Hugging Face's logo

furproxy
/
9b-47

Image-Text-to-Text
Transformers
Safetensors
qwen3_5
llama-factory
full
Generated from Trainer
conversational
Model card Files
xet
 Community
furproxy commited on
Commit
be28cf9
·
verified ·
1 Parent(s): ea97291

Upload folder using huggingface_hub

Browse files
Files changed (16) hide show
  1. .gitattributes +1 -0
  2. .ipynb_checkpoints/README-checkpoint.md +69 -0
  3. README.md +69 -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 +349 -0
  14. trainer_state.json +2479 -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 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_muon_lr = {
39
+ "language": _fallback(getattr(training_args, "language_muon_lr", 2e-5), language_lr),
40
+ "vision": _fallback(getattr(training_args, "vision_muon_lr", 2e-5), vision_lr),
41
+ "merger": _fallback(getattr(training_args, "merger_muon_lr", 2e-5), merger_lr),
42
+ }
43
+
44
+ family_to_adamw_lr = {
45
+ "language": _fallback(getattr(training_args, "language_adamw_lr", 2e-5), language_lr),
46
+ "vision": _fallback(getattr(training_args, "vision_adamw_lr", 3e-6), vision_lr),
47
+ "merger": _fallback(getattr(training_args, "merger_adamw_lr", 2e-5), merger_lr),
48
+ }
49
+ - train_batch_size: 2
50
+ - eval_batch_size: 8
51
+ - seed: 42
52
+ - distributed_type: multi-GPU
53
+ - gradient_accumulation_steps: 30
54
+ - total_train_batch_size: 60
55
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
56
+ - lr_scheduler_type: cosine_with_min_lr
57
+ - lr_scheduler_warmup_steps: 0.05
58
+ - num_epochs: 3
59
+
60
+ ### Training results
61
+
62
+
63
+
64
+ ### Framework versions
65
+
66
+ - Transformers 5.5.3
67
+ - Pytorch 2.11.0+cu130
68
+ - Datasets 4.0.0
69
+ - Tokenizers 0.22.2
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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_muon_lr = {
39
+ "language": _fallback(getattr(training_args, "language_muon_lr", 2e-5), language_lr),
40
+ "vision": _fallback(getattr(training_args, "vision_muon_lr", 2e-5), vision_lr),
41
+ "merger": _fallback(getattr(training_args, "merger_muon_lr", 2e-5), merger_lr),
42
+ }
43
+
44
+ family_to_adamw_lr = {
45
+ "language": _fallback(getattr(training_args, "language_adamw_lr", 2e-5), language_lr),
46
+ "vision": _fallback(getattr(training_args, "vision_adamw_lr", 3e-6), vision_lr),
47
+ "merger": _fallback(getattr(training_args, "merger_adamw_lr", 2e-5), merger_lr),
48
+ }
49
+ - train_batch_size: 2
50
+ - eval_batch_size: 8
51
+ - seed: 42
52
+ - distributed_type: multi-GPU
53
+ - gradient_accumulation_steps: 30
54
+ - total_train_batch_size: 60
55
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
56
+ - lr_scheduler_type: cosine_with_min_lr
57
+ - lr_scheduler_warmup_steps: 0.05
58
+ - num_epochs: 3
59
+
60
+ ### Training results
61
+
62
+
63
+
64
+ ### Framework versions
65
+
66
+ - Transformers 5.5.3
67
+ - Pytorch 2.11.0+cu130
68
+ - Datasets 4.0.0
69
+ - Tokenizers 0.22.2
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "effective_tokens_per_sec": 4889.172491836145,
3
+ "epoch": 3.0,
4
+ "total_flos": 2.7548743231466373e+18,
5
+ "train_loss": 1.0648414060130886,
6
+ "train_runtime": 11016.7324,
7
+ "train_samples_per_second": 3.791,
8
+ "train_steps_per_second": 0.063
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 sha256:1a4a1b07edef74dfe279457447150d0595517c5dbb9d238452ec8121428872b9
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": 4889.172491836145,
3
+ "epoch": 3.0,
4
+ "total_flos": 2.7548743231466373e+18,
5
+ "train_loss": 1.0648414060130886,
6
+ "train_runtime": 11016.7324,
7
+ "train_samples_per_second": 3.791,
8
+ "train_steps_per_second": 0.063
9
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,349 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 2, "total_steps": 696, "loss": 1.9584773778915405, "lr": 5.714285714285715e-07, "epoch": 0.008620689655172414, "percentage": 0.29, "elapsed_time": "0:00:47", "remaining_time": "4:33:05"}
2
+ {"current_steps": 4, "total_steps": 696, "loss": 6.022436618804932, "lr": 1.7142857142857145e-06, "epoch": 0.017241379310344827, "percentage": 0.57, "elapsed_time": "0:01:18", "remaining_time": "3:45:58"}
3
+ {"current_steps": 6, "total_steps": 696, "loss": 1.9596091508865356, "lr": 2.8571428571428573e-06, "epoch": 0.02586206896551724, "percentage": 0.86, "elapsed_time": "0:01:58", "remaining_time": "3:46:19"}
4
+ {"current_steps": 8, "total_steps": 696, "loss": 1.622937560081482, "lr": 4.000000000000001e-06, "epoch": 0.034482758620689655, "percentage": 1.15, "elapsed_time": "0:02:27", "remaining_time": "3:31:01"}
5
+ {"current_steps": 10, "total_steps": 696, "loss": 2.567897319793701, "lr": 5.142857142857142e-06, "epoch": 0.04310344827586207, "percentage": 1.44, "elapsed_time": "0:02:57", "remaining_time": "3:23:29"}
6
+ {"current_steps": 12, "total_steps": 696, "loss": 1.8926677703857422, "lr": 6.285714285714286e-06, "epoch": 0.05172413793103448, "percentage": 1.72, "elapsed_time": "0:03:38", "remaining_time": "3:27:35"}
7
+ {"current_steps": 14, "total_steps": 696, "loss": 1.7960844039916992, "lr": 7.428571428571429e-06, "epoch": 0.0603448275862069, "percentage": 2.01, "elapsed_time": "0:04:18", "remaining_time": "3:29:48"}
8
+ {"current_steps": 16, "total_steps": 696, "loss": 1.7497979402542114, "lr": 8.571428571428571e-06, "epoch": 0.06896551724137931, "percentage": 2.3, "elapsed_time": "0:05:02", "remaining_time": "3:34:16"}
9
+ {"current_steps": 18, "total_steps": 696, "loss": 2.5326366424560547, "lr": 9.714285714285715e-06, "epoch": 0.07758620689655173, "percentage": 2.59, "elapsed_time": "0:05:29", "remaining_time": "3:27:07"}
10
+ {"current_steps": 20, "total_steps": 696, "loss": 1.7976326942443848, "lr": 1.0857142857142858e-05, "epoch": 0.08620689655172414, "percentage": 2.87, "elapsed_time": "0:06:10", "remaining_time": "3:28:49"}
11
+ {"current_steps": 22, "total_steps": 696, "loss": 3.13413667678833, "lr": 1.2e-05, "epoch": 0.09482758620689655, "percentage": 3.16, "elapsed_time": "0:06:29", "remaining_time": "3:18:56"}
12
+ {"current_steps": 24, "total_steps": 696, "loss": 1.9254695177078247, "lr": 1.3142857142857145e-05, "epoch": 0.10344827586206896, "percentage": 3.45, "elapsed_time": "0:06:54", "remaining_time": "3:13:37"}
13
+ {"current_steps": 26, "total_steps": 696, "loss": 2.0263671875, "lr": 1.4285714285714287e-05, "epoch": 0.11206896551724138, "percentage": 3.74, "elapsed_time": "0:07:32", "remaining_time": "3:14:28"}
14
+ {"current_steps": 28, "total_steps": 696, "loss": 2.157352924346924, "lr": 1.542857142857143e-05, "epoch": 0.1206896551724138, "percentage": 4.02, "elapsed_time": "0:08:00", "remaining_time": "3:11:12"}
15
+ {"current_steps": 30, "total_steps": 696, "loss": 1.6684068441390991, "lr": 1.6571428571428574e-05, "epoch": 0.12931034482758622, "percentage": 4.31, "elapsed_time": "0:08:41", "remaining_time": "3:13:00"}
16
+ {"current_steps": 32, "total_steps": 696, "loss": 1.6966811418533325, "lr": 1.7714285714285717e-05, "epoch": 0.13793103448275862, "percentage": 4.6, "elapsed_time": "0:09:16", "remaining_time": "3:12:25"}
17
+ {"current_steps": 34, "total_steps": 696, "loss": 1.5750758647918701, "lr": 1.885714285714286e-05, "epoch": 0.14655172413793102, "percentage": 4.89, "elapsed_time": "0:09:56", "remaining_time": "3:13:34"}
18
+ {"current_steps": 36, "total_steps": 696, "loss": 1.5068325996398926, "lr": 2e-05, "epoch": 0.15517241379310345, "percentage": 5.17, "elapsed_time": "0:10:19", "remaining_time": "3:09:22"}
19
+ {"current_steps": 38, "total_steps": 696, "loss": 1.582505702972412, "lr": 1.9995836718255602e-05, "epoch": 0.16379310344827586, "percentage": 5.46, "elapsed_time": "0:10:55", "remaining_time": "3:09:13"}
20
+ {"current_steps": 40, "total_steps": 696, "loss": 1.5262649059295654, "lr": 1.9983350724781264e-05, "epoch": 0.1724137931034483, "percentage": 5.75, "elapsed_time": "0:11:23", "remaining_time": "3:06:49"}
21
+ {"current_steps": 42, "total_steps": 696, "loss": 1.858609914779663, "lr": 1.996255357129003e-05, "epoch": 0.1810344827586207, "percentage": 6.03, "elapsed_time": "0:11:58", "remaining_time": "3:06:22"}
22
+ {"current_steps": 44, "total_steps": 696, "loss": 1.520866870880127, "lr": 1.9933464498761785e-05, "epoch": 0.1896551724137931, "percentage": 6.32, "elapsed_time": "0:12:37", "remaining_time": "3:07:11"}
23
+ {"current_steps": 46, "total_steps": 696, "loss": 1.286926507949829, "lr": 1.989611041964199e-05, "epoch": 0.19827586206896552, "percentage": 6.61, "elapsed_time": "0:13:05", "remaining_time": "3:04:54"}
24
+ {"current_steps": 48, "total_steps": 696, "loss": 0.9975602030754089, "lr": 1.985052589294303e-05, "epoch": 0.20689655172413793, "percentage": 6.9, "elapsed_time": "0:13:21", "remaining_time": "3:00:17"}
25
+ {"current_steps": 50, "total_steps": 696, "loss": 1.4037001132965088, "lr": 1.9796753092271085e-05, "epoch": 0.21551724137931033, "percentage": 7.18, "elapsed_time": "0:13:54", "remaining_time": "2:59:47"}
26
+ {"current_steps": 52, "total_steps": 696, "loss": 1.5926319360733032, "lr": 1.973484176680824e-05, "epoch": 0.22413793103448276, "percentage": 7.47, "elapsed_time": "0:14:24", "remaining_time": "2:58:26"}
27
+ {"current_steps": 54, "total_steps": 696, "loss": 1.0562762022018433, "lr": 1.9664849195285844e-05, "epoch": 0.23275862068965517, "percentage": 7.76, "elapsed_time": "0:14:49", "remaining_time": "2:56:16"}
28
+ {"current_steps": 56, "total_steps": 696, "loss": 0.927044689655304, "lr": 1.958684013299173e-05, "epoch": 0.2413793103448276, "percentage": 8.05, "elapsed_time": "0:15:17", "remaining_time": "2:54:42"}
29
+ {"current_steps": 58, "total_steps": 696, "loss": 0.9711346626281738, "lr": 1.950088675186034e-05, "epoch": 0.25, "percentage": 8.33, "elapsed_time": "0:15:44", "remaining_time": "2:53:07"}
30
+ {"current_steps": 60, "total_steps": 696, "loss": 1.0954338312149048, "lr": 1.9407068573701116e-05, "epoch": 0.25862068965517243, "percentage": 8.62, "elapsed_time": "0:16:10", "remaining_time": "2:51:28"}
31
+ {"current_steps": 62, "total_steps": 696, "loss": 1.4207227230072021, "lr": 1.9305472396627044e-05, "epoch": 0.2672413793103448, "percentage": 8.91, "elapsed_time": "0:16:50", "remaining_time": "2:52:13"}
32
+ {"current_steps": 64, "total_steps": 696, "loss": 0.9308554530143738, "lr": 1.919619221475131e-05, "epoch": 0.27586206896551724, "percentage": 9.2, "elapsed_time": "0:17:06", "remaining_time": "2:48:59"}
33
+ {"current_steps": 66, "total_steps": 696, "loss": 1.4012649059295654, "lr": 1.90793291312264e-05, "epoch": 0.28448275862068967, "percentage": 9.48, "elapsed_time": "0:17:46", "remaining_time": "2:49:42"}
34
+ {"current_steps": 68, "total_steps": 696, "loss": 0.8479315638542175, "lr": 1.895499126470615e-05, "epoch": 0.29310344827586204, "percentage": 9.77, "elapsed_time": "0:18:11", "remaining_time": "2:47:57"}
35
+ {"current_steps": 70, "total_steps": 696, "loss": 1.381568193435669, "lr": 1.8823293649317186e-05, "epoch": 0.3017241379310345, "percentage": 10.06, "elapsed_time": "0:18:46", "remaining_time": "2:47:58"}
36
+ {"current_steps": 72, "total_steps": 696, "loss": 0.6094793677330017, "lr": 1.868435812823238e-05, "epoch": 0.3103448275862069, "percentage": 10.34, "elapsed_time": "0:19:03", "remaining_time": "2:45:07"}
37
+ {"current_steps": 74, "total_steps": 696, "loss": 0.8938958644866943, "lr": 1.853831324094475e-05, "epoch": 0.31896551724137934, "percentage": 10.63, "elapsed_time": "0:19:30", "remaining_time": "2:43:59"}
38
+ {"current_steps": 76, "total_steps": 696, "loss": 1.2029653787612915, "lr": 1.8385294104346103e-05, "epoch": 0.3275862068965517, "percentage": 10.92, "elapsed_time": "0:20:06", "remaining_time": "2:44:06"}
39
+ {"current_steps": 78, "total_steps": 696, "loss": 1.3596869707107544, "lr": 1.8225442287720417e-05, "epoch": 0.33620689655172414, "percentage": 11.21, "elapsed_time": "0:20:44", "remaining_time": "2:44:17"}
40
+ {"current_steps": 80, "total_steps": 696, "loss": 1.4085049629211426, "lr": 1.805890568176769e-05, "epoch": 0.3448275862068966, "percentage": 11.49, "elapsed_time": "0:21:11", "remaining_time": "2:43:14"}
41
+ {"current_steps": 82, "total_steps": 696, "loss": 1.2587382793426514, "lr": 1.7885838361779333e-05, "epoch": 0.35344827586206895, "percentage": 11.78, "elapsed_time": "0:21:52", "remaining_time": "2:43:50"}
42
+ {"current_steps": 84, "total_steps": 696, "loss": 1.3189198970794678, "lr": 1.7706400445091755e-05, "epoch": 0.3620689655172414, "percentage": 12.07, "elapsed_time": "0:22:25", "remaining_time": "2:43:20"}
43
+ {"current_steps": 86, "total_steps": 696, "loss": 1.327932596206665, "lr": 1.7520757942950026e-05, "epoch": 0.3706896551724138, "percentage": 12.36, "elapsed_time": "0:23:02", "remaining_time": "2:43:28"}
44
+ {"current_steps": 88, "total_steps": 696, "loss": 1.0215771198272705, "lr": 1.7329082606918616e-05, "epoch": 0.3793103448275862, "percentage": 12.64, "elapsed_time": "0:23:31", "remaining_time": "2:42:28"}
45
+ {"current_steps": 90, "total_steps": 696, "loss": 1.5617601871490479, "lr": 1.7131551769981392e-05, "epoch": 0.3879310344827586, "percentage": 12.93, "elapsed_time": "0:24:10", "remaining_time": "2:42:48"}
46
