 1.4051724137931034, "percentage": 35.13, "elapsed_time": "1:00:19", "remaining_time": "1:51:23"}
164
+ {"current_steps": 328, "total_steps": 928, "loss": 0.8372330069541931, "lr": 1.5874354729042023e-05, "epoch": 1.4137931034482758, "percentage": 35.34, "elapsed_time": "1:00:39", "remaining_time": "1:50:56"}
165
+ {"current_steps": 330, "total_steps": 928, "loss": 0.7255274057388306, "lr": 1.5820273166585947e-05, "epoch": 1.4224137931034484, "percentage": 35.56, "elapsed_time": "1:00:59", "remaining_time": "1:50:31"}
166
+ {"current_steps": 332, "total_steps": 928, "loss": 0.6583790183067322, "lr": 1.576594642825729e-05, "epoch": 1.4310344827586206, "percentage": 35.78, "elapsed_time": "1:01:19", "remaining_time": "1:50:04"}
167
+ {"current_steps": 334, "total_steps": 928, "loss": 0.5846832990646362, "lr": 1.571137727730307e-05, "epoch": 1.4396551724137931, "percentage": 35.99, "elapsed_time": "1:01:37", "remaining_time": "1:49:35"}
168
+ {"current_steps": 336, "total_steps": 928, "loss": 0.6714300513267517, "lr": 1.5656568489300265e-05, "epoch": 1.4482758620689655, "percentage": 36.21, "elapsed_time": "1:01:55", "remaining_time": "1:49:05"}
169
+ {"current_steps": 338, "total_steps": 928, "loss": 1.0285781621932983, "lr": 1.560152285201462e-05, "epoch": 1.456896551724138, "percentage": 36.42, "elapsed_time": "1:02:20", "remaining_time": "1:48:49"}
170
+ {"current_steps": 340, "total_steps": 928, "loss": 1.34324049949646, "lr": 1.5546243165258857e-05, "epoch": 1.4655172413793103, "percentage": 36.64, "elapsed_time": "1:02:46", "remaining_time": "1:48:33"}
171
+ {"current_steps": 342, "total_steps": 928, "loss": 1.1209205389022827, "lr": 1.5490732240750275e-05, "epoch": 1.4741379310344827, "percentage": 36.85, "elapsed_time": "1:03:12", "remaining_time": "1:48:18"}
172
+ {"current_steps": 344, "total_steps": 928, "loss": 0.944566547870636, "lr": 1.5434992901967735e-05, "epoch": 1.4827586206896552, "percentage": 37.07, "elapsed_time": "1:03:37", "remaining_time": "1:48:00"}
173
+ {"current_steps": 346, "total_steps": 928, "loss": 0.4405466318130493, "lr": 1.5379027984008024e-05, "epoch": 1.4913793103448276, "percentage": 37.28, "elapsed_time": "1:03:50", "remaining_time": "1:47:22"}
174
+ {"current_steps": 348, "total_steps": 928, "loss": 0.8692161440849304, "lr": 1.5322840333441687e-05, "epoch": 1.5, "percentage": 37.5, "elapsed_time": "1:04:15", "remaining_time": "1:47:05"}
175
+ {"current_steps": 350, "total_steps": 928, "loss": 1.0229344367980957, "lr": 1.526643280816822e-05, "epoch": 1.5086206896551724, "percentage": 37.72, "elapsed_time": "1:04:41", "remaining_time": "1:46:49"}
176
+ {"current_steps": 352, "total_steps": 928, "loss": 0.9797338843345642, "lr": 1.5209808277270714e-05, "epoch": 1.5172413793103448, "percentage": 37.93, "elapsed_time": "1:05:03", "remaining_time": "1:46:27"}
177
+ {"current_steps": 354, "total_steps": 928, "loss": 1.0694770812988281, "lr": 1.515296962086992e-05, "epoch": 1.5258620689655173, "percentage": 38.15, "elapsed_time": "1:05:28", "remaining_time": "1:46:10"}
178
+ {"current_steps": 356, "total_steps": 928, "loss": 0.8981337547302246, "lr": 1.5095919729977757e-05, "epoch": 1.5344827586206895, "percentage": 38.36, "elapsed_time": "1:05:54", "remaining_time": "1:45:53"}
179
+ {"current_steps": 358, "total_steps": 928, "loss": 0.7348160743713379, "lr": 1.5038661506350267e-05, "epoch": 1.543103448275862, "percentage": 38.58, "elapsed_time": "1:06:10", "remaining_time": "1:45:21"}
180
+ {"current_steps": 360, "total_steps": 928, "loss": 1.3365731239318848, "lr": 1.4981197862340015e-05, "epoch": 1.5517241379310345, "percentage": 38.79, "elapsed_time": "1:06:33", "remaining_time": "1:45:00"}
181
+ {"current_steps": 362, "total_steps": 928, "loss": 1.0422999858856201, "lr": 1.4923531720747975e-05, "epoch": 1.5603448275862069, "percentage": 39.01, "elapsed_time": "1:06:56", "remaining_time": "1:44:39"}
182
+ {"current_steps": 364, "total_steps": 928, "loss": 1.064208984375, "lr": 1.4865666014674833e-05, "epoch": 1.5689655172413794, "percentage": 39.22, "elapsed_time": "1:07:23", "remaining_time": "1:44:24"}
183
+ {"current_steps": 366, "total_steps": 928, "loss": 0.7693328857421875, "lr": 1.4807603687371835e-05, "epoch": 1.5775862068965516, "percentage": 39.44, "elapsed_time": "1:07:43", "remaining_time": "1:43:58"}
184
+ {"current_steps": 368, "total_steps": 928, "loss": 1.014492154121399, "lr": 1.4749347692091061e-05, "epoch": 1.5862068965517242, "percentage": 39.66, "elapsed_time": "1:08:10", "remaining_time": "1:43:44"}
185
+ {"current_steps": 370, "total_steps": 928, "loss": 0.9833155274391174, "lr": 1.4690900991935218e-05, "epoch": 1.5948275862068966, "percentage": 39.87, "elapsed_time": "1:08:32", "remaining_time": "1:43:22"}
186
+ {"current_steps": 372, "total_steps": 928, "loss": 1.0599433183670044, "lr": 1.4632266559706933e-05, "epoch": 1.603448275862069, "percentage": 40.09, "elapsed_time": "1:08:58", "remaining_time": "1:43:04"}
187
+ {"current_steps": 374, "total_steps": 928, "loss": 0.7715789675712585, "lr": 1.457344737775753e-05, "epoch": 1.6120689655172413, "percentage": 40.3, "elapsed_time": "1:09:20", "remaining_time": "1:42:42"}
188
+ {"current_steps": 376, "total_steps": 928, "loss": 1.008685827255249, "lr": 1.451444643783536e-05, "epoch": 1.6206896551724137, "percentage": 40.52, "elapsed_time": "1:09:46", "remaining_time": "1:42:26"}
189
+ {"current_steps": 378, "total_steps": 928, "loss": 1.1693607568740845, "lr": 1.4455266740933598e-05, "epoch": 1.6293103448275863, "percentage": 40.73, "elapsed_time": "1:10:12", "remaining_time": "1:42:09"}
190
+ {"current_steps": 380, "total_steps": 928, "loss": 0.7295485734939575, "lr": 1.439591129713764e-05, "epoch": 1.6379310344827587, "percentage": 40.95, "elapsed_time": "1:10:29", "remaining_time": "1:41:38"}
191
+ {"current_steps": 382, "total_steps": 928, "loss": 1.3163925409317017, "lr": 1.4336383125471974e-05, "epoch": 1.646551724137931, "percentage": 41.16, "elapsed_time": "1:10:54", "remaining_time": "1:41:21"}
192
+ {"current_steps": 384, "total_steps": 928, "loss": 1.00065279006958, "lr": 1.4276685253746623e-05, "epoch": 1.6551724137931034, "percentage": 41.38, "elapsed_time": "1:11:16", "remaining_time": "1:40:58"}
193
+ {"current_steps": 386, "total_steps": 928, "loss": 0.9677169322967529, "lr": 1.4216820718403153e-05, "epoch": 1.6637931034482758, "percentage": 41.59, "elapsed_time": "1:11:39", "remaining_time": "1:40:37"}
194
+ {"current_steps": 388, "total_steps": 928, "loss": 1.3287030458450317, "lr": 1.4156792564360222e-05, "epoch": 1.6724137931034484, "percentage": 41.81, "elapsed_time": "1:12:01", "remaining_time": "1:40:14"}
195
+ {"current_steps": 390, "total_steps": 928, "loss": 0.8481764197349548, "lr": 1.4096603844858696e-05, "epoch": 1.6810344827586206, "percentage": 42.03, "elapsed_time": "1:12:22", "remaining_time": "1:39:51"}
196
+ {"current_steps": 392, "total_steps": 928, "loss": 0.41724586486816406, "lr": 1.4036257621306375e-05, "epoch": 1.6896551724137931, "percentage": 42.24, "elapsed_time": "1:12:36", "remaining_time": "1:39:16"}
197
+ {"current_steps": 394, "total_steps": 928, "loss": 1.0509485006332397, "lr": 1.3975756963122242e-05, "epoch": 1.6982758620689655, "percentage": 42.46, "elapsed_time": "1:13:02", "remaining_time": "1:38:59"}
198
+ {"current_steps": 396, "total_steps": 928, "loss": 1.0144211053848267, "lr": 1.3915104947580382e-05, "epoch": 1.706896551724138, "percentage": 42.67, "elapsed_time": "1:13:28", "remaining_time": "1:38:43"}
199
+ {"current_steps": 398, "total_steps": 928, "loss": 1.0040127038955688, "lr": 1.3854304659653435e-05, "epoch": 1.7155172413793105, "percentage": 42.89, "elapsed_time": "1:13:50", "remaining_time": "1:38:20"}
200
+ {"current_steps": 400, "total_steps": 928, "loss": 1.0181949138641357, "lr": 1.3793359191855684e-05, "epoch": 1.7241379310344827, "percentage": 43.1, "elapsed_time": "1:14:11", "remaining_time": "1:37:56"}
201
+ {"current_steps": 402, "total_steps": 928, "loss": 0.7873343229293823, "lr": 1.3732271644085779e-05, "epoch": 1.7327586206896552, "percentage": 43.32, "elapsed_time": "1:14:30", "remaining_time": "1:37:29"}
202
+ {"current_steps": 404, "total_steps": 928, "loss": 0.594070553779602, "lr": 1.3671045123469038e-05, "epoch": 1.7413793103448276, "percentage": 43.53, "elapsed_time": "1:14:49", "remaining_time": "1:37:03"}
203
+ {"current_steps": 406, "total_steps": 928, "loss": 0.6642279028892517, "lr": 1.3609682744199434e-05, "epoch": 1.75, "percentage": 43.75, "elapsed_time": "1:15:09", "remaining_time": "1:36:37"}
204
+ {"current_steps": 408, "total_steps": 928, "loss": 0.7117359042167664, "lr": 1.354818762738118e-05, "epoch": 1.7586206896551724, "percentage": 43.97, "elapsed_time": "1:15:24", "remaining_time": "1:36:06"}
205
+ {"current_steps": 410, "total_steps": 928, "loss": 0.7889873385429382, "lr": 1.3486562900869975e-05, "epoch": 1.7672413793103448, "percentage": 44.18, "elapsed_time": "1:15:43", "remaining_time": "1:35:40"}
206
+ {"current_steps": 412, "total_steps": 928, "loss": 0.7825038433074951, "lr": 1.3424811699113924e-05, "epoch": 1.7758620689655173, "percentage": 44.4, "elapsed_time": "1:16:08", "remaining_time": "1:35:22"}
207
+ {"current_steps": 414, "total_steps": 928, "loss": 0.6144068837165833, "lr": 1.3362937162994115e-05, "epoch": 1.7844827586206895, "percentage": 44.61, "elapsed_time": "1:16:28", "remaining_time": "1:34:57"}
208
+ {"current_steps": 416, "total_steps": 928, "loss": 0.7337537407875061, "lr": 1.3300942439664826e-05, "epoch": 1.793103448275862, "percentage": 44.83, "elapsed_time": "1:16:46", "remaining_time": "1:34:29"}
209
+ {"current_steps": 418, "total_steps": 928, "loss": 0.9792155027389526, "lr": 1.3238830682393498e-05, "epoch": 1.8017241379310345, "percentage": 45.04, "elapsed_time": "1:17:12", "remaining_time": "1:34:12"}
210
+ {"current_steps": 420, "total_steps": 928, "loss": 1.0592936277389526, "lr": 1.317660505040031e-05, "epoch": 1.8103448275862069, "percentage": 45.26, "elapsed_time": "1:17:37", "remaining_time": "1:33:53"}
211
+ {"current_steps": 422, "total_steps": 928, "loss": 1.1423420906066895, "lr": 1.3114268708697517e-05, "epoch": 1.8189655172413794, "percentage": 45.47, "elapsed_time": "1:18:04", "remaining_time": "1:33:36"}
212
+ {"current_steps": 424, "total_steps": 928, "loss": 1.0336556434631348, "lr": 1.305182482792845e-05, "epoch": 1.8275862068965516, "percentage": 45.69, "elapsed_time": "1:18:23", "remaining_time": "1:33:11"}
213
+ {"current_steps": 426, "total_steps": 928, "loss": 0.633764386177063, "lr": 1.298927658420625e-05, "epoch": 1.8362068965517242, "percentage": 45.91, "elapsed_time": "1:18:37", "remaining_time": "1:32:39"}
214
+ {"current_steps": 428, "total_steps": 928, "loss": 1.000012993812561, "lr": 1.292662715895232e-05, "epoch": 1.8448275862068966, "percentage": 46.12, "elapsed_time": "1:19:03", "remaining_time": "1:32:21"}
215
+ {"current_steps": 430, "total_steps": 928, "loss": 0.972018301486969, "lr": 1.2863879738734515e-05, "epoch": 1.853448275862069, "percentage": 46.34, "elapsed_time": "1:19:31", "remaining_time": "1:32:06"}
216
+ {"current_steps": 432, "total_steps": 928, "loss": 1.0186474323272705, "lr": 1.2801037515105048e-05, "epoch": 1.8620689655172413, "percentage": 46.55, "elapsed_time": "1:19:53", "remaining_time": "1:31:43"}
217
+ {"current_steps": 434, "total_steps": 928, "loss": 1.1112819910049438, "lr": 1.2738103684438167e-05, "epoch": 1.8706896551724137, "percentage": 46.77, "elapsed_time": "1:20:16", "remaining_time": "1:31:22"}
218
+ {"current_steps": 436, "total_steps": 928, "loss": 0.6823700070381165, "lr": 1.267508144776757e-05, "epoch": 1.8793103448275863, "percentage": 46.98, "elapsed_time": "1:20:30", "remaining_time": "1:30:50"}
219
+ {"current_steps": 438, "total_steps": 928, "loss": 0.9664647579193115, "lr": 1.2611974010623598e-05, "epoch": 1.8879310344827587, "percentage": 47.2, "elapsed_time": "1:20:53", "remaining_time": "1:30:30"}
220
+ {"current_steps": 440, "total_steps": 928, "loss": 0.9077426791191101, "lr": 1.2548784582870175e-05, "epoch": 1.896551724137931, "percentage": 47.41, "elapsed_time": "1:21:15", "remaining_time": "1:30:07"}
221
+ {"current_steps": 442, "total_steps": 928, "loss": 1.095183253288269, "lr": 1.2485516378541562e-05, "epoch": 1.9051724137931034, "percentage": 47.63, "elapsed_time": "1:21:41", "remaining_time": "1:29:49"}
222
+ {"current_steps": 444, "total_steps": 928, "loss": 0.2406305968761444, "lr": 1.2422172615678867e-05, "epoch": 1.9137931034482758, "percentage": 47.84, "elapsed_time": "1:21:55", "remaining_time": "1:29:18"}
223
+ {"current_steps": 446, "total_steps": 928, "loss": 0.7011273503303528, "lr": 1.2358756516166368e-05, "epoch": 1.9224137931034484, "percentage": 48.06, "elapsed_time": "1:22:16", "remaining_time": "1:28:54"}
224
+ {"current_steps": 448, "total_steps": 928, "loss": 1.040317177772522, "lr": 1.2295271305567637e-05, "epoch": 1.9310344827586206, "percentage": 48.28, "elapsed_time": "1:22:39", "remaining_time": "1:28:33"}
225
+ {"current_steps": 450, "total_steps": 928, "loss": 1.0238956212997437, "lr": 1.2231720212961476e-05, "epoch": 1.9396551724137931, "percentage": 48.49, "elapsed_time": "1:23:06", "remaining_time": "1:28:16"}
226
+ {"current_steps": 452, "total_steps": 928, "loss": 0.8691494464874268, "lr": 1.2168106470777676e-05, "epoch": 1.9482758620689655, "percentage": 48.71, "elapsed_time": "1:23:26", "remaining_time": "1:27:52"}
227
+ {"current_steps": 454, "total_steps": 928, "loss": 1.03307044506073, "lr": 1.2104433314632605e-05, "epoch": 1.956896551724138, "percentage": 48.92, "elapsed_time": "1:23:48", "remaining_time": "1:27:30"}
228
+ {"current_steps": 456, "total_steps": 928, "loss": 1.0065069198608398, "lr": 1.2040703983164628e-05, "epoch": 1.9655172413793105, "percentage": 49.14, "elapsed_time": "1:24:15", "remaining_time": "1:27:12"}
229
+ {"current_steps": 458, "total_steps": 928, "loss": 0.3249227702617645, "lr": 1.1976921717869387e-05, "epoch": 1.9741379310344827, "percentage": 49.35, "elapsed_time": "1:24:25", "remaining_time": "1:26:38"}
230
+ {"current_steps": 460, "total_steps": 928, "loss": 0.9767223596572876, "lr": 1.1913089762934918e-05, "epoch": 1.9827586206896552, "percentage": 49.57, "elapsed_time": "1:24:52", "remaining_time": "1:26:20"}
231
+ {"current_steps": 462, "total_steps": 928, "loss": 0.6757299304008484, "lr": 1.184921136507665e-05, "epoch": 1.9913793103448276, "percentage": 49.78, "elapsed_time": "1:25:12", "remaining_time": "1:25:56"}
232
+ {"current_steps": 464, "total_steps": 928, "loss": 0.8873772621154785, "lr": 1.1785289773372252e-05, "epoch": 2.0, "percentage": 50.0, "elapsed_time": "1:25:35", "remaining_time": "1:25:35"}
233
+ {"current_steps": 466, "total_steps": 928, "loss": 0.4577694237232208, "lr": 1.1721328239096397e-05, "epoch": 2.0086206896551726, "percentage": 50.22, "elapsed_time": "1:26:04", "remaining_time": "1:25:20"}
234
+ {"current_steps": 468, "total_steps": 928, "loss": 0.729465663433075, "lr": 1.1657330015555365e-05, "epoch": 2.0172413793103448, "percentage": 50.43, "elapsed_time": "1:26:28", "remaining_time": "1:24:59"}
235
+ {"current_steps": 470, "total_steps": 928, "loss": 0.58217453956604, "lr": 1.159329835792158e-05, "epoch": 2.0258620689655173, "percentage": 50.65, "elapsed_time": "1:26:48", "remaining_time": "1:24:35"}
236
+ {"current_steps": 472, "total_steps": 928, "loss": 0.5115339159965515, "lr": 1.1529236523068046e-05, "epoch": 2.0344827586206895, "percentage": 50.86, "elapsed_time": "1:27:04", "remaining_time": "1:24:07"}
237
+ {"current_steps": 474, "total_steps": 928, "loss": 0.5147908329963684, "lr": 1.146514776940268e-05, "epoch": 2.043103448275862, "percentage": 51.08, "elapsed_time": "1:27:25", "remaining_time": "1:23:43"}
238
+ {"current_steps": 476, "total_steps": 928, "loss": 0.5318877100944519, "lr": 1.1401035356702585e-05, "epoch": 2.0517241379310347, "percentage": 51.29, "elapsed_time": "1:27:44", "remaining_time": "1:23:18"}
239
+ {"current_steps": 478, "total_steps": 928, "loss": 0.3974412977695465, "lr": 1.133690254594825e-05, "epoch": 2.060344827586207, "percentage": 51.51, "elapsed_time": "1:28:02", "remaining_time": "1:22:53"}
240
+ {"current_steps": 480, "total_steps": 928, "loss": 0.7926555871963501, "lr": 1.1272752599157673e-05, "epoch": 2.0689655172413794, "percentage": 51.72, "elapsed_time": "1:28:29", "remaining_time": "1:22:35"}
241
+ {"current_steps": 482, "total_steps": 928, "loss": 0.4678128659725189, "lr": 1.1208588779220454e-05, "epoch": 2.0775862068965516, "percentage": 51.94, "elapsed_time": "1:28:48", "remaining_time": "1:22:10"}
242
+ {"current_steps": 484, "total_steps": 928, "loss": 0.6200590133666992, "lr": 1.1144414349731826e-05, "epoch": 2.086206896551724, "percentage": 52.16, "elapsed_time": "1:29:13", "remaining_time": "1:21:51"}
243
+ {"current_steps": 486, "total_steps": 928, "loss": 0.5513038039207458, "lr": 1.1080232574826669e-05, "epoch": 2.0948275862068964, "percentage": 52.37, "elapsed_time": "1:29:36", "remaining_time": "1:21:29"}
244
+ {"current_steps": 488, "total_steps": 928, "loss": 0.6488674283027649, "lr": 1.101604671901346e-05, "epoch": 2.103448275862069, "percentage": 52.59, "elapsed_time": "1:30:00", "remaining_time": "1:21:09"}
245
+ {"current_steps": 490, "total_steps": 928, "loss": 0.6901264786720276, "lr": 1.0951860047008268e-05, "epoch": 2.1120689655172415, "percentage": 52.8, "elapsed_time": "1:30:26", "remaining_time": "1:20:50"}
246
+ {"current_steps": 492, "total_steps": 928, "loss": 0.6872219443321228, "lr": 1.0887675823568662e-05, "epoch": 2.1206896551724137, "percentage": 53.02, "elapsed_time": "1:30:50", "remaining_time": "1:20:29"}
247
+ {"current_steps": 494, "total_steps": 928, "loss": 0.7422826886177063, "lr": 1.0823497313327666e-05, "epoch": 2.1293103448275863, "percentage": 53.23, "elapsed_time": "1:31:16", "remaining_time": "1:20:11"}
248
+ {"current_steps": 496, "total_steps": 928, "loss": 0.6594029664993286, "lr": 1.075932778062772e-05, "epoch": 2.1379310344827585, "percentage": 53.45, "elapsed_time": "1:31:37", "remaining_time": "1:19:48"}
249
+ {"current_steps": 498, "total_steps": 928, "loss": 0.3747510313987732, "lr": 1.0695170489354627e-05, "epoch": 2.146551724137931, "percentage": 53.66, "elapsed_time": "1:31:52", "remaining_time": "1:19:20"}
250
+ {"current_steps": 500, "total_steps": 928, "loss": 0.3904026746749878, "lr": 1.0631028702771558e-05, "epoch": 2.1551724137931036, "percentage": 53.88, "elapsed_time": "1:32:11", "remaining_time": "1:18:55"}
251
+ {"current_steps": 502, "total_steps": 928, "loss": 0.42320704460144043, "lr": 1.0566905683353048e-05, "epoch": 2.163793103448276, "percentage": 54.09, "elapsed_time": "1:32:30", "remaining_time": "1:18:30"}
252
+ {"current_steps": 504, "total_steps": 928, "loss": 0.5195796489715576, "lr": 1.0502804692619085e-05, "epoch": 2.1724137931034484, "percentage": 54.31, "elapsed_time": "1:32:45", "remaining_time": "1:18:02"}
253
+ {"current_steps": 506, "total_steps": 928, "loss": 0.7010811567306519, "lr": 1.0438728990969182e-05, "epoch": 2.1810344827586206, "percentage": 54.53, "elapsed_time": "1:33:11", "remaining_time": "1:17:43"}
254
+ {"current_steps": 508, "total_steps": 928, "loss": 0.47865381836891174, "lr": 1.0374681837516585e-05, "epoch": 2.189655172413793, "percentage": 54.74, "elapsed_time": "1:33:32", "remaining_time": "1:17:19"}
255
+ {"current_steps": 510, "total_steps": 928, "loss": 0.7136699557304382, "lr": 1.0310666489922452e-05, "epoch": 2.1982758620689653, "percentage": 54.96, "elapsed_time": "1:33:54", "remaining_time": "1:16:58"}
256
+ {"current_steps": 512, "total_steps": 928, "loss": 0.5507462024688721, "lr": 1.0246686204230213e-05, "epoch": 2.206896551724138, "percentage": 55.17, "elapsed_time": "1:34:08", "remaining_time": "1:16:29"}
257
+ {"current_steps": 514, "total_steps": 928, "loss": 0.5582562685012817, "lr": 1.0182744234699897e-05, "epoch": 2.2155172413793105, "percentage": 55.39, "elapsed_time": "1:34:26", "remaining_time": "1:16:04"}
258
+ {"current_steps": 516, "total_steps": 928, "loss": 0.403819739818573, "lr": 1.0118843833642672e-05, "epoch": 2.2241379310344827, "percentage": 55.6, "elapsed_time": "1:34:46", "remaining_time": "1:15:40"}
259
+ {"current_steps": 518, "total_steps": 928, "loss": 0.0842498317360878, "lr": 1.005498825125536e-05, "epoch": 2.2327586206896552, "percentage": 55.82, "elapsed_time": "1:34:58", "remaining_time": "1:15:10"}
260
+ {"current_steps": 520, "total_steps": 928, "loss": 0.14793738722801208, "lr": 9.991180735455171e-06, "epoch": 2.2413793103448274, "percentage": 56.03, "elapsed_time": "1:35:16", "remaining_time": "1:14:45"}
261
+ {"current_steps": 522, "total_steps": 928, "loss": 0.7687300443649292, "lr": 9.927424531714463e-06, "epoch": 2.25, "percentage": 56.25, "elapsed_time": "1:35:42", "remaining_time": "1:14:26"}
262
+ {"current_steps": 524, "total_steps": 928, "loss": 0.5136147141456604, "lr": 9.863722882895703e-06, "epoch": 2.2586206896551726, "percentage": 56.47, "elapsed_time": "1:35:59", "remaining_time": "1:14:00"}
263
+ {"current_steps": 526, "total_steps": 928, "loss": 0.48577654361724854, "lr": 9.800079029086487e-06, "epoch": 2.2672413793103448, "percentage": 56.68, "elapsed_time": "1:36:19", "remaining_time": "1:13:36"}
264
+ {"current_steps": 528, "total_steps": 928, "loss": 0.5481493473052979, "lr": 9.736496207434767e-06, "epoch": 2.2758620689655173, "percentage": 56.9, "elapsed_time": "1:36:39", "remaining_time": "1:13:13"}
265
+ {"current_steps": 530, "total_steps": 928, "loss": 0.4460916519165039, "lr": 9.672977651984183e-06, "epoch": 2.2844827586206895, "percentage": 57.11, "elapsed_time": "1:36:58", "remaining_time": "1:12:49"}
266
+ {"current_steps": 532, "total_steps": 928, "loss": 0.44660893082618713, "lr": 9.609526593509572e-06, "epoch": 2.293103448275862, "percentage": 57.33, "elapsed_time": "1:37:16", "remaining_time": "1:12:24"}
267
+ {"current_steps": 534, "total_steps": 928, "loss": 0.4977568984031677, "lr": 9.54614625935264e-06, "epoch": 2.3017241379310347, "percentage": 57.54, "elapsed_time": "1:37:37", "remaining_time": "1:12:01"}
268
+ {"current_steps": 536, "total_steps": 928, "loss": 0.44564345479011536, "lr": 9.482839873257814e-06, "epoch": 2.310344827586207, "percentage": 57.76, "elapsed_time": "1:37:58", "remaining_time": "1:11:39"}
269
+ {"current_steps": 538, "total_steps": 928, "loss": 0.34829026460647583, "lr": 9.41961065520825e-06, "epoch": 2.3189655172413794, "percentage": 57.97, "elapsed_time": "1:38:10", "remaining_time": "1:11:10"}
270
+ {"current_steps": 540, "total_steps": 928, "loss": 0.42181143164634705, "lr": 9.356461821262088e-06, "epoch": 2.3275862068965516, "percentage": 58.19, "elapsed_time": "1:38:28", "remaining_time": "1:10:45"}
271
+ {"current_steps": 542, "total_steps": 928, "loss": 0.6050075888633728, "lr": 9.293396583388835e-06, "epoch": 2.336206896551724, "percentage": 58.41, "elapsed_time": "1:38:53", "remaining_time": "1:10:25"}
272
+ {"current_steps": 544, "total_steps": 928, "loss": 0.7277567982673645, "lr": 9.230418149306028e-06, "epoch": 2.344827586206897, "percentage": 58.62, "elapsed_time": "1:39:17", "remaining_time": "1:10:05"}
273
+ {"current_steps": 546, "total_steps": 928, "loss": 0.25289204716682434, "lr": 9.167529722316039e-06, "epoch": 2.353448275862069, "percentage": 58.84, "elapsed_time": "1:39:31", "remaining_time": "1:09:37"}
274
+ {"current_steps": 548, "total_steps": 928, "loss": 0.732049822807312, "lr": 9.104734501143195e-06, "epoch": 2.3620689655172415, "percentage": 59.05, "elapsed_time": "1:39:53", "remaining_time": "1:09:16"}
275
+ {"current_steps": 550, "total_steps": 928, "loss": 0.561985969543457, "lr": 9.04203567977102e-06, "epoch": 2.3706896551724137, "percentage": 59.27, "elapsed_time": "1:40:11", "remaining_time": "1:08:51"}
276
+ {"current_steps": 552, "total_steps": 928, "loss": 0.7146962285041809, "lr": 8.979436447279833e-06, "epoch": 2.3793103448275863, "percentage": 59.48, "elapsed_time": "1:40:36", "remaining_time": "1:08:32"}
277
+ {"current_steps": 554, "total_steps": 928, "loss": 0.5743066072463989, "lr": 8.916939987684497e-06, "epoch": 2.3879310344827585, "percentage": 59.7, "elapsed_time": "1:40:59", "remaining_time": "1:08:10"}
278
+ {"current_steps": 556, "total_steps": 928, "loss": 0.6506761312484741, "lr": 8.854549479772508e-06, "epoch": 2.396551724137931, "percentage": 59.91, "elapsed_time": "1:41:24", "remaining_time": "1:07:51"}
279
+ {"current_steps": 558, "total_steps": 928, "loss": 0.4716731607913971, "lr": 8.792268096942265e-06, "epoch": 2.405172413793103, "percentage": 60.13, "elapsed_time": "1:41:41", "remaining_time": "1:07:25"}
280
+ {"current_steps": 560, "total_steps": 928, "loss": 0.6651344299316406, "lr": 8.730099007041712e-06, "epoch": 2.413793103448276, "percentage": 60.34, "elapsed_time": "1:41:59", "remaining_time": "1:07:01"}
281
+ {"current_steps": 562, "total_steps": 928, "loss": 0.5968424081802368, "lr": 8.668045372207165e-06, "epoch": 2.4224137931034484, "percentage": 60.56, "elapsed_time": "1:42:25", "remaining_time": "1:06:42"}
282
+ {"current_steps": 564, "total_steps": 928, "loss": 0.656068742275238, "lr": 8.606110348702502e-06, "epoch": 2.4310344827586206, "percentage": 60.78, "elapsed_time": "1:42:43", "remaining_time": "1:06:18"}
283
+ {"current_steps": 566, "total_steps": 928, "loss": 0.6287741661071777, "lr": 8.544297086758612e-06, "epoch": 2.439655172413793, "percentage": 60.99, "elapsed_time": "1:43:08", "remaining_time": "1:05:58"}
284
+ {"current_steps": 568, "total_steps": 928, "loss": 0.3020762503147125, "lr": 8.482608730413173e-06, "epoch": 2.4482758620689653, "percentage": 61.21, "elapsed_time": "1:43:23", "remaining_time": "1:05:31"}
285
+ {"current_steps": 570, "total_steps": 928, "loss": 1.0642119646072388, "lr": 8.421048417350722e-06, "epoch": 2.456896551724138, "percentage": 61.42, "elapsed_time": "1:43:49", "remaining_time": "1:05:12"}
286
+ {"current_steps": 572, "total_steps": 928, "loss": 0.46571430563926697, "lr": 8.359619278743072e-06, "epoch": 2.4655172413793105, "percentage": 61.64, "elapsed_time": "1:44:08", "remaining_time": "1:04:48"}
287
+ {"current_steps": 574, "total_steps": 928, "loss": 0.5474472641944885, "lr": 8.298324439090044e-06, "epoch": 2.4741379310344827, "percentage": 61.85, "elapsed_time": "1:44:31", "remaining_time": "1:04:27"}
288
+ {"current_steps": 576, "total_steps": 928, "loss": 0.7401635050773621, "lr": 8.23716701606056e-06, "epoch": 2.4827586206896552, "percentage": 62.07, "elapsed_time": "1:44:57", "remaining_time": "1:04:08"}
289
+ {"current_steps": 578, "total_steps": 928, "loss": 0.4356614053249359, "lr": 8.176150120334035e-06, "epoch": 2.4913793103448274, "percentage": 62.28, "elapsed_time": "1:45:17", "remaining_time": "1:03:45"}
290
+ {"current_steps": 580, "total_steps": 928, "loss": 0.7132285237312317, "lr": 8.115276855442186e-06, "epoch": 2.5, "percentage": 62.5, "elapsed_time": "1:45:44", "remaining_time": "1:03:26"}
291
+ {"current_steps": 582, "total_steps": 928, "loss": 0.6939521431922913, "lr": 8.054550317611166e-06, "epoch": 2.5086206896551726, "percentage": 62.72, "elapsed_time": "1:46:10", "remaining_time": "1:03:07"}
292
+ {"current_steps": 584, "total_steps": 928, "loss": 0.5915881395339966, "lr": 7.99397359560408e-06, "epoch": 2.5172413793103448, "percentage": 62.93, "elapsed_time": "1:46:28", "remaining_time": "1:02:43"}
293
+ {"current_steps": 586, "total_steps": 928, "loss": 0.5379845499992371, "lr": 7.933549770563868e-06, "epoch": 2.5258620689655173, "percentage": 63.15, "elapsed_time": "1:46:53", "remaining_time": "1:02:23"}
294
+ {"current_steps": 588, "total_steps": 928, "loss": 0.6815016269683838, "lr": 7.873281915856618e-06, "epoch": 2.5344827586206895, "percentage": 63.36, "elapsed_time": "1:47:20", "remaining_time": "1:02:03"}
295
+ {"current_steps": 590, "total_steps": 928, "loss": 0.7555568218231201, "lr": 7.8131730969152e-06, "epoch": 2.543103448275862, "percentage": 63.58, "elapsed_time": "1:47:43", "remaining_time": "1:01:43"}
296
+ {"current_steps": 592, "total_steps": 928, "loss": 0.6491318941116333, "lr": 7.753226371083398e-06, "epoch": 2.5517241379310347, "percentage": 63.79, "elapsed_time": "1:48:05", "remaining_time": "1:01:21"}
297
+ {"current_steps": 594, "total_steps": 928, "loss": 0.38945457339286804, "lr": 7.693444787460346e-06, "epoch": 2.560344827586207, "percentage": 64.01, "elapsed_time": "1:48:25", "remaining_time": "1:00:58"}
298
+ {"current_steps": 596, "total_steps": 928, "loss": 0.5126080513000488, "lr": 7.6338313867455e-06, "epoch": 2.5689655172413794, "percentage": 64.22, "elapsed_time": "1:48:40", "remaining_time": "1:00:32"}
299
+ {"current_steps": 598, "total_steps": 928, "loss": 0.6170058250427246, "lr": 7.574389201083925e-06, "epoch": 2.5775862068965516, "percentage": 64.44, "elapsed_time": "1:49:06", "remaining_time": "1:00:12"}
300
+ {"current_steps": 600, "total_steps": 928, "loss": 0.6505869030952454, "lr": 7.515121253912107e-06, "epoch": 2.586206896551724, "percentage": 64.66, "elapsed_time": "1:49:31", "remaining_time": "0:59:52"}
301
+ {"current_steps": 602, "total_steps": 928, "loss": 0.47480010986328125, "lr": 7.456030559804146e-06, "epoch": 2.594827586206897, "percentage": 64.87, "elapsed_time": "1:49:49", "remaining_time": "0:59:28"}
302
+ {"current_steps": 604, "total_steps": 928, "loss": 0.5783487558364868, "lr": 7.3971201243184425e-06, "epoch": 2.603448275862069, "percentage": 65.09, "elapsed_time": "1:50:15", "remaining_time": "0:59:08"}
303
+ {"current_steps": 606, "total_steps": 928, "loss": 0.5066084265708923, "lr": 7.338392943844807e-06, "epoch": 2.612068965517241, "percentage": 65.3, "elapsed_time": "1:50:34", "remaining_time": "0:58:45"}
304
+ {"current_steps": 608, "total_steps": 928, "loss": 0.7519516348838806, "lr": 7.279852005452072e-06, "epoch": 2.6206896551724137, "percentage": 65.52, "elapsed_time": "1:51:01", "remaining_time": "0:58:26"}
305
+ {"current_steps": 610, "total_steps": 928, "loss": 0.6951370239257812, "lr": 7.221500286736145e-06, "epoch": 2.6293103448275863, "percentage": 65.73, "elapsed_time": "1:51:19", "remaining_time": "0:58:02"}
306
+ {"current_steps": 612, "total_steps": 928, "loss": 0.7426925301551819, "lr": 7.16334075566856e-06, "epoch": 2.637931034482759, "percentage": 65.95, "elapsed_time": "1:51:42", "remaining_time": "0:57:40"}
307
+ {"current_steps": 614, "total_steps": 928, "loss": 0.6338915824890137, "lr": 7.105376370445516e-06, "epoch": 2.646551724137931, "percentage": 66.16, "elapsed_time": "1:52:07", "remaining_time": "0:57:20"}
308
+ {"current_steps": 616, "total_steps": 928, "loss": 0.43320754170417786, "lr": 7.047610079337426e-06, "epoch": 2.655172413793103, "percentage": 66.38, "elapsed_time": "1:52:27", "remaining_time": "0:56:57"}
309
+ {"current_steps": 618, "total_steps": 928, "loss": 0.7193683981895447, "lr": 6.990044820538932e-06, "epoch": 2.663793103448276, "percentage": 66.59, "elapsed_time": "1:52:54", "remaining_time": "0:56:38"}
310
+ {"current_steps": 620, "total_steps": 928, "loss": 0.4045131802558899, "lr": 6.932683522019486e-06, "epoch": 2.6724137931034484, "percentage": 66.81, "elapsed_time": "1:53:12", "remaining_time": "0:56:14"}
311
+ {"current_steps": 622, "total_steps": 928, "loss": 0.7818766832351685, "lr": 6.875529101374407e-06, "epoch": 2.6810344827586206, "percentage": 67.03, "elapsed_time": "1:53:38", "remaining_time": "0:55:54"}
312
+ {"current_steps": 624, "total_steps": 928, "loss": 0.4748293161392212, "lr": 6.818584465676476e-06, "epoch": 2.689655172413793, "percentage": 67.24, "elapsed_time": "1:53:59", "remaining_time": "0:55:32"}
313
+ {"current_steps": 626, "total_steps": 928, "loss": 0.6603924036026001, "lr": 6.761852511328097e-06, "epoch": 2.6982758620689653, "percentage": 67.46, "elapsed_time": "1:54:22", "remaining_time": "0:55:10"}
314
+ {"current_steps": 628, "total_steps": 928, "loss": 0.4325425326824188, "lr": 6.705336123913943e-06, "epoch": 2.706896551724138, "percentage": 67.67, "elapsed_time": "1:54:42", "remaining_time": "0:54:47"}
315
+ {"current_steps": 630, "total_steps": 928, "loss": 0.8396017551422119, "lr": 6.649038178054216e-06, "epoch": 2.7155172413793105, "percentage": 67.89, "elapsed_time": "1:55:10", "remaining_time": "0:54:28"}
316
+ {"current_steps": 632, "total_steps": 928, "loss": 0.34978973865509033, "lr": 6.592961537258415e-06, "epoch": 2.7241379310344827, "percentage": 68.1, "elapsed_time": "1:55:27", "remaining_time": "0:54:04"}
317
+ {"current_steps": 634, "total_steps": 928, "loss": 0.7622851133346558, "lr": 6.53710905377969e-06, "epoch": 2.7327586206896552, "percentage": 68.32, "elapsed_time": "1:55:49", "remaining_time": "0:53:42"}
318
+ {"current_steps": 636, "total_steps": 928, "loss": 0.5066753625869751, "lr": 6.48148356846978e-06, "epoch": 2.7413793103448274, "percentage": 68.53, "elapsed_time": "1:56:09", "remaining_time": "0:53:20"}
319
+ {"current_steps": 638, "total_steps": 928, "loss": 0.7104685306549072, "lr": 6.426087910634488e-06, "epoch": 2.75, "percentage": 68.75, "elapsed_time": "1:56:32", "remaining_time": "0:52:58"}
320
+ {"current_steps": 640, "total_steps": 928, "loss": 0.45420828461647034, "lr": 6.370924897889815e-06, "epoch": 2.7586206896551726, "percentage": 68.97, "elapsed_time": "1:56:50", "remaining_time": "0:52:34"}
321
+ {"current_steps": 642, "total_steps": 928, "loss": 0.6258912086486816, "lr": 6.315997336018602e-06, "epoch": 2.7672413793103448, "percentage": 69.18, "elapsed_time": "1:57:11", "remaining_time": "0:52:12"}
322
+ {"current_steps": 644, "total_steps": 928, "loss": 0.8025404214859009, "lr": 6.261308018827857e-06, "epoch": 2.7758620689655173, "percentage": 69.4, "elapsed_time": "1:57:37", "remaining_time": "0:51:52"}
323
+ {"current_steps": 646, "total_steps": 928, "loss": 0.6612002849578857, "lr": 6.206859728006616e-06, "epoch": 2.7844827586206895, "percentage": 69.61, "elapsed_time": "1:57:59", "remaining_time": "0:51:30"}
324
+ {"current_steps": 648, "total_steps": 928, "loss": 0.3472132980823517, "lr": 6.152655232984493e-06, "epoch": 2.793103448275862, "percentage": 69.83, "elapsed_time": "1:58:19", "remaining_time": "0:51:07"}
325
+ {"current_steps": 650, "total_steps": 928, "loss": 0.5596845746040344, "lr": 6.0986972907907805e-06, "epoch": 2.8017241379310347, "percentage": 70.04, "elapsed_time": "1:58:41", "remaining_time": "0:50:45"}
326
+ {"current_steps": 652, "total_steps": 928, "loss": 0.7260704636573792, "lr": 6.044988645914256e-06, "epoch": 2.810344827586207, "percentage": 70.26, "elapsed_time": "1:59:05", "remaining_time": "0:50:24"}
327
+ {"current_steps": 654, "total_steps": 928, "loss": 0.7274701595306396, "lr": 5.991532030163549e-06, "epoch": 2.8189655172413794, "percentage": 70.47, "elapsed_time": "1:59:32", "remaining_time": "0:50:04"}
328
+ {"current_steps": 656, "total_steps": 928, "loss": 0.5958088636398315, "lr": 5.938330162528225e-06, "epoch": 2.8275862068965516, "percentage": 70.69, "elapsed_time": "1:59:52", "remaining_time": "0:49:42"}
329
+ {"current_steps": 658, "total_steps": 928, "loss": 0.7417409420013428, "lr": 5.8853857490404605e-06, "epoch": 2.836206896551724, "percentage": 70.91, "elapsed_time": "2:00:18", "remaining_time": "0:49:22"}
330
+ {"current_steps": 660, "total_steps": 928, "loss": 0.19964106380939484, "lr": 5.832701482637433e-06, "epoch": 2.844827586206897, "percentage": 71.12, "elapsed_time": "2:00:33", "remaining_time": "0:48:57"}
331
+ {"current_steps": 662, "total_steps": 928, "loss": 0.3988272547721863, "lr": 5.780280043024316e-06, "epoch": 2.853448275862069, "percentage": 71.34, "elapsed_time": "2:00:46", "remaining_time": "0:48:31"}
332
+ {"current_steps": 664, "total_steps": 928, "loss": 0.4928115904331207, "lr": 5.728124096538014e-06, "epoch": 2.862068965517241, "percentage": 71.55, "elapsed_time": "2:01:09", "remaining_time": "0:48:10"}
333
+ {"current_steps": 666, "total_steps": 928, "loss": 0.42225298285484314, "lr": 5.676236296011513e-06, "epoch": 2.8706896551724137, "percentage": 71.77, "elapsed_time": "2:01:29", "remaining_time": "0:47:47"}
334
+ {"current_steps": 668, "total_steps": 928, "loss": 0.36263778805732727, "lr": 5.624619280638979e-06, "epoch": 2.8793103448275863, "percentage": 71.98, "elapsed_time": "2:01:49", "remaining_time": "0:47:24"}
335
+ {"current_steps": 670, "total_steps": 928, "loss": 0.720154345035553, "lr": 5.573275675841481e-06, "epoch": 2.887931034482759, "percentage": 72.2, "elapsed_time": "2:02:15", "remaining_time": "0:47:04"}
336
+ {"current_steps": 672, "total_steps": 928, "loss": 0.44341281056404114, "lr": 5.522208093133492e-06, "epoch": 2.896551724137931, "percentage": 72.41, "elapsed_time": "2:02:30", "remaining_time": "0:46:40"}
337
+ {"current_steps": 674, "total_steps": 928, "loss": 0.712740957736969, "lr": 5.4714191299900325e-06, "epoch": 2.905172413793103, "percentage": 72.63, "elapsed_time": "2:02:48", "remaining_time": "0:46:16"}
338
+ {"current_steps": 676, "total_steps": 928, "loss": 0.7482072710990906, "lr": 5.4209113697145656e-06, "epoch": 2.913793103448276, "percentage": 72.84, "elapsed_time": "2:03:14", "remaining_time": "0:45:56"}
339
+ {"current_steps": 678, "total_steps": 928, "loss": 0.6825791001319885, "lr": 5.370687381307591e-06, "epoch": 2.9224137931034484, "percentage": 73.06, "elapsed_time": "2:03:38", "remaining_time": "0:45:35"}
340
+ {"current_steps": 680, "total_steps": 928, "loss": 0.5156247615814209, "lr": 5.320749719335997e-06, "epoch": 2.9310344827586206, "percentage": 73.28, "elapsed_time": "2:03:57", "remaining_time": "0:45:12"}
341
+ {"current_steps": 682, "total_steps": 928, "loss": 0.29097798466682434, "lr": 5.271100923803095e-06, "epoch": 2.939655172413793, "percentage": 73.49, "elapsed_time": "2:04:07", "remaining_time": "0:44:46"}
342
+ {"current_steps": 684, "total_steps": 928, "loss": 0.42494314908981323, "lr": 5.2217435200194595e-06, "epoch": 2.9482758620689653, "percentage": 73.71, "elapsed_time": "2:04:26", "remaining_time": "0:44:23"}
343
+ {"current_steps": 686, "total_steps": 928, "loss": 0.6773520708084106, "lr": 5.172680018474456e-06, "epoch": 2.956896551724138, "percentage": 73.92, "elapsed_time": "2:04:49", "remaining_time": "0:44:02"}
344
+ {"current_steps": 688, "total_steps": 928, "loss": 0.6991716027259827, "lr": 5.123912914708566e-06, "epoch": 2.9655172413793105, "percentage": 74.14, "elapsed_time": "2:05:11", "remaining_time": "0:43:40"}
345
+ {"current_steps": 690, "total_steps": 928, "loss": 0.9952356815338135, "lr": 5.075444689186439e-06, "epoch": 2.9741379310344827, "percentage": 74.35, "elapsed_time": "2:05:37", "remaining_time": "0:43:19"}