+ {"current_steps": 92, "total_steps": 696, "loss": 1.3252155780792236, "lr": 1.6928348182477765e-05, "epoch": 0.39655172413793105, "percentage": 13.22, "elapsed_time": "0:24:46", "remaining_time": "2:42:38"}
47
+ {"current_steps": 94, "total_steps": 696, "loss": 1.335471272468567, "lr": 1.67196598430269e-05, "epoch": 0.4051724137931034, "percentage": 13.51, "elapsed_time": "0:25:15", "remaining_time": "2:41:46"}
48
+ {"current_steps": 96, "total_steps": 696, "loss": 1.4329684972763062, "lr": 1.6505679824596338e-05, "epoch": 0.41379310344827586, "percentage": 13.79, "elapsed_time": "0:25:49", "remaining_time": "2:41:26"}
49
+ {"current_steps": 98, "total_steps": 696, "loss": 1.302343487739563, "lr": 1.6286606095875938e-05, "epoch": 0.4224137931034483, "percentage": 14.08, "elapsed_time": "0:26:30", "remaining_time": "2:41:44"}
50
+ {"current_steps": 100, "total_steps": 696, "loss": 1.3569573163986206, "lr": 1.6062641338122474e-05, "epoch": 0.43103448275862066, "percentage": 14.37, "elapsed_time": "0:27:05", "remaining_time": "2:41:27"}
51
+ {"current_steps": 102, "total_steps": 696, "loss": 1.309763789176941, "lr": 1.583399275764424e-05, "epoch": 0.4396551724137931, "percentage": 14.66, "elapsed_time": "0:27:46", "remaining_time": "2:41:43"}
52
+ {"current_steps": 104, "total_steps": 696, "loss": 0.9164461493492126, "lr": 1.5600871894099243e-05, "epoch": 0.4482758620689655, "percentage": 14.94, "elapsed_time": "0:28:14", "remaining_time": "2:40:43"}
53
+ {"current_steps": 106, "total_steps": 696, "loss": 0.8275090456008911, "lr": 1.5363494424784238e-05, "epoch": 0.45689655172413796, "percentage": 15.23, "elapsed_time": "0:28:45", "remaining_time": "2:40:05"}
54
+ {"current_steps": 108, "total_steps": 696, "loss": 0.7831904292106628, "lr": 1.5122079965095786e-05, "epoch": 0.46551724137931033, "percentage": 15.52, "elapsed_time": "0:29:13", "remaining_time": "2:39:07"}
55
+ {"current_steps": 110, "total_steps": 696, "loss": 1.0925136804580688, "lr": 1.4876851865347857e-05, "epoch": 0.47413793103448276, "percentage": 15.8, "elapsed_time": "0:29:41", "remaining_time": "2:38:12"}
56
+ {"current_steps": 112, "total_steps": 696, "loss": 0.9729475378990173, "lr": 1.4628037004133986e-05, "epoch": 0.4827586206896552, "percentage": 16.09, "elapsed_time": "0:30:05", "remaining_time": "2:36:55"}
57
+ {"current_steps": 114, "total_steps": 696, "loss": 0.6473991274833679, "lr": 1.4375865578425157e-05, "epoch": 0.49137931034482757, "percentage": 16.38, "elapsed_time": "0:30:24", "remaining_time": "2:35:15"}
58
+ {"current_steps": 116, "total_steps": 696, "loss": 1.1790492534637451, "lr": 1.4120570890597611e-05, "epoch": 0.5, "percentage": 16.67, "elapsed_time": "0:30:55", "remaining_time": "2:34:37"}
59
+ {"current_steps": 118, "total_steps": 696, "loss": 1.2727322578430176, "lr": 1.3862389132587639e-05, "epoch": 0.5086206896551724, "percentage": 16.95, "elapsed_time": "0:31:23", "remaining_time": "2:33:47"}
60
+ {"current_steps": 120, "total_steps": 696, "loss": 0.7860153913497925, "lr": 1.3601559167372983e-05, "epoch": 0.5172413793103449, "percentage": 17.24, "elapsed_time": "0:31:51", "remaining_time": "2:32:54"}
61
+ {"current_steps": 122, "total_steps": 696, "loss": 1.043981671333313, "lr": 1.3338322307983105e-05, "epoch": 0.5258620689655172, "percentage": 17.53, "elapsed_time": "0:32:23", "remaining_time": "2:32:24"}
62
+ {"current_steps": 124, "total_steps": 696, "loss": 1.278051733970642, "lr": 1.3072922094242703e-05, "epoch": 0.5344827586206896, "percentage": 17.82, "elapsed_time": "0:32:58", "remaining_time": "2:32:07"}
63
+ {"current_steps": 126, "total_steps": 696, "loss": 1.255641222000122, "lr": 1.2805604067455076e-05, "epoch": 0.5431034482758621, "percentage": 18.1, "elapsed_time": "0:33:39", "remaining_time": "2:32:15"}
64
+ {"current_steps": 128, "total_steps": 696, "loss": 1.255129337310791, "lr": 1.2536615543233748e-05, "epoch": 0.5517241379310345, "percentage": 18.39, "elapsed_time": "0:34:13", "remaining_time": "2:31:52"}
65
+ {"current_steps": 130, "total_steps": 696, "loss": 1.5166298151016235, "lr": 1.2266205382692574e-05, "epoch": 0.5603448275862069, "percentage": 18.68, "elapsed_time": "0:34:52", "remaining_time": "2:31:49"}
66
+ {"current_steps": 132, "total_steps": 696, "loss": 0.8270851373672485, "lr": 1.199462376220597e-05, "epoch": 0.5689655172413793, "percentage": 18.97, "elapsed_time": "0:35:10", "remaining_time": "2:30:15"}
67
+ {"current_steps": 134, "total_steps": 696, "loss": 0.6701672077178955, "lr": 1.1722121941952313e-05, "epoch": 0.5775862068965517, "percentage": 19.25, "elapsed_time": "0:35:37", "remaining_time": "2:29:25"}
68
+ {"current_steps": 136, "total_steps": 696, "loss": 1.077882170677185, "lr": 1.144895203345462e-05, "epoch": 0.5862068965517241, "percentage": 19.54, "elapsed_time": "0:36:16", "remaining_time": "2:29:23"}
69
+ {"current_steps": 138, "total_steps": 696, "loss": 0.8770455718040466, "lr": 1.11753667663336e-05, "epoch": 0.5948275862068966, "percentage": 19.83, "elapsed_time": "0:36:43", "remaining_time": "2:28:30"}
70
+ {"current_steps": 140, "total_steps": 696, "loss": 0.9084359407424927, "lr": 1.0901619254488831e-05, "epoch": 0.603448275862069, "percentage": 20.11, "elapsed_time": "0:37:04", "remaining_time": "2:27:15"}
71
+ {"current_steps": 142, "total_steps": 696, "loss": 1.370908260345459, "lr": 1.0627962761924462e-05, "epoch": 0.6120689655172413, "percentage": 20.4, "elapsed_time": "0:37:43", "remaining_time": "2:27:12"}
72
+ {"current_steps": 144, "total_steps": 696, "loss": 0.7351585030555725, "lr": 1.0354650468435985e-05, "epoch": 0.6206896551724138, "percentage": 20.69, "elapsed_time": "0:38:12", "remaining_time": "2:26:27"}
73
+ {"current_steps": 146, "total_steps": 696, "loss": 1.2708691358566284, "lr": 1.0081935235374955e-05, "epoch": 0.6293103448275862, "percentage": 20.98, "elapsed_time": "0:38:46", "remaining_time": "2:26:05"}
74
+ {"current_steps": 148, "total_steps": 696, "loss": 1.2674932479858398, "lr": 9.810069371708313e-06, "epoch": 0.6379310344827587, "percentage": 21.26, "elapsed_time": "0:39:12", "remaining_time": "2:25:11"}
75
+ {"current_steps": 150, "total_steps": 696, "loss": 1.0046426057815552, "lr": 9.539304400588752e-06, "epoch": 0.646551724137931, "percentage": 21.55, "elapsed_time": "0:39:51", "remaining_time": "2:25:04"}
76
+ {"current_steps": 152, "total_steps": 696, "loss": 0.664213240146637, "lr": 9.269890826652113e-06, "epoch": 0.6551724137931034, "percentage": 21.84, "elapsed_time": "0:40:19", "remaining_time": "2:24:17"}
77
+ {"current_steps": 154, "total_steps": 696, "loss": 1.5197818279266357, "lr": 9.00207790425708e-06, "epoch": 0.6637931034482759, "percentage": 22.13, "elapsed_time": "0:40:58", "remaining_time": "2:24:12"}
78
+ {"current_steps": 156, "total_steps": 696, "loss": 1.062941312789917, "lr": 8.73611340688159e-06, "epoch": 0.6724137931034483, "percentage": 22.41, "elapsed_time": "0:41:21", "remaining_time": "2:23:08"}
79
+ {"current_steps": 158, "total_steps": 696, "loss": 0.7909774780273438, "lr": 8.472243397889327e-06, "epoch": 0.6810344827586207, "percentage": 22.7, "elapsed_time": "0:41:48", "remaining_time": "2:22:22"}
80
+ {"current_steps": 160, "total_steps": 696, "loss": 0.3939180374145508, "lr": 8.210712002878366e-06, "epoch": 0.6896551724137931, "percentage": 22.99, "elapsed_time": "0:42:03", "remaining_time": "2:20:52"}
81
+ {"current_steps": 162, "total_steps": 696, "loss": 1.134759545326233, "lr": 7.951761183822578e-06, "epoch": 0.6982758620689655, "percentage": 23.28, "elapsed_time": "0:42:42", "remaining_time": "2:20:48"}
82
+ {"current_steps": 164, "total_steps": 696, "loss": 0.9670589566230774, "lr": 7.695630515214783e-06, "epoch": 0.7068965517241379, "percentage": 23.56, "elapsed_time": "0:43:06", "remaining_time": "2:19:51"}
83
+ {"current_steps": 166, "total_steps": 696, "loss": 1.3202883005142212, "lr": 7.442556962418712e-06, "epoch": 0.7155172413793104, "percentage": 23.85, "elapsed_time": "0:43:46", "remaining_time": "2:19:44"}
84
+ {"current_steps": 168, "total_steps": 696, "loss": 0.7861534357070923, "lr": 7.192774662434887e-06, "epoch": 0.7241379310344828, "percentage": 24.14, "elapsed_time": "0:44:13", "remaining_time": "2:18:59"}
85
+ {"current_steps": 170, "total_steps": 696, "loss": 0.8559544682502747, "lr": 6.94651470728322e-06, "epoch": 0.7327586206896551, "percentage": 24.43, "elapsed_time": "0:44:36", "remaining_time": "2:18:02"}
86
+ {"current_steps": 172, "total_steps": 696, "loss": 1.349075198173523, "lr": 6.7040049302027445e-06, "epoch": 0.7413793103448276, "percentage": 24.71, "elapsed_time": "0:45:16", "remaining_time": "2:17:55"}
87
+ {"current_steps": 174, "total_steps": 696, "loss": 1.2694404125213623, "lr": 6.465469694866295e-06, "epoch": 0.75, "percentage": 25.0, "elapsed_time": "0:45:56", "remaining_time": "2:17:50"}
88
+ {"current_steps": 176, "total_steps": 696, "loss": 1.2068278789520264, "lr": 6.2311296878051355e-06, "epoch": 0.7586206896551724, "percentage": 25.29, "elapsed_time": "0:46:26", "remaining_time": "2:17:13"}
89
+ {"current_steps": 178, "total_steps": 696, "loss": 1.311326503753662, "lr": 6.001201714235567e-06, "epoch": 0.7672413793103449, "percentage": 25.57, "elapsed_time": "0:47:00", "remaining_time": "2:16:46"}
90
+ {"current_steps": 180, "total_steps": 696, "loss": 0.768395185470581, "lr": 5.7758984974764586e-06, "epoch": 0.7758620689655172, "percentage": 25.86, "elapsed_time": "0:47:37", "remaining_time": "2:16:31"}
91
+ {"current_steps": 182, "total_steps": 696, "loss": 1.1966248750686646, "lr": 5.555428482143206e-06, "epoch": 0.7844827586206896, "percentage": 26.15, "elapsed_time": "0:48:16", "remaining_time": "2:16:20"}
92
+ {"current_steps": 184, "total_steps": 696, "loss": 1.1476389169692993, "lr": 5.339995641300257e-06, "epoch": 0.7931034482758621, "percentage": 26.44, "elapsed_time": "0:48:43", "remaining_time": "2:15:35"}
93
+ {"current_steps": 186, "total_steps": 696, "loss": 0.9965859651565552, "lr": 5.129799287750601e-06, "epoch": 0.8017241379310345, "percentage": 26.72, "elapsed_time": "0:49:11", "remaining_time": "2:14:52"}
94
+ {"current_steps": 188, "total_steps": 696, "loss": 0.8189845085144043, "lr": 4.9250338896367814e-06, "epoch": 0.8103448275862069, "percentage": 27.01, "elapsed_time": "0:49:38", "remaining_time": "2:14:07"}
95
+ {"current_steps": 190, "total_steps": 696, "loss": 1.2450443506240845, "lr": 4.725888890524101e-06, "epoch": 0.8189655172413793, "percentage": 27.3, "elapsed_time": "0:50:11", "remaining_time": "2:13:41"}
96
+ {"current_steps": 192, "total_steps": 696, "loss": 0.9530048966407776, "lr": 4.5325485341324026e-06, "epoch": 0.8275862068965517, "percentage": 27.59, "elapsed_time": "0:50:50", "remaining_time": "2:13:28"}
97
+ {"current_steps": 194, "total_steps": 696, "loss": 1.2201794385910034, "lr": 4.345191693878623e-06, "epoch": 0.8362068965517241, "percentage": 27.87, "elapsed_time": "0:51:28", "remaining_time": "2:13:12"}
98
+ {"current_steps": 196, "total_steps": 696, "loss": 0.8178768754005432, "lr": 4.163991707387815e-06, "epoch": 0.8448275862068966, "percentage": 28.16, "elapsed_time": "0:51:56", "remaining_time": "2:12:30"}
99
+ {"current_steps": 198, "total_steps": 696, "loss": 1.228522539138794, "lr": 3.989116216125717e-06, "epoch": 0.853448275862069, "percentage": 28.45, "elapsed_time": "0:52:35", "remaining_time": "2:12:15"}
100
+ {"current_steps": 200, "total_steps": 696, "loss": 1.3021351099014282, "lr": 3.8207270103012905e-06, "epoch": 0.8620689655172413, "percentage": 28.74, "elapsed_time": "0:53:10", "remaining_time": "2:11:53"}
101
+ {"current_steps": 202, "total_steps": 696, "loss": 0.6399316787719727, "lr": 3.6589798791826593e-06, "epoch": 0.8706896551724138, "percentage": 29.02, "elapsed_time": "0:53:26", "remaining_time": "2:10:42"}
102
+ {"current_steps": 204, "total_steps": 696, "loss": 1.600140929222107, "lr": 3.5040244669649804e-06, "epoch": 0.8793103448275862, "percentage": 29.31, "elapsed_time": "0:54:06", "remaining_time": "2:10:29"}
103
+ {"current_steps": 206, "total_steps": 696, "loss": 1.0438815355300903, "lr": 3.3560041343235544e-06, "epoch": 0.8879310344827587, "percentage": 29.6, "elapsed_time": "0:54:34", "remaining_time": "2:09:48"}
104
+ {"current_steps": 208, "total_steps": 696, "loss": 1.339942216873169, "lr": 3.2150558257803178e-06, "epoch": 0.896551724137931, "percentage": 29.89, "elapsed_time": "0:55:13", "remaining_time": "2:09:34"}
105
+ {"current_steps": 210, "total_steps": 696, "loss": 0.9232547879219055, "lr": 3.081309943006349e-06, "epoch": 0.9051724137931034, "percentage": 30.17, "elapsed_time": "0:55:41", "remaining_time": "2:08:53"}
106
+ {"current_steps": 212, "total_steps": 696, "loss": 1.1493092775344849, "lr": 2.9548902241776783e-06, "epoch": 0.9137931034482759, "percentage": 30.46, "elapsed_time": "0:56:11", "remaining_time": "2:08:16"}
107
+ {"current_steps": 214, "total_steps": 696, "loss": 1.5147000551223755, "lr": 2.8359136294959814e-06, "epoch": 0.9224137931034483, "percentage": 30.75, "elapsed_time": "0:56:50", "remaining_time": "2:08:01"}
108
+ {"current_steps": 216, "total_steps": 696, "loss": 1.3095285892486572, "lr": 2.7244902329800697e-06, "epoch": 0.9310344827586207, "percentage": 31.03, "elapsed_time": "0:57:23", "remaining_time": "2:07:32"}
109