346
+ {"current_steps": 692, "total_steps": 928, "loss": 0.9628766179084778, "lr": 5.02727780717075e-06, "epoch": 2.9827586206896552, "percentage": 74.57, "elapsed_time": "2:06:03", "remaining_time": "0:42:59"}
347
+ {"current_steps": 694, "total_steps": 928, "loss": 0.46527406573295593, "lr": 4.979414718596779e-06, "epoch": 2.9913793103448274, "percentage": 74.78, "elapsed_time": "2:06:21", "remaining_time": "0:42:36"}
348
+ {"current_steps": 696, "total_steps": 928, "loss": 0.8153132200241089, "lr": 4.931857857947829e-06, "epoch": 3.0, "percentage": 75.0, "elapsed_time": "2:06:47", "remaining_time": "0:42:15"}
349
+ {"current_steps": 698, "total_steps": 928, "loss": 0.2907155454158783, "lr": 4.884609644131373e-06, "epoch": 3.0086206896551726, "percentage": 75.22, "elapsed_time": "2:07:08", "remaining_time": "0:41:53"}
350
+ {"current_steps": 700, "total_steps": 928, "loss": 0.4242869019508362, "lr": 4.837672480356037e-06, "epoch": 3.0172413793103448, "percentage": 75.43, "elapsed_time": "2:07:35", "remaining_time": "0:41:33"}
351
+ {"current_steps": 702, "total_steps": 928, "loss": 0.3136082589626312, "lr": 4.791048754009362e-06, "epoch": 3.0258620689655173, "percentage": 75.65, "elapsed_time": "2:07:56", "remaining_time": "0:41:11"}
352
+ {"current_steps": 704, "total_steps": 928, "loss": 0.6226381063461304, "lr": 4.7447408365363616e-06, "epoch": 3.0344827586206895, "percentage": 75.86, "elapsed_time": "2:08:20", "remaining_time": "0:40:50"}
353
+ {"current_steps": 706, "total_steps": 928, "loss": 0.0870320051908493, "lr": 4.698751083318918e-06, "epoch": 3.043103448275862, "percentage": 76.08, "elapsed_time": "2:08:35", "remaining_time": "0:40:25"}
354
+ {"current_steps": 708, "total_steps": 928, "loss": 0.379879891872406, "lr": 4.653081833555975e-06, "epoch": 3.0517241379310347, "percentage": 76.29, "elapsed_time": "2:09:01", "remaining_time": "0:40:05"}
355
+ {"current_steps": 710, "total_steps": 928, "loss": 0.10685593634843826, "lr": 4.60773541014454e-06, "epoch": 3.060344827586207, "percentage": 76.51, "elapsed_time": "2:09:21", "remaining_time": "0:39:43"}
356
+ {"current_steps": 712, "total_steps": 928, "loss": 0.40309441089630127, "lr": 4.562714119561566e-06, "epoch": 3.0689655172413794, "percentage": 76.72, "elapsed_time": "2:09:42", "remaining_time": "0:39:21"}
357
+ {"current_steps": 714, "total_steps": 928, "loss": 0.6567556262016296, "lr": 4.518020251746607e-06, "epoch": 3.0775862068965516, "percentage": 76.94, "elapsed_time": "2:10:11", "remaining_time": "0:39:01"}
358
+ {"current_steps": 716, "total_steps": 928, "loss": 0.14177639782428741, "lr": 4.473656079985362e-06, "epoch": 3.086206896551724, "percentage": 77.16, "elapsed_time": "2:10:29", "remaining_time": "0:38:38"}
359
+ {"current_steps": 718, "total_steps": 928, "loss": 0.3785516917705536, "lr": 4.429623860794032e-06, "epoch": 3.0948275862068964, "percentage": 77.37, "elapsed_time": "2:10:54", "remaining_time": "0:38:17"}
360
+ {"current_steps": 720, "total_steps": 928, "loss": 0.18618693947792053, "lr": 4.38592583380457e-06, "epoch": 3.103448275862069, "percentage": 77.59, "elapsed_time": "2:11:07", "remaining_time": "0:37:52"}
361
+ {"current_steps": 722, "total_steps": 928, "loss": 0.20070935785770416, "lr": 4.3425642216507325e-06, "epoch": 3.1120689655172415, "percentage": 77.8, "elapsed_time": "2:11:26", "remaining_time": "0:37:30"}
362
+ {"current_steps": 724, "total_steps": 928, "loss": 0.3287600874900818, "lr": 4.2995412298550624e-06, "epoch": 3.1206896551724137, "percentage": 78.02, "elapsed_time": "2:11:52", "remaining_time": "0:37:09"}
363
+ {"current_steps": 726, "total_steps": 928, "loss": 0.36046263575553894, "lr": 4.256859046716677e-06, "epoch": 3.1293103448275863, "percentage": 78.23, "elapsed_time": "2:12:15", "remaining_time": "0:36:47"}
364
+ {"current_steps": 728, "total_steps": 928, "loss": 0.4107988774776459, "lr": 4.214519843199995e-06, "epoch": 3.1379310344827585, "percentage": 78.45, "elapsed_time": "2:12:40", "remaining_time": "0:36:27"}
365
+ {"current_steps": 730, "total_steps": 928, "loss": 0.2075681984424591, "lr": 4.172525772824283e-06, "epoch": 3.146551724137931, "percentage": 78.66, "elapsed_time": "2:12:59", "remaining_time": "0:36:04"}
366
+ {"current_steps": 732, "total_steps": 928, "loss": 0.17175163328647614, "lr": 4.130878971554148e-06, "epoch": 3.1551724137931036, "percentage": 78.88, "elapsed_time": "2:13:16", "remaining_time": "0:35:41"}
367
+ {"current_steps": 734, "total_steps": 928, "loss": 0.3509948253631592, "lr": 4.0895815576908686e-06, "epoch": 3.163793103448276, "percentage": 79.09, "elapsed_time": "2:13:43", "remaining_time": "0:35:20"}
368
+ {"current_steps": 736, "total_steps": 928, "loss": 0.22303853929042816, "lr": 4.048635631764673e-06, "epoch": 3.1724137931034484, "percentage": 79.31, "elapsed_time": "2:14:02", "remaining_time": "0:34:58"}
369
+ {"current_steps": 738, "total_steps": 928, "loss": 0.523650050163269, "lr": 4.008043276427886e-06, "epoch": 3.1810344827586206, "percentage": 79.53, "elapsed_time": "2:14:28", "remaining_time": "0:34:37"}
370
+ {"current_steps": 740, "total_steps": 928, "loss": 0.08064466714859009, "lr": 3.967806556349001e-06, "epoch": 3.189655172413793, "percentage": 79.74, "elapsed_time": "2:14:40", "remaining_time": "0:34:12"}
371
+ {"current_steps": 742, "total_steps": 928, "loss": 0.17742744088172913, "lr": 3.927927518107664e-06, "epoch": 3.1982758620689653, "percentage": 79.96, "elapsed_time": "2:15:01", "remaining_time": "0:33:50"}
372
+ {"current_steps": 744, "total_steps": 928, "loss": 0.2840654253959656, "lr": 3.888408190090583e-06, "epoch": 3.206896551724138, "percentage": 80.17, "elapsed_time": "2:15:21", "remaining_time": "0:33:28"}
373
+ {"current_steps": 746, "total_steps": 928, "loss": 0.27302879095077515, "lr": 3.8492505823883394e-06, "epoch": 3.2155172413793105, "percentage": 80.39, "elapsed_time": "2:15:43", "remaining_time": "0:33:06"}
374
+ {"current_steps": 748, "total_steps": 928, "loss": 0.38051924109458923, "lr": 3.810456686693178e-06, "epoch": 3.2241379310344827, "percentage": 80.6, "elapsed_time": "2:16:06", "remaining_time": "0:32:45"}
375
+ {"current_steps": 750, "total_steps": 928, "loss": 0.33380451798439026, "lr": 3.7720284761976657e-06, "epoch": 3.2327586206896552, "percentage": 80.82, "elapsed_time": "2:16:32", "remaining_time": "0:32:24"}
376
+ {"current_steps": 752, "total_steps": 928, "loss": 0.2860594093799591, "lr": 3.7339679054943645e-06, "epoch": 3.2413793103448274, "percentage": 81.03, "elapsed_time": "2:16:50", "remaining_time": "0:32:01"}
377
+ {"current_steps": 754, "total_steps": 928, "loss": 0.36991697549819946, "lr": 3.696276910476382e-06, "epoch": 3.25, "percentage": 81.25, "elapsed_time": "2:17:16", "remaining_time": "0:31:40"}
378
+ {"current_steps": 756, "total_steps": 928, "loss": 0.3221489191055298, "lr": 3.6589574082389324e-06, "epoch": 3.2586206896551726, "percentage": 81.47, "elapsed_time": "2:17:37", "remaining_time": "0:31:18"}
379
+ {"current_steps": 758, "total_steps": 928, "loss": 0.37856197357177734, "lr": 3.6220112969818045e-06, "epoch": 3.2672413793103448, "percentage": 81.68, "elapsed_time": "2:18:03", "remaining_time": "0:30:57"}
380
+ {"current_steps": 760, "total_steps": 928, "loss": 0.24704982340335846, "lr": 3.5854404559128287e-06, "epoch": 3.2758620689655173, "percentage": 81.9, "elapsed_time": "2:18:21", "remaining_time": "0:30:35"}
381
+ {"current_steps": 762, "total_steps": 928, "loss": 0.36558791995048523, "lr": 3.549246745152283e-06, "epoch": 3.2844827586206895, "percentage": 82.11, "elapsed_time": "2:18:47", "remaining_time": "0:30:14"}
382
+ {"current_steps": 764, "total_steps": 928, "loss": 0.3346759080886841, "lr": 3.5134320056382877e-06, "epoch": 3.293103448275862, "percentage": 82.33, "elapsed_time": "2:19:10", "remaining_time": "0:29:52"}
383
+ {"current_steps": 766, "total_steps": 928, "loss": 0.2471802681684494, "lr": 3.477998059033163e-06, "epoch": 3.3017241379310347, "percentage": 82.54, "elapsed_time": "2:19:28", "remaining_time": "0:29:29"}
384
+ {"current_steps": 768, "total_steps": 928, "loss": 0.34330496191978455, "lr": 3.4429467076307823e-06, "epoch": 3.310344827586207, "percentage": 82.76, "elapsed_time": "2:19:50", "remaining_time": "0:29:08"}
385
+ {"current_steps": 770, "total_steps": 928, "loss": 0.10211128741502762, "lr": 3.4082797342648867e-06, "epoch": 3.3189655172413794, "percentage": 82.97, "elapsed_time": "2:20:10", "remaining_time": "0:28:45"}
386
+ {"current_steps": 772, "total_steps": 928, "loss": 0.22283262014389038, "lr": 3.373998902218423e-06, "epoch": 3.3275862068965516, "percentage": 83.19, "elapsed_time": "2:20:27", "remaining_time": "0:28:22"}
387
+ {"current_steps": 774, "total_steps": 928, "loss": 0.3982272148132324, "lr": 3.3401059551338326e-06, "epoch": 3.336206896551724, "percentage": 83.41, "elapsed_time": "2:20:49", "remaining_time": "0:28:01"}
388
+ {"current_steps": 776, "total_steps": 928, "loss": 0.35799640417099, "lr": 3.3066026169243916e-06, "epoch": 3.344827586206897, "percentage": 83.62, "elapsed_time": "2:21:16", "remaining_time": "0:27:40"}
389
+ {"current_steps": 778, "total_steps": 928, "loss": 0.33809182047843933, "lr": 3.2734905916865045e-06, "epoch": 3.353448275862069, "percentage": 83.84, "elapsed_time": "2:21:39", "remaining_time": "0:27:18"}
390
+ {"current_steps": 780, "total_steps": 928, "loss": 0.24870137870311737, "lr": 3.2407715636130365e-06, "epoch": 3.3620689655172415, "percentage": 84.05, "elapsed_time": "2:22:04", "remaining_time": "0:26:57"}
391
+ {"current_steps": 782, "total_steps": 928, "loss": 0.25115785002708435, "lr": 3.20844719690765e-06, "epoch": 3.3706896551724137, "percentage": 84.27, "elapsed_time": "2:22:21", "remaining_time": "0:26:34"}
392
+ {"current_steps": 784, "total_steps": 928, "loss": 0.34631893038749695, "lr": 3.1765191357001596e-06, "epoch": 3.3793103448275863, "percentage": 84.48, "elapsed_time": "2:22:44", "remaining_time": "0:26:12"}
393
+ {"current_steps": 786, "total_steps": 928, "loss": 0.21534167230129242, "lr": 3.1449890039628956e-06, "epoch": 3.3879310344827585, "percentage": 84.7, "elapsed_time": "2:23:02", "remaining_time": "0:25:50"}
394
+ {"current_steps": 788, "total_steps": 928, "loss": 0.38198593258857727, "lr": 3.1138584054281177e-06, "epoch": 3.396551724137931, "percentage": 84.91, "elapsed_time": "2:23:29", "remaining_time": "0:25:29"}
395
+ {"current_steps": 790, "total_steps": 928, "loss": 0.3233962655067444, "lr": 3.0831289235064297e-06, "epoch": 3.405172413793103, "percentage": 85.13, "elapsed_time": "2:23:51", "remaining_time": "0:25:07"}
396
+ {"current_steps": 792, "total_steps": 928, "loss": 0.39972445368766785, "lr": 3.0528021212062536e-06, "epoch": 3.413793103448276, "percentage": 85.34, "elapsed_time": "2:24:04", "remaining_time": "0:24:44"}
397
+ {"current_steps": 794, "total_steps": 928, "loss": 0.2797224819660187, "lr": 3.0228795410543182e-06, "epoch": 3.4224137931034484, "percentage": 85.56, "elapsed_time": "2:24:22", "remaining_time": "0:24:21"}
398
+ {"current_steps": 796, "total_steps": 928, "loss": 0.4961601793766022, "lr": 2.9933627050172152e-06, "epoch": 3.4310344827586206, "percentage": 85.78, "elapsed_time": "2:24:49", "remaining_time": "0:24:01"}
399
+ {"current_steps": 798, "total_steps": 928, "loss": 0.2608637511730194, "lr": 2.96425311442397e-06, "epoch": 3.439655172413793, "percentage": 85.99, "elapsed_time": "2:25:15", "remaining_time": "0:23:39"}
400
+ {"current_steps": 800, "total_steps": 928, "loss": 0.389303594827652, "lr": 2.9355522498896914e-06, "epoch": 3.4482758620689653, "percentage": 86.21, "elapsed_time": "2:25:41", "remaining_time": "0:23:18"}
401
+ {"current_steps": 802, "total_steps": 928, "loss": 0.416867733001709, "lr": 2.907261571240257e-06, "epoch": 3.456896551724138, "percentage": 86.42, "elapsed_time": "2:26:07", "remaining_time": "0:22:57"}
402
+ {"current_steps": 804, "total_steps": 928, "loss": 0.28168269991874695, "lr": 2.879382517438062e-06, "epoch": 3.4655172413793105, "percentage": 86.64, "elapsed_time": "2:26:29", "remaining_time": "0:22:35"}
403
+ {"current_steps": 806, "total_steps": 928, "loss": 0.39084625244140625, "lr": 2.8519165065088293e-06, "epoch": 3.4741379310344827, "percentage": 86.85, "elapsed_time": "2:26:51", "remaining_time": "0:22:13"}
404
+ {"current_steps": 808, "total_steps": 928, "loss": 0.3399796187877655, "lr": 2.824864935469487e-06, "epoch": 3.4827586206896552, "percentage": 87.07, "elapsed_time": "2:27:11", "remaining_time": "0:21:51"}
405
+ {"current_steps": 810, "total_steps": 928, "loss": 0.10816089808940887, "lr": 2.7982291802570983e-06, "epoch": 3.4913793103448274, "percentage": 87.28, "elapsed_time": "2:27:25", "remaining_time": "0:21:28"}
406
+ {"current_steps": 812, "total_steps": 928, "loss": 0.3328501880168915, "lr": 2.7720105956588983e-06, "epoch": 3.5, "percentage": 87.5, "elapsed_time": "2:27:50", "remaining_time": "0:21:07"}
407
+ {"current_steps": 814, "total_steps": 928, "loss": 0.4947846531867981, "lr": 2.7462105152433617e-06, "epoch": 3.5086206896551726, "percentage": 87.72, "elapsed_time": "2:28:12", "remaining_time": "0:20:45"}
408
+ {"current_steps": 816, "total_steps": 928, "loss": 0.3645175099372864, "lr": 2.720830251292391e-06, "epoch": 3.5172413793103448, "percentage": 87.93, "elapsed_time": "2:28:34", "remaining_time": "0:20:23"}
409
+ {"current_steps": 818, "total_steps": 928, "loss": 0.4159550368785858, "lr": 2.695871094734559e-06, "epoch": 3.5258620689655173, "percentage": 88.15, "elapsed_time": "2:29:01", "remaining_time": "0:20:02"}
410
+ {"current_steps": 820, "total_steps": 928, "loss": 0.31124040484428406, "lr": 2.6713343150794506e-06, "epoch": 3.5344827586206895, "percentage": 88.36, "elapsed_time": "2:29:23", "remaining_time": "0:19:40"}
411
+ {"current_steps": 822, "total_steps": 928, "loss": 0.2729651927947998, "lr": 2.6472211603530894e-06, "epoch": 3.543103448275862, "percentage": 88.58, "elapsed_time": "2:29:49", "remaining_time": "0:19:19"}
412
+ {"current_steps": 824, "total_steps": 928, "loss": 0.27673032879829407, "lr": 2.6235328570344666e-06, "epoch": 3.5517241379310347, "percentage": 88.79, "elapsed_time": "2:30:03", "remaining_time": "0:18:56"}
413
+ {"current_steps": 826, "total_steps": 928, "loss": 0.0557853989303112, "lr": 2.600270609993143e-06, "epoch": 3.560344827586207, "percentage": 89.01, "elapsed_time": "2:30:17", "remaining_time": "0:18:33"}
414
+ {"current_steps": 828, "total_steps": 928, "loss": 0.18158602714538574, "lr": 2.5774356024279827e-06, "epoch": 3.5689655172413794, "percentage": 89.22, "elapsed_time": "2:30:30", "remaining_time": "0:18:10"}
415
+ {"current_steps": 830, "total_steps": 928, "loss": 0.18923550844192505, "lr": 2.555028995806956e-06, "epoch": 3.5775862068965516, "percentage": 89.44, "elapsed_time": "2:30:52", "remaining_time": "0:17:48"}
416
+ {"current_steps": 832, "total_steps": 928, "loss": 0.22178103029727936, "lr": 2.5330519298080757e-06, "epoch": 3.586206896551724, "percentage": 89.66, "elapsed_time": "2:31:10", "remaining_time": "0:17:26"}
417
+ {"current_steps": 834, "total_steps": 928, "loss": 0.43088600039482117, "lr": 2.511505522261417e-06, "epoch": 3.594827586206897, "percentage": 89.87, "elapsed_time": "2:31:34", "remaining_time": "0:17:05"}
418
+ {"current_steps": 836, "total_steps": 928, "loss": 0.3035900592803955, "lr": 2.4903908690922746e-06, "epoch": 3.603448275862069, "percentage": 90.09, "elapsed_time": "2:31:53", "remaining_time": "0:16:42"}
419
+ {"current_steps": 838, "total_steps": 928, "loss": 0.3073073923587799, "lr": 2.4697090442654055e-06, "epoch": 3.612068965517241, "percentage": 90.3, "elapsed_time": "2:32:19", "remaining_time": "0:16:21"}
420
+ {"current_steps": 840, "total_steps": 928, "loss": 0.1451842337846756, "lr": 2.4494610997304156e-06, "epoch": 3.6206896551724137, "percentage": 90.52, "elapsed_time": "2:32:39", "remaining_time": "0:15:59"}
421
+ {"current_steps": 842, "total_steps": 928, "loss": 0.19056405127048492, "lr": 2.4296480653682444e-06, "epoch": 3.6293103448275863, "percentage": 90.73, "elapsed_time": "2:32:51", "remaining_time": "0:15:36"}
422
+ {"current_steps": 844, "total_steps": 928, "loss": 0.3310995101928711, "lr": 2.4102709489387925e-06, "epoch": 3.637931034482759, "percentage": 90.95, "elapsed_time": "2:33:16", "remaining_time": "0:15:15"}
423
+ {"current_steps": 846, "total_steps": 928, "loss": 0.27216413617134094, "lr": 2.391330736029649e-06, "epoch": 3.646551724137931, "percentage": 91.16, "elapsed_time": "2:33:40", "remaining_time": "0:14:53"}
424
+ {"current_steps": 848, "total_steps": 928, "loss": 0.13349619507789612, "lr": 2.3728283900059756e-06, "epoch": 3.655172413793103, "percentage": 91.38, "elapsed_time": "2:33:53", "remaining_time": "0:14:31"}
425
+ {"current_steps": 850, "total_steps": 928, "loss": 0.20533311367034912, "lr": 2.3547648519614967e-06, "epoch": 3.663793103448276, "percentage": 91.59, "elapsed_time": "2:34:12", "remaining_time": "0:14:09"}
426
+ {"current_steps": 852, "total_steps": 928, "loss": 0.38371843099594116, "lr": 2.3371410406706356e-06, "epoch": 3.6724137931034484, "percentage": 91.81, "elapsed_time": "2:34:38", "remaining_time": "0:13:47"}
427
+ {"current_steps": 854, "total_steps": 928, "loss": 0.2169187366962433, "lr": 2.319957852541786e-06, "epoch": 3.6810344827586206, "percentage": 92.03, "elapsed_time": "2:34:57", "remaining_time": "0:13:25"}
428
+ {"current_steps": 856, "total_steps": 928, "loss": 0.37597841024398804, "lr": 2.303216161571706e-06, "epoch": 3.689655172413793, "percentage": 92.24, "elapsed_time": "2:35:23", "remaining_time": "0:13:04"}
429
+ {"current_steps": 858, "total_steps": 928, "loss": 0.4011721611022949, "lr": 2.2869168193010793e-06, "epoch": 3.6982758620689653, "percentage": 92.46, "elapsed_time": "2:35:51", "remaining_time": "0:12:42"}
430
+ {"current_steps": 860, "total_steps": 928, "loss": 0.18101970851421356, "lr": 2.2710606547711917e-06, "epoch": 3.706896551724138, "percentage": 92.67, "elapsed_time": "2:36:11", "remaining_time": "0:12:20"}
431
+ {"current_steps": 862, "total_steps": 928, "loss": 0.23301751911640167, "lr": 2.2556484744817635e-06, "epoch": 3.7155172413793105, "percentage": 92.89, "elapsed_time": "2:36:33", "remaining_time": "0:11:59"}
432
+ {"current_steps": 864, "total_steps": 928, "loss": 0.1944858729839325, "lr": 2.2406810623499382e-06, "epoch": 3.7241379310344827, "percentage": 93.1, "elapsed_time": "2:36:53", "remaining_time": "0:11:37"}
433
+ {"current_steps": 866, "total_steps": 928, "loss": 0.11213114112615585, "lr": 2.226159179670395e-06, "epoch": 3.7327586206896552, "percentage": 93.32, "elapsed_time": "2:37:07", "remaining_time": "0:11:14"}
434
+ {"current_steps": 868, "total_steps": 928, "loss": 0.25432637333869934, "lr": 2.2120835650766386e-06, "epoch": 3.7413793103448274, "percentage": 93.53, "elapsed_time": "2:37:26", "remaining_time": "0:10:52"}
435
+ {"current_steps": 870, "total_steps": 928, "loss": 0.2917807102203369, "lr": 2.1984549345034234e-06, "epoch": 3.75, "percentage": 93.75, "elapsed_time": "2:37:44", "remaining_time": "0:10:30"}
436
+ {"current_steps": 872, "total_steps": 928, "loss": 0.1030154600739479, "lr": 2.1852739811503433e-06, "epoch": 3.7586206896551726, "percentage": 93.97, "elapsed_time": "2:38:03", "remaining_time": "0:10:09"}
437
+ {"current_steps": 874, "total_steps": 928, "loss": 0.24243952333927155, "lr": 2.172541375446566e-06, "epoch": 3.7672413793103448, "percentage": 94.18, "elapsed_time": "2:38:23", "remaining_time": "0:09:47"}
438
+ {"current_steps": 876, "total_steps": 928, "loss": 0.4455357789993286, "lr": 2.1602577650167436e-06, "epoch": 3.7758620689655173, "percentage": 94.4, "elapsed_time": "2:38:48", "remaining_time": "0:09:25"}
439
+ {"current_steps": 878, "total_steps": 928, "loss": 0.5109292268753052, "lr": 2.1484237746480574e-06, "epoch": 3.7844827586206895, "percentage": 94.61, "elapsed_time": "2:39:14", "remaining_time": "0:09:04"}
440
+ {"current_steps": 880, "total_steps": 928, "loss": 0.1860283464193344, "lr": 2.1370400062584544e-06, "epoch": 3.793103448275862, "percentage": 94.83, "elapsed_time": "2:39:30", "remaining_time": "0:08:42"}
441
+ {"current_steps": 882, "total_steps": 928, "loss": 0.20024971663951874, "lr": 2.1261070388660202e-06, "epoch": 3.8017241379310347, "percentage": 95.04, "elapsed_time": "2:39:49", "remaining_time": "0:08:20"}
442
+ {"current_steps": 884, "total_steps": 928, "loss": 0.43951112031936646, "lr": 2.115625428559535e-06, "epoch": 3.810344827586207, "percentage": 95.26, "elapsed_time": "2:40:15", "remaining_time": "0:07:58"}
443
+ {"current_steps": 886, "total_steps": 928, "loss": 0.16899777948856354, "lr": 2.105595708470185e-06, "epoch": 3.8189655172413794, "percentage": 95.47, "elapsed_time": "2:40:35", "remaining_time": "0:07:36"}
444
+ {"current_steps": 888, "total_steps": 928, "loss": 0.26339343190193176, "lr": 2.096018388744448e-06, "epoch": 3.8275862068965516, "percentage": 95.69, "elapsed_time": "2:41:01", "remaining_time": "0:07:15"}
445
+ {"current_steps": 890, "total_steps": 928, "loss": 0.11539773643016815, "lr": 2.086893956518142e-06, "epoch": 3.836206896551724, "percentage": 95.91, "elapsed_time": "2:41:13", "remaining_time": "0:06:53"}
446
+ {"current_steps": 892, "total_steps": 928, "loss": 0.3523973822593689, "lr": 2.078222875891654e-06, "epoch": 3.844827586206897, "percentage": 96.12, "elapsed_time": "2:41:38", "remaining_time": "0:06:31"}
447
+ {"current_steps": 894, "total_steps": 928, "loss": 0.33388248085975647, "lr": 2.070005587906328e-06, "epoch": 3.853448275862069, "percentage": 96.34, "elapsed_time": "2:41:59", "remaining_time": "0:06:09"}
448
+ {"current_steps": 896, "total_steps": 928, "loss": 0.2652340531349182, "lr": 2.0622425105220346e-06, "epoch": 3.862068965517241, "percentage": 96.55, "elapsed_time": "2:42:20", "remaining_time": "0:05:47"}
449
+ {"current_steps": 898, "total_steps": 928, "loss": 0.1966915875673294, "lr": 2.0549340385959123e-06, "epoch": 3.8706896551724137, "percentage": 96.77, "elapsed_time": "2:42:38", "remaining_time": "0:05:26"}
450
+ {"current_steps": 900, "total_steps": 928, "loss": 0.391522079706192, "lr": 2.0480805438622865e-06, "epoch": 3.8793103448275863, "percentage": 96.98, "elapsed_time": "2:43:05", "remaining_time": "0:05:04"}
451
+ {"current_steps": 902, "total_steps": 928, "loss": 0.3690534234046936, "lr": 2.0416823749137542e-06, "epoch": 3.887931034482759, "percentage": 97.2, "elapsed_time": "2:43:21", "remaining_time": "0:04:42"}
452
+ {"current_steps": 904, "total_steps": 928, "loss": 0.35606420040130615, "lr": 2.03573985718346e-06, "epoch": 3.896551724137931, "percentage": 97.41, "elapsed_time": "2:43:40", "remaining_time": "0:04:20"}
453
+ {"current_steps": 906, "total_steps": 928, "loss": 0.37323978543281555, "lr": 2.0302532929285423e-06, "epoch": 3.905172413793103, "percentage": 97.63, "elapsed_time": "2:44:07", "remaining_time": "0:03:59"}
454
+ {"current_steps": 908, "total_steps": 928, "loss": 0.22919489443302155, "lr": 2.0252229612147587e-06, "epoch": 3.913793103448276, "percentage": 97.84, "elapsed_time": "2:44:25", "remaining_time": "0:03:37"}
455
+ {"current_steps": 910, "total_steps": 928, "loss": 0.17647181451320648, "lr": 2.0206491179022887e-06, "epoch": 3.9224137931034484, "percentage": 98.06, "elapsed_time": "2:44:37", "remaining_time": "0:03:15"}
456
+ {"current_steps": 912, "total_steps": 928, "loss": 0.23643004894256592, "lr": 2.0165319956327258e-06, "epoch": 3.9310344827586206, "percentage": 98.28, "elapsed_time": "2:44:57", "remaining_time": "0:02:53"}
457
+ {"current_steps": 914, "total_steps": 928, "loss": 0.3691279888153076, "lr": 2.01287180381724e-06, "epoch": 3.939655172413793, "percentage": 98.49, "elapsed_time": "2:45:24", "remaining_time": "0:02:32"}
458
+ {"current_steps": 916, "total_steps": 928, "loss": 0.3844853937625885, "lr": 2.0096687286259286e-06, "epoch": 3.9482758620689653, "percentage": 98.71, "elapsed_time": "2:45:47", "remaining_time": "0:02:10"}
459
+ {"current_steps": 918, "total_steps": 928, "loss": 0.28372105956077576, "lr": 2.006922932978346e-06, "epoch": 3.956896551724138, "percentage": 98.92, "elapsed_time": "2:46:03", "remaining_time": "0:01:48"}
460
+ {"current_steps": 920, "total_steps": 928, "loss": 0.36251726746559143, "lr": 2.004634556535215e-06, "epoch": 3.9655172413793105, "percentage": 99.14, "elapsed_time": "2:46:21", "remaining_time": "0:01:26"}
461
+ {"current_steps": 922, "total_steps": 928, "loss": 0.2959789037704468, "lr": 2.0028037156913294e-06, "epoch": 3.9741379310344827, "percentage": 99.35, "elapsed_time": "2:46:41", "remaining_time": "0:01:05"}
462
+ {"current_steps": 924, "total_steps": 928, "loss": 0.3481578230857849, "lr": 2.0014305035696265e-06, "epoch": 3.9827586206896552, "percentage": 99.57, "elapsed_time": "2:47:04", "remaining_time": "0:00:43"}
463
+ {"current_steps": 926, "total_steps": 928, "loss": 0.36124011874198914, "lr": 2.0005149900164546e-06, "epoch": 3.9913793103448274, "percentage": 99.78, "elapsed_time": "2:47:26", "remaining_time": "0:00:21"}
464
+ {"current_steps": 928, "total_steps": 928, "loss": 0.39811837673187256, "lr": 2.000057221598022e-06, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "2:47:51", "remaining_time": "0:00:00"}
465
+ {"current_steps": 928, "total_steps": 928, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "2:47:51", "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": 0.752173662185669,
15
+ "learning_rate": 4.2553191489361704e-07,
16
+ "loss": 3.6935362815856934,
17
+ "step": 2
18
+ },
19
+ {
20
+ "epoch": 0.017241379310344827,
21
+ "grad_norm": 3.0686516761779785,
22
+ "learning_rate": 1.276595744680851e-06,
23
+ "loss": 2.4514498710632324,
24
+ "step": 4
25
+ },
26
+ {
27
+ "epoch": 0.02586206896551724,
28
+ "grad_norm": 3.670806407928467,
29
+ "learning_rate": 2.1276595744680853e-06,
30
+ "loss": 2.2222416400909424,
31
+ "step": 6
32
+ },
33
+ {
34
+ "epoch": 0.034482758620689655,
35
+ "grad_norm": 0.4838178753852844,
36
+ "learning_rate": 2.978723404255319e-06,
37
+ "loss": 1.9781272411346436,
38
+ "step": 8
39
+ },
40
+ {
41
+ "epoch": 0.04310344827586207,
42
+ "grad_norm": 3.7360541820526123,
43
+ "learning_rate": 3.8297872340425535e-06,
44
+ "loss": 1.768671989440918,
45
+ "step": 10
46
+ },
47
+ {
48
+ "epoch": 0.05172413793103448,
49
+ "grad_norm": 0.6814174056053162,
50
+ "learning_rate": 4.680851063829788e-06,
51
+ "loss": 1.7612519264221191,
52
+ "step": 12
53
+ },
54
+ {
55
+ "epoch": 0.0603448275862069,
56
+ "grad_norm": 0.6352251768112183,
57
+ "learning_rate": 5.531914893617022e-06,
58
+ "loss": 1.8349356651306152,
59
+ "step": 14
60
+ },
61
+ {
62
+ "epoch": 0.06896551724137931,
63
+ "grad_norm": 0.4332002103328705,
64
+ "learning_rate": 6.382978723404256e-06,
65
+ "loss": 0.766176164150238,
66
+ "step": 16
67
+ },
68
+ {
69
+ "epoch": 0.07758620689655173,
70
+ "grad_norm": 0.5920107364654541,
71
+ "learning_rate": 7.234042553191491e-06,
72
+ "loss": 0.9314699769020081,
73
+ "step": 18
74
+ },
75
+ {
76
+ "epoch": 0.08620689655172414,
77
+ "grad_norm": 0.40042662620544434,
78
+ "learning_rate": 8.085106382978723e-06,
79
+ "loss": 1.5076656341552734,
80
+ "step": 20
81
+ },
82
+ {
83
+ "epoch": 0.09482758620689655,
84
+ "grad_norm": 0.33145302534103394,
85
+ "learning_rate": 8.936170212765958e-06,
86
+ "loss": 1.6648786067962646,
87
+ "step": 22
88
+ },
89
+ {
90
+ "epoch": 0.10344827586206896,
91
+ "grad_norm": 0.7193792462348938,
92
+ "learning_rate": 9.787234042553192e-06,
93
+ "loss": 1.508145809173584,
94
+ "step": 24
95
+ },
96
+ {
97
+ "epoch": 0.11206896551724138,
98
+ "grad_norm": 2.580322027206421,
99
+ "learning_rate": 1.0638297872340426e-05,
100
+ "loss": 1.3233157396316528,
101
+ "step": 26
102
+ },
103
+ {
104
+ "epoch": 0.1206896551724138,
105
+ "grad_norm": 1.6866916418075562,
106
+ "learning_rate": 1.1489361702127662e-05,
107
+ "loss": 0.5060393214225769,
108
+ "step": 28
109
+ },
110
+ {
111
+ "epoch": 0.12931034482758622,
112
+ "grad_norm": 0.40954306721687317,
113
+ "learning_rate": 1.2340425531914895e-05,
114
+ "loss": 1.489959716796875,
115
+ "step": 30
116
+ },
117
+ {
118
+ "epoch": 0.13793103448275862,
119
+ "grad_norm": 0.9065203666687012,
120
+ "learning_rate": 1.3191489361702127e-05,
121
+ "loss": 1.14947509765625,
122
+ "step": 32
123
+ },
124
+ {
125
+ "epoch": 0.14655172413793102,
126
+ "grad_norm": 0.3966251611709595,
127
+ "learning_rate": 1.4042553191489363e-05,
128
+ "loss": 1.3647483587265015,
129
+ "step": 34
130
+ },
131
+ {
132
+ "epoch": 0.15517241379310345,
133
+ "grad_norm": 0.3009808659553528,
134
+ "learning_rate": 1.4893617021276596e-05,
135
+ "loss": 1.4109106063842773,
136
+ "step": 36
137
+ },
138
+ {
139
+ "epoch": 0.16379310344827586,
140
+ "grad_norm": 0.33264291286468506,
141
+ "learning_rate": 1.5744680851063832e-05,
142
+ "loss": 1.3988299369812012,
143
+ "step": 38
144
+ },
145
+ {
146
+ "epoch": 0.1724137931034483,
147
+ "grad_norm": 0.3600396513938904,
148
+ "learning_rate": 1.6595744680851064e-05,
149
+ "loss": 1.4429720640182495,
150
+ "step": 40
151
+ },
152
+ {
153
+ "epoch": 0.1810344827586207,
154
+ "grad_norm": 0.65086430311203,
155
+ "learning_rate": 1.74468085106383e-05,
156
+ "loss": 1.5474897623062134,
157
+ "step": 42
158
+ },
159
+ {
160
+ "epoch": 0.1896551724137931,
161
+ "grad_norm": 0.8449164628982544,
162
+ "learning_rate": 1.8297872340425533e-05,
163
+ "loss": 0.9270882606506348,
164
+ "step": 44
165
+ },
166
+ {
167
+ "epoch": 0.19827586206896552,
168
+ "grad_norm": 0.561241626739502,
169
+ "learning_rate": 1.914893617021277e-05,
170
+ "loss": 1.1169898509979248,
171
+ "step": 46
172
+ },
173
+ {
174
+ "epoch": 0.20689655172413793,
175
+ "grad_norm": 0.2715710401535034,
176
+ "learning_rate": 2e-05,
177
+ "loss": 1.3745737075805664,
178
+ "step": 48
179
+ },
180
+ {
181
+ "epoch": 0.21551724137931033,
182
+ "grad_norm": 0.3442396819591522,
183
+ "learning_rate": 1.999977111433554e-05,
184
+ "loss": 1.4785292148590088,
185
+ "step": 50
186
+ },
187
+ {
188
+ "epoch": 0.22413793103448276,
189
+ "grad_norm": 0.6824942827224731,
190
+ "learning_rate": 1.9999084468984077e-05,
191
+ "loss": 0.7019240260124207,
192
+ "step": 52
193
+ },
194
+ {
195
+ "epoch": 0.23275862068965517,
196
+ "grad_norm": 0.31582292914390564,
197
+ "learning_rate": 1.9997940098870788e-05,
198
+ "loss": 1.4191925525665283,
199
+ "step": 54
200
+ },
201
+ {
202
+ "epoch": 0.2413793103448276,
203
+ "grad_norm": 0.27809372544288635,
204
+ "learning_rate": 1.9996338062202316e-05,
205
+ "loss": 1.3712444305419922,
206
+ "step": 56
207
+ },
208
+ {
209
+ "epoch": 0.25,
210
+ "grad_norm": 0.5872206091880798,
211
+ "learning_rate": 1.9994278440463823e-05,
212
+ "loss": 1.4150975942611694,
213
+ "step": 58
214
+ },
215
+ {
216
+ "epoch": 0.25862068965517243,
217
+ "grad_norm": 0.2935428321361542,
218
+ "learning_rate": 1.9991761338414844e-05,
219
+ "loss": 1.3828074932098389,
220
+ "step": 60
221
+ },
222
+ {
223
+ "epoch": 0.2672413793103448,
224
+ "grad_norm": 1.7192302942276,
225
+ "learning_rate": 1.998878688408395e-05,
226
+ "loss": 1.0158110857009888,
227
+ "step": 62
228
+ },
229
+ {
230
+ "epoch": 0.27586206896551724,
231
+ "grad_norm": 0.22481945157051086,
232
+ "learning_rate": 1.9985355228762237e-05,
233
+ "loss": 1.3290989398956299,
234
+ "step": 64
235
+ },
236
+ {
237
+ "epoch": 0.28448275862068967,
238
+ "grad_norm": 0.2971683144569397,
239
+ "learning_rate": 1.9981466546995648e-05,
240
+ "loss": 1.303815484046936,
241
+ "step": 66
242
+ },
243
+ {
244
+ "epoch": 0.29310344827586204,
245
+ "grad_norm": 0.23132647573947906,
246
+ "learning_rate": 1.997712103657607e-05,
247
+ "loss": 1.2936116456985474,
248
+ "step": 68
249
+ },
250
+ {
251
+ "epoch": 0.3017241379310345,
252
+ "grad_norm": 0.2712137699127197,
253
+ "learning_rate": 1.9972318918531294e-05,
254
+ "loss": 1.0591557025909424,
255
+ "step": 70
256
+ },
257
+ {
258
+ "epoch": 0.3103448275862069,
259
+ "grad_norm": 1.0306100845336914,
260
+ "learning_rate": 1.9967060437113752e-05,
261
+ "loss": 0.9161477088928223,
262
+ "step": 72
263
+ },
264
+ {
265
+ "epoch": 0.31896551724137934,
266
+ "grad_norm": 0.7370746731758118,
267
+ "learning_rate": 1.996134585978812e-05,
268
+ "loss": 1.130851149559021,
269
+ "step": 74
270
+ },
271
+ {
272
+ "epoch": 0.3275862068965517,
273
+ "grad_norm": 0.36369481682777405,
274
+ "learning_rate": 1.995517547721769e-05,
275
+ "loss": 1.167314052581787,
276
+ "step": 76
277
+ },
278
+ {
279
+ "epoch": 0.33620689655172414,
280
+ "grad_norm": 0.31680384278297424,
281
+ "learning_rate": 1.9948549603249602e-05,
282
+ "loss": 1.1086735725402832,
283
+ "step": 78
284
+ },
285
+ {
286
+ "epoch": 0.3448275862068966,
287
+ "grad_norm": 1.57016921043396,
288
+ "learning_rate": 1.9941468574898867e-05,
289
+ "loss": 1.0062172412872314,
290
+ "step": 80
291
+ },
292
+ {
293
+ "epoch": 0.35344827586206895,
294
+ "grad_norm": 0.39987874031066895,
295
+ "learning_rate": 1.993393275233123e-05,
296
+ "loss": 1.3131077289581299,
297
+ "step": 82
298
+ },
299
+ {
300
+ "epoch": 0.3620689655172414,
301
+ "grad_norm": 0.738694965839386,
302
+ "learning_rate": 1.9925942518844875e-05,
303
+ "loss": 1.1570929288864136,
304
+ "step": 84
305
+ },
306
+ {
307
+ "epoch": 0.3706896551724138,
308
+ "grad_norm": 0.4991850256919861,
309
+ "learning_rate": 1.991749828085088e-05,
310
+ "loss": 1.3317445516586304,
311
+ "step": 86
312
+ },
313
+ {
314
+ "epoch": 0.3793103448275862,
315
+ "grad_norm": 0.496286541223526,
316
+ "learning_rate": 1.9908600467852585e-05,
317
+ "loss": 1.2484678030014038,
318
+ "step": 88
319
+ },
320
+ {
321
+ "epoch": 0.3879310344827586,
322
+ "grad_norm": 0.6867333054542542,
323
+ "learning_rate": 1.9899249532423733e-05,
324
+ "loss": 1.0302324295043945,
325
+ "step": 90
326
+ },
327
+ {
328
+ "epoch": 0.39655172413793105,
329
+ "grad_norm": 0.2537033259868622,
330
+ "learning_rate": 1.9889445950185452e-05,
331
+ "loss": 0.8809990882873535,
332
+ "step": 92
333
+ },
334
+ {
335
+ "epoch": 0.4051724137931034,
336
+ "grad_norm": 0.7086202502250671,
337
+ "learning_rate": 1.9879190219782058e-05,
338
+ "loss": 1.0718659162521362,
339
+ "step": 94
340
+ },
341
+ {
342
+ "epoch": 0.41379310344827586,
343
+ "grad_norm": 0.6780977845191956,
344
+ "learning_rate": 1.98684828628557e-05,
345
+ "loss": 1.2635523080825806,
346
+ "step": 96
347
+ },
348
+ {
349
+ "epoch": 0.4224137931034483,
350
+ "grad_norm": 0.9746212363243103,
351
+ "learning_rate": 1.9857324424019827e-05,
352
+ "loss": 1.186186671257019,
353
+ "step": 98
354
+ },
355
+ {
356
+ "epoch": 0.43103448275862066,
357
+ "grad_norm": 1.5146777629852295,
358
+ "learning_rate": 1.9845715470831476e-05,
359
+ "loss": 1.1258275508880615,
360
+ "step": 100
361
+ },
362
+ {
363
+ "epoch": 0.4396551724137931,
364
+ "grad_norm": 0.6449897289276123,
365
+ "learning_rate": 1.983365659376242e-05,
366
+ "loss": 1.2144218683242798,
367
+ "step": 102
368
+ },
369
+ {
370
+ "epoch": 0.4482758620689655,
371
+ "grad_norm": 0.3542836606502533,
372
+ "learning_rate": 1.9821148406169124e-05,
373
+ "loss": 1.2573832273483276,
374
+ "step": 104
375
+ },
376
+ {
377
+ "epoch": 0.45689655172413796,
378
+ "grad_norm": 0.7458034753799438,
379
+ "learning_rate": 1.9808191544261545e-05,
380
+ "loss": 0.9863552451133728,
381
+ "step": 106
382
+ },
383
+ {
384
+ "epoch": 0.46551724137931033,
385
+ "grad_norm": 1.6855500936508179,
386
+ "learning_rate": 1.9794786667070786e-05,
387
+ "loss": 1.1030296087265015,
388
+ "step": 108
389
+ },
390
+ {
391
+ "epoch": 0.47413793103448276,
392
+ "grad_norm": 1.344683289527893,
393
+ "learning_rate": 1.9780934456415567e-05,
394
+ "loss": 1.1802244186401367,
395
+ "step": 110
396
+ },
397
+ {
398
+ "epoch": 0.4827586206896552,
399
+ "grad_norm": 0.39590156078338623,
400
+ "learning_rate": 1.976663561686753e-05,
401
+ "loss": 1.410926103591919,
402
+ "step": 112
403
+ },
404
+ {
405
+ "epoch": 0.49137931034482757,
406
+ "grad_norm": 0.6388300657272339,
407
+ "learning_rate": 1.975189087571544e-05,
408
+ "loss": 0.8382052183151245,
409
+ "step": 114
410
+ },
411
+ {
412
+ "epoch": 0.5,
413
+ "grad_norm": 1.3271642923355103,
414
+ "learning_rate": 1.9736700982928156e-05,
415
+ "loss": 0.8902297616004944,
416
+ "step": 116
417
+ },
418
+ {
419
+ "epoch": 0.5086206896551724,
420
+ "grad_norm": 0.28911092877388,
421
+ "learning_rate": 1.9721066711116496e-05,
422
+ "loss": 1.2678132057189941,
423
+ "step": 118
424
+ },
425
+ {
426
+ "epoch": 0.5172413793103449,
427
+ "grad_norm": 0.41628220677375793,
428
+ "learning_rate": 1.9704988855493948e-05,
429
+ "loss": 1.1739540100097656,
430
+ "step": 120
431
+ },
432
+ {
433
+ "epoch": 0.5258620689655172,
434
+ "grad_norm": 0.2931750416755676,
435
+ "learning_rate": 1.9688468233836224e-05,
436
+ "loss": 1.2742866277694702,
437
+ "step": 122
438
+ },
439
+ {
440
+ "epoch": 0.5344827586206896,
441
+ "grad_norm": 0.3954867422580719,
442
+ "learning_rate": 1.9671505686439637e-05,
443
+ "loss": 1.3943136930465698,
444
+ "step": 124
445
+ },
446
+ {
447
+ "epoch": 0.5431034482758621,
448
+ "grad_norm": 0.45475512742996216,
449
+ "learning_rate": 1.9654102076078406e-05,
450
+ "loss": 1.564052939414978,
451
+ "step": 126
452
+ },
453
+ {
454
+ "epoch": 0.5517241379310345,
455
+ "grad_norm": 0.2707035541534424,
456
+ "learning_rate": 1.963625828796073e-05,
457
+ "loss": 1.28196382522583,
458
+ "step": 128
459
+ },
460
+ {
461
+ "epoch": 0.5603448275862069,
462
+ "grad_norm": 0.32626017928123474,
463
+ "learning_rate": 1.9617975229683784e-05,
464
+ "loss": 1.2830570936203003,
465
+ "step": 130
466
+ },
467
+ {
468
+ "epoch": 0.5689655172413793,
469
+ "grad_norm": 0.9617651104927063,
470
+ "learning_rate": 1.959925383118756e-05,
471
+ "loss": 1.1657885313034058,
472
+ "step": 132
473
+ },
474
+ {
475
+ "epoch": 0.5775862068965517,
476
+ "grad_norm": 0.7409307956695557,
477
+ "learning_rate": 1.9580095044707553e-05,
478
+ "loss": 0.7719605565071106,
479
+ "step": 134
480
+ },
481
+ {
482
+ "epoch": 0.5862068965517241,
483
+ "grad_norm": 0.3257262110710144,
484
+ "learning_rate": 1.956049984472634e-05,
485
+ "loss": 0.9887097477912903,
486
+ "step": 136
487
+ },
488
+ {
489
+ "epoch": 0.5948275862068966,
490
+ "grad_norm": 0.13194787502288818,
491
+ "learning_rate": 1.954046922792399e-05,
492
+ "loss": 0.9446215033531189,
493
+ "step": 138
494
+ },
495
+ {
496
+ "epoch": 0.603448275862069,
497
+ "grad_norm": 0.3538772761821747,
498
+ "learning_rate": 1.9520004213127412e-05,
499
+ "loss": 0.9638822078704834,
500
+ "step": 140
501
+ },
502
+ {
503
+ "epoch": 0.6120689655172413,
504
+ "grad_norm": 0.307935893535614,
505