+ {"current_steps": 218, "total_steps": 696, "loss": 1.230783462524414, "lr": 2.620723120628298e-06, "epoch": 0.9396551724137931, "percentage": 31.32, "elapsed_time": "0:58:05", "remaining_time": "2:07:22"}
110
+ {"current_steps": 220, "total_steps": 696, "loss": 1.3510628938674927, "lr": 2.5247082950461204e-06, "epoch": 0.9482758620689655, "percentage": 31.61, "elapsed_time": "0:58:46", "remaining_time": "2:07:10"}
111
+ {"current_steps": 222, "total_steps": 696, "loss": 1.1361650228500366, "lr": 2.4365345866269884e-06, "epoch": 0.9568965517241379, "percentage": 31.9, "elapsed_time": "0:59:18", "remaining_time": "2:06:38"}
112
+ {"current_steps": 224, "total_steps": 696, "loss": 1.2898920774459839, "lr": 2.35628357136881e-06, "epoch": 0.9655172413793104, "percentage": 32.18, "elapsed_time": "0:59:59", "remaining_time": "2:06:24"}
113
+ {"current_steps": 226, "total_steps": 696, "loss": 1.3678969144821167, "lr": 2.284029495401986e-06, "epoch": 0.9741379310344828, "percentage": 32.47, "elapsed_time": "1:00:38", "remaining_time": "2:06:07"}
114
+ {"current_steps": 228, "total_steps": 696, "loss": 0.8566112518310547, "lr": 2.219839206298835e-06, "epoch": 0.9827586206896551, "percentage": 32.76, "elapsed_time": "1:00:58", "remaining_time": "2:05:08"}
115
+ {"current_steps": 230, "total_steps": 696, "loss": 0.8546447157859802, "lr": 2.163772091227977e-06, "epoch": 0.9913793103448276, "percentage": 33.05, "elapsed_time": "1:01:19", "remaining_time": "2:04:14"}
116
+ {"current_steps": 232, "total_steps": 696, "loss": 1.1330009698867798, "lr": 2.115880022010884e-06, "epoch": 1.0, "percentage": 33.33, "elapsed_time": "1:01:45", "remaining_time": "2:03:31"}
117
+ {"current_steps": 234, "total_steps": 696, "loss": 1.1593165397644043, "lr": 2.076207307131444e-06, "epoch": 1.0086206896551724, "percentage": 33.62, "elapsed_time": "1:02:26", "remaining_time": "2:03:17"}
118
+ {"current_steps": 236, "total_steps": 696, "loss": 0.6197730898857117, "lr": 2.0447906507428975e-06, "epoch": 1.0172413793103448, "percentage": 33.91, "elapsed_time": "1:02:51", "remaining_time": "2:02:31"}
119
+ {"current_steps": 238, "total_steps": 696, "loss": 1.2394027709960938, "lr": 2.021659118710141e-06, "epoch": 1.0258620689655173, "percentage": 34.2, "elapsed_time": "1:03:32", "remaining_time": "2:02:16"}
120
+ {"current_steps": 240, "total_steps": 696, "loss": 1.2157349586486816, "lr": 2.0068341117187347e-06, "epoch": 1.0344827586206897, "percentage": 34.48, "elapsed_time": "1:04:06", "remaining_time": "2:01:49"}
121
+ {"current_steps": 242, "total_steps": 696, "loss": 1.2545582056045532, "lr": 2.00032934547556e-06, "epoch": 1.043103448275862, "percentage": 34.77, "elapsed_time": "1:04:40", "remaining_time": "2:01:20"}
122
+ {"current_steps": 244, "total_steps": 696, "loss": 1.173262596130371, "lr": 2.002150838019407e-06, "epoch": 1.0517241379310345, "percentage": 35.06, "elapsed_time": "1:05:20", "remaining_time": "2:01:02"}
123
+ {"current_steps": 246, "total_steps": 696, "loss": 1.1772663593292236, "lr": 2.012296904153243e-06, "epoch": 1.0603448275862069, "percentage": 35.34, "elapsed_time": "1:06:00", "remaining_time": "2:00:45"}
124
+ {"current_steps": 248, "total_steps": 696, "loss": 1.1834620237350464, "lr": 2.030758157003306e-06, "epoch": 1.0689655172413792, "percentage": 35.63, "elapsed_time": "1:06:41", "remaining_time": "2:00:29"}
125
+ {"current_steps": 250, "total_steps": 696, "loss": 0.8118759393692017, "lr": 2.057517516703605e-06, "epoch": 1.0775862068965518, "percentage": 35.92, "elapsed_time": "1:07:09", "remaining_time": "1:59:48"}
126
+ {"current_steps": 252, "total_steps": 696, "loss": 0.7529575824737549, "lr": 2.0925502261977544e-06, "epoch": 1.0862068965517242, "percentage": 36.21, "elapsed_time": "1:07:37", "remaining_time": "1:59:08"}
127
+ {"current_steps": 254, "total_steps": 696, "loss": 0.7137669324874878, "lr": 2.1358238741435576e-06, "epoch": 1.0948275862068966, "percentage": 36.49, "elapsed_time": "1:08:04", "remaining_time": "1:58:27"}
128
+ {"current_steps": 256, "total_steps": 696, "loss": 1.1082866191864014, "lr": 2.1872984248991204e-06, "epoch": 1.103448275862069, "percentage": 36.78, "elapsed_time": "1:08:37", "remaining_time": "1:57:57"}
129
+ {"current_steps": 258, "total_steps": 696, "loss": 1.020033359527588, "lr": 2.246926255562788e-06, "epoch": 1.1120689655172413, "percentage": 37.07, "elapsed_time": "1:09:04", "remaining_time": "1:57:15"}
130
+ {"current_steps": 260, "total_steps": 696, "loss": 0.4955134391784668, "lr": 2.3146522000325895e-06, "epoch": 1.1206896551724137, "percentage": 37.36, "elapsed_time": "1:09:19", "remaining_time": "1:56:15"}
131
+ {"current_steps": 262, "total_steps": 696, "loss": 1.212654709815979, "lr": 2.3904136000444754e-06, "epoch": 1.1293103448275863, "percentage": 37.64, "elapsed_time": "1:10:00", "remaining_time": "1:55:57"}
132
+ {"current_steps": 264, "total_steps": 696, "loss": 1.2140189409255981, "lr": 2.4741403631420874e-06, "epoch": 1.1379310344827587, "percentage": 37.93, "elapsed_time": "1:10:30", "remaining_time": "1:55:22"}
133
+ {"current_steps": 266, "total_steps": 696, "loss": 1.3088136911392212, "lr": 2.5657550275244664e-06, "epoch": 1.146551724137931, "percentage": 38.22, "elapsed_time": "1:11:10", "remaining_time": "1:55:02"}
134
+ {"current_steps": 268, "total_steps": 696, "loss": 0.9128588438034058, "lr": 2.66517283371167e-06, "epoch": 1.1551724137931034, "percentage": 38.51, "elapsed_time": "1:11:47", "remaining_time": "1:54:38"}
135
+ {"current_steps": 270, "total_steps": 696, "loss": 1.2309114933013916, "lr": 2.7723018029620165e-06, "epoch": 1.1637931034482758, "percentage": 38.79, "elapsed_time": "1:12:22", "remaining_time": "1:54:11"}
136
+ {"current_steps": 272, "total_steps": 696, "loss": 0.8446346521377563, "lr": 2.8870428223683972e-06, "epoch": 1.1724137931034484, "percentage": 39.08, "elapsed_time": "1:12:50", "remaining_time": "1:53:33"}
137
+ {"current_steps": 274, "total_steps": 696, "loss": 1.3168221712112427, "lr": 3.009289736554944e-06, "epoch": 1.1810344827586208, "percentage": 39.37, "elapsed_time": "1:13:31", "remaining_time": "1:53:14"}
138
+ {"current_steps": 276, "total_steps": 696, "loss": 0.7556365132331848, "lr": 3.1389294458891755e-06, "epoch": 1.1896551724137931, "percentage": 39.66, "elapsed_time": "1:13:54", "remaining_time": "1:52:27"}
139
+ {"current_steps": 278, "total_steps": 696, "loss": 1.229781985282898, "lr": 3.2758420111188274e-06, "epoch": 1.1982758620689655, "percentage": 39.94, "elapsed_time": "1:14:27", "remaining_time": "1:51:57"}
140
+ {"current_steps": 280, "total_steps": 696, "loss": 0.888131320476532, "lr": 3.419900764336468e-06, "epoch": 1.206896551724138, "percentage": 40.23, "elapsed_time": "1:15:06", "remaining_time": "1:51:35"}
141
+ {"current_steps": 282, "total_steps": 696, "loss": 1.2144626379013062, "lr": 3.5709724261693514e-06, "epoch": 1.2155172413793103, "percentage": 40.52, "elapsed_time": "1:15:42", "remaining_time": "1:51:08"}
142
+ {"current_steps": 284, "total_steps": 696, "loss": 0.8701605200767517, "lr": 3.7289172290859634e-06, "epoch": 1.2241379310344827, "percentage": 40.8, "elapsed_time": "1:16:10", "remaining_time": "1:50:30"}
143
+ {"current_steps": 286, "total_steps": 696, "loss": 0.5924324989318848, "lr": 3.893589046705279e-06, "epoch": 1.2327586206896552, "percentage": 41.09, "elapsed_time": "1:16:34", "remaining_time": "1:49:47"}
144
+ {"current_steps": 288, "total_steps": 696, "loss": 1.1794689893722534, "lr": 4.064835528989056e-06, "epoch": 1.2413793103448276, "percentage": 41.38, "elapsed_time": "1:17:15", "remaining_time": "1:49:27"}
145
+ {"current_steps": 290, "total_steps": 696, "loss": 0.7260119915008545, "lr": 4.242498243192066e-06, "epoch": 1.25, "percentage": 41.67, "elapsed_time": "1:17:42", "remaining_time": "1:48:47"}
146
+ {"current_steps": 292, "total_steps": 696, "loss": 1.0264055728912354, "lr": 4.426412820439941e-06, "epoch": 1.2586206896551724, "percentage": 41.95, "elapsed_time": "1:18:03", "remaining_time": "1:48:00"}
147
+ {"current_steps": 294, "total_steps": 696, "loss": 1.4605242013931274, "lr": 4.616409107798898e-06, "epoch": 1.2672413793103448, "percentage": 42.24, "elapsed_time": "1:18:37", "remaining_time": "1:47:30"}
148
+ {"current_steps": 296, "total_steps": 696, "loss": 0.7666136622428894, "lr": 4.812311325696798e-06, "epoch": 1.2758620689655173, "percentage": 42.53, "elapsed_time": "1:19:04", "remaining_time": "1:46:51"}
149
+ {"current_steps": 298, "total_steps": 696, "loss": 1.0257295370101929, "lr": 5.013938230549751e-06, "epoch": 1.2844827586206897, "percentage": 42.82, "elapsed_time": "1:19:38", "remaining_time": "1:46:21"}
150
+ {"current_steps": 300, "total_steps": 696, "loss": 1.0583548545837402, "lr": 5.221103282443953e-06, "epoch": 1.293103448275862, "percentage": 43.1, "elapsed_time": "1:20:10", "remaining_time": "1:45:49"}
151
+ {"current_steps": 302, "total_steps": 696, "loss": 0.9327218532562256, "lr": 5.433614817717478e-06, "epoch": 1.3017241379310345, "percentage": 43.39, "elapsed_time": "1:20:38", "remaining_time": "1:45:12"}
152
+ {"current_steps": 304, "total_steps": 696, "loss": 1.2481307983398438, "lr": 5.6512762262824986e-06, "epoch": 1.3103448275862069, "percentage": 43.68, "elapsed_time": "1:21:17", "remaining_time": "1:44:49"}
153
+ {"current_steps": 306, "total_steps": 696, "loss": 1.1840277910232544, "lr": 5.873886133523725e-06, "epoch": 1.3189655172413794, "percentage": 43.97, "elapsed_time": "1:21:58", "remaining_time": "1:44:28"}
154
+ {"current_steps": 308, "total_steps": 696, "loss": 0.6794692277908325, "lr": 6.101238586604963e-06, "epoch": 1.3275862068965516, "percentage": 44.25, "elapsed_time": "1:22:17", "remaining_time": "1:43:39"}
155
+ {"current_steps": 310, "total_steps": 696, "loss": 0.9396043419837952, "lr": 6.333123245011208e-06, "epoch": 1.3362068965517242, "percentage": 44.54, "elapsed_time": "1:22:46", "remaining_time": "1:43:03"}
156
+ {"current_steps": 312, "total_steps": 696, "loss": 1.292694091796875, "lr": 6.569325575150216e-06, "epoch": 1.3448275862068966, "percentage": 44.83, "elapsed_time": "1:23:20", "remaining_time": "1:42:34"}
157
+ {"current_steps": 314, "total_steps": 696, "loss": 1.1025623083114624, "lr": 6.809627048833306e-06, "epoch": 1.353448275862069, "percentage": 45.11, "elapsed_time": "1:23:51", "remaining_time": "1:42:01"}
158
+ {"current_steps": 316, "total_steps": 696, "loss": 0.7764808535575867, "lr": 7.053805345451903e-06, "epoch": 1.3620689655172413, "percentage": 45.4, "elapsed_time": "1:24:19", "remaining_time": "1:41:24"}
159
+ {"current_steps": 318, "total_steps": 696, "loss": 1.2299896478652954, "lr": 7.3016345576627605e-06, "epoch": 1.3706896551724137, "percentage": 45.69, "elapsed_time": "1:24:54", "remaining_time": "1:40:55"}
160
+ {"current_steps": 320, "total_steps": 696, "loss": 0.8323980569839478, "lr": 7.552885400391432e-06, "epoch": 1.3793103448275863, "percentage": 45.98, "elapsed_time": "1:25:22", "remaining_time": "1:40:18"}
161
+ {"current_steps": 322, "total_steps": 696, "loss": 1.1044297218322754, "lr": 7.80732542296087e-06, "epoch": 1.3879310344827587, "percentage": 46.26, "elapsed_time": "1:26:02", "remaining_time": "1:39:56"}
162
+ {"current_steps": 324, "total_steps": 696, "loss": 1.518593430519104, "lr": 8.064719224148607e-06, "epoch": 1.396551724137931, "percentage": 46.55, "elapsed_time": "1:26:41", "remaining_time": "1:39:32"}
163
+ {"current_steps": 326, "total_steps": 696, "loss": 1.134735345840454, "lr": 8.324828669973863e-06, "epoch": 1.4051724137931034, "percentage": 46.84, "elapsed_time": "1:27:18", "remaining_time": "1:39:05"}
164
+ {"current_steps": 328, "total_steps": 696, "loss": 0.9491145610809326, "lr": 8.587413114012767e-06, "epoch": 1.4137931034482758, "percentage": 47.13, "elapsed_time": "1:27:56", "remaining_time": "1:38:40"}
165
+ {"current_steps": 330, "total_steps": 696, "loss": 0.7093353271484375, "lr": 8.852229620038169e-06, "epoch": 1.4224137931034484, "percentage": 47.41, "elapsed_time": "1:28:21", "remaining_time": "1:38:00"}
166
+ {"current_steps": 332, "total_steps": 696, "loss": 1.178290605545044, "lr": 9.119033186777797e-06, "epoch": 1.4310344827586206, "percentage": 47.7, "elapsed_time": "1:29:02", "remaining_time": "1:37:37"}
167
+ {"current_steps": 334, "total_steps": 696, "loss": 1.2559309005737305, "lr": 9.387576974583059e-06, "epoch": 1.4396551724137931, "percentage": 47.99, "elapsed_time": "1:29:37", "remaining_time": "1:37:07"}
168
+ {"current_steps": 336, "total_steps": 696, "loss": 1.1395691633224487, "lr": 9.657612533798527e-06, "epoch": 1.4482758620689655, "percentage": 48.28, "elapsed_time": "1:30:14", "remaining_time": "1:36:40"}
169
+ {"current_steps": 338, "total_steps": 696, "loss": 0.824805498123169, "lr": 9.928890034621103e-06, "epoch": 1.456896551724138, "percentage": 48.56, "elapsed_time": "1:30:41", "remaining_time": "1:36:03"}
170
+ {"current_steps": 340, "total_steps": 696, "loss": 1.2302451133728027, "lr": 1.020115849823592e-05, "epoch": 1.4655172413793103, "percentage": 48.85, "elapsed_time": "1:31:21", "remaining_time": "1:35:39"}
171
+ {"current_steps": 342, "total_steps": 696, "loss": 1.2292989492416382, "lr": 1.04741660290154e-05, "epoch": 1.4741379310344827, "percentage": 49.14, "elapsed_time": "1:31:56", "remaining_time": "1:35:10"}