+ "learning_rate": 1.9499105841258488e-05,
506
+ "loss": 1.028863549232483,
507
+ "step": 142
508
+ },
509
+ {
510
+ "epoch": 0.6206896551724138,
511
+ "grad_norm": 0.2776618003845215,
512
+ "learning_rate": 1.9477775175281166e-05,
513
+ "loss": 1.0197412967681885,
514
+ "step": 144
515
+ },
516
+ {
517
+ "epoch": 0.6293103448275862,
518
+ "grad_norm": 0.2247132509946823,
519
+ "learning_rate": 1.945601330014737e-05,
520
+ "loss": 1.2361186742782593,
521
+ "step": 146
522
+ },
523
+ {
524
+ "epoch": 0.6379310344827587,
525
+ "grad_norm": 0.47068798542022705,
526
+ "learning_rate": 1.9433821322741814e-05,
527
+ "loss": 1.2686266899108887,
528
+ "step": 148
529
+ },
530
+ {
531
+ "epoch": 0.646551724137931,
532
+ "grad_norm": 0.35479265451431274,
533
+ "learning_rate": 1.9411200371825724e-05,
534
+ "loss": 1.1161521673202515,
535
+ "step": 150
536
+ },
537
+ {
538
+ "epoch": 0.6551724137931034,
539
+ "grad_norm": 0.9881391525268555,
540
+ "learning_rate": 1.9388151597979407e-05,
541
+ "loss": 0.8658434152603149,
542
+ "step": 152
543
+ },
544
+ {
545
+ "epoch": 0.6637931034482759,
546
+ "grad_norm": 1.1565440893173218,
547
+ "learning_rate": 1.9364676173543734e-05,
548
+ "loss": 1.0216702222824097,
549
+ "step": 154
550
+ },
551
+ {
552
+ "epoch": 0.6724137931034483,
553
+ "grad_norm": 1.531794548034668,
554
+ "learning_rate": 1.934077529256051e-05,
555
+ "loss": 1.0728117227554321,
556
+ "step": 156
557
+ },
558
+ {
559
+ "epoch": 0.6810344827586207,
560
+ "grad_norm": 0.47022947669029236,
561
+ "learning_rate": 1.9316450170711732e-05,
562
+ "loss": 1.2310281991958618,
563
+ "step": 158
564
+ },
565
+ {
566
+ "epoch": 0.6896551724137931,
567
+ "grad_norm": 0.5281792283058167,
568
+ "learning_rate": 1.929170204525779e-05,
569
+ "loss": 1.2321813106536865,
570
+ "step": 160
571
+ },
572
+ {
573
+ "epoch": 0.6982758620689655,
574
+ "grad_norm": 0.5046831369400024,
575
+ "learning_rate": 1.9266532174974476e-05,
576
+ "loss": 1.5227911472320557,
577
+ "step": 162
578
+ },
579
+ {
580
+ "epoch": 0.7068965517241379,
581
+ "grad_norm": 0.4423198103904724,
582
+ "learning_rate": 1.9240941840089015e-05,
583
+ "loss": 0.9460239410400391,
584
+ "step": 164
585
+ },
586
+ {
587
+ "epoch": 0.7155172413793104,
588
+ "grad_norm": 0.4108419120311737,
589
+ "learning_rate": 1.9214932342214923e-05,
590
+ "loss": 1.24778151512146,
591
+ "step": 166
592
+ },
593
+ {
594
+ "epoch": 0.7241379310344828,
595
+ "grad_norm": 0.40896740555763245,
596
+ "learning_rate": 1.9188505004285795e-05,
597
+ "loss": 1.3147329092025757,
598
+ "step": 168
599
+ },
600
+ {
601
+ "epoch": 0.7327586206896551,
602
+ "grad_norm": 0.5323073863983154,
603
+ "learning_rate": 1.916166117048803e-05,
604
+ "loss": 1.223319411277771,
605
+ "step": 170
606
+ },
607
+ {
608
+ "epoch": 0.7413793103448276,
609
+ "grad_norm": 0.5909634232521057,
610
+ "learning_rate": 1.9134402206192468e-05,
611
+ "loss": 1.2373883724212646,
612
+ "step": 172
613
+ },
614
+ {
615
+ "epoch": 0.75,
616
+ "grad_norm": 6.457983016967773,
617
+ "learning_rate": 1.910672949788491e-05,
618
+ "loss": 0.937210202217102,
619
+ "step": 174
620
+ },
621
+ {
622
+ "epoch": 0.7586206896551724,
623
+ "grad_norm": 0.8323431015014648,
624
+ "learning_rate": 1.9078644453095634e-05,
625
+ "loss": 0.6929088234901428,
626
+ "step": 176
627
+ },
628
+ {
629
+ "epoch": 0.7672413793103449,
630
+ "grad_norm": 1.8793997764587402,
631
+ "learning_rate": 1.9050148500327783e-05,
632
+ "loss": 1.1937777996063232,
633
+ "step": 178
634
+ },
635
+ {
636
+ "epoch": 0.7758620689655172,
637
+ "grad_norm": 0.40397757291793823,
638
+ "learning_rate": 1.9021243088984705e-05,
639
+ "loss": 0.9337822198867798,
640
+ "step": 180
641
+ },
642
+ {
643
+ "epoch": 0.7844827586206896,
644
+ "grad_norm": 0.3196965754032135,
645
+ "learning_rate": 1.8991929689296245e-05,
646
+ "loss": 0.9228455424308777,
647
+ "step": 182
648
+ },
649
+ {
650
+ "epoch": 0.7931034482758621,
651
+ "grad_norm": 0.24245284497737885,
652
+ "learning_rate": 1.896220979224395e-05,
653
+ "loss": 0.77471923828125,
654
+ "step": 184
655
+ },
656
+ {
657
+ "epoch": 0.8017241379310345,
658
+ "grad_norm": 0.35483860969543457,
659
+ "learning_rate": 1.8932084909485233e-05,
660
+ "loss": 1.335769772529602,
661
+ "step": 186
662
+ },
663
+ {
664
+ "epoch": 0.8103448275862069,
665
+ "grad_norm": 1.2779498100280762,
666
+ "learning_rate": 1.89015565732765e-05,
667
+ "loss": 1.4579006433486938,
668
+ "step": 188
669
+ },
670
+ {
671
+ "epoch": 0.8189655172413793,
672
+ "grad_norm": 0.5677570104598999,
673
+ "learning_rate": 1.8870626336395204e-05,
674
+ "loss": 1.2512203454971313,
675
+ "step": 190
676
+ },
677
+ {
678
+ "epoch": 0.8275862068965517,
679
+ "grad_norm": 0.3824968934059143,
680
+ "learning_rate": 1.8839295772060856e-05,
681
+ "loss": 1.217123031616211,
682
+ "step": 192
683
+ },
684
+ {
685
+ "epoch": 0.8362068965517241,
686
+ "grad_norm": 0.3712068200111389,
687
+ "learning_rate": 1.8807566473855018e-05,
688
+ "loss": 1.2150022983551025,
689
+ "step": 194
690
+ },
691
+ {
692
+ "epoch": 0.8448275862068966,
693
+ "grad_norm": 0.4604543447494507,
694
+ "learning_rate": 1.8775440055640245e-05,
695
+ "loss": 1.2615362405776978,
696
+ "step": 196
697
+ },
698
+ {
699
+ "epoch": 0.853448275862069,
700
+ "grad_norm": 0.6100837588310242,
701
+ "learning_rate": 1.8742918151478008e-05,
702
+ "loss": 0.9542128443717957,
703
+ "step": 198
704
+ },
705
+ {
706
+ "epoch": 0.8620689655172413,
707
+ "grad_norm": 0.43527752161026,
708
+ "learning_rate": 1.8710002415545552e-05,
709
+ "loss": 1.2213218212127686,
710
+ "step": 200
711
+ },
712
+ {
713
+ "epoch": 0.8706896551724138,
714
+ "grad_norm": 0.37527814507484436,
715
+ "learning_rate": 1.867669452205179e-05,
716
+ "loss": 1.2830454111099243,
717
+ "step": 202
718
+ },
719
+ {
720
+ "epoch": 0.8793103448275862,
721
+ "grad_norm": 1.642726182937622,
722
+ "learning_rate": 1.8642996165152126e-05,
723
+ "loss": 0.8405295610427856,
724
+ "step": 204
725
+ },
726
+ {
727
+ "epoch": 0.8879310344827587,
728
+ "grad_norm": 0.20496216416358948,
729
+ "learning_rate": 1.86089090588623e-05,
730
+ "loss": 1.2193654775619507,
731
+ "step": 206
732
+ },
733
+ {
734
+ "epoch": 0.896551724137931,
735
+ "grad_norm": 0.3890923261642456,
736
+ "learning_rate": 1.857443493697119e-05,
737
+ "loss": 1.2507728338241577,
738
+ "step": 208
739
+ },
740
+ {
741
+ "epoch": 0.9051724137931034,
742
+ "grad_norm": 0.6140430569648743,
743
+ "learning_rate": 1.853957555295265e-05,
744
+ "loss": 0.5853071212768555,
745
+ "step": 210
746
+ },
747
+ {
748
+ "epoch": 0.9137931034482759,
749
+ "grad_norm": 1.1236848831176758,
750
+ "learning_rate": 1.850433267987628e-05,
751
+ "loss": 0.9337871074676514,
752
+ "step": 212
753
+ },
754
+ {
755
+ "epoch": 0.9224137931034483,
756
+ "grad_norm": 1.0812921524047852,
757
+ "learning_rate": 1.84687081103173e-05,
758
+ "loss": 1.1255706548690796,
759
+ "step": 214
760
+ },
761
+ {
762
+ "epoch": 0.9310344827586207,
763
+ "grad_norm": 0.34510231018066406,
764
+ "learning_rate": 1.8432703656265317e-05,
765
+ "loss": 1.2182952165603638,
766
+ "step": 216
767
+ },
768
+ {
769
+ "epoch": 0.9396551724137931,
770
+ "grad_norm": 1.5022380352020264,
771
+ "learning_rate": 1.83963211490322e-05,
772
+ "loss": 0.755200207233429,
773
+ "step": 218
774
+ },
775
+ {
776
+ "epoch": 0.9482758620689655,
777
+ "grad_norm": 1.6012593507766724,
778
+ "learning_rate": 1.8359562439158915e-05,
779
+ "loss": 0.9116122722625732,
780
+ "step": 220
781
+ },
782
+ {
783
+ "epoch": 0.9568965517241379,
784
+ "grad_norm": 0.5603123903274536,
785
+ "learning_rate": 1.8322429396321395e-05,
786
+ "loss": 1.314684271812439,
787
+ "step": 222
788
+ },
789
+ {
790
+ "epoch": 0.9655172413793104,
791
+ "grad_norm": 1.2745404243469238,
792
+ "learning_rate": 1.828492390923547e-05,
793
+ "loss": 0.8659108877182007,
794
+ "step": 224
795
+ },
796
+ {
797
+ "epoch": 0.9741379310344828,
798
+ "grad_norm": 0.2233542650938034,
799
+ "learning_rate": 1.824704788556076e-05,
800
+ "loss": 0.8610473275184631,
801
+ "step": 226
802
+ },
803
+ {
804
+ "epoch": 0.9827586206896551,
805
+ "grad_norm": 0.27494016289711,
806
+ "learning_rate": 1.820880325180368e-05,
807
+ "loss": 1.2119630575180054,
808
+ "step": 228
809
+ },
810
+ {
811
+ "epoch": 0.9913793103448276,
812
+ "grad_norm": 0.8948925733566284,
813
+ "learning_rate": 1.817019195321943e-05,
814
+ "loss": 0.886069118976593,
815
+ "step": 230
816
+ },
817
+ {
818
+ "epoch": 1.0,
819
+ "grad_norm": 0.4077295958995819,
820
+ "learning_rate": 1.8131215953713065e-05,
821
+ "loss": 0.7485522031784058,
822
+ "step": 232
823
+ },
824
+ {
825
+ "epoch": 1.0086206896551724,
826
+ "grad_norm": 0.6140213012695312,
827
+ "learning_rate": 1.809187723573959e-05,
828
+ "loss": 0.8965163826942444,
829
+ "step": 234
830
+ },
831
+ {
832
+ "epoch": 1.0172413793103448,
833
+ "grad_norm": 0.9723572731018066,
834
+ "learning_rate": 1.8052177800203142e-05,
835
+ "loss": 0.745244026184082,
836
+ "step": 236
837
+ },
838
+ {
839
+ "epoch": 1.0258620689655173,
840
+ "grad_norm": 1.8601301908493042,
841
+ "learning_rate": 1.8012119666355207e-05,
842
+ "loss": 0.8928873538970947,
843
+ "step": 238
844
+ },
845
+ {
846
+ "epoch": 1.0344827586206897,
847
+ "grad_norm": 0.7225033044815063,
848
+ "learning_rate": 1.7971704871691913e-05,
849
+ "loss": 0.5402289628982544,
850
+ "step": 240
851
+ },
852
+ {
853
+ "epoch": 1.043103448275862,
854
+ "grad_norm": 0.3545401394367218,
855
+ "learning_rate": 1.7930935471850397e-05,
856
+ "loss": 1.0471925735473633,
857
+ "step": 242
858
+ },
859
+ {
860
+ "epoch": 1.0517241379310345,
861
+ "grad_norm": 0.36405056715011597,
862
+ "learning_rate": 1.7889813540504255e-05,
863
+ "loss": 0.3569323420524597,
864
+ "step": 244
865
+ },
866
+ {
867
+ "epoch": 1.0603448275862069,
868
+ "grad_norm": 0.44639429450035095,
869
+ "learning_rate": 1.7848341169258066e-05,
870
+ "loss": 1.0155720710754395,
871
+ "step": 246
872
+ },
873
+ {
874
+ "epoch": 1.0689655172413792,
875
+ "grad_norm": 0.3922642469406128,
876
+ "learning_rate": 1.780652046754099e-05,
877
+ "loss": 0.9613431096076965,
878
+ "step": 248
879
+ },
880
+ {
881
+ "epoch": 1.0775862068965518,
882
+ "grad_norm": 0.46293577551841736,
883
+ "learning_rate": 1.776435356249949e-05,
884
+ "loss": 0.9193274974822998,
885
+ "step": 250
886
+ },
887
+ {
888
+ "epoch": 1.0862068965517242,
889
+ "grad_norm": 0.6368116736412048,
890
+ "learning_rate": 1.772184259888915e-05,
891
+ "loss": 0.42583054304122925,
892
+ "step": 252
893
+ },
894
+ {
895
+ "epoch": 1.0948275862068966,
896
+ "grad_norm": 1.0385651588439941,
897
+ "learning_rate": 1.7678989738965554e-05,
898
+ "loss": 0.9037254452705383,
899
+ "step": 254
900
+ },
901
+ {
902
+ "epoch": 1.103448275862069,
903
+ "grad_norm": 0.548372209072113,
904
+ "learning_rate": 1.7635797162374325e-05,
905
+ "loss": 0.6561298370361328,
906
+ "step": 256
907
+ },
908
+ {
909
+ "epoch": 1.1120689655172413,
910
+ "grad_norm": 0.8821052312850952,
911
+ "learning_rate": 1.7592267066040266e-05,
912
+ "loss": 0.5268462896347046,
913
+ "step": 258
914
+ },
915
+ {
916
+ "epoch": 1.1206896551724137,
917
+ "grad_norm": 16.59650993347168,
918
+ "learning_rate": 1.7548401664055605e-05,
919
+ "loss": 0.6683530211448669,
920
+ "step": 260
921
+ },
922
+ {
923
+ "epoch": 1.1293103448275863,
924
+ "grad_norm": 0.24353168904781342,
925
+ "learning_rate": 1.7504203187567378e-05,
926
+ "loss": 1.1267895698547363,
927
+ "step": 262
928
+ },
929
+ {
930
+ "epoch": 1.1379310344827587,
931
+ "grad_norm": 0.23041954636573792,
932
+ "learning_rate": 1.7459673884663956e-05,
933
+ "loss": 0.864532470703125,
934
+ "step": 264
935
+ },
936
+ {
937
+ "epoch": 1.146551724137931,
938
+ "grad_norm": 0.24124076962471008,
939
+ "learning_rate": 1.7414816020260694e-05,
940
+ "loss": 1.043933629989624,
941
+ "step": 266
942
+ },
943
+ {
944
+ "epoch": 1.1551724137931034,
945
+ "grad_norm": 0.36801600456237793,
946
+ "learning_rate": 1.7369631875984715e-05,
947
+ "loss": 1.0288755893707275,
948
+ "step": 268
949
+ },
950
+ {
951
+ "epoch": 1.1637931034482758,
952
+ "grad_norm": 0.2811622619628906,
953
+ "learning_rate": 1.7324123750058895e-05,
954
+ "loss": 1.0703890323638916,
955
+ "step": 270
956
+ },
957
+ {
958
+ "epoch": 1.1724137931034484,
959
+ "grad_norm": 0.8666506409645081,
960
+ "learning_rate": 1.7278293957184924e-05,
961
+ "loss": 1.0030405521392822,
962
+ "step": 272
963
+ },
964
+ {
965
+ "epoch": 1.1810344827586208,
966
+ "grad_norm": 0.8689898252487183,
967
+ "learning_rate": 1.7232144828425606e-05,
968
+ "loss": 0.8162426352500916,
969
+ "step": 274
970
+ },
971
+ {
972
+ "epoch": 1.1896551724137931,
973
+ "grad_norm": 0.28137898445129395,
974
+ "learning_rate": 1.718567871108627e-05,
975
+ "loss": 0.5113853216171265,
976
+ "step": 276
977
+ },
978
+ {
979
+ "epoch": 1.1982758620689655,
980
+ "grad_norm": 0.3475714921951294,
981
+ "learning_rate": 1.7138897968595394e-05,
982
+ "loss": 1.0502547025680542,
983
+ "step": 278
984
+ },
985
+ {
986
+ "epoch": 1.206896551724138,
987
+ "grad_norm": 0.31786224246025085,
988
+ "learning_rate": 1.7091804980384382e-05,
989
+ "loss": 0.9660123586654663,
990
+ "step": 280
991
+ },
992
+ {
993
+ "epoch": 1.2155172413793103,
994
+ "grad_norm": 0.297315388917923,
995
+ "learning_rate": 1.7044402141766547e-05,
996
+ "loss": 0.5680896043777466,
997
+ "step": 282
998
+ },
999
+ {
1000
+ "epoch": 1.2241379310344827,
1001
+ "grad_norm": 3.397669553756714,
1002
+ "learning_rate": 1.6996691863815274e-05,
1003
+ "loss": 0.6948413848876953,
1004
+ "step": 284
1005
+ },
1006
+ {
1007
+ "epoch": 1.2327586206896552,
1008
+ "grad_norm": 0.8230817914009094,
1009
+ "learning_rate": 1.694867657324137e-05,
1010
+ "loss": 0.8091163635253906,
1011
+ "step": 286
1012
+ },
1013
+ {
1014
+ "epoch": 1.2413793103448276,
1015
+ "grad_norm": 0.5911218523979187,
1016
+ "learning_rate": 1.6900358712269656e-05,
1017
+ "loss": 0.4585968852043152,
1018
+ "step": 288
1019
+ },
1020
+ {
1021
+ "epoch": 1.25,
1022
+ "grad_norm": 0.30154550075531006,
1023
+ "learning_rate": 1.6851740738514745e-05,
1024
+ "loss": 0.9265656471252441,
1025
+ "step": 290
1026
+ },
1027
+ {
1028
+ "epoch": 1.2586206896551724,
1029
+ "grad_norm": 0.33858662843704224,
1030
+ "learning_rate": 1.680282512485601e-05,
1031
+ "loss": 0.8860665559768677,
1032
+ "step": 292
1033
+ },
1034
+ {
1035
+ "epoch": 1.2672413793103448,
1036
+ "grad_norm": 0.6046611070632935,
1037
+ "learning_rate": 1.675361435931184e-05,
1038
+ "loss": 1.003743052482605,
1039
+ "step": 294
1040
+ },
1041
+ {
1042
+ "epoch": 1.2758620689655173,
1043
+ "grad_norm": 0.265019029378891,
1044
+ "learning_rate": 1.670411094491308e-05,
1045
+ "loss": 0.7721095085144043,
1046
+ "step": 296
1047
+ },
1048
+ {
1049
+ "epoch": 1.2844827586206897,
1050
+ "grad_norm": 0.1970166265964508,
1051
+ "learning_rate": 1.665431739957569e-05,
1052
+ "loss": 1.0421152114868164,
1053
+ "step": 298
1054
+ },
1055
+ {
1056
+ "epoch": 1.293103448275862,
1057
+ "grad_norm": 0.391703337430954,
1058
+ "learning_rate": 1.660423625597273e-05,
1059
+ "loss": 0.9568123817443848,
1060
+ "step": 300
1061
+ },
1062
+ {
1063
+ "epoch": 1.3017241379310345,
1064
+ "grad_norm": 0.28192439675331116,
1065
+ "learning_rate": 1.6553870061405496e-05,
1066
+ "loss": 0.610411524772644,
1067
+ "step": 302
1068
+ },
1069
+ {
1070
+ "epoch": 1.3103448275862069,
1071
+ "grad_norm": 0.2877638339996338,
1072
+ "learning_rate": 1.6503221377673965e-05,
1073
+ "loss": 0.5043181777000427,
1074
+ "step": 304
1075
+ },
1076
+ {
1077
+ "epoch": 1.3189655172413794,
1078
+ "grad_norm": 0.3669877350330353,
1079
+ "learning_rate": 1.6452292780946495e-05,
1080
+ "loss": 1.1974719762802124,
1081
+ "step": 306
1082
+ },
1083
+ {
1084
+ "epoch": 1.3275862068965516,
1085
+ "grad_norm": 0.2513487637042999,
1086
+ "learning_rate": 1.6401086861628802e-05,
1087
+ "loss": 0.6518297791481018,
1088
+ "step": 308
1089
+ },
1090
+ {
1091
+ "epoch": 1.3362068965517242,
1092
+ "grad_norm": 0.6886904835700989,
1093
+ "learning_rate": 1.6349606224232193e-05,
1094
+ "loss": 0.9824197292327881,
1095
+ "step": 310
1096
+ },
1097
+ {
1098
+ "epoch": 1.3448275862068966,
1099
+ "grad_norm": 0.27164769172668457,
1100
+ "learning_rate": 1.6297853487241087e-05,
1101
+ "loss": 0.9464894533157349,
1102
+ "step": 312
1103
+ },
1104
+ {
1105
+ "epoch": 1.353448275862069,
1106
+ "grad_norm": 0.6101466417312622,
1107
+ "learning_rate": 1.6245831282979834e-05,
1108
+ "loss": 1.1367565393447876,
1109
+ "step": 314
1110
+ },
1111
+ {
1112
+ "epoch": 1.3620689655172413,
1113
+ "grad_norm": 0.4131507873535156,
1114
+ "learning_rate": 1.6193542257478837e-05,
1115
+ "loss": 1.3186949491500854,
1116
+ "step": 316
1117
+ },
1118
+ {
1119
+ "epoch": 1.3706896551724137,
1120
+ "grad_norm": 0.39837753772735596,
1121
+ "learning_rate": 1.6140989070339947e-05,
1122
+ "loss": 0.9798977375030518,
1123
+ "step": 318
1124
+ },
1125
+ {
1126
+ "epoch": 1.3793103448275863,
1127
+ "grad_norm": 1.0206124782562256,
1128
+ "learning_rate": 1.60881743946012e-05,
1129
+ "loss": 0.7535231709480286,
1130
+ "step": 320
1131
+ },
1132
+ {
1133
+ "epoch": 1.3879310344827587,
1134
+ "grad_norm": 0.7909939289093018,
1135
+ "learning_rate": 1.6035100916600856e-05,
1136
+ "loss": 1.412162184715271,
1137
+ "step": 322
1138
+ },
1139
+ {
1140
+ "epoch": 1.396551724137931,
1141
+ "grad_norm": 0.604773759841919,
1142
+ "learning_rate": 1.5981771335840744e-05,
1143
+ "loss": 0.5099858641624451,
1144
+ "step": 324
1145
+ },
1146
+ {
1147
+ "epoch": 1.4051724137931034,
1148
+ "grad_norm": 0.4729211926460266,
1149
+ "learning_rate": 1.5928188364848995e-05,
1150
+ "loss": 0.5017666220664978,
1151
+ "step": 326
1152
+ },
1153
+ {
1154
+ "epoch": 1.4137931034482758,
1155
+ "grad_norm": 0.7915529012680054,
1156
+ "learning_rate": 1.5874354729042023e-05,
1157
+ "loss": 0.8372330069541931,
1158
+ "step": 328
1159
+ },
1160
+ {
1161
+ "epoch": 1.4224137931034484,
1162
+ "grad_norm": 0.37416186928749084,
1163
+ "learning_rate": 1.5820273166585947e-05,
1164
+ "loss": 0.7255274057388306,
1165
+ "step": 330
1166
+ },
1167
+ {
1168
+ "epoch": 1.4310344827586206,
1169
+ "grad_norm": 0.6319824457168579,
1170
+ "learning_rate": 1.576594642825729e-05,
1171
+ "loss": 0.6583790183067322,
1172
+ "step": 332
1173
+ },
1174
+ {
1175
+ "epoch": 1.4396551724137931,
1176
+ "grad_norm": 0.2038448005914688,
1177
+ "learning_rate": 1.571137727730307e-05,
1178
+ "loss": 0.5846832990646362,
1179
+ "step": 334
1180
+ },
1181
+ {
1182
+ "epoch": 1.4482758620689655,
1183
+ "grad_norm": 0.442600816488266,
1184
+ "learning_rate": 1.5656568489300265e-05,
1185
+ "loss": 0.6714300513267517,
1186
+ "step": 336
1187
+ },
1188
+ {
1189
+ "epoch": 1.456896551724138,
1190
+ "grad_norm": 1.4567723274230957,
1191
+ "learning_rate": 1.560152285201462e-05,
1192
+ "loss": 1.0285781621932983,
1193
+ "step": 338
1194
+ },
1195
+ {
1196
+ "epoch": 1.4655172413793103,
1197
+ "grad_norm": 0.9464113712310791,
1198
+ "learning_rate": 1.5546243165258857e-05,
1199
+ "loss": 1.34324049949646,
1200
+ "step": 340
1201
+ },
1202
+ {
1203
+ "epoch": 1.4741379310344827,
1204
+ "grad_norm": 0.3753282129764557,
1205
+ "learning_rate": 1.5490732240750275e-05,
1206
+ "loss": 1.1209205389022827,
1207
+ "step": 342
1208
+ },
1209
+ {
1210
+ "epoch": 1.4827586206896552,
1211
+ "grad_norm": 0.3289780020713806,
1212
+ "learning_rate": 1.5434992901967735e-05,
1213
+ "loss": 0.944566547870636,
1214
+ "step": 344
1215
+ },
1216
+ {
1217
+ "epoch": 1.4913793103448276,
1218
+ "grad_norm": 0.5948997735977173,
1219
+ "learning_rate": 1.5379027984008024e-05,
1220
+ "loss": 0.4405466318130493,
1221
+ "step": 346
1222
+ },
1223
+ {
1224
+ "epoch": 1.5,
1225
+ "grad_norm": 1.1603318452835083,
1226
+ "learning_rate": 1.5322840333441687e-05,
1227
+ "loss": 0.8692161440849304,
1228
+ "step": 348
1229
+ },
1230
+ {
1231
+ "epoch": 1.5086206896551724,
1232
+ "grad_norm": 0.3296312093734741,
1233
+ "learning_rate": 1.526643280816822e-05,
1234
+ "loss": 1.0229344367980957,
1235
+ "step": 350
1236
+ },
1237
+ {
1238
+ "epoch": 1.5172413793103448,
1239
+ "grad_norm": 1.4501453638076782,
1240
+ "learning_rate": 1.5209808277270714e-05,
1241
+ "loss": 0.9797338843345642,
1242
+ "step": 352
1243
+ },
1244
+ {
1245
+ "epoch": 1.5258620689655173,
1246
+ "grad_norm": 0.7663815021514893,
1247
+ "learning_rate": 1.515296962086992e-05,
1248
+ "loss": 1.0694770812988281,
1249
+ "step": 354
1250
+ },
1251
+ {
1252
+ "epoch": 1.5344827586206895,
1253
+ "grad_norm": 0.7203207612037659,
1254
+ "learning_rate": 1.5095919729977757e-05,
1255
+ "loss": 0.8981337547302246,
1256
+ "step": 356
1257
+ },
1258
+ {
1259
+ "epoch": 1.543103448275862,
1260
+ "grad_norm": 0.17651817202568054,
1261
+ "learning_rate": 1.5038661506350267e-05,
1262
+ "loss": 0.7348160743713379,
1263
+ "step": 358
1264
+ },
1265
+ {
1266
+ "epoch": 1.5517241379310345,
1267
+ "grad_norm": 0.3057866096496582,
1268
+ "learning_rate": 1.4981197862340015e-05,
1269
+ "loss": 1.3365731239318848,
1270
+ "step": 360
1271
+ },
1272
+ {
1273
+ "epoch": 1.5603448275862069,
1274
+ "grad_norm": 0.2411855012178421,
1275
+ "learning_rate": 1.4923531720747975e-05,
1276
+ "loss": 1.0422999858856201,
1277
+ "step": 362
1278
+ },
1279
+ {
1280
+ "epoch": 1.5689655172413794,
1281
+ "grad_norm": 0.30987802147865295,
1282
+ "learning_rate": 1.4865666014674833e-05,
1283
+ "loss": 1.064208984375,
1284
+ "step": 364
1285
+ },
1286
+ {
1287
+ "epoch": 1.5775862068965516,
1288
+ "grad_norm": 0.4095076024532318,
1289
+ "learning_rate": 1.4807603687371835e-05,
1290
+ "loss": 0.7693328857421875,
1291
+ "step": 366
1292
+ },
1293
+ {
1294
+ "epoch": 1.5862068965517242,
1295
+ "grad_norm": 0.2905380129814148,
1296
+ "learning_rate": 1.4749347692091061e-05,
1297
+ "loss": 1.014492154121399,
1298
+ "step": 368
1299
+ },
1300
+ {
1301
+ "epoch": 1.5948275862068966,
1302
+ "grad_norm": 0.2323673665523529,
1303
+ "learning_rate": 1.4690900991935218e-05,
1304
+ "loss": 0.9833155274391174,
1305
+ "step": 370
1306
+ },
1307
+ {
1308
+ "epoch": 1.603448275862069,
1309
+ "grad_norm": 1.5103213787078857,
1310
+ "learning_rate": 1.4632266559706933e-05,
1311
+ "loss": 1.0599433183670044,
1312
+ "step": 372
1313
+ },
1314
+ {
1315
+ "epoch": 1.6120689655172413,
1316
+ "grad_norm": 0.3394826352596283,
1317
+ "learning_rate": 1.457344737775753e-05,
1318
+ "loss": 0.7715789675712585,
1319
+ "step": 374
1320
+ },
1321
+ {
1322
+ "epoch": 1.6206896551724137,
1323
+ "grad_norm": 0.24792537093162537,
1324
+ "learning_rate": 1.451444643783536e-05,
1325
+ "loss": 1.008685827255249,
1326
+ "step": 376
1327
+ },
1328
+ {
1329
+ "epoch": 1.6293103448275863,
1330
+ "grad_norm": 0.2639085054397583,
1331
+ "learning_rate": 1.4455266740933598e-05,
1332
+ "loss": 1.1693607568740845,
1333
+ "step": 378
1334
+ },
1335
+ {
1336
+ "epoch": 1.6379310344827587,
1337
+ "grad_norm": 0.27107664942741394,
1338
+ "learning_rate": 1.439591129713764e-05,
1339
+ "loss": 0.7295485734939575,
1340
+ "step": 380
1341
+ },
1342
+ {
1343
+ "epoch": 1.646551724137931,
1344
+ "grad_norm": 2.177382707595825,
1345
+ "learning_rate": 1.4336383125471974e-05,
1346
+ "loss": 1.3163925409317017,
1347
+ "step": 382
1348
+ },
1349
+ {
1350
+ "epoch": 1.6551724137931034,
1351
+ "grad_norm": 0.5485315322875977,
1352
+ "learning_rate": 1.4276685253746623e-05,
1353
+ "loss": 1.00065279006958,
1354
+ "step": 384
1355
+ },
1356
+ {
1357
+ "epoch": 1.6637931034482758,
1358
+ "grad_norm": 0.25248005986213684,
1359
+ "learning_rate": 1.4216820718403153e-05,
1360
+ "loss": 0.9677169322967529,
1361
+ "step": 386
1362
+ },
1363
+ {
1364
+ "epoch": 1.6724137931034484,
1365
+ "grad_norm": 0.3703874945640564,
1366
+ "learning_rate": 1.4156792564360222e-05,
1367
+ "loss": 1.3287030458450317,
1368
+ "step": 388
1369
+ },
1370
+ {
1371
+ "epoch": 1.6810344827586206,
1372
+ "grad_norm": 1.0064659118652344,
1373
+ "learning_rate": 1.4096603844858696e-05,
1374
+ "loss": 0.8481764197349548,
1375
+ "step": 390
1376
+ },
1377
+ {
1378
+ "epoch": 1.6896551724137931,
1379
+ "grad_norm": 0.3846883773803711,
1380
+ "learning_rate": 1.4036257621306375e-05,
1381
+ "loss": 0.41724586486816406,
1382
+ "step": 392
1383
+ },
1384
+ {
1385
+ "epoch": 1.6982758620689655,
1386
+ "grad_norm": 1.3841114044189453,
1387
+ "learning_rate": 1.3975756963122242e-05,
1388
+ "loss": 1.0509485006332397,
1389
+ "step": 394
1390
+ },
1391
+ {
1392
+ "epoch": 1.706896551724138,
1393
+ "grad_norm": 0.7592343688011169,
1394
+ "learning_rate": 1.3915104947580382e-05,
1395
+ "loss": 1.0144211053848267,
1396
+ "step": 396
1397
+ },
1398
+ {
1399
+ "epoch": 1.7155172413793105,
1400
+ "grad_norm": 1.6238418817520142,
1401
+ "learning_rate": 1.3854304659653435e-05,
1402
+ "loss": 1.0040127038955688,
1403
+ "step": 398
1404
+ },
1405
+ {
1406
+ "epoch": 1.7241379310344827,
1407
+ "grad_norm": 0.42679479718208313,
1408
+ "learning_rate": 1.3793359191855684e-05,
1409
+ "loss": 1.0181949138641357,
1410
+ "step": 400
1411
+ },
1412
+ {
1413
+ "epoch": 1.7327586206896552,
1414
+ "grad_norm": 0.29004621505737305,
1415
+ "learning_rate": 1.3732271644085779e-05,
1416
+ "loss": 0.7873343229293823,
1417
+ "step": 402
1418
+ },
1419
+ {
1420
+ "epoch": 1.7413793103448276,
1421
+ "grad_norm": 0.2753044366836548,
1422
+ "learning_rate": 1.3671045123469038e-05,
1423
+ "loss": 0.594070553779602,
1424
+ "step": 404
1425
+ },
1426
+ {
1427
+ "epoch": 1.75,
1428
+ "grad_norm": 0.860210657119751,
1429
+ "learning_rate": 1.3609682744199434e-05,
1430
+ "loss": 0.6642279028892517,
1431
+ "step": 406
1432
+ },
1433
+ {
1434
+ "epoch": 1.7586206896551724,
1435
+ "grad_norm": 0.3674435615539551,
1436
+ "learning_rate": 1.354818762738118e-05,
1437
+ "loss": 0.7117359042167664,
1438
+ "step": 408
1439
+ },
1440
+ {
1441
+ "epoch": 1.7672413793103448,
1442
+ "grad_norm": 0.23552259802818298,
1443
+ "learning_rate": 1.3486562900869975e-05,
1444
+ "loss": 0.7889873385429382,
1445
+ "step": 410
1446
+ },
1447
+ {
1448
+ "epoch": 1.7758620689655173,
1449
+ "grad_norm": 0.3450159728527069,
1450
+ "learning_rate": 1.3424811699113924e-05,
1451
+ "loss": 0.7825038433074951,
1452
+ "step": 412
1453
+ },
1454
+ {
1455
+ "epoch": 1.7844827586206895,
1456
+ "grad_norm": 0.6731851100921631,
1457
+ "learning_rate": 1.3362937162994115e-05,
1458
+ "loss": 0.6144068837165833,
1459
+ "step": 414
1460
+ },
1461
+ {
1462
+ "epoch": 1.793103448275862,
1463
+ "grad_norm": 0.29475781321525574,
1464
+ "learning_rate": 1.3300942439664826e-05,
1465
+ "loss": 0.7337537407875061,
1466
+ "step": 416
1467
+ },
1468
+ {
1469
+ "epoch": 1.8017241379310345,
1470
+ "grad_norm": 2.000307083129883,
1471
+ "learning_rate": 1.3238830682393498e-05,
1472
+ "loss": 0.9792155027389526,
1473
+ "step": 418
1474
+ },
1475
+ {
1476
+ "epoch": 1.8103448275862069,
1477
+ "grad_norm": 0.6585050821304321,
1478
+ "learning_rate": 1.317660505040031e-05,
1479
+ "loss": 1.0592936277389526,
1480
+ "step": 420
1481
+ },
1482
+ {
1483
+ "epoch": 1.8189655172413794,
1484
+ "grad_norm": 0.6050742268562317,
1485
+ "learning_rate": 1.3114268708697517e-05,
1486
+ "loss": 1.1423420906066895,
1487
+ "step": 422
1488
+ },
1489
+ {
1490
+ "epoch": 1.8275862068965516,
1491
+ "grad_norm": 0.22784169018268585,
1492
+ "learning_rate": 1.305182482792845e-05,
1493
+ "loss": 1.0336556434631348,
1494
+ "step": 424
1495
+ },
1496
+ {
1497
+ "epoch": 1.8362068965517242,
1498
+ "grad_norm": 0.28103554248809814,
1499
+ "learning_rate": 1.298927658420625e-05,
1500
+ "loss": 0.633764386177063,
1501
+ "step": 426
1502
+ },
1503
+ {
1504
+ "epoch": 1.8448275862068966,
1505
+ "grad_norm": 0.3265269696712494,
1506
+ "learning_rate": 1.292662715895232e-05,
1507
+ "loss": 1.000012993812561,
1508
+ "step": 428
1509
+ },
1510
+ {
1511
+ "epoch": 1.853448275862069,
1512
+ "grad_norm": 0.25722262263298035,
1513
+ "learning_rate": 1.2863879738734515e-05,
1514
+ "loss": 0.972018301486969,
1515
+ "step": 430
1516
+ },
1517
+ {
1518
+ "epoch": 1.8620689655172413,
1519
+ "grad_norm": 0.3606722056865692,
1520
+ "learning_rate": 1.2801037515105048e-05,
1521
+ "loss": 1.0186474323272705,
1522
+ "step": 432
1523
+ },
1524
+ {
1525
+ "epoch": 1.8706896551724137,
1526
+ "grad_norm": 0.5681195855140686,
1527
+ "learning_rate": 1.2738103684438167e-05,
1528
+ "loss": 1.1112819910049438,
1529
+ "step": 434
1530
+ },
1531
+ {
1532
+ "epoch": 1.8793103448275863,
1533
+ "grad_norm": 0.2608093321323395,
1534
+ "learning_rate": 1.267508144776757e-05,
1535
+ "loss": 0.6823700070381165,
1536
+ "step": 436
1537
+ },
1538
+ {
1539
+ "epoch": 1.8879310344827587,
1540
+ "grad_norm": 0.5412386059761047,
1541
+ "learning_rate": 1.2611974010623598e-05,
1542
+ "loss": 0.9664647579193115,
1543
+ "step": 438
1544
+ },
1545
+ {
1546
+ "epoch": 1.896551724137931,
1547
+ "grad_norm": 0.2498435080051422,
1548
+ "learning_rate": 1.2548784582870175e-05,
1549
+ "loss": 0.9077426791191101,
1550
+ "step": 440
1551
+ },
1552
+ {
1553
+ "epoch": 1.9051724137931034,
1554
+ "grad_norm": 0.25406646728515625,
1555
+ "learning_rate": 1.2485516378541562e-05,
1556
+ "loss": 1.095183253288269,
1557
+ "step": 442
1558
+ },
1559
+ {
1560
+ "epoch": 1.9137931034482758,
1561
+ "grad_norm": 0.3728492558002472,
1562
+ "learning_rate": 1.2422172615678867e-05,
1563
+ "loss": 0.2406305968761444,
1564
+ "step": 444
1565
+ },
1566
+ {
1567
+ "epoch": 1.9224137931034484,
1568
+ "grad_norm": 1.3624857664108276,
1569
+ "learning_rate": 1.2358756516166368e-05,
1570
+ "loss": 0.7011273503303528,
1571
+ "step": 446
1572
+ },
1573
+ {
1574
+ "epoch": 1.9310344827586206,
1575
+ "grad_norm": 0.31631484627723694,
1576
+ "learning_rate": 1.2295271305567637e-05,
1577
+ "loss": 1.040317177772522,
1578
+ "step": 448
1579
+ },
1580
+ {
1581
+ "epoch": 1.9396551724137931,
1582
+ "grad_norm": 1.0889196395874023,
1583
+ "learning_rate": 1.2231720212961476e-05,
1584
+ "loss": 1.0238956212997437,
1585
+ "step": 450
1586
+ },
1587
+ {
1588
+ "epoch": 1.9482758620689655,
1589
+ "grad_norm": 0.17187516391277313,
1590
+ "learning_rate": 1.2168106470777676e-05,
1591
+ "loss": 0.8691494464874268,
1592
+ "step": 452
1593
+ },
1594
+ {
1595
+ "epoch": 1.956896551724138,
1596
+ "grad_norm": 0.3179636299610138,
1597
+ "learning_rate": 1.2104433314632605e-05,
1598
+ "loss": 1.03307044506073,
1599
+ "step": 454
1600
+ },
1601
+ {
1602
+ "epoch": 1.9655172413793105,
1603
+ "grad_norm": 0.2639262080192566,
1604
+ "learning_rate": 1.2040703983164628e-05,
1605
+ "loss": 1.0065069198608398,
1606
+ "step": 456
1607
+ },
1608
+ {
1609
+ "epoch": 1.9741379310344827,
1610
+ "grad_norm": 8.715909004211426,
1611
+ "learning_rate": 1.1976921717869387e-05,
1612
+ "loss": 0.3249227702617645,
1613
+ "step": 458
1614
+ },
1615
+ {
1616
+ "epoch": 1.9827586206896552,
1617
+ "grad_norm": 0.4118382930755615,
1618
+ "learning_rate": 1.1913089762934918e-05,
1619
+ "loss": 0.9767223596572876,
1620
+ "step": 460
1621
+ },
1622
+ {
1623
+ "epoch": 1.9913793103448276,
1624
+ "grad_norm": 0.8583071827888489,
1625
+ "learning_rate": 1.184921136507665e-05,
1626
+ "loss": 0.6757299304008484,
1627
+ "step": 462
1628
+ },
1629
+ {
1630
+ "epoch": 2.0,
1631
+ "grad_norm": 1.8066082000732422,
1632
+ "learning_rate": 1.1785289773372252e-05,
1633
+ "loss": 0.8873772621154785,
1634
+ "step": 464
1635
+ },
1636
+ {
1637
+ "epoch": 2.0086206896551726,
1638
+ "grad_norm": 0.31068357825279236,
1639
+ "learning_rate": 1.1721328239096397e-05,
1640
+ "loss": 0.4577694237232208,
1641
+ "step": 466
1642
+ },
1643
+ {
1644
+ "epoch": 2.0172413793103448,
1645
+ "grad_norm": 1.1041500568389893,
1646
+ "learning_rate": 1.1657330015555365e-05,
1647
+ "loss": 0.729465663433075,
1648
+ "step": 468
1649
+ },
1650
+ {
1651
+ "epoch": 2.0258620689655173,
1652
+ "grad_norm": 0.2672136723995209,
1653
+ "learning_rate": 1.159329835792158e-05,
1654
+ "loss": 0.58217453956604,
1655
+ "step": 470
1656
+ },
1657
+ {
1658
+ "epoch": 2.0344827586206895,
1659
+ "grad_norm": 2.1604278087615967,
1660
+ "learning_rate": 1.1529236523068046e-05,
1661
+ "loss": 0.5115339159965515,
1662
+ "step": 472
1663
+ },
1664
+ {
1665
+ "epoch": 2.043103448275862,
1666
+ "grad_norm": 0.3200737237930298,
1667
+ "learning_rate": 1.146514776940268e-05,
1668
+ "loss": 0.5147908329963684,
1669
+ "step": 474
1670
+ },
1671
+ {
1672
+ "epoch": 2.0517241379310347,
1673
+ "grad_norm": 0.6346356272697449,
1674
+ "learning_rate": 1.1401035356702585e-05,
1675
+ "loss": 0.5318877100944519,
1676
+ "step": 476
1677
+ },
1678
+ {
1679
+ "epoch": 2.060344827586207,
1680
+ "grad_norm": 0.5195797681808472,
1681
+ "learning_rate": 1.133690254594825e-05,
1682
+ "loss": 0.3974412977695465,
1683
+ "step": 478
1684
+ },
1685
+ {
1686
+ "epoch": 2.0689655172413794,
1687
+ "grad_norm": 0.3041704297065735,
1688
+ "learning_rate": 1.1272752599157673e-05,
1689
+ "loss": 0.7926555871963501,
1690
+ "step": 480
1691
+ },
1692
+ {
1693
+ "epoch": 2.0775862068965516,
1694
+ "grad_norm": 0.3499699831008911,
1695
+ "learning_rate": 1.1208588779220454e-05,
1696
+ "loss": 0.4678128659725189,
1697
+ "step": 482
1698
+ },
1699
+ {
1700
+ "epoch": 2.086206896551724,
1701
+ "grad_norm": 0.33471551537513733,
1702
+ "learning_rate": 1.1144414349731826e-05,
1703
+ "loss": 0.6200590133666992,
1704
+ "step": 484
1705
+ },
1706
+ {
1707
+ "epoch": 2.0948275862068964,
1708
+ "grad_norm": 0.2929646372795105,
1709
+ "learning_rate": 1.1080232574826669e-05,
1710
+ "loss": 0.5513038039207458,
1711
+ "step": 486
1712
+ },
1713
+ {
1714
+ "epoch": 2.103448275862069,
1715
+ "grad_norm": 1.0158162117004395,
1716
+ "learning_rate": 1.101604671901346e-05,
1717
+ "loss": 0.6488674283027649,
1718
+ "step": 488
1719
+ },
1720
+ {
1721
+ "epoch": 2.1120689655172415,
1722
+ "grad_norm": 0.3108230531215668,
1723
+ "learning_rate": 1.0951860047008268e-05,
1724
+ "loss": 0.6901264786720276,
1725
+ "step": 490
1726
+ },
1727
+ {
1728
+ "epoch": 2.1206896551724137,
1729
+ "grad_norm": 1.0879099369049072,
1730
+ "learning_rate": 1.0887675823568662e-05,
1731
+ "loss": 0.6872219443321228,
1732
+ "step": 492
1733
+ },
1734
+ {
1735
+ "epoch": 2.1293103448275863,
1736
+ "grad_norm": 0.6477283835411072,
1737
+ "learning_rate": 1.0823497313327666e-05,
1738
+ "loss": 0.7422826886177063,
1739
+ "step": 494
1740
+ },
1741
+ {
1742
+ "epoch": 2.1379310344827585,
1743
+ "grad_norm": 0.44943997263908386,
1744
+ "learning_rate": 1.075932778062772e-05,
1745
+ "loss": 0.6594029664993286,
1746
+ "step": 496
1747
+ },
1748
+ {
1749
+ "epoch": 2.146551724137931,
1750
+ "grad_norm": 0.9052010774612427,
1751
+ "learning_rate": 1.0695170489354627e-05,
1752
+ "loss": 0.3747510313987732,
1753
+ "step": 498
1754
+ },
1755
+ {
1756
+ "epoch": 2.1551724137931036,
1757
+ "grad_norm": 0.7915022969245911,
1758
+ "learning_rate": 1.0631028702771558e-05,
1759
+ "loss": 0.3904026746749878,
1760
+ "step": 500
1761
+ },
1762
+ {
1763
+ "epoch": 2.163793103448276,
1764
+ "grad_norm": 0.3351036012172699,
1765
+ "learning_rate": 1.0566905683353048e-05,
1766
+ "loss": 0.42320704460144043,
1767
+ "step": 502
1768
+ },
1769
+ {
1770
+ "epoch": 2.1724137931034484,
1771
+ "grad_norm": 0.39706820249557495,
1772
+ "learning_rate": 1.0502804692619085e-05,
1773
+ "loss": 0.5195796489715576,
1774
+ "step": 504
1775
+ },
1776
+ {
1777
+ "epoch": 2.1810344827586206,
1778
+ "grad_norm": 0.7268723249435425,
1779
+ "learning_rate": 1.0438728990969182e-05,
1780
+ "loss": 0.7010811567306519,
1781
+ "step": 506
1782
+ },
1783
+ {
1784
+ "epoch": 2.189655172413793,
1785
+ "grad_norm": 2.3827953338623047,
1786
+ "learning_rate": 1.0374681837516585e-05,
1787
+ "loss": 0.47865381836891174,
1788
+ "step": 508
1789
+ },
1790
+ {
1791
+ "epoch": 2.1982758620689653,
1792
+ "grad_norm": 0.3070574998855591,
1793
+ "learning_rate": 1.0310666489922452e-05,
1794
+ "loss": 0.7136699557304382,
1795
+ "step": 510
1796
+ },
1797
+ {
1798
+ "epoch": 2.206896551724138,
1799
+ "grad_norm": 0.36808905005455017,
1800
+ "learning_rate": 1.0246686204230213e-05,
1801
+ "loss": 0.5507462024688721,
1802
+ "step": 512
1803
+ },
1804
+ {
1805
+ "epoch": 2.2155172413793105,
1806
+ "grad_norm": 0.49329182505607605,
1807
+ "learning_rate": 1.0182744234699897e-05,
1808
+ "loss": 0.5582562685012817,
1809
+ "step": 514
1810
+ },
1811
+ {
1812
+ "epoch": 2.2241379310344827,
1813
+ "grad_norm": 1.3508820533752441,
1814
+ "learning_rate": 1.0118843833642672e-05,
1815
+ "loss": 0.403819739818573,
1816
+ "step": 516
1817
+ },
1818
+ {
1819
+ "epoch": 2.2327586206896552,
1820
+ "grad_norm": 0.1710726022720337,
1821
+ "learning_rate": 1.005498825125536e-05,
1822
+ "loss": 0.0842498317360878,
1823
+ "step": 518
1824
+ },
1825
+ {
1826
+ "epoch": 2.2413793103448274,
1827
+ "grad_norm": 0.8797479867935181,
1828
+ "learning_rate": 9.991180735455171e-06,
1829
+ "loss": 0.14793738722801208,
1830
+ "step": 520
1831
+ },
1832
+ {
1833
+ "epoch": 2.25,
1834
+ "grad_norm": 0.6412666440010071,
1835
+ "learning_rate": 9.927424531714463e-06,
1836
+ "loss": 0.7687300443649292,
1837
+ "step": 522
1838
+ },
1839
+ {
1840
+ "epoch": 2.2586206896551726,
1841
+ "grad_norm": 0.3679327070713043,
1842
+ "learning_rate": 9.863722882895703e-06,
1843
+ "loss": 0.5136147141456604,
1844
+ "step": 524
1845
+ },
1846
+ {
1847
+ "epoch": 2.2672413793103448,
1848
+ "grad_norm": 0.66014564037323,
1849
+ "learning_rate": 9.800079029086487e-06,
1850
+ "loss": 0.48577654361724854,
1851
+ "step": 526
1852
+ },
1853
+ {
1854
+ "epoch": 2.2758620689655173,
1855
+ "grad_norm": 0.36126595735549927,
1856
+ "learning_rate": 9.736496207434767e-06,
1857
+ "loss": 0.5481493473052979,
1858
+ "step": 528
1859
+ },
1860
+ {
1861
+ "epoch": 2.2844827586206895,
1862
+ "grad_norm": 0.18358245491981506,
1863
+ "learning_rate": 9.672977651984183e-06,
1864
+ "loss": 0.4460916519165039,
1865
+ "step": 530
1866
+ },
1867
+ {
1868
+ "epoch": 2.293103448275862,
1869
+ "grad_norm": 0.5238314867019653,
1870
+ "learning_rate": 9.609526593509572e-06,
1871
+ "loss": 0.44660893082618713,
1872
+ "step": 532
1873
+ },
1874
+ {
1875
+ "epoch": 2.3017241379310347,
1876
+ "grad_norm": 0.45486119389533997,
1877
+ "learning_rate": 9.54614625935264e-06,
1878
+ "loss": 0.4977568984031677,
1879
+ "step": 534
1880
+ },
1881
+ {
1882
+ "epoch": 2.310344827586207,
1883
+ "grad_norm": 0.3980855643749237,
1884
+ "learning_rate": 9.482839873257814e-06,
1885
+ "loss": 0.44564345479011536,
1886
+ "step": 536
1887
+ },
1888
+ {
1889
+ "epoch": 2.3189655172413794,
1890
+ "grad_norm": 0.1288735568523407,
1891
+ "learning_rate": 9.41961065520825e-06,
1892
+ "loss": 0.34829026460647583,
1893
+ "step": 538
1894
+ },
1895
+ {
1896
+ "epoch": 2.3275862068965516,
1897
+ "grad_norm": 0.3644492030143738,
1898
+ "learning_rate": 9.356461821262088e-06,
1899
+ "loss": 0.42181143164634705,