172
+ {"current_steps": 344, "total_steps": 696, "loss": 0.9929612278938293, "lr": 1.0747660047566425e-05, "epoch": 1.4827586206896552, "percentage": 49.43, "elapsed_time": "1:32:26", "remaining_time": "1:34:35"}
173
+ {"current_steps": 346, "total_steps": 696, "loss": 0.8662385940551758, "lr": 1.1021387524410204e-05, "epoch": 1.4913793103448276, "percentage": 49.71, "elapsed_time": "1:32:44", "remaining_time": "1:33:48"}
174
+ {"current_steps": 348, "total_steps": 696, "loss": 0.1633024513721466, "lr": 1.129509521407843e-05, "epoch": 1.5, "percentage": 50.0, "elapsed_time": "1:32:58", "remaining_time": "1:32:58"}
175
+ {"current_steps": 350, "total_steps": 696, "loss": 0.7720933556556702, "lr": 1.1568529889409417e-05, "epoch": 1.5086206896551724, "percentage": 50.29, "elapsed_time": "1:33:26", "remaining_time": "1:32:22"}
176
+ {"current_steps": 352, "total_steps": 696, "loss": 0.43289288878440857, "lr": 1.1841438575827132e-05, "epoch": 1.5172413793103448, "percentage": 50.57, "elapsed_time": "1:33:41", "remaining_time": "1:31:34"}
177
+ {"current_steps": 354, "total_steps": 696, "loss": 0.7231047749519348, "lr": 1.2113568785386663e-05, "epoch": 1.5258620689655173, "percentage": 50.86, "elapsed_time": "1:34:04", "remaining_time": "1:30:52"}
178
+ {"current_steps": 356, "total_steps": 696, "loss": 0.5929489731788635, "lr": 1.2384668750369462e-05, "epoch": 1.5344827586206895, "percentage": 51.15, "elapsed_time": "1:34:19", "remaining_time": "1:30:05"}
179
+ {"current_steps": 358, "total_steps": 696, "loss": 1.3106164932250977, "lr": 1.2654487656212154e-05, "epoch": 1.543103448275862, "percentage": 51.44, "elapsed_time": "1:34:59", "remaining_time": "1:29:40"}
180
+ {"current_steps": 360, "total_steps": 696, "loss": 0.6820691823959351, "lr": 1.2922775873553643e-05, "epoch": 1.5517241379310345, "percentage": 51.72, "elapsed_time": "1:35:15", "remaining_time": "1:28:54"}
181
+ {"current_steps": 362, "total_steps": 696, "loss": 1.2352055311203003, "lr": 1.3189285189185581e-05, "epoch": 1.5603448275862069, "percentage": 52.01, "elapsed_time": "1:35:55", "remaining_time": "1:28:30"}
182
+ {"current_steps": 364, "total_steps": 696, "loss": 1.2412972450256348, "lr": 1.3453769035692767e-05, "epoch": 1.5689655172413794, "percentage": 52.3, "elapsed_time": "1:36:33", "remaining_time": "1:28:04"}
183
+ {"current_steps": 366, "total_steps": 696, "loss": 1.192505955696106, "lr": 1.3715982719570773e-05, "epoch": 1.5775862068965516, "percentage": 52.59, "elapsed_time": "1:37:13", "remaining_time": "1:27:39"}
184
+ {"current_steps": 368, "total_steps": 696, "loss": 1.2814549207687378, "lr": 1.397568364761001e-05, "epoch": 1.5862068965517242, "percentage": 52.87, "elapsed_time": "1:37:53", "remaining_time": "1:27:15"}
185
+ {"current_steps": 370, "total_steps": 696, "loss": 1.2581911087036133, "lr": 1.4232631551336535e-05, "epoch": 1.5948275862068966, "percentage": 53.16, "elapsed_time": "1:38:20", "remaining_time": "1:26:38"}
186
+ {"current_steps": 372, "total_steps": 696, "loss": 1.0574073791503906, "lr": 1.4486588709302146e-05, "epoch": 1.603448275862069, "percentage": 53.45, "elapsed_time": "1:38:59", "remaining_time": "1:26:13"}
187
+ {"current_steps": 374, "total_steps": 696, "loss": 0.6223809123039246, "lr": 1.4737320167018063e-05, "epoch": 1.6120689655172413, "percentage": 53.74, "elapsed_time": "1:39:26", "remaining_time": "1:25:36"}
188
+ {"current_steps": 376, "total_steps": 696, "loss": 0.8683441281318665, "lr": 1.4984593954328579e-05, "epoch": 1.6206896551724137, "percentage": 54.02, "elapsed_time": "1:40:04", "remaining_time": "1:25:10"}
189
+ {"current_steps": 378, "total_steps": 696, "loss": 1.1772154569625854, "lr": 1.5228181300023775e-05, "epoch": 1.6293103448275863, "percentage": 54.31, "elapsed_time": "1:40:45", "remaining_time": "1:24:45"}
190
+ {"current_steps": 380, "total_steps": 696, "loss": 1.0002669095993042, "lr": 1.5467856843492662e-05, "epoch": 1.6379310344827587, "percentage": 54.6, "elapsed_time": "1:41:11", "remaining_time": "1:24:08"}
191
+ {"current_steps": 382, "total_steps": 696, "loss": 0.7394900321960449, "lr": 1.5703398843220852e-05, "epoch": 1.646551724137931, "percentage": 54.89, "elapsed_time": "1:41:33", "remaining_time": "1:23:28"}
192
+ {"current_steps": 384, "total_steps": 696, "loss": 0.7356396913528442, "lr": 1.5934589381940026e-05, "epoch": 1.6551724137931034, "percentage": 55.17, "elapsed_time": "1:42:00", "remaining_time": "1:22:52"}
193
+ {"current_steps": 386, "total_steps": 696, "loss": 1.3296085596084595, "lr": 1.6161214568239207e-05, "epoch": 1.6637931034482758, "percentage": 55.46, "elapsed_time": "1:42:39", "remaining_time": "1:22:26"}
194
+ {"current_steps": 388, "total_steps": 696, "loss": 1.1994720697402954, "lr": 1.638306473445153e-05, "epoch": 1.6724137931034484, "percentage": 55.75, "elapsed_time": "1:43:16", "remaining_time": "1:21:59"}
195
+ {"current_steps": 390, "total_steps": 696, "loss": 1.0203474760055542, "lr": 1.6599934630633217e-05, "epoch": 1.6810344827586206, "percentage": 56.03, "elapsed_time": "1:43:39", "remaining_time": "1:21:20"}
196
+ {"current_steps": 392, "total_steps": 696, "loss": 0.8025456666946411, "lr": 1.681162361445544e-05, "epoch": 1.6896551724137931, "percentage": 56.32, "elapsed_time": "1:44:01", "remaining_time": "1:20:40"}
197
+ {"current_steps": 394, "total_steps": 696, "loss": 1.2664248943328857, "lr": 1.7017935836833353e-05, "epoch": 1.6982758620689655, "percentage": 56.61, "elapsed_time": "1:44:41", "remaining_time": "1:20:14"}
198
+ {"current_steps": 396, "total_steps": 696, "loss": 0.8020282983779907, "lr": 1.7218680423120506e-05, "epoch": 1.706896551724138, "percentage": 56.9, "elapsed_time": "1:45:03", "remaining_time": "1:19:35"}
199
+ {"current_steps": 398, "total_steps": 696, "loss": 0.8401910066604614, "lr": 1.7413671649701026e-05, "epoch": 1.7155172413793105, "percentage": 57.18, "elapsed_time": "1:45:31", "remaining_time": "1:19:00"}
200
+ {"current_steps": 400, "total_steps": 696, "loss": 0.9220103621482849, "lr": 1.7602729115816268e-05, "epoch": 1.7241379310344827, "percentage": 57.47, "elapsed_time": "1:45:59", "remaining_time": "1:18:25"}
201
+ {"current_steps": 402, "total_steps": 696, "loss": 0.999250054359436, "lr": 1.77856779104668e-05, "epoch": 1.7327586206896552, "percentage": 57.76, "elapsed_time": "1:46:37", "remaining_time": "1:17:58"}
202
+ {"current_steps": 404, "total_steps": 696, "loss": 1.7149617671966553, "lr": 1.7962348774235565e-05, "epoch": 1.7413793103448276, "percentage": 58.05, "elapsed_time": "1:47:15", "remaining_time": "1:17:31"}
203
+ {"current_steps": 406, "total_steps": 696, "loss": 0.9877924919128418, "lr": 1.8132578255882124e-05, "epoch": 1.75, "percentage": 58.33, "elapsed_time": "1:47:47", "remaining_time": "1:16:59"}
204
+ {"current_steps": 408, "total_steps": 696, "loss": 1.2993067502975464, "lr": 1.8296208863563515e-05, "epoch": 1.7586206896551724, "percentage": 58.62, "elapsed_time": "1:48:24", "remaining_time": "1:16:31"}
205
+ {"current_steps": 410, "total_steps": 696, "loss": 0.8246227502822876, "lr": 1.8453089210541577e-05, "epoch": 1.7672413793103448, "percentage": 58.91, "elapsed_time": "1:48:52", "remaining_time": "1:15:57"}
206
+ {"current_steps": 412, "total_steps": 696, "loss": 1.2052366733551025, "lr": 1.8603074155241946e-05, "epoch": 1.7758620689655173, "percentage": 59.2, "elapsed_time": "1:49:32", "remaining_time": "1:15:30"}
207
+ {"current_steps": 414, "total_steps": 696, "loss": 1.217560887336731, "lr": 1.8746024935535238e-05, "epoch": 1.7844827586206895, "percentage": 59.48, "elapsed_time": "1:50:13", "remaining_time": "1:15:04"}
208
+ {"current_steps": 416, "total_steps": 696, "loss": 0.3758542239665985, "lr": 1.8881809297116126e-05, "epoch": 1.793103448275862, "percentage": 59.77, "elapsed_time": "1:50:27", "remaining_time": "1:14:21"}
209
+ {"current_steps": 418, "total_steps": 696, "loss": 1.4365136623382568, "lr": 1.901030161586157e-05, "epoch": 1.8017241379310345, "percentage": 60.06, "elapsed_time": "1:51:01", "remaining_time": "1:13:50"}
210
+ {"current_steps": 420, "total_steps": 696, "loss": 0.7461058497428894, "lr": 1.9131383014054904e-05, "epoch": 1.8103448275862069, "percentage": 60.34, "elapsed_time": "1:51:23", "remaining_time": "1:13:11"}
211
+ {"current_steps": 422, "total_steps": 696, "loss": 1.2452733516693115, "lr": 1.924494147036844e-05, "epoch": 1.8189655172413794, "percentage": 60.63, "elapsed_time": "1:52:03", "remaining_time": "1:12:45"}
212
+ {"current_steps": 424, "total_steps": 696, "loss": 1.1930017471313477, "lr": 1.9350871923502582e-05, "epoch": 1.8275862068965516, "percentage": 60.92, "elapsed_time": "1:52:34", "remaining_time": "1:12:12"}
213
+ {"current_steps": 426, "total_steps": 696, "loss": 1.0968483686447144, "lr": 1.944907636938585e-05, "epoch": 1.8362068965517242, "percentage": 61.21, "elapsed_time": "1:53:06", "remaining_time": "1:11:41"}
214
+ {"current_steps": 428, "total_steps": 696, "loss": 0.3261430561542511, "lr": 1.953946395184563e-05, "epoch": 1.8448275862068966, "percentage": 61.49, "elapsed_time": "1:53:20", "remaining_time": "1:10:58"}
215
+ {"current_steps": 430, "total_steps": 696, "loss": 0.6268512010574341, "lr": 1.962195104666592e-05, "epoch": 1.853448275862069, "percentage": 61.78, "elapsed_time": "1:53:46", "remaining_time": "1:10:22"}
216
+ {"current_steps": 432, "total_steps": 696, "loss": 0.832156240940094, "lr": 1.9696461338954278e-05, "epoch": 1.8620689655172413, "percentage": 62.07, "elapsed_time": "1:54:17", "remaining_time": "1:09:50"}
217
+ {"current_steps": 434, "total_steps": 696, "loss": 0.5735204815864563, "lr": 1.976292589374634e-05, "epoch": 1.8706896551724137, "percentage": 62.36, "elapsed_time": "1:54:32", "remaining_time": "1:09:09"}
218
+ {"current_steps": 436, "total_steps": 696, "loss": 0.5290496945381165, "lr": 1.9821283219782627e-05, "epoch": 1.8793103448275863, "percentage": 62.64, "elapsed_time": "1:54:59", "remaining_time": "1:08:34"}
219
+ {"current_steps": 438, "total_steps": 696, "loss": 1.232295036315918, "lr": 1.9871479326398668e-05, "epoch": 1.8879310344827587, "percentage": 62.93, "elapsed_time": "1:55:34", "remaining_time": "1:08:04"}
220
+ {"current_steps": 440, "total_steps": 696, "loss": 1.189243197441101, "lr": 1.9913467773475727e-05, "epoch": 1.896551724137931, "percentage": 63.22, "elapsed_time": "1:56:14", "remaining_time": "1:07:37"}
221
+ {"current_steps": 442, "total_steps": 696, "loss": 0.7931770086288452, "lr": 1.9947209714405984e-05, "epoch": 1.9051724137931034, "percentage": 63.51, "elapsed_time": "1:56:39", "remaining_time": "1:07:02"}
222
+ {"current_steps": 444, "total_steps": 696, "loss": 0.7982081174850464, "lr": 1.9972673932032394e-05, "epoch": 1.9137931034482758, "percentage": 63.79, "elapsed_time": "1:57:04", "remaining_time": "1:06:26"}
223
+ {"current_steps": 446, "total_steps": 696, "loss": 0.661638081073761, "lr": 1.9989836867529987e-05, "epoch": 1.9224137931034484, "percentage": 64.08, "elapsed_time": "1:57:31", "remaining_time": "1:05:52"}
224
+ {"current_steps": 448, "total_steps": 696, "loss": 1.3983592987060547, "lr": 1.999868264220186e-05, "epoch": 1.9310344827586206, "percentage": 64.37, "elapsed_time": "1:58:10", "remaining_time": "1:05:25"}
225
+ {"current_steps": 450, "total_steps": 696, "loss": 1.5112402439117432, "lr": 1.999920307216975e-05, "epoch": 1.9396551724137931, "percentage": 64.66, "elapsed_time": "1:58:50", "remaining_time": "1:04:58"}
226
+ {"current_steps": 452, "total_steps": 696, "loss": 1.2273776531219482, "lr": 1.9991397675945524e-05, "epoch": 1.9482758620689655, "percentage": 64.94, "elapsed_time": "1:59:21", "remaining_time": "1:04:25"}
227
+ {"current_steps": 454, "total_steps": 696, "loss": 0.7660670876502991, "lr": 1.9975273674876653e-05, "epoch": 1.956896551724138, "percentage": 65.23, "elapsed_time": "1:59:49", "remaining_time": "1:03:52"}
228
+ {"current_steps": 456, "total_steps": 696, "loss": 0.9497285485267639, "lr": 1.99508459864652e-05, "epoch": 1.9655172413793105, "percentage": 65.52, "elapsed_time": "2:00:13", "remaining_time": "1:03:16"}
229
+ {"current_steps": 458, "total_steps": 696, "loss": 1.0104745626449585, "lr": 1.991813721056655e-05, "epoch": 1.9741379310344827, "percentage": 65.8, "elapsed_time": "2:00:53", "remaining_time": "1:02:49"}
230
+ {"current_steps": 460, "total_steps": 696, "loss": 0.919417679309845, "lr": 1.987717760848061e-05, "epoch": 1.9827586206896552, "percentage": 66.09, "elapsed_time": "2:01:20", "remaining_time": "1:02:15"}
231
+ {"current_steps": 462, "total_steps": 696, "loss": 1.400168538093567, "lr": 1.982800507495485e-05, "epoch": 1.9913793103448276, "percentage": 66.38, "elapsed_time": "2:02:00", "remaining_time": "1:01:47"}
232
+ {"current_steps": 464, "total_steps": 696, "loss": 1.335584282875061, "lr": 1.977066510312508e-05, "epoch": 2.0, "percentage": 66.67, "elapsed_time": "2:02:38", "remaining_time": "1:01:19"}
233
+ {"current_steps": 466, "total_steps": 696, "loss": 1.1145689487457275, "lr": 1.9705210742426404e-05, "epoch": 2.0086206896551726, "percentage": 66.95, "elapsed_time": "2:03:14", "remaining_time": "1:00:49"}
234