1900
+ "step": 540
1901
+ },
1902
+ {
1903
+ "epoch": 2.336206896551724,
1904
+ "grad_norm": 0.45030930638313293,
1905
+ "learning_rate": 9.293396583388835e-06,
1906
+ "loss": 0.6050075888633728,
1907
+ "step": 542
1908
+ },
1909
+ {
1910
+ "epoch": 2.344827586206897,
1911
+ "grad_norm": 0.49424102902412415,
1912
+ "learning_rate": 9.230418149306028e-06,
1913
+ "loss": 0.7277567982673645,
1914
+ "step": 544
1915
+ },
1916
+ {
1917
+ "epoch": 2.353448275862069,
1918
+ "grad_norm": 0.2475537806749344,
1919
+ "learning_rate": 9.167529722316039e-06,
1920
+ "loss": 0.25289204716682434,
1921
+ "step": 546
1922
+ },
1923
+ {
1924
+ "epoch": 2.3620689655172415,
1925
+ "grad_norm": 0.5678421258926392,
1926
+ "learning_rate": 9.104734501143195e-06,
1927
+ "loss": 0.732049822807312,
1928
+ "step": 548
1929
+ },
1930
+ {
1931
+ "epoch": 2.3706896551724137,
1932
+ "grad_norm": 5.805917739868164,
1933
+ "learning_rate": 9.04203567977102e-06,
1934
+ "loss": 0.561985969543457,
1935
+ "step": 550
1936
+ },
1937
+ {
1938
+ "epoch": 2.3793103448275863,
1939
+ "grad_norm": 1.29057776927948,
1940
+ "learning_rate": 8.979436447279833e-06,
1941
+ "loss": 0.7146962285041809,
1942
+ "step": 552
1943
+ },
1944
+ {
1945
+ "epoch": 2.3879310344827585,
1946
+ "grad_norm": 0.24367839097976685,
1947
+ "learning_rate": 8.916939987684497e-06,
1948
+ "loss": 0.5743066072463989,
1949
+ "step": 554
1950
+ },
1951
+ {
1952
+ "epoch": 2.396551724137931,
1953
+ "grad_norm": 0.29842236638069153,
1954
+ "learning_rate": 8.854549479772508e-06,
1955
+ "loss": 0.6506761312484741,
1956
+ "step": 556
1957
+ },
1958
+ {
1959
+ "epoch": 2.405172413793103,
1960
+ "grad_norm": 0.4994705319404602,
1961
+ "learning_rate": 8.792268096942265e-06,
1962
+ "loss": 0.4716731607913971,
1963
+ "step": 558
1964
+ },
1965
+ {
1966
+ "epoch": 2.413793103448276,
1967
+ "grad_norm": 0.3811204135417938,
1968
+ "learning_rate": 8.730099007041712e-06,
1969
+ "loss": 0.6651344299316406,
1970
+ "step": 560
1971
+ },
1972
+ {
1973
+ "epoch": 2.4224137931034484,
1974
+ "grad_norm": 0.7616989016532898,
1975
+ "learning_rate": 8.668045372207165e-06,
1976
+ "loss": 0.5968424081802368,
1977
+ "step": 562
1978
+ },
1979
+ {
1980
+ "epoch": 2.4310344827586206,
1981
+ "grad_norm": 0.31395435333251953,
1982
+ "learning_rate": 8.606110348702502e-06,
1983
+ "loss": 0.656068742275238,
1984
+ "step": 564
1985
+ },
1986
+ {
1987
+ "epoch": 2.439655172413793,
1988
+ "grad_norm": 3.1607255935668945,
1989
+ "learning_rate": 8.544297086758612e-06,
1990
+ "loss": 0.6287741661071777,
1991
+ "step": 566
1992
+ },
1993
+ {
1994
+ "epoch": 2.4482758620689653,
1995
+ "grad_norm": 0.47689536213874817,
1996
+ "learning_rate": 8.482608730413173e-06,
1997
+ "loss": 0.3020762503147125,
1998
+ "step": 568
1999
+ },
2000
+ {
2001
+ "epoch": 2.456896551724138,
2002
+ "grad_norm": 0.2523362636566162,
2003
+ "learning_rate": 8.421048417350722e-06,
2004
+ "loss": 1.0642119646072388,
2005
+ "step": 570
2006
+ },
2007
+ {
2008
+ "epoch": 2.4655172413793105,
2009
+ "grad_norm": 0.746324896812439,
2010
+ "learning_rate": 8.359619278743072e-06,
2011
+ "loss": 0.46571430563926697,
2012
+ "step": 572
2013
+ },
2014
+ {
2015
+ "epoch": 2.4741379310344827,
2016
+ "grad_norm": 0.2691429555416107,
2017
+ "learning_rate": 8.298324439090044e-06,
2018
+ "loss": 0.5474472641944885,
2019
+ "step": 574
2020
+ },
2021
+ {
2022
+ "epoch": 2.4827586206896552,
2023
+ "grad_norm": 0.3174271881580353,
2024
+ "learning_rate": 8.23716701606056e-06,
2025
+ "loss": 0.7401635050773621,
2026
+ "step": 576
2027
+ },
2028
+ {
2029
+ "epoch": 2.4913793103448274,
2030
+ "grad_norm": 1.0416983366012573,
2031
+ "learning_rate": 8.176150120334035e-06,
2032
+ "loss": 0.4356614053249359,
2033
+ "step": 578
2034
+ },
2035
+ {
2036
+ "epoch": 2.5,
2037
+ "grad_norm": 0.4628841280937195,
2038
+ "learning_rate": 8.115276855442186e-06,
2039
+ "loss": 0.7132285237312317,
2040
+ "step": 580
2041
+ },
2042
+ {
2043
+ "epoch": 2.5086206896551726,
2044
+ "grad_norm": 0.2986053228378296,
2045
+ "learning_rate": 8.054550317611166e-06,
2046
+ "loss": 0.6939521431922913,
2047
+ "step": 582
2048
+ },
2049
+ {
2050
+ "epoch": 2.5172413793103448,
2051
+ "grad_norm": 0.4543410837650299,
2052
+ "learning_rate": 7.99397359560408e-06,
2053
+ "loss": 0.5915881395339966,
2054
+ "step": 584
2055
+ },
2056
+ {
2057
+ "epoch": 2.5258620689655173,
2058
+ "grad_norm": 0.42003390192985535,
2059
+ "learning_rate": 7.933549770563868e-06,
2060
+ "loss": 0.5379845499992371,
2061
+ "step": 586
2062
+ },
2063
+ {
2064
+ "epoch": 2.5344827586206895,
2065
+ "grad_norm": 0.28614842891693115,
2066
+ "learning_rate": 7.873281915856618e-06,
2067
+ "loss": 0.6815016269683838,
2068
+ "step": 588
2069
+ },
2070
+ {
2071
+ "epoch": 2.543103448275862,
2072
+ "grad_norm": 0.32113906741142273,
2073
+ "learning_rate": 7.8131730969152e-06,
2074
+ "loss": 0.7555568218231201,
2075
+ "step": 590
2076
+ },
2077
+ {
2078
+ "epoch": 2.5517241379310347,
2079
+ "grad_norm": 0.3256307542324066,
2080
+ "learning_rate": 7.753226371083398e-06,
2081
+ "loss": 0.6491318941116333,
2082
+ "step": 592
2083
+ },
2084
+ {
2085
+ "epoch": 2.560344827586207,
2086
+ "grad_norm": 0.3508186638355255,
2087
+ "learning_rate": 7.693444787460346e-06,
2088
+ "loss": 0.38945457339286804,
2089
+ "step": 594
2090
+ },
2091
+ {
2092
+ "epoch": 2.5689655172413794,
2093
+ "grad_norm": 0.19432027637958527,
2094
+ "learning_rate": 7.6338313867455e-06,
2095
+ "loss": 0.5126080513000488,
2096
+ "step": 596
2097
+ },
2098
+ {
2099
+ "epoch": 2.5775862068965516,
2100
+ "grad_norm": 0.3698360323905945,
2101
+ "learning_rate": 7.574389201083925e-06,
2102
+ "loss": 0.6170058250427246,
2103
+ "step": 598
2104
+ },
2105
+ {
2106
+ "epoch": 2.586206896551724,
2107
+ "grad_norm": 0.7802625894546509,
2108
+ "learning_rate": 7.515121253912107e-06,
2109
+ "loss": 0.6505869030952454,
2110
+ "step": 600
2111
+ },
2112
+ {
2113
+ "epoch": 2.594827586206897,
2114
+ "grad_norm": 0.12341656535863876,
2115
+ "learning_rate": 7.456030559804146e-06,
2116
+ "loss": 0.47480010986328125,
2117
+ "step": 602
2118
+ },
2119
+ {
2120
+ "epoch": 2.603448275862069,
2121
+ "grad_norm": 0.6200041174888611,
2122
+ "learning_rate": 7.3971201243184425e-06,
2123
+ "loss": 0.5783487558364868,
2124
+ "step": 604
2125
+ },
2126
+ {
2127
+ "epoch": 2.612068965517241,
2128
+ "grad_norm": 0.9620341062545776,
2129
+ "learning_rate": 7.338392943844807e-06,
2130
+ "loss": 0.5066084265708923,
2131
+ "step": 606
2132
+ },
2133
+ {
2134
+ "epoch": 2.6206896551724137,
2135
+ "grad_norm": 0.3163914382457733,
2136
+ "learning_rate": 7.279852005452072e-06,
2137
+ "loss": 0.7519516348838806,
2138
+ "step": 608
2139
+ },
2140
+ {
2141
+ "epoch": 2.6293103448275863,
2142
+ "grad_norm": 0.7371922731399536,
2143
+ "learning_rate": 7.221500286736145e-06,
2144
+ "loss": 0.6951370239257812,
2145
+ "step": 610
2146
+ },
2147
+ {
2148
+ "epoch": 2.637931034482759,
2149
+ "grad_norm": 0.87571781873703,
2150
+ "learning_rate": 7.16334075566856e-06,
2151
+ "loss": 0.7426925301551819,
2152
+ "step": 612
2153
+ },
2154
+ {
2155
+ "epoch": 2.646551724137931,
2156
+ "grad_norm": 0.34502312541007996,
2157
+ "learning_rate": 7.105376370445516e-06,
2158
+ "loss": 0.6338915824890137,
2159
+ "step": 614
2160
+ },
2161
+ {
2162
+ "epoch": 2.655172413793103,
2163
+ "grad_norm": 0.551734209060669,
2164
+ "learning_rate": 7.047610079337426e-06,
2165
+ "loss": 0.43320754170417786,
2166
+ "step": 616
2167
+ },
2168
+ {
2169
+ "epoch": 2.663793103448276,
2170
+ "grad_norm": 0.3614158034324646,
2171
+ "learning_rate": 6.990044820538932e-06,
2172
+ "loss": 0.7193683981895447,
2173
+ "step": 618
2174
+ },
2175
+ {
2176
+ "epoch": 2.6724137931034484,
2177
+ "grad_norm": 0.15170885622501373,
2178
+ "learning_rate": 6.932683522019486e-06,
2179
+ "loss": 0.4045131802558899,
2180
+ "step": 620
2181
+ },
2182
+ {
2183
+ "epoch": 2.6810344827586206,
2184
+ "grad_norm": 0.3657248318195343,
2185
+ "learning_rate": 6.875529101374407e-06,
2186
+ "loss": 0.7818766832351685,
2187
+ "step": 622
2188
+ },
2189
+ {
2190
+ "epoch": 2.689655172413793,
2191
+ "grad_norm": 0.34267038106918335,
2192
+ "learning_rate": 6.818584465676476e-06,
2193
+ "loss": 0.4748293161392212,
2194
+ "step": 624
2195
+ },
2196
+ {
2197
+ "epoch": 2.6982758620689653,
2198
+ "grad_norm": 0.2588467001914978,
2199
+ "learning_rate": 6.761852511328097e-06,
2200
+ "loss": 0.6603924036026001,
2201
+ "step": 626
2202
+ },
2203
+ {
2204
+ "epoch": 2.706896551724138,
2205
+ "grad_norm": 0.5171433687210083,
2206
+ "learning_rate": 6.705336123913943e-06,
2207
+ "loss": 0.4325425326824188,
2208
+ "step": 628
2209
+ },
2210
+ {
2211
+ "epoch": 2.7155172413793105,
2212
+ "grad_norm": 0.5215042233467102,
2213
+ "learning_rate": 6.649038178054216e-06,
2214
+ "loss": 0.8396017551422119,
2215
+ "step": 630
2216
+ },
2217
+ {
2218
+ "epoch": 2.7241379310344827,
2219
+ "grad_norm": 1.0772958993911743,
2220
+ "learning_rate": 6.592961537258415e-06,
2221
+ "loss": 0.34978973865509033,
2222
+ "step": 632
2223
+ },
2224
+ {
2225
+ "epoch": 2.7327586206896552,
2226
+ "grad_norm": 0.2992818057537079,
2227
+ "learning_rate": 6.53710905377969e-06,
2228
+ "loss": 0.7622851133346558,
2229
+ "step": 634
2230
+ },
2231
+ {
2232
+ "epoch": 2.7413793103448274,
2233
+ "grad_norm": 0.5735934972763062,
2234
+ "learning_rate": 6.48148356846978e-06,
2235
+ "loss": 0.5066753625869751,
2236
+ "step": 636
2237
+ },
2238
+ {
2239
+ "epoch": 2.75,
2240
+ "grad_norm": 0.30223897099494934,
2241
+ "learning_rate": 6.426087910634488e-06,
2242
+ "loss": 0.7104685306549072,
2243
+ "step": 638
2244
+ },
2245
+ {
2246
+ "epoch": 2.7586206896551726,
2247
+ "grad_norm": 0.575935959815979,
2248
+ "learning_rate": 6.370924897889815e-06,
2249
+ "loss": 0.45420828461647034,
2250
+ "step": 640
2251
+ },
2252
+ {
2253
+ "epoch": 2.7672413793103448,
2254
+ "grad_norm": 0.6481202244758606,
2255
+ "learning_rate": 6.315997336018602e-06,
2256
+ "loss": 0.6258912086486816,
2257
+ "step": 642
2258
+ },
2259
+ {
2260
+ "epoch": 2.7758620689655173,
2261
+ "grad_norm": 0.3766607940196991,
2262
+ "learning_rate": 6.261308018827857e-06,
2263
+ "loss": 0.8025404214859009,
2264
+ "step": 644
2265
+ },
2266
+ {
2267
+ "epoch": 2.7844827586206895,
2268
+ "grad_norm": 0.47904494404792786,
2269
+ "learning_rate": 6.206859728006616e-06,
2270
+ "loss": 0.6612002849578857,
2271
+ "step": 646
2272
+ },
2273
+ {
2274
+ "epoch": 2.793103448275862,
2275
+ "grad_norm": 0.2501460015773773,
2276
+ "learning_rate": 6.152655232984493e-06,
2277
+ "loss": 0.3472132980823517,
2278
+ "step": 648
2279
+ },
2280
+ {
2281
+ "epoch": 2.8017241379310347,
2282
+ "grad_norm": 1.059028148651123,
2283
+ "learning_rate": 6.0986972907907805e-06,
2284
+ "loss": 0.5596845746040344,
2285
+ "step": 650
2286
+ },
2287
+ {
2288
+ "epoch": 2.810344827586207,
2289
+ "grad_norm": 1.5540813207626343,
2290
+ "learning_rate": 6.044988645914256e-06,
2291
+ "loss": 0.7260704636573792,
2292
+ "step": 652
2293
+ },
2294
+ {
2295
+ "epoch": 2.8189655172413794,
2296
+ "grad_norm": 0.5435662865638733,
2297
+ "learning_rate": 5.991532030163549e-06,
2298
+ "loss": 0.7274701595306396,
2299
+ "step": 654
2300
+ },
2301
+ {
2302
+ "epoch": 2.8275862068965516,
2303
+ "grad_norm": 0.24130015075206757,
2304
+ "learning_rate": 5.938330162528225e-06,
2305
+ "loss": 0.5958088636398315,
2306
+ "step": 656
2307
+ },
2308
+ {
2309
+ "epoch": 2.836206896551724,
2310
+ "grad_norm": 0.32772472500801086,
2311
+ "learning_rate": 5.8853857490404605e-06,
2312
+ "loss": 0.7417409420013428,
2313
+ "step": 658
2314
+ },
2315
+ {
2316
+ "epoch": 2.844827586206897,
2317
+ "grad_norm": 0.34546273946762085,
2318
+ "learning_rate": 5.832701482637433e-06,
2319
+ "loss": 0.19964106380939484,
2320
+ "step": 660
2321
+ },
2322
+ {
2323
+ "epoch": 2.853448275862069,
2324
+ "grad_norm": 0.9264662861824036,
2325
+ "learning_rate": 5.780280043024316e-06,
2326
+ "loss": 0.3988272547721863,
2327
+ "step": 662
2328
+ },
2329
+ {
2330
+ "epoch": 2.862068965517241,
2331
+ "grad_norm": 0.47499191761016846,
2332
+ "learning_rate": 5.728124096538014e-06,
2333
+ "loss": 0.4928115904331207,
2334
+ "step": 664
2335
+ },
2336
+ {
2337
+ "epoch": 2.8706896551724137,
2338
+ "grad_norm": 0.24683460593223572,
2339
+ "learning_rate": 5.676236296011513e-06,
2340
+ "loss": 0.42225298285484314,
2341
+ "step": 666
2342
+ },
2343
+ {
2344
+ "epoch": 2.8793103448275863,
2345
+ "grad_norm": 0.4660603106021881,
2346
+ "learning_rate": 5.624619280638979e-06,
2347
+ "loss": 0.36263778805732727,
2348
+ "step": 668
2349
+ },
2350
+ {
2351
+ "epoch": 2.887931034482759,
2352
+ "grad_norm": 0.8654661774635315,
2353
+ "learning_rate": 5.573275675841481e-06,
2354
+ "loss": 0.720154345035553,
2355
+ "step": 670
2356
+ },
2357
+ {
2358
+ "epoch": 2.896551724137931,
2359
+ "grad_norm": 0.5549874901771545,
2360
+ "learning_rate": 5.522208093133492e-06,
2361
+ "loss": 0.44341281056404114,
2362
+ "step": 672
2363
+ },
2364
+ {
2365
+ "epoch": 2.905172413793103,
2366
+ "grad_norm": 0.3595507740974426,
2367
+ "learning_rate": 5.4714191299900325e-06,
2368
+ "loss": 0.712740957736969,
2369
+ "step": 674
2370
+ },
2371
+ {
2372
+ "epoch": 2.913793103448276,
2373
+ "grad_norm": 0.898064374923706,
2374
+ "learning_rate": 5.4209113697145656e-06,
2375
+ "loss": 0.7482072710990906,
2376
+ "step": 676
2377
+ },
2378
+ {
2379
+ "epoch": 2.9224137931034484,
2380
+ "grad_norm": 0.32744261622428894,
2381
+ "learning_rate": 5.370687381307591e-06,
2382
+ "loss": 0.6825791001319885,
2383
+ "step": 678
2384
+ },
2385
+ {
2386
+ "epoch": 2.9310344827586206,
2387
+ "grad_norm": 0.3922792971134186,
2388
+ "learning_rate": 5.320749719335997e-06,
2389
+ "loss": 0.5156247615814209,
2390
+ "step": 680
2391
+ },
2392
+ {
2393
+ "epoch": 2.939655172413793,
2394
+ "grad_norm": 0.21706661581993103,
2395
+ "learning_rate": 5.271100923803095e-06,
2396
+ "loss": 0.29097798466682434,
2397
+ "step": 682
2398
+ },
2399
+ {
2400
+ "epoch": 2.9482758620689653,
2401
+ "grad_norm": 0.2500763237476349,
2402
+ "learning_rate": 5.2217435200194595e-06,
2403
+ "loss": 0.42494314908981323,
2404
+ "step": 684
2405
+ },
2406
+ {
2407
+ "epoch": 2.956896551724138,
2408
+ "grad_norm": 0.28385624289512634,
2409
+ "learning_rate": 5.172680018474456e-06,
2410
+ "loss": 0.6773520708084106,
2411
+ "step": 686
2412
+ },
2413
+ {
2414
+ "epoch": 2.9655172413793105,
2415
+ "grad_norm": 0.32070091366767883,
2416
+ "learning_rate": 5.123912914708566e-06,
2417
+ "loss": 0.6991716027259827,
2418
+ "step": 688
2419
+ },
2420
+ {
2421
+ "epoch": 2.9741379310344827,
2422
+ "grad_norm": 0.28078293800354004,
2423
+ "learning_rate": 5.075444689186439e-06,
2424
+ "loss": 0.9952356815338135,
2425
+ "step": 690
2426
+ },
2427
+ {
2428
+ "epoch": 2.9827586206896552,
2429
+ "grad_norm": 0.30109333992004395,
2430
+ "learning_rate": 5.02727780717075e-06,
2431
+ "loss": 0.9628766179084778,
2432
+ "step": 692
2433
+ },
2434
+ {
2435
+ "epoch": 2.9913793103448274,
2436
+ "grad_norm": 0.4280763864517212,
2437
+ "learning_rate": 4.979414718596779e-06,
2438
+ "loss": 0.46527406573295593,
2439
+ "step": 694
2440
+ },
2441
+ {
2442
+ "epoch": 3.0,
2443
+ "grad_norm": 0.32468336820602417,
2444
+ "learning_rate": 4.931857857947829e-06,
2445
+ "loss": 0.8153132200241089,
2446
+ "step": 696
2447
+ },
2448
+ {
2449
+ "epoch": 3.0086206896551726,
2450
+ "grad_norm": 0.38036271929740906,
2451
+ "learning_rate": 4.884609644131373e-06,
2452
+ "loss": 0.2907155454158783,
2453
+ "step": 698
2454
+ },
2455
+ {
2456
+ "epoch": 3.0172413793103448,
2457
+ "grad_norm": 0.28837043046951294,
2458
+ "learning_rate": 4.837672480356037e-06,
2459
+ "loss": 0.4242869019508362,
2460
+ "step": 700
2461
+ },
2462
+ {
2463
+ "epoch": 3.0258620689655173,
2464
+ "grad_norm": 0.582526683807373,
2465
+ "learning_rate": 4.791048754009362e-06,
2466
+ "loss": 0.3136082589626312,
2467
+ "step": 702
2468
+ },
2469
+ {
2470
+ "epoch": 3.0344827586206895,
2471
+ "grad_norm": 0.7384637594223022,
2472
+ "learning_rate": 4.7447408365363616e-06,
2473
+ "loss": 0.6226381063461304,
2474
+ "step": 704
2475
+ },
2476
+ {
2477
+ "epoch": 3.043103448275862,
2478
+ "grad_norm": 0.09337347000837326,
2479
+ "learning_rate": 4.698751083318918e-06,
2480
+ "loss": 0.0870320051908493,
2481
+ "step": 706
2482
+ },
2483
+ {
2484
+ "epoch": 3.0517241379310347,
2485
+ "grad_norm": 1.523980736732483,
2486
+ "learning_rate": 4.653081833555975e-06,
2487
+ "loss": 0.379879891872406,
2488
+ "step": 708
2489
+ },
2490
+ {
2491
+ "epoch": 3.060344827586207,
2492
+ "grad_norm": 0.2947269380092621,
2493
+ "learning_rate": 4.60773541014454e-06,
2494
+ "loss": 0.10685593634843826,
2495
+ "step": 710
2496
+ },
2497
+ {
2498
+ "epoch": 3.0689655172413794,
2499
+ "grad_norm": 0.9110690951347351,
2500
+ "learning_rate": 4.562714119561566e-06,
2501
+ "loss": 0.40309441089630127,
2502
+ "step": 712
2503
+ },
2504
+ {
2505
+ "epoch": 3.0775862068965516,
2506
+ "grad_norm": 0.4863778054714203,
2507
+ "learning_rate": 4.518020251746607e-06,
2508
+ "loss": 0.6567556262016296,
2509
+ "step": 714
2510
+ },
2511
+ {
2512
+ "epoch": 3.086206896551724,
2513
+ "grad_norm": 0.4967314898967743,
2514
+ "learning_rate": 4.473656079985362e-06,
2515
+ "loss": 0.14177639782428741,
2516
+ "step": 716
2517
+ },
2518
+ {
2519
+ "epoch": 3.0948275862068964,
2520
+ "grad_norm": 0.4328520596027374,
2521
+ "learning_rate": 4.429623860794032e-06,
2522
+ "loss": 0.3785516917705536,
2523
+ "step": 718
2524
+ },
2525
+ {
2526
+ "epoch": 3.103448275862069,
2527
+ "grad_norm": 0.7979432344436646,
2528
+ "learning_rate": 4.38592583380457e-06,
2529
+ "loss": 0.18618693947792053,
2530
+ "step": 720
2531
+ },
2532
+ {
2533
+ "epoch": 3.1120689655172415,
2534
+ "grad_norm": 0.3564269542694092,
2535
+ "learning_rate": 4.3425642216507325e-06,
2536
+ "loss": 0.20070935785770416,
2537
+ "step": 722
2538
+ },
2539
+ {
2540
+ "epoch": 3.1206896551724137,
2541
+ "grad_norm": 0.34848275780677795,
2542
+ "learning_rate": 4.2995412298550624e-06,
2543
+ "loss": 0.3287600874900818,
2544
+ "step": 724
2545
+ },
2546
+ {
2547
+ "epoch": 3.1293103448275863,
2548
+ "grad_norm": 0.4305904507637024,
2549
+ "learning_rate": 4.256859046716677e-06,
2550
+ "loss": 0.36046263575553894,
2551
+ "step": 726
2552
+ },
2553
+ {
2554
+ "epoch": 3.1379310344827585,
2555
+ "grad_norm": 0.708617627620697,
2556
+ "learning_rate": 4.214519843199995e-06,
2557
+ "loss": 0.4107988774776459,
2558
+ "step": 728
2559
+ },
2560
+ {
2561
+ "epoch": 3.146551724137931,
2562
+ "grad_norm": 0.0694451779127121,
2563
+ "learning_rate": 4.172525772824283e-06,
2564
+ "loss": 0.2075681984424591,
2565
+ "step": 730
2566
+ },
2567
+ {
2568
+ "epoch": 3.1551724137931036,
2569
+ "grad_norm": 0.164693221449852,
2570
+ "learning_rate": 4.130878971554148e-06,
2571
+ "loss": 0.17175163328647614,
2572
+ "step": 732
2573
+ },
2574
+ {
2575
+ "epoch": 3.163793103448276,
2576
+ "grad_norm": 0.36916273832321167,
2577
+ "learning_rate": 4.0895815576908686e-06,
2578
+ "loss": 0.3509948253631592,
2579
+ "step": 734
2580
+ },
2581
+ {
2582
+ "epoch": 3.1724137931034484,
2583
+ "grad_norm": 0.2566669285297394,
2584
+ "learning_rate": 4.048635631764673e-06,
2585
+ "loss": 0.22303853929042816,
2586
+ "step": 736
2587
+ },
2588
+ {
2589
+ "epoch": 3.1810344827586206,
2590
+ "grad_norm": 1.5648850202560425,
2591
+ "learning_rate": 4.008043276427886e-06,
2592
+ "loss": 0.523650050163269,
2593
+ "step": 738
2594
+ },
2595
+ {
2596
+ "epoch": 3.189655172413793,
2597
+ "grad_norm": 0.22325409948825836,
2598
+ "learning_rate": 3.967806556349001e-06,
2599
+ "loss": 0.08064466714859009,
2600
+ "step": 740
2601
+ },
2602
+ {
2603
+ "epoch": 3.1982758620689653,
2604
+ "grad_norm": 0.08049521595239639,
2605
+ "learning_rate": 3.927927518107664e-06,
2606
+ "loss": 0.17742744088172913,
2607
+ "step": 742
2608
+ },
2609
+ {
2610
+ "epoch": 3.206896551724138,
2611
+ "grad_norm": 0.3570193350315094,
2612
+ "learning_rate": 3.888408190090583e-06,
2613
+ "loss": 0.2840654253959656,
2614
+ "step": 744
2615
+ },
2616
+ {
2617
+ "epoch": 3.2155172413793105,
2618
+ "grad_norm": 0.5950610041618347,
2619
+ "learning_rate": 3.8492505823883394e-06,
2620
+ "loss": 0.27302879095077515,
2621
+ "step": 746
2622
+ },
2623
+ {
2624
+ "epoch": 3.2241379310344827,
2625
+ "grad_norm": 0.6008321046829224,
2626
+ "learning_rate": 3.810456686693178e-06,
2627
+ "loss": 0.38051924109458923,
2628
+ "step": 748
2629
+ },
2630
+ {
2631
+ "epoch": 3.2327586206896552,
2632
+ "grad_norm": 0.2667677104473114,
2633
+ "learning_rate": 3.7720284761976657e-06,
2634
+ "loss": 0.33380451798439026,
2635
+ "step": 750
2636
+ },
2637
+ {
2638
+ "epoch": 3.2413793103448274,
2639
+ "grad_norm": 1.0264617204666138,
2640
+ "learning_rate": 3.7339679054943645e-06,
2641
+ "loss": 0.2860594093799591,
2642
+ "step": 752
2643
+ },
2644
+ {
2645
+ "epoch": 3.25,
2646
+ "grad_norm": 0.5752837657928467,
2647
+ "learning_rate": 3.696276910476382e-06,
2648
+ "loss": 0.36991697549819946,
2649
+ "step": 754
2650
+ },
2651
+ {
2652
+ "epoch": 3.2586206896551726,
2653
+ "grad_norm": 0.29292166233062744,
2654
+ "learning_rate": 3.6589574082389324e-06,
2655
+ "loss": 0.3221489191055298,
2656
+ "step": 756
2657
+ },
2658
+ {
2659
+ "epoch": 3.2672413793103448,
2660
+ "grad_norm": 0.35932210087776184,
2661
+ "learning_rate": 3.6220112969818045e-06,
2662
+ "loss": 0.37856197357177734,
2663
+ "step": 758
2664
+ },
2665
+ {
2666
+ "epoch": 3.2758620689655173,
2667
+ "grad_norm": 0.11113092303276062,
2668
+ "learning_rate": 3.5854404559128287e-06,
2669
+ "loss": 0.24704982340335846,
2670
+ "step": 760
2671
+ },
2672
+ {
2673
+ "epoch": 3.2844827586206895,
2674
+ "grad_norm": 1.6768494844436646,
2675
+ "learning_rate": 3.549246745152283e-06,
2676
+ "loss": 0.36558791995048523,
2677
+ "step": 762
2678
+ },
2679
+ {
2680
+ "epoch": 3.293103448275862,
2681
+ "grad_norm": 0.4623495638370514,
2682
+ "learning_rate": 3.5134320056382877e-06,
2683
+ "loss": 0.3346759080886841,
2684
+ "step": 764
2685
+ },
2686
+ {
2687
+ "epoch": 3.3017241379310347,
2688
+ "grad_norm": 0.18975688517093658,
2689
+ "learning_rate": 3.477998059033163e-06,
2690
+ "loss": 0.2471802681684494,
2691
+ "step": 766
2692
+ },
2693
+ {
2694
+ "epoch": 3.310344827586207,
2695
+ "grad_norm": 0.2749144434928894,
2696
+ "learning_rate": 3.4429467076307823e-06,
2697
+ "loss": 0.34330496191978455,
2698
+ "step": 768
2699
+ },
2700
+ {
2701
+ "epoch": 3.3189655172413794,
2702
+ "grad_norm": 0.7870236039161682,
2703
+ "learning_rate": 3.4082797342648867e-06,
2704
+ "loss": 0.10211128741502762,
2705
+ "step": 770
2706
+ },
2707
+ {
2708
+ "epoch": 3.3275862068965516,
2709
+ "grad_norm": 0.4417794346809387,
2710
+ "learning_rate": 3.373998902218423e-06,
2711
+ "loss": 0.22283262014389038,
2712
+ "step": 772
2713
+ },
2714
+ {
2715
+ "epoch": 3.336206896551724,
2716
+ "grad_norm": 2.35225510597229,
2717
+ "learning_rate": 3.3401059551338326e-06,
2718
+ "loss": 0.3982272148132324,
2719
+ "step": 774
2720
+ },
2721
+ {
2722
+ "epoch": 3.344827586206897,
2723
+ "grad_norm": 0.32708537578582764,
2724
+ "learning_rate": 3.3066026169243916e-06,
2725
+ "loss": 0.35799640417099,
2726
+ "step": 776
2727
+ },
2728
+ {
2729
+ "epoch": 3.353448275862069,
2730
+ "grad_norm": 0.39861956238746643,
2731
+ "learning_rate": 3.2734905916865045e-06,
2732
+ "loss": 0.33809182047843933,
2733
+ "step": 778
2734
+ },
2735
+ {
2736
+ "epoch": 3.3620689655172415,
2737
+ "grad_norm": 2.7856783866882324,
2738
+ "learning_rate": 3.2407715636130365e-06,
2739
+ "loss": 0.24870137870311737,
2740
+ "step": 780
2741
+ },
2742
+ {
2743
+ "epoch": 3.3706896551724137,
2744
+ "grad_norm": 0.3867567777633667,
2745
+ "learning_rate": 3.20844719690765e-06,
2746
+ "loss": 0.25115785002708435,
2747
+ "step": 782
2748
+ },
2749
+ {
2750
+ "epoch": 3.3793103448275863,
2751
+ "grad_norm": 0.2818472385406494,
2752
+ "learning_rate": 3.1765191357001596e-06,
2753
+ "loss": 0.34631893038749695,
2754
+ "step": 784
2755
+ },
2756
+ {
2757
+ "epoch": 3.3879310344827585,
2758
+ "grad_norm": 0.6454493999481201,
2759
+ "learning_rate": 3.1449890039628956e-06,
2760
+ "loss": 0.21534167230129242,
2761
+ "step": 786
2762
+ },
2763
+ {
2764
+ "epoch": 3.396551724137931,
2765
+ "grad_norm": 0.2576633095741272,
2766
+ "learning_rate": 3.1138584054281177e-06,
2767
+ "loss": 0.38198593258857727,
2768
+ "step": 788
2769
+ },
2770
+ {
2771
+ "epoch": 3.405172413793103,
2772
+ "grad_norm": 0.3199053704738617,
2773
+ "learning_rate": 3.0831289235064297e-06,
2774
+ "loss": 0.3233962655067444,
2775
+ "step": 790
2776
+ },
2777
+ {
2778
+ "epoch": 3.413793103448276,
2779
+ "grad_norm": 0.24734322726726532,
2780
+ "learning_rate": 3.0528021212062536e-06,
2781
+ "loss": 0.39972445368766785,
2782
+ "step": 792
2783
+ },
2784
+ {
2785
+ "epoch": 3.4224137931034484,
2786
+ "grad_norm": 0.7525281310081482,
2787
+ "learning_rate": 3.0228795410543182e-06,
2788
+ "loss": 0.2797224819660187,
2789
+ "step": 794
2790
+ },
2791
+ {
2792
+ "epoch": 3.4310344827586206,
2793
+ "grad_norm": 0.38152387738227844,
2794
+ "learning_rate": 2.9933627050172152e-06,
2795
+ "loss": 0.4961601793766022,
2796
+ "step": 796
2797
+ },
2798
+ {
2799
+ "epoch": 3.439655172413793,
2800
+ "grad_norm": 0.6566803455352783,
2801
+ "learning_rate": 2.96425311442397e-06,
2802
+ "loss": 0.2608637511730194,
2803
+ "step": 798
2804
+ },
2805
+ {
2806
+ "epoch": 3.4482758620689653,
2807
+ "grad_norm": 0.2898580729961395,
2808
+ "learning_rate": 2.9355522498896914e-06,
2809
+ "loss": 0.389303594827652,
2810
+ "step": 800
2811
+ },
2812
+ {
2813
+ "epoch": 3.456896551724138,
2814
+ "grad_norm": 0.2688350975513458,
2815
+ "learning_rate": 2.907261571240257e-06,
2816
+ "loss": 0.416867733001709,
2817
+ "step": 802
2818
+ },
2819
+ {
2820
+ "epoch": 3.4655172413793105,
2821
+ "grad_norm": 0.3749980032444,
2822
+ "learning_rate": 2.879382517438062e-06,
2823
+ "loss": 0.28168269991874695,
2824
+ "step": 804
2825
+ },
2826
+ {
2827
+ "epoch": 3.4741379310344827,
2828
+ "grad_norm": 0.5589833855628967,
2829
+ "learning_rate": 2.8519165065088293e-06,
2830
+ "loss": 0.39084625244140625,
2831
+ "step": 806
2832
+ },
2833
+ {
2834
+ "epoch": 3.4827586206896552,
2835
+ "grad_norm": 0.2921109199523926,
2836
+ "learning_rate": 2.824864935469487e-06,
2837
+ "loss": 0.3399796187877655,
2838
+ "step": 808
2839
+ },
2840
+ {
2841
+ "epoch": 3.4913793103448274,
2842
+ "grad_norm": 0.22237980365753174,
2843
+ "learning_rate": 2.7982291802570983e-06,
2844
+ "loss": 0.10816089808940887,
2845
+ "step": 810
2846
+ },
2847
+ {
2848
+ "epoch": 3.5,
2849
+ "grad_norm": 0.608170747756958,
2850
+ "learning_rate": 2.7720105956588983e-06,
2851
+ "loss": 0.3328501880168915,
2852
+ "step": 812
2853
+ },
2854
+ {
2855
+ "epoch": 3.5086206896551726,
2856
+ "grad_norm": 1.1251869201660156,
2857
+ "learning_rate": 2.7462105152433617e-06,
2858
+ "loss": 0.4947846531867981,
2859
+ "step": 814
2860
+ },
2861
+ {
2862
+ "epoch": 3.5172413793103448,
2863
+ "grad_norm": 0.5909225344657898,
2864
+ "learning_rate": 2.720830251292391e-06,
2865
+ "loss": 0.3645175099372864,
2866
+ "step": 816
2867
+ },
2868
+ {
2869
+ "epoch": 3.5258620689655173,
2870
+ "grad_norm": 0.3970613181591034,
2871
+ "learning_rate": 2.695871094734559e-06,
2872
+ "loss": 0.4159550368785858,
2873
+ "step": 818
2874
+ },
2875
+ {
2876
+ "epoch": 3.5344827586206895,
2877
+ "grad_norm": 0.3088339567184448,
2878
+ "learning_rate": 2.6713343150794506e-06,
2879
+ "loss": 0.31124040484428406,
2880
+ "step": 820
2881
+ },
2882
+ {
2883
+ "epoch": 3.543103448275862,
2884
+ "grad_norm": 0.8232600688934326,
2885
+ "learning_rate": 2.6472211603530894e-06,
2886
+ "loss": 0.2729651927947998,
2887
+ "step": 822
2888
+ },
2889
+ {
2890
+ "epoch": 3.5517241379310347,
2891
+ "grad_norm": 0.47030189633369446,
2892
+ "learning_rate": 2.6235328570344666e-06,
2893
+ "loss": 0.27673032879829407,
2894
+ "step": 824
2895
+ },
2896
+ {
2897
+ "epoch": 3.560344827586207,
2898
+ "grad_norm": 0.07642904669046402,
2899
+ "learning_rate": 2.600270609993143e-06,
2900
+ "loss": 0.0557853989303112,
2901
+ "step": 826
2902
+ },
2903
+ {
2904
+ "epoch": 3.5689655172413794,
2905
+ "grad_norm": 1.2752304077148438,
2906
+ "learning_rate": 2.5774356024279827e-06,
2907
+ "loss": 0.18158602714538574,
2908
+ "step": 828
2909
+ },
2910
+ {
2911
+ "epoch": 3.5775862068965516,
2912
+ "grad_norm": 0.2896014153957367,
2913
+ "learning_rate": 2.555028995806956e-06,
2914
+ "loss": 0.18923550844192505,
2915
+ "step": 830
2916
+ },
2917
+ {
2918
+ "epoch": 3.586206896551724,
2919
+ "grad_norm": 0.3300286829471588,
2920
+ "learning_rate": 2.5330519298080757e-06,
2921
+ "loss": 0.22178103029727936,
2922
+ "step": 832
2923
+ },
2924
+ {
2925
+ "epoch": 3.594827586206897,
2926
+ "grad_norm": 0.47191181778907776,
2927
+ "learning_rate": 2.511505522261417e-06,
2928
+ "loss": 0.43088600039482117,
2929
+ "step": 834
2930
+ },
2931
+ {
2932
+ "epoch": 3.603448275862069,
2933
+ "grad_norm": 0.5750867128372192,
2934
+ "learning_rate": 2.4903908690922746e-06,
2935
+ "loss": 0.3035900592803955,
2936
+ "step": 836
2937
+ },
2938
+ {
2939
+ "epoch": 3.612068965517241,
2940
+ "grad_norm": 0.31900304555892944,
2941
+ "learning_rate": 2.4697090442654055e-06,
2942
+ "loss": 0.3073073923587799,
2943
+ "step": 838
2944
+ },
2945
+ {
2946
+ "epoch": 3.6206896551724137,
2947
+ "grad_norm": 0.9884761571884155,
2948
+ "learning_rate": 2.4494610997304156e-06,
2949
+ "loss": 0.1451842337846756,
2950
+ "step": 840
2951
+ },
2952
+ {
2953
+ "epoch": 3.6293103448275863,
2954
+ "grad_norm": 0.5963663458824158,
2955
+ "learning_rate": 2.4296480653682444e-06,
2956
+ "loss": 0.19056405127048492,
2957
+ "step": 842
2958
+ },
2959
+ {
2960
+ "epoch": 3.637931034482759,
2961
+ "grad_norm": 0.4518829882144928,
2962
+ "learning_rate": 2.4102709489387925e-06,
2963
+ "loss": 0.3310995101928711,
2964
+ "step": 844
2965
+ },
2966
+ {
2967
+ "epoch": 3.646551724137931,
2968
+ "grad_norm": 13.950098991394043,
2969
+ "learning_rate": 2.391330736029649e-06,
2970
+ "loss": 0.27216413617134094,
2971
+ "step": 846
2972
+ },
2973
+ {
2974
+ "epoch": 3.655172413793103,
2975
+ "grad_norm": 1.5341519117355347,
2976
+ "learning_rate": 2.3728283900059756e-06,
2977
+ "loss": 0.13349619507789612,
2978
+ "step": 848
2979
+ },
2980
+ {
2981
+ "epoch": 3.663793103448276,
2982
+ "grad_norm": 0.07572399079799652,
2983
+ "learning_rate": 2.3547648519614967e-06,
2984
+ "loss": 0.20533311367034912,
2985
+ "step": 850
2986
+ },
2987
+ {
2988
+ "epoch": 3.6724137931034484,
2989
+ "grad_norm": 0.36925849318504333,
2990
+ "learning_rate": 2.3371410406706356e-06,
2991
+ "loss": 0.38371843099594116,
2992
+ "step": 852
2993
+ },
2994
+ {
2995
+ "epoch": 3.6810344827586206,
2996
+ "grad_norm": 0.9557275176048279,
2997
+ "learning_rate": 2.319957852541786e-06,
2998
+ "loss": 0.2169187366962433,
2999
+ "step": 854
3000
+ },
3001
+ {
3002
+ "epoch": 3.689655172413793,
3003
+ "grad_norm": 0.5187146663665771,
3004
+ "learning_rate": 2.303216161571706e-06,
3005
+ "loss": 0.37597841024398804,
3006
+ "step": 856
3007
+ },
3008
+ {
3009
+ "epoch": 3.6982758620689653,
3010
+ "grad_norm": 0.2956836223602295,
3011
+ "learning_rate": 2.2869168193010793e-06,
3012
+ "loss": 0.4011721611022949,
3013
+ "step": 858
3014
+ },
3015
+ {
3016
+ "epoch": 3.706896551724138,
3017
+ "grad_norm": 0.09843911975622177,
3018
+ "learning_rate": 2.2710606547711917e-06,
3019
+ "loss": 0.18101970851421356,
3020
+ "step": 860
3021
+ },
3022
+ {
3023
+ "epoch": 3.7155172413793105,
3024
+ "grad_norm": 0.6112839579582214,
3025
+ "learning_rate": 2.2556484744817635e-06,
3026
+ "loss": 0.23301751911640167,
3027
+ "step": 862
3028
+ },
3029
+ {
3030
+ "epoch": 3.7241379310344827,
3031
+ "grad_norm": 0.03941560536623001,
3032
+ "learning_rate": 2.2406810623499382e-06,
3033
+ "loss": 0.1944858729839325,
3034
+ "step": 864
3035
+ },
3036
+ {
3037
+ "epoch": 3.7327586206896552,
3038
+ "grad_norm": 0.2406272292137146,
3039
+ "learning_rate": 2.226159179670395e-06,
3040
+ "loss": 0.11213114112615585,
3041
+ "step": 866
3042
+ },
3043
+ {
3044
+ "epoch": 3.7413793103448274,
3045
+ "grad_norm": 0.9250267148017883,
3046
+ "learning_rate": 2.2120835650766386e-06,
3047
+ "loss": 0.25432637333869934,
3048
+ "step": 868
3049
+ },
3050
+ {
3051
+ "epoch": 3.75,
3052
+ "grad_norm": 0.3792153596878052,
3053
+ "learning_rate": 2.1984549345034234e-06,
3054
+ "loss": 0.2917807102203369,
3055
+ "step": 870
3056
+ },
3057
+ {
3058
+ "epoch": 3.7586206896551726,
3059
+ "grad_norm": 0.050461042672395706,
3060
+ "learning_rate": 2.1852739811503433e-06,
3061
+ "loss": 0.1030154600739479,
3062
+ "step": 872
3063
+ },
3064
+ {
3065
+ "epoch": 3.7672413793103448,
3066
+ "grad_norm": 0.03659242019057274,
3067
+ "learning_rate": 2.172541375446566e-06,
3068
+ "loss": 0.24243952333927155,
3069
+ "step": 874
3070
+ },
3071
+ {
3072
+ "epoch": 3.7758620689655173,
3073
+ "grad_norm": 0.3581734895706177,
3074
+ "learning_rate": 2.1602577650167436e-06,
3075
+ "loss": 0.4455357789993286,
3076
+ "step": 876
3077
+ },
3078
+ {
3079
+ "epoch": 3.7844827586206895,
3080
+ "grad_norm": 0.6977261304855347,
3081
+ "learning_rate": 2.1484237746480574e-06,
3082
+ "loss": 0.5109292268753052,
3083
+ "step": 878
3084
+ },
3085
+ {
3086
+ "epoch": 3.793103448275862,
3087
+ "grad_norm": 0.25545695424079895,
3088
+ "learning_rate": 2.1370400062584544e-06,
3089
+ "loss": 0.1860283464193344,
3090
+ "step": 880
3091
+ },
3092
+ {
3093
+ "epoch": 3.8017241379310347,
3094
+ "grad_norm": 0.5381384491920471,
3095
+ "learning_rate": 2.1261070388660202e-06,
3096
+ "loss": 0.20024971663951874,
3097
+ "step": 882
3098
+ },
3099
+ {
3100
+ "epoch": 3.810344827586207,
3101
+ "grad_norm": 0.3756001591682434,
3102
+ "learning_rate": 2.115625428559535e-06,
3103
+ "loss": 0.43951112031936646,
3104
+ "step": 884
3105
+ },
3106
+ {
3107
+ "epoch": 3.8189655172413794,
3108
+ "grad_norm": 0.5115352869033813,
3109
+ "learning_rate": 2.105595708470185e-06,
3110
+ "loss": 0.16899777948856354,
3111
+ "step": 886
3112
+ },
3113
+ {
3114
+ "epoch": 3.8275862068965516,
3115
+ "grad_norm": 0.2846692204475403,
3116
+ "learning_rate": 2.096018388744448e-06,
3117
+ "loss": 0.26339343190193176,
3118
+ "step": 888
3119
+ },
3120
+ {
3121
+ "epoch": 3.836206896551724,
3122
+ "grad_norm": 0.7379516959190369,
3123
+ "learning_rate": 2.086893956518142e-06,
3124
+ "loss": 0.11539773643016815,
3125
+ "step": 890
3126
+ },
3127
+ {
3128
+ "epoch": 3.844827586206897,
3129
+ "grad_norm": 0.5147801041603088,
3130
+ "learning_rate": 2.078222875891654e-06,
3131
+ "loss": 0.3523973822593689,
3132
+ "step": 892
3133
+ },
3134
+ {
3135
+ "epoch": 3.853448275862069,
3136
+ "grad_norm": 0.15211860835552216,
3137
+ "learning_rate": 2.070005587906328e-06,
3138
+ "loss": 0.33388248085975647,
3139
+ "step": 894
3140
+ },
3141
+ {
3142
+ "epoch": 3.862068965517241,
3143
+ "grad_norm": 0.3375397324562073,
3144
+ "learning_rate": 2.0622425105220346e-06,
3145
+ "loss": 0.2652340531349182,
3146
+ "step": 896
3147
+ },
3148
+ {
3149
+ "epoch": 3.8706896551724137,
3150
+ "grad_norm": 0.3318890333175659,
3151
+ "learning_rate": 2.0549340385959123e-06,
3152
+ "loss": 0.1966915875673294,
3153
+ "step": 898
3154
+ },
3155
+ {
3156
+ "epoch": 3.8793103448275863,
3157
+ "grad_norm": 0.3057389259338379,
3158
+ "learning_rate": 2.0480805438622865e-06,
3159
+ "loss": 0.391522079706192,
3160
+ "step": 900
3161
+ },
3162
+ {
3163
+ "epoch": 3.887931034482759,
3164
+ "grad_norm": 0.5989668369293213,
3165
+ "learning_rate": 2.0416823749137542e-06,
3166
+ "loss": 0.3690534234046936,
3167
+ "step": 902
3168
+ },
3169
+ {
3170
+ "epoch": 3.896551724137931,
3171
+ "grad_norm": 0.3611077666282654,
3172
+ "learning_rate": 2.03573985718346e-06,
3173
+ "loss": 0.35606420040130615,
3174
+ "step": 904
3175
+ },
3176
+ {
3177
+ "epoch": 3.905172413793103,
3178
+ "grad_norm": 0.518487274646759,
3179
+ "learning_rate": 2.0302532929285423e-06,
3180
+ "loss": 0.37323978543281555,
3181
+ "step": 906
3182
+ },
3183
+ {
3184
+ "epoch": 3.913793103448276,
3185
+ "grad_norm": 0.3315095603466034,
3186
+ "learning_rate": 2.0252229612147587e-06,
3187
+ "loss": 0.22919489443302155,
3188
+ "step": 908
3189
+ },
3190
+ {
3191
+ "epoch": 3.9224137931034484,
3192
+ "grad_norm": 0.22296565771102905,
3193
+ "learning_rate": 2.0206491179022887e-06,
3194
+ "loss": 0.17647181451320648,
3195
+ "step": 910
3196
+ },
3197
+ {
3198
+ "epoch": 3.9310344827586206,
3199
+ "grad_norm": 0.2577214539051056,
3200
+ "learning_rate": 2.0165319956327258e-06,
3201
+ "loss": 0.23643004894256592,
3202
+ "step": 912
3203
+ },
3204
+ {
3205
+ "epoch": 3.939655172413793,
3206
+ "grad_norm": 0.37906980514526367,
3207
+ "learning_rate": 2.01287180381724e-06,
3208
+ "loss": 0.3691279888153076,
3209
+ "step": 914
3210
+ },
3211
+ {
3212
+ "epoch": 3.9482758620689653,
3213
+ "grad_norm": 0.30090105533599854,
3214
+ "learning_rate": 2.0096687286259286e-06,
3215
+ "loss": 0.3844853937625885,
3216
+ "step": 916
3217
+ },
3218
+ {
3219
+ "epoch": 3.956896551724138,
3220
+ "grad_norm": 0.3276263475418091,
3221
+ "learning_rate": 2.006922932978346e-06,
3222
+ "loss": 0.28372105956077576,
3223
+ "step": 918
3224
+ },
3225
+ {
3226
+ "epoch": 3.9655172413793105,
3227
+ "grad_norm": 2.499201536178589,
3228
+ "learning_rate": 2.004634556535215e-06,
3229
+ "loss": 0.36251726746559143,
3230
+ "step": 920
3231
+ },
3232
+ {
3233
+ "epoch": 3.9741379310344827,
3234
+ "grad_norm": 0.3088098466396332,
3235
+ "learning_rate": 2.0028037156913294e-06,
3236
+ "loss": 0.2959789037704468,
3237
+ "step": 922
3238
+ },
3239
+ {
3240
+ "epoch": 3.9827586206896552,
3241
+ "grad_norm": 0.5072318911552429,
3242
+ "learning_rate": 2.0014305035696265e-06,
3243
+ "loss": 0.3481578230857849,
3244
+ "step": 924
3245
+ },
3246
+ {
3247
+ "epoch": 3.9913793103448274,
3248
+ "grad_norm": 0.469334214925766,
3249
+ "learning_rate": 2.0005149900164546e-06,
3250
+ "loss": 0.36124011874198914,
3251
+ "step": 926
3252
+ },
3253
+ {
3254
+ "epoch": 4.0,
3255
+ "grad_norm": 0.7579672932624817,
3256
+ "learning_rate": 2.000057221598022e-06,
3257
+ "loss": 0.39811837673187256,
3258
+ "step": 928
3259
+ },
3260
+ {
3261
+ "epoch": 4.0,
3262
+ "step": 928,
3263
+ "total_flos": 3.6942184710227886e+18,
3264
+ "train_loss": 0.7407096255137103,
3265
+ "train_runtime": 10071.2083,
3266
+ "train_samples_per_second": 5.529,
3267
+ "train_steps_per_second": 0.092
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": 3.6942184710227886e+18,
3288
+ "train_batch_size": 3,
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 sha256:98ca1a3153d0bb381eded7d815a0d6c846f865dfd739454844594d8ea37ee83d
3
+ size 5649
training_loss.png ADDED