+ {"current_steps": 468, "total_steps": 696, "loss": 1.1242371797561646, "lr": 1.9631702549513264e-05, "epoch": 2.0172413793103448, "percentage": 67.24, "elapsed_time": "2:03:54", "remaining_time": "1:00:21"}
235
+ {"current_steps": 470, "total_steps": 696, "loss": 1.022644281387329, "lr": 1.9550208532234013e-05, "epoch": 2.0258620689655173, "percentage": 67.53, "elapsed_time": "2:04:33", "remaining_time": "0:59:53"}
236
+ {"current_steps": 472, "total_steps": 696, "loss": 0.6067448258399963, "lr": 1.9460804086711857e-05, "epoch": 2.0344827586206895, "percentage": 67.82, "elapsed_time": "2:04:56", "remaining_time": "0:59:17"}
237
+ {"current_steps": 474, "total_steps": 696, "loss": 1.1570570468902588, "lr": 1.936357192759033e-05, "epoch": 2.043103448275862, "percentage": 68.1, "elapsed_time": "2:05:29", "remaining_time": "0:58:46"}
238
+ {"current_steps": 476, "total_steps": 696, "loss": 1.0996088981628418, "lr": 1.9258602011507896e-05, "epoch": 2.0517241379310347, "percentage": 68.39, "elapsed_time": "2:06:06", "remaining_time": "0:58:17"}
239
+ {"current_steps": 478, "total_steps": 696, "loss": 0.6901102066040039, "lr": 1.9145991453872398e-05, "epoch": 2.060344827586207, "percentage": 68.68, "elapsed_time": "2:06:31", "remaining_time": "0:57:42"}
240
+ {"current_steps": 480, "total_steps": 696, "loss": 1.4176607131958008, "lr": 1.902584443901246e-05, "epoch": 2.0689655172413794, "percentage": 68.97, "elapsed_time": "2:07:04", "remaining_time": "0:57:11"}
241
+ {"current_steps": 482, "total_steps": 696, "loss": 1.0099068880081177, "lr": 1.8898272123788897e-05, "epoch": 2.0775862068965516, "percentage": 69.25, "elapsed_time": "2:07:40", "remaining_time": "0:56:41"}
242
+ {"current_steps": 484, "total_steps": 696, "loss": 0.9407973289489746, "lr": 1.8763392534755276e-05, "epoch": 2.086206896551724, "percentage": 69.54, "elapsed_time": "2:08:19", "remaining_time": "0:56:12"}
243
+ {"current_steps": 486, "total_steps": 696, "loss": 1.2682446241378784, "lr": 1.862133045896286e-05, "epoch": 2.0948275862068964, "percentage": 69.83, "elapsed_time": "2:08:59", "remaining_time": "0:55:44"}
244
+ {"current_steps": 488, "total_steps": 696, "loss": 0.2819199562072754, "lr": 1.847221732851092e-05, "epoch": 2.103448275862069, "percentage": 70.11, "elapsed_time": "2:09:14", "remaining_time": "0:55:05"}
245
+ {"current_steps": 490, "total_steps": 696, "loss": 0.4681054651737213, "lr": 1.831619109894921e-05, "epoch": 2.1120689655172415, "percentage": 70.4, "elapsed_time": "2:09:30", "remaining_time": "0:54:26"}
246
+ {"current_steps": 492, "total_steps": 696, "loss": 0.9479153156280518, "lr": 1.81533961216451e-05, "epoch": 2.1206896551724137, "percentage": 70.69, "elapsed_time": "2:10:01", "remaining_time": "0:53:54"}
247
+ {"current_steps": 494, "total_steps": 696, "loss": 1.2515679597854614, "lr": 1.79839830102335e-05, "epoch": 2.1293103448275863, "percentage": 70.98, "elapsed_time": "2:10:40", "remaining_time": "0:53:26"}
248
+ {"current_steps": 496, "total_steps": 696, "loss": 1.110595464706421, "lr": 1.780810850127307e-05, "epoch": 2.1379310344827585, "percentage": 71.26, "elapsed_time": "2:11:15", "remaining_time": "0:52:55"}
249
+ {"current_steps": 498, "total_steps": 696, "loss": 0.6906008720397949, "lr": 1.7625935309237705e-05, "epoch": 2.146551724137931, "percentage": 71.55, "elapsed_time": "2:11:43", "remaining_time": "0:52:22"}
250
+ {"current_steps": 500, "total_steps": 696, "loss": 0.8901933431625366, "lr": 1.7437631975977327e-05, "epoch": 2.1551724137931036, "percentage": 71.84, "elapsed_time": "2:12:12", "remaining_time": "0:51:49"}
251
+ {"current_steps": 502, "total_steps": 696, "loss": 1.1632163524627686, "lr": 1.7243372714787447e-05, "epoch": 2.163793103448276, "percentage": 72.13, "elapsed_time": "2:12:53", "remaining_time": "0:51:21"}
252
+ {"current_steps": 504, "total_steps": 696, "loss": 1.0225820541381836, "lr": 1.704333724923158e-05, "epoch": 2.1724137931034484, "percentage": 72.41, "elapsed_time": "2:13:21", "remaining_time": "0:50:48"}
253
+ {"current_steps": 506, "total_steps": 696, "loss": 0.596871018409729, "lr": 1.683771064686572e-05, "epoch": 2.1810344827586206, "percentage": 72.7, "elapsed_time": "2:13:50", "remaining_time": "0:50:15"}
254
+ {"current_steps": 508, "total_steps": 696, "loss": 1.1479698419570923, "lr": 1.6626683148018704e-05, "epoch": 2.189655172413793, "percentage": 72.99, "elapsed_time": "2:14:30", "remaining_time": "0:49:46"}
255
+ {"current_steps": 510, "total_steps": 696, "loss": 1.113614559173584, "lr": 1.6410449989786864e-05, "epoch": 2.1982758620689653, "percentage": 73.28, "elapsed_time": "2:15:05", "remaining_time": "0:49:15"}
256
+ {"current_steps": 512, "total_steps": 696, "loss": 0.7449208498001099, "lr": 1.6189211225405815e-05, "epoch": 2.206896551724138, "percentage": 73.56, "elapsed_time": "2:15:32", "remaining_time": "0:48:42"}
257
+ {"current_steps": 514, "total_steps": 696, "loss": 1.1159427165985107, "lr": 1.596317153916643e-05, "epoch": 2.2155172413793105, "percentage": 73.85, "elapsed_time": "2:16:13", "remaining_time": "0:48:14"}
258
+ {"current_steps": 516, "total_steps": 696, "loss": 0.7390024662017822, "lr": 1.5732540057046285e-05, "epoch": 2.2241379310344827, "percentage": 74.14, "elapsed_time": "2:16:40", "remaining_time": "0:47:40"}
259
+ {"current_steps": 518, "total_steps": 696, "loss": 0.7803560495376587, "lr": 1.5497530153231886e-05, "epoch": 2.2327586206896552, "percentage": 74.43, "elapsed_time": "2:17:11", "remaining_time": "0:47:08"}
260
+ {"current_steps": 520, "total_steps": 696, "loss": 1.3723326921463013, "lr": 1.5258359252710415e-05, "epoch": 2.2413793103448274, "percentage": 74.71, "elapsed_time": "2:17:50", "remaining_time": "0:46:39"}
261
+ {"current_steps": 522, "total_steps": 696, "loss": 0.7922369241714478, "lr": 1.5015248630113865e-05, "epoch": 2.25, "percentage": 75.0, "elapsed_time": "2:18:29", "remaining_time": "0:46:09"}
262
+ {"current_steps": 524, "total_steps": 696, "loss": 0.7873719334602356, "lr": 1.476842320500153e-05, "epoch": 2.2586206896551726, "percentage": 75.29, "elapsed_time": "2:18:56", "remaining_time": "0:45:36"}
263
+ {"current_steps": 526, "total_steps": 696, "loss": 0.7667290568351746, "lr": 1.4518111333770401e-05, "epoch": 2.2672413793103448, "percentage": 75.57, "elapsed_time": "2:19:24", "remaining_time": "0:45:03"}
264
+ {"current_steps": 528, "total_steps": 696, "loss": 1.106222152709961, "lr": 1.4264544598385801e-05, "epoch": 2.2758620689655173, "percentage": 75.86, "elapsed_time": "2:20:00", "remaining_time": "0:44:32"}
265
+ {"current_steps": 530, "total_steps": 696, "loss": 1.0902438163757324, "lr": 1.4007957592127807e-05, "epoch": 2.2844827586206895, "percentage": 76.15, "elapsed_time": "2:20:40", "remaining_time": "0:44:03"}
266
+ {"current_steps": 532, "total_steps": 696, "loss": 0.8156579732894897, "lr": 1.374858770255177e-05, "epoch": 2.293103448275862, "percentage": 76.44, "elapsed_time": "2:21:05", "remaining_time": "0:43:29"}
267
+ {"current_steps": 534, "total_steps": 696, "loss": 1.139012098312378, "lr": 1.3486674891863518e-05, "epoch": 2.3017241379310347, "percentage": 76.72, "elapsed_time": "2:21:33", "remaining_time": "0:42:56"}
268
+ {"current_steps": 536, "total_steps": 696, "loss": 1.191633701324463, "lr": 1.322246147491271e-05, "epoch": 2.310344827586207, "percentage": 77.01, "elapsed_time": "2:22:13", "remaining_time": "0:42:27"}
269
+ {"current_steps": 538, "total_steps": 696, "loss": 0.7655143141746521, "lr": 1.2956191895009415e-05, "epoch": 2.3189655172413794, "percentage": 77.3, "elapsed_time": "2:22:40", "remaining_time": "0:41:53"}
270
+ {"current_steps": 540, "total_steps": 696, "loss": 0.9563652276992798, "lr": 1.268811249777167e-05, "epoch": 2.3275862068965516, "percentage": 77.59, "elapsed_time": "2:23:13", "remaining_time": "0:41:22"}
271
+ {"current_steps": 542, "total_steps": 696, "loss": 0.7211173176765442, "lr": 1.2418471303212933e-05, "epoch": 2.336206896551724, "percentage": 77.87, "elapsed_time": "2:23:33", "remaining_time": "0:40:47"}
272
+ {"current_steps": 544, "total_steps": 696, "loss": 1.0490473508834839, "lr": 1.2147517776280517e-05, "epoch": 2.344827586206897, "percentage": 78.16, "elapsed_time": "2:24:09", "remaining_time": "0:40:16"}
273
+ {"current_steps": 546, "total_steps": 696, "loss": 0.8839560151100159, "lr": 1.1875502596057116e-05, "epoch": 2.353448275862069, "percentage": 78.45, "elapsed_time": "2:24:37", "remaining_time": "0:39:43"}
274
+ {"current_steps": 548, "total_steps": 696, "loss": 1.127523422241211, "lr": 1.1602677423839195e-05, "epoch": 2.3620689655172415, "percentage": 78.74, "elapsed_time": "2:25:17", "remaining_time": "0:39:14"}
275
+ {"current_steps": 550, "total_steps": 696, "loss": 0.6361532807350159, "lr": 1.1329294670306513e-05, "epoch": 2.3706896551724137, "percentage": 79.02, "elapsed_time": "2:25:43", "remaining_time": "0:38:40"}
276
+ {"current_steps": 552, "total_steps": 696, "loss": 0.31083130836486816, "lr": 1.1055607261998416e-05, "epoch": 2.3793103448275863, "percentage": 79.31, "elapsed_time": "2:25:58", "remaining_time": "0:38:04"}
277
+ {"current_steps": 554, "total_steps": 696, "loss": 1.2638636827468872, "lr": 1.078186840731281e-05, "epoch": 2.3879310344827585, "percentage": 79.6, "elapsed_time": "2:26:38", "remaining_time": "0:37:35"}
278
+ {"current_steps": 556, "total_steps": 696, "loss": 1.079472303390503, "lr": 1.0508331362244472e-05, "epoch": 2.396551724137931, "percentage": 79.89, "elapsed_time": "2:27:18", "remaining_time": "0:37:05"}
279
+ {"current_steps": 558, "total_steps": 696, "loss": 0.4111257493495941, "lr": 1.0235249196079217e-05, "epoch": 2.405172413793103, "percentage": 80.17, "elapsed_time": "2:27:34", "remaining_time": "0:36:29"}
280
+ {"current_steps": 560, "total_steps": 696, "loss": 0.9900328516960144, "lr": 9.96287455726086e-06, "epoch": 2.413793103448276, "percentage": 80.46, "elapsed_time": "2:28:09", "remaining_time": "0:35:58"}
281
+ {"current_steps": 562, "total_steps": 696, "loss": 0.8341249823570251, "lr": 9.691459439647469e-06, "epoch": 2.4224137931034484, "percentage": 80.75, "elapsed_time": "2:28:38", "remaining_time": "0:35:26"}
282
+ {"current_steps": 564, "total_steps": 696, "loss": 0.8743176460266113, "lr": 9.421254949373347e-06, "epoch": 2.4310344827586206, "percentage": 81.03, "elapsed_time": "2:29:17", "remaining_time": "0:34:56"}
283
+ {"current_steps": 566, "total_steps": 696, "loss": 1.056296467781067, "lr": 9.152511072532206e-06, "epoch": 2.439655172413793, "percentage": 81.32, "elapsed_time": "2:29:57", "remaining_time": "0:34:26"}
284
+ {"current_steps": 568, "total_steps": 696, "loss": 1.413284182548523, "lr": 8.88547644389654e-06, "epoch": 2.4482758620689653, "percentage": 81.61, "elapsed_time": "2:30:36", "remaining_time": "0:33:56"}
285
+ {"current_steps": 570, "total_steps": 696, "loss": 0.1686408817768097, "lr": 8.62039811688737e-06, "epoch": 2.456896551724138, "percentage": 81.9, "elapsed_time": "2:30:50", "remaining_time": "0:33:20"}
286
+ {"current_steps": 572, "total_steps": 696, "loss": 0.9291278123855591, "lr": 8.357521335006843e-06, "epoch": 2.4655172413793105, "percentage": 82.18, "elapsed_time": "2:31:23", "remaining_time": "0:32:49"}
287
+ {"current_steps": 574, "total_steps": 696, "loss": 1.1657003164291382, "lr": 8.097089304945431e-06, "epoch": 2.4741379310344827, "percentage": 82.47, "elapsed_time": "2:31:58", "remaining_time": "0:32:17"}
288
+ {"current_steps": 576, "total_steps": 696, "loss": 1.1208579540252686, "lr": 7.839342971573412e-06, "epoch": 2.4827586206896552, "percentage": 82.76, "elapsed_time": "2:32:38", "remaining_time": "0:31:47"}
289
+ {"current_steps": 578, "total_steps": 696, "loss": 0.6583477258682251, "lr": 7.584520795025116e-06, "epoch": 2.4913793103448274, "percentage": 83.05, "elapsed_time": "2:33:04", "remaining_time": "0:31:15"}
290
+ {"current_steps": 580, "total_steps": 696, "loss": 1.1409282684326172, "lr": 7.332858530081812e-06, "epoch": 2.5, "percentage": 83.33, "elapsed_time": "2:33:44", "remaining_time": "0:30:44"}
291
+ {"current_steps": 582, "total_steps": 696, "loss": 0.9950229525566101, "lr": 7.084589008057583e-06, "epoch": 2.5086206896551726, "percentage": 83.62, "elapsed_time": "2:34:19", "remaining_time": "0:30:13"}
292
+ {"current_steps": 584, "total_steps": 696, "loss": 0.677608847618103, "lr": 6.839941921389941e-06, "epoch": 2.5172413793103448, "percentage": 83.91, "elapsed_time": "2:34:45", "remaining_time": "0:29:40"}
293
+ {"current_steps": 586, "total_steps": 696, "loss": 0.21669866144657135, "lr": 6.599143611134356e-06, "epoch": 2.5258620689655173, "percentage": 84.2, "elapsed_time": "2:35:00", "remaining_time": "0:29:05"}
294
+ {"current_steps": 588, "total_steps": 696, "loss": 0.7037891745567322, "lr": 6.362416857559549e-06, "epoch": 2.5344827586206895, "percentage": 84.48, "elapsed_time": "2:35:21", "remaining_time": "0:28:32"}
295
+ {"current_steps": 590, "total_steps": 696, "loss": 0.9864909648895264, "lr": 6.129980674036977e-06, "epoch": 2.543103448275862, "percentage": 84.77, "elapsed_time": "2:35:52", "remaining_time": "0:28:00"}
296
+ {"current_steps": 592, "total_steps": 696, "loss": 0.8599920272827148, "lr": 5.90205010441544e-06, "epoch": 2.5517241379310347, "percentage": 85.06, "elapsed_time": "2:36:25", "remaining_time": "0:27:28"}
297
+ {"current_steps": 594, "total_steps": 696, "loss": 0.7931281328201294, "lr": 5.678836024068174e-06, "epoch": 2.560344827586207, "percentage": 85.34, "elapsed_time": "2:36:58", "remaining_time": "0:26:57"}
298
+ {"current_steps": 596, "total_steps": 696, "loss": 0.7438328862190247, "lr": 5.4605449447964406e-06, "epoch": 2.5689655172413794, "percentage": 85.63, "elapsed_time": "2:37:29", "remaining_time": "0:26:25"}
299
+ {"current_steps": 598, "total_steps": 696, "loss": 0.7791666984558105, "lr": 5.24737882377032e-06, "epoch": 2.5775862068965516, "percentage": 85.92, "elapsed_time": "2:38:08", "remaining_time": "0:25:54"}
300
+ {"current_steps": 600, "total_steps": 696, "loss": 1.1297264099121094, "lr": 5.039534876683188e-06, "epoch": 2.586206896551724, "percentage": 86.21, "elapsed_time": "2:38:43", "remaining_time": "0:25:23"}
301
+ {"current_steps": 602, "total_steps": 696, "loss": 0.6758660674095154, "lr": 4.837205395292953e-06, "epoch": 2.594827586206897, "percentage": 86.49, "elapsed_time": "2:39:10", "remaining_time": "0:24:51"}
302
+ {"current_steps": 604, "total_steps": 696, "loss": 1.114516019821167, "lr": 4.640577569518789e-06, "epoch": 2.603448275862069, "percentage": 86.78, "elapsed_time": "2:39:50", "remaining_time": "0:24:20"}
303
+ {"current_steps": 606, "total_steps": 696, "loss": 0.6865538954734802, "lr": 4.449833314257892e-06, "epoch": 2.612068965517241, "percentage": 87.07, "elapsed_time": "2:40:18", "remaining_time": "0:23:48"}
304
+ {"current_steps": 608, "total_steps": 696, "loss": 0.667411744594574, "lr": 4.265149101082668e-06, "epoch": 2.6206896551724137, "percentage": 87.36, "elapsed_time": "2:40:45", "remaining_time": "0:23:16"}
305
+ {"current_steps": 610, "total_steps": 696, "loss": 0.999786376953125, "lr": 4.086695794973827e-06, "epoch": 2.6293103448275863, "percentage": 87.64, "elapsed_time": "2:41:24", "remaining_time": "0:22:45"}
306
+ {"current_steps": 612, "total_steps": 696, "loss": 0.8935456275939941, "lr": 3.914638496240613e-06, "epoch": 2.637931034482759, "percentage": 87.93, "elapsed_time": "2:41:53", "remaining_time": "0:22:13"}
307
+ {"current_steps": 614, "total_steps": 696, "loss": 0.293642520904541, "lr": 3.7491363877742946e-06, "epoch": 2.646551724137931, "percentage": 88.22, "elapsed_time": "2:42:08", "remaining_time": "0:21:39"}
308
+ {"current_steps": 616, "total_steps": 696, "loss": 0.860302209854126, "lr": 3.5903425877764024e-06, "epoch": 2.655172413793103, "percentage": 88.51, "elapsed_time": "2:42:48", "remaining_time": "0:21:08"}
309
+ {"current_steps": 618, "total_steps": 696, "loss": 0.679624617099762, "lr": 3.4384040080977486e-06, "epoch": 2.663793103448276, "percentage": 88.79, "elapsed_time": "2:43:09", "remaining_time": "0:20:35"}
310
+ {"current_steps": 620, "total_steps": 696, "loss": 0.6118066310882568, "lr": 3.2934612183194636e-06, "epoch": 2.6724137931034484, "percentage": 89.08, "elapsed_time": "2:43:37", "remaining_time": "0:20:03"}
311
+ {"current_steps": 622, "total_steps": 696, "loss": 1.0104352235794067, "lr": 3.1556483157016765e-06, "epoch": 2.6810344827586206, "percentage": 89.37, "elapsed_time": "2:44:17", "remaining_time": "0:19:32"}
312
+ {"current_steps": 624, "total_steps": 696, "loss": 0.7297549843788147, "lr": 3.0250928011203114e-06, "epoch": 2.689655172413793, "percentage": 89.66, "elapsed_time": "2:44:45", "remaining_time": "0:19:00"}
313
+ {"current_steps": 626, "total_steps": 696, "loss": 1.0180128812789917, "lr": 2.9019154611065938e-06, "epoch": 2.6982758620689653, "percentage": 89.94, "elapsed_time": "2:45:23", "remaining_time": "0:18:29"}
314
+ {"current_steps": 628, "total_steps": 696, "loss": 0.729310154914856, "lr": 2.7862302560985105e-06, "epoch": 2.706896551724138, "percentage": 90.23, "elapsed_time": "2:45:51", "remaining_time": "0:17:57"}
315
+ {"current_steps": 630, "total_steps": 696, "loss": 0.5759393572807312, "lr": 2.6781442150076485e-06, "epoch": 2.7155172413793105, "percentage": 90.52, "elapsed_time": "2:46:18", "remaining_time": "0:17:25"}
316
+ {"current_steps": 632, "total_steps": 696, "loss": 0.8420278429985046, "lr": 2.577757336198829e-06, "epoch": 2.7241379310344827, "percentage": 90.8, "elapsed_time": "2:46:41", "remaining_time": "0:16:52"}
317
+ {"current_steps": 634, "total_steps": 696, "loss": 1.203715443611145, "lr": 2.4851624949742624e-06, "epoch": 2.7327586206896552, "percentage": 91.09, "elapsed_time": "2:47:20", "remaining_time": "0:16:21"}
318
+ {"current_steps": 636, "total_steps": 696, "loss": 1.251405119895935, "lr": 2.4004453576477363e-06, "epoch": 2.7413793103448274, "percentage": 91.38, "elapsed_time": "2:48:01", "remaining_time": "0:15:51"}
319
+ {"current_steps": 638, "total_steps": 696, "loss": 0.5981881022453308, "lr": 2.323684302288429e-06, "epoch": 2.75, "percentage": 91.67, "elapsed_time": "2:48:23", "remaining_time": "0:15:18"}
320
+ {"current_steps": 640, "total_steps": 696, "loss": 0.4858713448047638, "lr": 2.2549503462075524e-06, "epoch": 2.7586206896551726, "percentage": 91.95, "elapsed_time": "2:48:49", "remaining_time": "0:14:46"}
321
+ {"current_steps": 642, "total_steps": 696, "loss": 1.0236581563949585, "lr": 2.1943070802550147e-06, "epoch": 2.7672413793103448, "percentage": 92.24, "elapsed_time": "2:49:26", "remaining_time": "0:14:15"}
322
+ {"current_steps": 644, "total_steps": 696, "loss": 0.7582760453224182, "lr": 2.141810609986821e-06, "epoch": 2.7758620689655173, "percentage": 92.53, "elapsed_time": "2:49:49", "remaining_time": "0:13:42"}
323
+ {"current_steps": 646, "total_steps": 696, "loss": 0.7772106528282166, "lr": 2.0975095037577143e-06, "epoch": 2.7844827586206895, "percentage": 92.82, "elapsed_time": "2:50:28", "remaining_time": "0:13:11"}
324
+ {"current_steps": 648, "total_steps": 696, "loss": 0.9372343420982361, "lr": 2.0614447477869856e-06, "epoch": 2.793103448275862, "percentage": 93.1, "elapsed_time": "2:50:59", "remaining_time": "0:12:39"}
325
+ {"current_steps": 650, "total_steps": 696, "loss": 1.1293443441390991, "lr": 2.0336497082391155e-06, "epoch": 2.8017241379310347, "percentage": 93.39, "elapsed_time": "2:51:40", "remaining_time": "0:12:08"}
326
+ {"current_steps": 652, "total_steps": 696, "loss": 0.21890364587306976, "lr": 2.0141501003542617e-06, "epoch": 2.810344827586207, "percentage": 93.68, "elapsed_time": "2:51:55", "remaining_time": "0:11:36"}
327
+ {"current_steps": 654, "total_steps": 696, "loss": 0.9752658605575562, "lr": 2.002963964657213e-06, "epoch": 2.8189655172413794, "percentage": 93.97, "elapsed_time": "2:52:27", "remaining_time": "0:11:04"}
328
+ {"current_steps": 656, "total_steps": 696, "loss": 1.158425211906433, "lr": 2.0001016502667552e-06, "epoch": 2.8275862068965516, "percentage": 94.25, "elapsed_time": "2:52:58", "remaining_time": "0:10:32"}
329
+ {"current_steps": 658, "total_steps": 696, "loss": 0.09864208102226257, "lr": 2.0055658053209428e-06, "epoch": 2.836206896551724, "percentage": 94.54, "elapsed_time": "2:53:13", "remaining_time": "0:10:00"}
330
+ {"current_steps": 660, "total_steps": 696, "loss": 1.1222366094589233, "lr": 2.0193513745271104e-06, "epoch": 2.844827586206897, "percentage": 94.83, "elapsed_time": "2:53:49", "remaining_time": "0:09:28"}
331
+ {"current_steps": 662, "total_steps": 696, "loss": 0.7978563904762268, "lr": 2.041445603838903e-06, "epoch": 2.853448275862069, "percentage": 95.11, "elapsed_time": "2:54:12", "remaining_time": "0:08:56"}
332
+ {"current_steps": 664, "total_steps": 696, "loss": 1.0888785123825073, "lr": 2.071828052255974e-06, "epoch": 2.862068965517241, "percentage": 95.4, "elapsed_time": "2:54:45", "remaining_time": "0:08:25"}
333
+ {"current_steps": 666, "total_steps": 696, "loss": 0.8604229688644409, "lr": 2.110470610735471e-06, "epoch": 2.8706896551724137, "percentage": 95.69, "elapsed_time": "2:55:23", "remaining_time": "0:07:54"}
334
+ {"current_steps": 668, "total_steps": 696, "loss": 0.6924110054969788, "lr": 2.157337528197775e-06, "epoch": 2.8793103448275863, "percentage": 95.98, "elapsed_time": "2:55:51", "remaining_time": "0:07:22"}
335
+ {"current_steps": 670, "total_steps": 696, "loss": 1.1259899139404297, "lr": 2.2123854446024656e-06, "epoch": 2.887931034482759, "percentage": 96.26, "elapsed_time": "2:56:31", "remaining_time": "0:06:51"}
336
+ {"current_steps": 672, "total_steps": 696, "loss": 0.5768133997917175, "lr": 2.275563431063895e-06, "epoch": 2.896551724137931, "percentage": 96.55, "elapsed_time": "2:57:00", "remaining_time": "0:06:19"}
337
+ {"current_steps": 674, "total_steps": 696, "loss": 0.655694842338562, "lr": 2.346813036969242e-06, "epoch": 2.905172413793103, "percentage": 96.84, "elapsed_time": "2:57:29", "remaining_time": "0:05:47"}
338
+ {"current_steps": 676, "total_steps": 696, "loss": 0.6339759230613708, "lr": 2.4260683440554954e-06, "epoch": 2.913793103448276, "percentage": 97.13, "elapsed_time": "2:57:57", "remaining_time": "0:05:15"}
339
+ {"current_steps": 678, "total_steps": 696, "loss": 0.9285910129547119, "lr": 2.5132560273952874e-06, "epoch": 2.9224137931034484, "percentage": 97.41, "elapsed_time": "2:58:27", "remaining_time": "0:04:44"}
340
+ {"current_steps": 680, "total_steps": 696, "loss": 0.8067626953125, "lr": 2.6082954232351966e-06, "epoch": 2.9310344827586206, "percentage": 97.7, "elapsed_time": "2:58:53", "remaining_time": "0:04:12"}
341
+ {"current_steps": 682, "total_steps": 696, "loss": 1.2526490688323975, "lr": 2.711098603623742e-06, "epoch": 2.939655172413793, "percentage": 97.99, "elapsed_time": "2:59:34", "remaining_time": "0:03:41"}
342
+ {"current_steps": 684, "total_steps": 696, "loss": 1.1392483711242676, "lr": 2.821570457759993e-06, "epoch": 2.9482758620689653, "percentage": 98.28, "elapsed_time": "3:00:14", "remaining_time": "0:03:09"}
343
+ {"current_steps": 686, "total_steps": 696, "loss": 0.7813910841941833, "lr": 2.9396087799875866e-06, "epoch": 2.956896551724138, "percentage": 98.56, "elapsed_time": "3:00:41", "remaining_time": "0:02:38"}
344
+ {"current_steps": 688, "total_steps": 696, "loss": 0.9301526546478271, "lr": 3.065104364352743e-06, "epoch": 2.9655172413793105, "percentage": 98.85, "elapsed_time": "3:01:13", "remaining_time": "0:02:06"}
345
+ {"current_steps": 690, "total_steps": 696, "loss": 1.025092601776123, "lr": 3.197941105638702e-06, "epoch": 2.9741379310344827, "percentage": 99.14, "elapsed_time": "3:01:53", "remaining_time": "0:01:34"}
346
+ {"current_steps": 692, "total_steps": 696, "loss": 0.6867604851722717, "lr": 3.337996106783247e-06, "epoch": 2.9827586206896552, "percentage": 99.43, "elapsed_time": "3:02:21", "remaining_time": "0:01:03"}
347
+ {"current_steps": 694, "total_steps": 696, "loss": 1.19440495967865, "lr": 3.4851397925798167e-06, "epoch": 2.9913793103448274, "percentage": 99.71, "elapsed_time": "3:03:01", "remaining_time": "0:00:31"}
348
+ {"current_steps": 696, "total_steps": 696, "loss": 1.113608717918396, "lr": 3.6392360295571228e-06, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "3:03:36", "remaining_time": "0:00:00"}
349
+ {"current_steps": 696, "total_steps": 696, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "3:03:36", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,2479 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 696,
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": 0.5692610740661621,
15
+ "learning_rate": 5.714285714285715e-07,
16
+ "loss": 1.9584773778915405,
17
+ "step": 2
18
+ },
19
+ {
20
+ "epoch": 0.017241379310344827,
21
+ "grad_norm": 0.726763904094696,
22
+ "learning_rate": 1.7142857142857145e-06,
23
+ "loss": 6.022436618804932,
24
+ "step": 4
25
+ },
26
+ {
27
+ "epoch": 0.02586206896551724,
28
+ "grad_norm": 1.3075555562973022,
29
+ "learning_rate": 2.8571428571428573e-06,
30
+ "loss": 1.9596091508865356,
31
+ "step": 6
32
+ },
33
+ {
34
+ "epoch": 0.034482758620689655,
35
+ "grad_norm": 0.4725358784198761,
36
+ "learning_rate": 4.000000000000001e-06,
37
+ "loss": 1.622937560081482,
38
+ "step": 8
39
+ },
40
+ {
41
+ "epoch": 0.04310344827586207,
42
+ "grad_norm": 3.1994080543518066,
43
+ "learning_rate": 5.142857142857142e-06,
44
+ "loss": 2.567897319793701,
45
+ "step": 10
46
+ },
47
+ {
48
+ "epoch": 0.05172413793103448,
49
+ "grad_norm": 0.3738505244255066,
50
+ "learning_rate": 6.285714285714286e-06,
51
+ "loss": 1.8926677703857422,
52
+ "step": 12
53
+ },
54
+ {
55
+ "epoch": 0.0603448275862069,
56
+ "grad_norm": 0.6169027090072632,
57
+ "learning_rate": 7.428571428571429e-06,
58
+ "loss": 1.7960844039916992,
59
+ "step": 14
60
+ },
61
+ {
62
+ "epoch": 0.06896551724137931,
63
+ "grad_norm": 0.7442793250083923,
64
+ "learning_rate": 8.571428571428571e-06,
65
+ "loss": 1.7497979402542114,
66
+ "step": 16
67
+ },
68
+ {
69
+ "epoch": 0.07758620689655173,
70
+ "grad_norm": 3.6816675662994385,
71
+ "learning_rate": 9.714285714285715e-06,
72
+ "loss": 2.5326366424560547,
73
+ "step": 18
74
+ },
75
+ {
76
+ "epoch": 0.08620689655172414,
77
+ "grad_norm": 0.4809480607509613,
78
+ "learning_rate": 1.0857142857142858e-05,
79
+ "loss": 1.7976326942443848,
80
+ "step": 20
81
+ },
82
+ {
83
+ "epoch": 0.09482758620689655,
84
+ "grad_norm": 1.6670607328414917,
85
+ "learning_rate": 1.2e-05,
86
+ "loss": 3.13413667678833,
87
+ "step": 22
88
+ },
89
+ {
90
+ "epoch": 0.10344827586206896,
91
+ "grad_norm": 2.1275734901428223,
92
+ "learning_rate": 1.3142857142857145e-05,
93
+ "loss": 1.9254695177078247,
94
+ "step": 24
95
+ },
96
+ {
97
+ "epoch": 0.11206896551724138,
98
+ "grad_norm": 5.445179462432861,
99
+ "learning_rate": 1.4285714285714287e-05,
100
+ "loss": 2.0263671875,
101
+ "step": 26
102
+ },
103
+ {
104
+ "epoch": 0.1206896551724138,
105
+ "grad_norm": 3.1292130947113037,
106
+ "learning_rate": 1.542857142857143e-05,
107
+ "loss": 2.157352924346924,
108
+ "step": 28
109
+ },
110
+ {
111
+ "epoch": 0.12931034482758622,
112
+ "grad_norm": 0.3002319037914276,
113
+ "learning_rate": 1.6571428571428574e-05,
114
+ "loss": 1.6684068441390991,
115
+ "step": 30
116
+ },
117
+ {
118
+ "epoch": 0.13793103448275862,
119
+ "grad_norm": 0.3610835671424866,
120
+ "learning_rate": 1.7714285714285717e-05,
121
+ "loss": 1.6966811418533325,
122
+ "step": 32
123
+ },
124
+ {
125
+ "epoch": 0.14655172413793102,
126
+ "grad_norm": 2.521209955215454,
127
+ "learning_rate": 1.885714285714286e-05,
128
+ "loss": 1.5750758647918701,
129
+ "step": 34
130
+ },
131
+ {
132
+ "epoch": 0.15517241379310345,
133
+ "grad_norm": 2.5947937965393066,
134
+ "learning_rate": 2e-05,
135
+ "loss": 1.5068325996398926,
136
+ "step": 36
137
+ },
138
+ {
139
+ "epoch": 0.16379310344827586,
140
+ "grad_norm": 0.5535445809364319,
141
+ "learning_rate": 1.9995836718255602e-05,
142
+ "loss": 1.582505702972412,
143
+ "step": 38
144
+ },
145
+ {
146
+ "epoch": 0.1724137931034483,
147
+ "grad_norm": 6.105097770690918,
148
+ "learning_rate": 1.9983350724781264e-05,
149
+ "loss": 1.5262649059295654,
150
+ "step": 40
151
+ },
152
+ {
153
+ "epoch": 0.1810344827586207,
154
+ "grad_norm": 0.7298107743263245,
155
+ "learning_rate": 1.996255357129003e-05,
156
+ "loss": 1.858609914779663,
157
+ "step": 42
158
+ },
159
+ {
160
+ "epoch": 0.1896551724137931,
161
+ "grad_norm": 0.6571905612945557,
162
+ "learning_rate": 1.9933464498761785e-05,
163
+ "loss": 1.520866870880127,
164
+ "step": 44
165
+ },
166
+ {
167
+ "epoch": 0.19827586206896552,
168
+ "grad_norm": 0.5775193572044373,
169
+ "learning_rate": 1.989611041964199e-05,
170
+ "loss": 1.286926507949829,
171
+ "step": 46
172
+ },
173
+ {
174
+ "epoch": 0.20689655172413793,
175
+ "grad_norm": 0.757246196269989,
176
+ "learning_rate": 1.985052589294303e-05,
177
+ "loss": 0.9975602030754089,
178
+ "step": 48
179
+ },
180
+ {
181
+ "epoch": 0.21551724137931033,
182
+ "grad_norm": 0.2879011929035187,
183
+ "learning_rate": 1.9796753092271085e-05,
184
+ "loss": 1.4037001132965088,
185
+ "step": 50
186
+ },
187
+ {
188
+ "epoch": 0.22413793103448276,
189
+ "grad_norm": 0.39832451939582825,
190
+ "learning_rate": 1.973484176680824e-05,
191
+ "loss": 1.5926319360733032,
192
+ "step": 52
193
+ },
194
+ {
195
+ "epoch": 0.23275862068965517,
196
+ "grad_norm": 0.3482792377471924,
197
+ "learning_rate": 1.9664849195285844e-05,
198
+ "loss": 1.0562762022018433,
199
+ "step": 54
200
+ },
201
+ {
202
+ "epoch": 0.2413793103448276,
203
+ "grad_norm": 0.24125751852989197,
204
+ "learning_rate": 1.958684013299173e-05,
205
+ "loss": 0.927044689655304,
206
+ "step": 56
207
+ },
208
+ {
209
+ "epoch": 0.25,
210
+ "grad_norm": 0.5061832070350647,
211
+ "learning_rate": 1.950088675186034e-05,
212
+ "loss": 0.9711346626281738,
213
+ "step": 58
214
+ },
215
+ {
216
+ "epoch": 0.25862068965517243,
217
+ "grad_norm": 0.7697155475616455,
218
+ "learning_rate": 1.9407068573701116e-05,
219
+ "loss": 1.0954338312149048,
220
+ "step": 60
221
+ },
222
+ {
223
+ "epoch": 0.2672413793103448,
224
+ "grad_norm": 0.31880345940589905,
225
+ "learning_rate": 1.9305472396627044e-05,
226
+ "loss": 1.4207227230072021,
227
+ "step": 62
228
+ },
229
+ {
230
+ "epoch": 0.27586206896551724,
231
+ "grad_norm": 0.3055737316608429,
232
+ "learning_rate": 1.919619221475131e-05,
233
+ "loss": 0.9308554530143738,
234
+ "step": 64
235
+ },
236
+ {
237
+ "epoch": 0.28448275862068967,
238
+ "grad_norm": 0.2709406316280365,
239
+ "learning_rate": 1.90793291312264e-05,
240
+ "loss": 1.4012649059295654,
241
+ "step": 66
242
+ },
243
+ {
244
+ "epoch": 0.29310344827586204,
245
+ "grad_norm": 0.3806637227535248,
246
+ "learning_rate": 1.895499126470615e-05,
247
+ "loss": 0.8479315638542175,
248
+ "step": 68
249
+ },
250
+ {
251
+ "epoch": 0.3017241379310345,
252
+ "grad_norm": 0.3242409825325012,
253
+ "learning_rate": 1.8823293649317186e-05,
254
+ "loss": 1.381568193435669,
255
+ "step": 70
256
+ },
257
+ {
258
+ "epoch": 0.3103448275862069,
259
+ "grad_norm": 0.13824906945228577,
260
+ "learning_rate": 1.868435812823238e-05,
261
+ "loss": 0.6094793677330017,
262
+ "step": 72
263
+ },
264
+ {
265
+ "epoch": 0.31896551724137934,
266
+ "grad_norm": 0.27344992756843567,
267
+ "learning_rate": 1.853831324094475e-05,
268
+ "loss": 0.8938958644866943,
269
+ "step": 74
270
+ },
271
+ {
272
+ "epoch": 0.3275862068965517,
273
+ "grad_norm": 0.660277783870697,
274
+ "learning_rate": 1.8385294104346103e-05,
275
+ "loss": 1.2029653787612915,
276
+ "step": 76
277
+ },
278
+ {
279
+ "epoch": 0.33620689655172414,
280
+ "grad_norm": 0.30026018619537354,
281
+ "learning_rate": 1.8225442287720417e-05,
282
+ "loss": 1.3596869707107544,
283
+ "step": 78
284
+ },
285
+ {
286
+ "epoch": 0.3448275862068966,
287
+ "grad_norm": 0.3138439953327179,
288
+ "learning_rate": 1.805890568176769e-05,
289
+ "loss": 1.4085049629211426,
290
+ "step": 80
291
+ },
292
+ {
293
+ "epoch": 0.35344827586206895,
294
+ "grad_norm": 0.5029515624046326,
295
+ "learning_rate": 1.7885838361779333e-05,
296
+ "loss": 1.2587382793426514,
297
+ "step": 82
298
+ },
299
+ {
300
+ "epoch": 0.3620689655172414,
301
+ "grad_norm": 0.3618926405906677,
302
+ "learning_rate": 1.7706400445091755e-05,
303
+ "loss": 1.3189198970794678,
304
+ "step": 84
305
+ },
306
+ {
307
+ "epoch": 0.3706896551724138,
308
+ "grad_norm": 0.7304700016975403,
309
+ "learning_rate": 1.7520757942950026e-05,
310
+ "loss": 1.327932596206665,
311
+ "step": 86
312
+ },
313
+ {
314
+ "epoch": 0.3793103448275862,
315
+ "grad_norm": 0.5044423937797546,
316
+ "learning_rate": 1.7329082606918616e-05,
317
+ "loss": 1.0215771198272705,
318
+ "step": 88
319
+ },
320
+ {
321
+ "epoch": 0.3879310344827586,
322
+ "grad_norm": 0.38377389311790466,
323
+ "learning_rate": 1.7131551769981392e-05,
324
+ "loss": 1.5617601871490479,
325
+ "step": 90
326
+ },
327
+ {
328
+ "epoch": 0.39655172413793105,
329
+ "grad_norm": 0.2912510931491852,
330
+ "learning_rate": 1.6928348182477765e-05,
331
+ "loss": 1.3252155780792236,
332
+ "step": 92
333
+ },
334
+ {
335
+ "epoch": 0.4051724137931034,
336
+ "grad_norm": 0.3194431960582733,
337
+ "learning_rate": 1.67196598430269e-05,
338
+ "loss": 1.335471272468567,
339
+ "step": 94
340
+ },
341
+ {
342
+ "epoch": 0.41379310344827586,
343
+ "grad_norm": 0.3167501986026764,
344
+ "learning_rate": 1.6505679824596338e-05,
345
+ "loss": 1.4329684972763062,
346
+ "step": 96
347
+ },
348
+ {
349
+ "epoch": 0.4224137931034483,
350
+ "grad_norm": 0.27103397250175476,
351
+ "learning_rate": 1.6286606095875938e-05,
352
+ "loss": 1.302343487739563,
353
+ "step": 98
354
+ },
355
+ {
356
+ "epoch": 0.43103448275862066,
357
+ "grad_norm": 0.3172292113304138,
358
+ "learning_rate": 1.6062641338122474e-05,
359
+ "loss": 1.3569573163986206,
360
+ "step": 100
361
+ },
362
+ {
363
+ "epoch": 0.4396551724137931,
364
+ "grad_norm": 0.2844351530075073,
365
+ "learning_rate": 1.583399275764424e-05,
366
+ "loss": 1.309763789176941,
367
+ "step": 102
368
+ },
369
+ {
370
+ "epoch": 0.4482758620689655,
371
+ "grad_norm": 0.4061580002307892,
372
+ "learning_rate": 1.5600871894099243e-05,
373
+ "loss": 0.9164461493492126,
374
+ "step": 104
375
+ },
376
+ {
377
+ "epoch": 0.45689655172413796,
378
+ "grad_norm": 1.267133355140686,
379
+ "learning_rate": 1.5363494424784238e-05,
380
+ "loss": 0.8275090456008911,
381
+ "step": 106
382
+ },
383
+ {
384
+ "epoch": 0.46551724137931033,
385
+ "grad_norm": 0.3342474102973938,
386
+ "learning_rate": 1.5122079965095786e-05,
387
+ "loss": 0.7831904292106628,
388
+ "step": 108
389
+ },
390
+ {
391
+ "epoch": 0.47413793103448276,
392
+ "grad_norm": 0.33141425251960754,
393
+ "learning_rate": 1.4876851865347857e-05,
394
+ "loss": 1.0925136804580688,
395
+ "step": 110
396
+ },
397
+ {
398
+ "epoch": 0.4827586206896552,
399
+ "grad_norm": 0.296645849943161,
400
+ "learning_rate": 1.4628037004133986e-05,
401
+ "loss": 0.9729475378990173,
402
+ "step": 112
403
+ },
404
+ {
405
+ "epoch": 0.49137931034482757,
406
+ "grad_norm": 0.7777587175369263,
407
+ "learning_rate": 1.4375865578425157e-05,
408
+ "loss": 0.6473991274833679,
409
+ "step": 114
410
+ },
411
+ {
412
+ "epoch": 0.5,
413
+ "grad_norm": 0.38521984219551086,
414
+ "learning_rate": 1.4120570890597611e-05,
415
+ "loss": 1.1790492534637451,
416
+ "step": 116
417
+ },
418
+ {
419
+ "epoch": 0.5086206896551724,
420
+ "grad_norm": 0.4623865485191345,
421
+ "learning_rate": 1.3862389132587639e-05,
422
+ "loss": 1.2727322578430176,
423
+ "step": 118
424
+ },
425
+ {
426
+ "epoch": 0.5172413793103449,
427
+ "grad_norm": 0.5431216955184937,
428
+ "learning_rate": 1.3601559167372983e-05,
429
+ "loss": 0.7860153913497925,
430
+ "step": 120
431
+ },
432
+ {
433
+ "epoch": 0.5258620689655172,
434
+ "grad_norm": 0.5915956497192383,
435
+ "learning_rate": 1.3338322307983105e-05,
436
+ "loss": 1.043981671333313,
437
+ "step": 122
438
+ },
439
+ {
440
+ "epoch": 0.5344827586206896,
441
+ "grad_norm": 0.24551472067832947,
442
+ "learning_rate": 1.3072922094242703e-05,
443
+ "loss": 1.278051733970642,
444
+ "step": 124
445
+ },
446
+ {
447
+ "epoch": 0.5431034482758621,
448
+ "grad_norm": 0.2816060781478882,
449
+ "learning_rate": 1.2805604067455076e-05,
450
+ "loss": 1.255641222000122,
451
+ "step": 126
452
+ },
453
+ {
454
+ "epoch": 0.5517241379310345,
455
+ "grad_norm": 0.4124661684036255,
456
+ "learning_rate": 1.2536615543233748e-05,
457
+ "loss": 1.255129337310791,
458
+ "step": 128
459
+ },
460
+ {
461
+ "epoch": 0.5603448275862069,
462
+ "grad_norm": 0.5298042893409729,
463
+ "learning_rate": 1.2266205382692574e-05,
464
+ "loss": 1.5166298151016235,
465
+ "step": 130
466
+ },
467
+ {
468
+ "epoch": 0.5689655172413793,
469
+ "grad_norm": 0.2111591249704361,
470
+ "learning_rate": 1.199462376220597e-05,
471
+ "loss": 0.8270851373672485,
472
+ "step": 132
473
+ },
474
+ {
475
+ "epoch": 0.5775862068965517,
476
+ "grad_norm": 0.889839768409729,
477
+ "learning_rate": 1.1722121941952313e-05,
478
+ "loss": 0.6701672077178955,
479
+ "step": 134
480
+ },
481
+ {
482
+ "epoch": 0.5862068965517241,
483
+ "grad_norm": 0.2590690553188324,
484
+ "learning_rate": 1.144895203345462e-05,
485
+ "loss": 1.077882170677185,
486
+ "step": 136
487
+ },
488
+ {
489
+ "epoch": 0.5948275862068966,
490
+ "grad_norm": 0.298973947763443,
491
+ "learning_rate": 1.11753667663336e-05,
492
+ "loss": 0.8770455718040466,
493
+ "step": 138
494
+ },
495
+ {
496
+ "epoch": 0.603448275862069,
497
+ "grad_norm": 0.6260126233100891,
498
+ "learning_rate": 1.0901619254488831e-05,
499
+ "loss": 0.9084359407424927,
500
+ "step": 140
501
+ },
502
+ {
503
+ "epoch": 0.6120689655172413,
504
+ "grad_norm": 0.2761947512626648,
505
+ "learning_rate": 1.0627962761924462e-05,
506
+ "loss": 1.370908260345459,
507
+ "step": 142
508
+ },
509
+ {
510
+ "epoch": 0.6206896551724138,
511
+ "grad_norm": 0.3692832291126251,
512
+ "learning_rate": 1.0354650468435985e-05,
513
+ "loss": 0.7351585030555725,
514
+ "step": 144
515
+ },
516
+ {
517
+ "epoch": 0.6293103448275862,
518
+ "grad_norm": 0.4929635524749756,
519
+ "learning_rate": 1.0081935235374955e-05,
520
+ "loss": 1.2708691358566284,
521
+ "step": 146
522
+ },
523
+ {
524
+ "epoch": 0.6379310344827587,
525
+ "grad_norm": 0.3487298786640167,
526
+ "learning_rate": 9.810069371708313e-06,
527
+ "loss": 1.2674932479858398,
528
+ "step": 148
529
+ },
530
+ {
531
+ "epoch": 0.646551724137931,
532
+ "grad_norm": 0.36564555764198303,
533
+ "learning_rate": 9.539304400588752e-06,
534
+ "loss": 1.0046426057815552,
535
+ "step": 150
536
+ },
537
+ {
538
+ "epoch": 0.6551724137931034,
539
+ "grad_norm": 0.2302585393190384,
540
+ "learning_rate": 9.269890826652113e-06,
541
+ "loss": 0.664213240146637,
542
+ "step": 152
543
+ },
544
+ {
545
+ "epoch": 0.6637931034482759,
546
+ "grad_norm": 0.3203712999820709,
547
+ "learning_rate": 9.00207790425708e-06,
548
+ "loss": 1.5197818279266357,
549
+ "step": 154
550
+ },
551
+ {
552
+ "epoch": 0.6724137931034483,
553
+ "grad_norm": 0.6393405199050903,
554
+ "learning_rate": 8.73611340688159e-06,
555
+ "loss": 1.062941312789917,
556
+ "step": 156
557
+ },
558
+ {
559
+ "epoch": 0.6810344827586207,
560
+ "grad_norm": 0.33355581760406494,
561
+ "learning_rate": 8.472243397889327e-06,
562
+ "loss": 0.7909774780273438,
563
+ "step": 158
564
+ },
565
+ {
566
+ "epoch": 0.6896551724137931,
567
+ "grad_norm": 0.34659793972969055,
568
+ "learning_rate": 8.210712002878366e-06,
569
+ "loss": 0.3939180374145508,
570
+ "step": 160
571
+ },
572
+ {
573
+ "epoch": 0.6982758620689655,
574
+ "grad_norm": 0.995121955871582,
575
+ "learning_rate": 7.951761183822578e-06,
576
+ "loss": 1.134759545326233,
577
+ "step": 162
578
+ },
579
+ {
580
+ "epoch": 0.7068965517241379,
581
+ "grad_norm": 1.618648886680603,
582
+ "learning_rate": 7.695630515214783e-06,
583
+ "loss": 0.9670589566230774,
584
+ "step": 164
585
+ },
586
+ {
587
+ "epoch": 0.7155172413793104,
588
+ "grad_norm": 0.22626011073589325,
589
+ "learning_rate": 7.442556962418712e-06,
590
+ "loss": 1.3202883005142212,
591
+ "step": 166
592
+ },
593
+ {
594
+ "epoch": 0.7241379310344828,
595
+ "grad_norm": 0.32568663358688354,
596
+ "learning_rate": 7.192774662434887e-06,
597
+ "loss": 0.7861534357070923,
598
+ "step": 168
599
+ },
600
+ {
601
+ "epoch": 0.7327586206896551,
602
+ "grad_norm": 1.0204657316207886,
603
+ "learning_rate": 6.94651470728322e-06,
604
+ "loss": 0.8559544682502747,
605
+ "step": 170
606
+ },
607
+ {
608
+ "epoch": 0.7413793103448276,
609
+ "grad_norm": 0.46275442838668823,
610
+ "learning_rate": 6.7040049302027445e-06,
611
+ "loss": 1.349075198173523,
612
+ "step": 172
613
+ },
614
+ {
615
+ "epoch": 0.75,
616
+ "grad_norm": 0.3042738139629364,
617
+ "learning_rate": 6.465469694866295e-06,
618
+ "loss": 1.2694404125213623,
619
+ "step": 174
620
+ },
621
+ {
622
+ "epoch": 0.7586206896551724,
623
+ "grad_norm": 0.24299518764019012,
624
+ "learning_rate": 6.2311296878051355e-06,
625
+ "loss": 1.2068278789520264,
626
+ "step": 176
627
+ },
628
+ {
629
+ "epoch": 0.7672413793103449,
630
+ "grad_norm": 0.3234924376010895,
631
+ "learning_rate": 6.001201714235567e-06,
632
+ "loss": 1.311326503753662,
633
+ "step": 178
634
+ },
635
+ {
636
+ "epoch": 0.7758620689655172,
637
+ "grad_norm": 0.7479709982872009,
638
+ "learning_rate": 5.7758984974764586e-06,
639
+ "loss": 0.768395185470581,
640
+ "step": 180
641
+ },
642
+ {
643
+ "epoch": 0.7844827586206896,
644
+ "grad_norm": 0.35784727334976196,
645
+ "learning_rate": 5.555428482143206e-06,
646
+ "loss": 1.1966248750686646,
647
+ "step": 182
648
+ },
649
+ {
650
+ "epoch": 0.7931034482758621,
651
+ "grad_norm": 0.4064531922340393,
652
+ "learning_rate": 5.339995641300257e-06,
653
+ "loss": 1.1476389169692993,
654
+ "step": 184
655
+ },
656
+ {
657
+ "epoch": 0.8017241379310345,
658
+ "grad_norm": 0.31240594387054443,
659
+ "learning_rate": 5.129799287750601e-06,
660
+ "loss": 0.9965859651565552,
661
+ "step": 186
662
+ },
663
+ {
664
+ "epoch": 0.8103448275862069,
665
+ "grad_norm": 0.3878633975982666,
666
+ "learning_rate": 4.9250338896367814e-06,
667
+ "loss": 0.8189845085144043,
668
+ "step": 188
669
+ },
670
+ {
671
+ "epoch": 0.8189655172413793,
672
+ "grad_norm": 0.29073813557624817,
673
+ "learning_rate": 4.725888890524101e-06,
674
+ "loss": 1.2450443506240845,
675
+ "step": 190
676
+ },
677
+ {
678
+ "epoch": 0.8275862068965517,
679
+ "grad_norm": 1.2100752592086792,
680
+ "learning_rate": 4.5325485341324026e-06,
681
+ "loss": 0.9530048966407776,
682
+ "step": 192
683
+ },
684
+ {
685
+ "epoch": 0.8362068965517241,
686
+ "grad_norm": 0.4809584319591522,
687
+ "learning_rate": 4.345191693878623e-06,
688
+ "loss": 1.2201794385910034,
689
+ "step": 194
690
+ },
691
+ {
692
+ "epoch": 0.8448275862068966,
693
+ "grad_norm": 0.9147862792015076,
694
+ "learning_rate": 4.163991707387815e-06,
695
+ "loss": 0.8178768754005432,
696
+ "step": 196
697
+ },
698
+ {
699
+ "epoch": 0.853448275862069,
700
+ "grad_norm": 0.27828192710876465,
701
+ "learning_rate": 3.989116216125717e-06,
702
+ "loss": 1.228522539138794,
703
+ "step": 198
704
+ },
705
+ {
706
+ "epoch": 0.8620689655172413,
707
+ "grad_norm": 0.7031840085983276,
708
+ "learning_rate": 3.8207270103012905e-06,
709
+ "loss": 1.3021351099014282,
710
+ "step": 200
711
+ },
712
+ {
713
+ "epoch": 0.8706896551724138,
714
+ "grad_norm": 0.6149956583976746,
715
+ "learning_rate": 3.6589798791826593e-06,
716
+ "loss": 0.6399316787719727,
717
+ "step": 202
718
+ },
719
+ {
720
+ "epoch": 0.8793103448275862,
721
+ "grad_norm": 0.2772592604160309,
722
+ "learning_rate": 3.5040244669649804e-06,
723
+ "loss": 1.600140929222107,
724
+ "step": 204
725
+ },
726
+ {
727
+ "epoch": 0.8879310344827587,
728
+ "grad_norm": 0.1492106020450592,
729
+ "learning_rate": 3.3560041343235544e-06,
730
+ "loss": 1.0438815355300903,
731
+ "step": 206
732
+ },
733
+ {
734
+ "epoch": 0.896551724137931,
735
+ "grad_norm": 0.3777860701084137,
736
+ "learning_rate": 3.2150558257803178e-06,
737
+ "loss": 1.339942216873169,
738
+ "step": 208
739
+ },
740
+ {
741
+ "epoch": 0.9051724137931034,
742
+ "grad_norm": 0.6718384027481079,
743
+ "learning_rate": 3.081309943006349e-06,
744
+ "loss": 0.9232547879219055,
745
+ "step": 210
746
+ },
747
+ {
748
+ "epoch": 0.9137931034482759,
749
+ "grad_norm": 0.9841593503952026,
750
+ "learning_rate": 2.9548902241776783e-06,
751
+ "loss": 1.1493092775344849,
752
+ "step": 212
753
+ },
754
+ {
755
+ "epoch": 0.9224137931034483,
756
+ "grad_norm": 0.8034397959709167,
757
+ "learning_rate": 2.8359136294959814e-06,
758
+ "loss": 1.5147000551223755,
759
+ "step": 214
760
+ },
761
+ {
762
+ "epoch": 0.9310344827586207,
763
+ "grad_norm": 1.0949209928512573,
764
+ "learning_rate": 2.7244902329800697e-06,
765
+ "loss": 1.3095285892486572,
766
+ "step": 216
767
+ },
768
+ {
769
+ "epoch": 0.9396551724137931,
770
+ "grad_norm": 0.22381076216697693,
771
+ "learning_rate": 2.620723120628298e-06,
772
+ "loss": 1.230783462524414,
773
+ "step": 218
774
+ },
775
+ {
776
+ "epoch": 0.9482758620689655,
777
+ "grad_norm": 0.3169087767601013,
778
+ "learning_rate": 2.5247082950461204e-06,
779
+ "loss": 1.3510628938674927,
780
+ "step": 220
781
+ },
782
+ {
783
+ "epoch": 0.9568965517241379,
784
+ "grad_norm": 0.23972630500793457,
785
+ "learning_rate": 2.4365345866269884e-06,
786
+ "loss": 1.1361650228500366,
787
+ "step": 222
788
+ },
789
+ {
790
+ "epoch": 0.9655172413793104,
791
+ "grad_norm": 0.32792967557907104,
792
+ "learning_rate": 2.35628357136881e-06,
793
+ "loss": 1.2898920774459839,
794
+ "step": 224
795
+ },
796
+ {
797
+ "epoch": 0.9741379310344828,
798
+ "grad_norm": 0.4268532693386078,
799
+ "learning_rate": 2.284029495401986e-06,
800
+ "loss": 1.3678969144821167,
801
+ "step": 226
802
+ },
803
+ {
804
+ "epoch": 0.9827586206896551,
805
+ "grad_norm": 0.7636785507202148,
806
+ "learning_rate": 2.219839206298835e-06,
807
+ "loss": 0.8566112518310547,
808
+ "step": 228
809
+ },
810
+ {
811
+ "epoch": 0.9913793103448276,
812
+ "grad_norm": 0.9001679420471191,
813
+ "learning_rate": 2.163772091227977e-06,
814
+ "loss": 0.8546447157859802,
815
+ "step": 230
816
+ },
817
+ {
818
+ "epoch": 1.0,
819
+ "grad_norm": 0.9462145566940308,
820
+ "learning_rate": 2.115880022010884e-06,
821
+ "loss": 1.1330009698867798,
822
+ "step": 232
823
+ },
824
+ {
825
+ "epoch": 1.0086206896551724,
826
+ "grad_norm": 0.3172188103199005,
827
+ "learning_rate": 2.076207307131444e-06,
828
+ "loss": 1.1593165397644043,
829
+ "step": 234
830
+ },
831
+ {
832
+ "epoch": 1.0172413793103448,
833
+ "grad_norm": 0.3420685827732086,
834
+ "learning_rate": 2.0447906507428975e-06,
835
+ "loss": 0.6197730898857117,
836
+ "step": 236
837
+ },
838
+ {
839
+ "epoch": 1.0258620689655173,
840
+ "grad_norm": 0.32782989740371704,
841
+ "learning_rate": 2.021659118710141e-06,
842
+ "loss": 1.2394027709960938,
843
+ "step": 238
844
+ },
845
+ {
846
+ "epoch": 1.0344827586206897,
847
+ "grad_norm": 0.2151000201702118,
848
+ "learning_rate": 2.0068341117187347e-06,
849
+ "loss": 1.2157349586486816,
850
+ "step": 240
851
+ },
852
+ {
853
+ "epoch": 1.043103448275862,
854
+ "grad_norm": 0.23503771424293518,
855
+ "learning_rate": 2.00032934547556e-06,
856
+ "loss": 1.2545582056045532,
857
+ "step": 242
858
+ },
859
+ {
860
+ "epoch": 1.0517241379310345,
861
+ "grad_norm": 0.33568423986434937,
862
+ "learning_rate": 2.002150838019407e-06,
863
+ "loss": 1.173262596130371,
864
+ "step": 244
865
+ },
866
+ {
867
+ "epoch": 1.0603448275862069,
868
+ "grad_norm": 0.23418180644512177,
869
+ "learning_rate": 2.012296904153243e-06,
870
+ "loss": 1.1772663593292236,
871
+ "step": 246
872
+ },
873
+ {
874
+ "epoch": 1.0689655172413792,
875
+ "grad_norm": 0.2445734143257141,
876
+ "learning_rate": 2.030758157003306e-06,
877
+ "loss": 1.1834620237350464,
878
+ "step": 248
879
+ },
880
+ {
881
+ "epoch": 1.0775862068965518,
882
+ "grad_norm": 1.2816580533981323,
883
+ "learning_rate": 2.057517516703605e-06,
884
+ "loss": 0.8118759393692017,
885
+ "step": 250
886
+ },
887
+ {
888
+ "epoch": 1.0862068965517242,
889
+ "grad_norm": 0.30513474345207214,
890
+ "learning_rate": 2.0925502261977544e-06,
891
+ "loss": 0.7529575824737549,
892
+ "step": 252
893
+ },
894
+ {
895
+ "epoch": 1.0948275862068966,
896
+ "grad_norm": 0.35363683104515076,
897
+ "learning_rate": 2.1358238741435576e-06,
898
+ "loss": 0.7137669324874878,
899
+ "step": 254
900
+ },
901
+ {
902
+ "epoch": 1.103448275862069,
903
+ "grad_norm": 0.20758303999900818,
904
+ "learning_rate": 2.1872984248991204e-06,
905
+ "loss": 1.1082866191864014,
906
+ "step": 256
907
+ },
908
+ {
909
+ "epoch": 1.1120689655172413,
910
+ "grad_norm": 0.44538792967796326,
911
+ "learning_rate": 2.246926255562788e-06,
912
+ "loss": 1.020033359527588,
913
+ "step": 258
914
+ },
915
+ {
916
+ "epoch": 1.1206896551724137,
917
+ "grad_norm": 0.32762372493743896,
918
+ "learning_rate": 2.3146522000325895e-06,
919
+ "loss": 0.4955134391784668,
920
+ "step": 260
921
+ },
922
+ {
923
+ "epoch": 1.1293103448275863,
924
+ "grad_norm": 0.4477788507938385,
925
+ "learning_rate": 2.3904136000444754e-06,
926
+ "loss": 1.212654709815979,
927
+ "step": 262
928
+ },
929
+ {
930
+ "epoch": 1.1379310344827587,
931
+ "grad_norm": 0.2977157235145569,
932
+ "learning_rate": 2.4741403631420874e-06,
933
+ "loss": 1.2140189409255981,
934
+ "step": 264
935
+ },
936
+ {
937
+ "epoch": 1.146551724137931,
938
+ "grad_norm": 0.49766677618026733,
939
+ "learning_rate": 2.5657550275244664e-06,
940
+ "loss": 1.3088136911392212,
941
+ "step": 266
942
+ },
943
+ {
944
+ "epoch": 1.1551724137931034,
945
+ "grad_norm": 1.1656190156936646,
946
+ "learning_rate": 2.66517283371167e-06,
947
+ "loss": 0.9128588438034058,
948
+ "step": 268
949
+ },
950
+ {
951
+ "epoch": 1.1637931034482758,
952
+ "grad_norm": 0.25955432653427124,
953
+ "learning_rate": 2.7723018029620165e-06,
954
+ "loss": 1.2309114933013916,
955
+ "step": 270
956
+ },
957
+ {
958
+ "epoch": 1.1724137931034484,
959
+ "grad_norm": 1.4371591806411743,
960
+ "learning_rate": 2.8870428223683972e-06,
961
+ "loss": 0.8446346521377563,
962
+ "step": 272
963
+ },
964
+ {
965
+ "epoch": 1.1810344827586208,
966
+ "grad_norm": 0.7202423810958862,
967
+ "learning_rate": 3.009289736554944e-06,
968
+ "loss": 1.3168221712112427,
969
+ "step": 274
970
+ },
971
+ {
972
+ "epoch": 1.1896551724137931,
973
+ "grad_norm": 0.2864224314689636,
974
+ "learning_rate": 3.1389294458891755e-06,
975
+ "loss": 0.7556365132331848,
976
+ "step": 276
977
+ },
978
+ {
979
+ "epoch": 1.1982758620689655,
980
+ "grad_norm": 0.3141017556190491,
981
+ "learning_rate": 3.2758420111188274e-06,
982
+ "loss": 1.229781985282898,
983
+ "step": 278
984
+ },
985
+ {
986
+ "epoch": 1.206896551724138,
987
+ "grad_norm": 0.650051474571228,
988
+ "learning_rate": 3.419900764336468e-06,
989
+ "loss": 0.888131320476532,
990
+ "step": 280
991
+ },
992
+ {
993
+ "epoch": 1.2155172413793103,
994
+ "grad_norm": 0.5339681506156921,
995
+ "learning_rate": 3.5709724261693514e-06,
996
+ "loss": 1.2144626379013062,
997
+ "step": 282
998
+ },
999
+ {
1000
+ "epoch": 1.2241379310344827,
1001
+ "grad_norm": 0.27003324031829834,
1002
+ "learning_rate": 3.7289172290859634e-06,
1003
+ "loss": 0.8701605200767517,
1004
+ "step": 284
1005
+ },
1006
+ {
1007
+ "epoch": 1.2327586206896552,
1008
+ "grad_norm": 0.10487275570631027,
1009
+ "learning_rate": 3.893589046705279e-06,
1010
+ "loss": 0.5924324989318848,
1011
+ "step": 286
1012
+ },
1013
+ {
1014
+ "epoch": 1.2413793103448276,
1015
+ "grad_norm": 0.5368720293045044,
1016
+ "learning_rate": 4.064835528989056e-06,
1017
+ "loss": 1.1794689893722534,
1018
+ "step": 288
1019
+ },
1020
+ {
1021
+ "epoch": 1.25,
1022
+ "grad_norm": 0.47765061259269714,
1023
+ "learning_rate": 4.242498243192066e-06,
1024
+ "loss": 0.7260119915008545,
1025
+ "step": 290
1026
+ },
1027
+ {
1028
+ "epoch": 1.2586206896551724,
1029
+ "grad_norm": 0.3096713423728943,
1030
+ "learning_rate": 4.426412820439941e-06,
1031
+ "loss": 1.0264055728912354,
1032
+ "step": 292
1033
+ },
1034
+ {
1035
+ "epoch": 1.2672413793103448,
1036
+ "grad_norm": 0.4803042411804199,
1037
+ "learning_rate": 4.616409107798898e-06,
1038
+ "loss": 1.4605242013931274,
1039
+ "step": 294
1040
+ },
1041
+ {
1042
+ "epoch": 1.2758620689655173,
1043
+ "grad_norm": 0.356592059135437,
1044
+ "learning_rate": 4.812311325696798e-06,
1045
+ "loss": 0.7666136622428894,
1046
+ "step": 296
1047
+ },
1048
+ {
1049
+ "epoch": 1.2844827586206897,
1050
+ "grad_norm": 0.5286064743995667,
1051
+ "learning_rate": 5.013938230549751e-06,
1052
+ "loss": 1.0257295370101929,
1053
+ "step": 298
1054
+ },
1055
+ {
1056
+ "epoch": 1.293103448275862,
1057
+ "grad_norm": 0.8442373275756836,
1058
+ "learning_rate": 5.221103282443953e-06,
1059
+ "loss": 1.0583548545837402,
1060
+ "step": 300
1061
+ },
1062
+ {
1063
+ "epoch": 1.3017241379310345,
1064
+ "grad_norm": 0.32554081082344055,
1065
+ "learning_rate": 5.433614817717478e-06,
1066
+ "loss": 0.9327218532562256,
1067
+ "step": 302
1068
+ },
1069
+ {
1070
+ "epoch": 1.3103448275862069,
1071
+ "grad_norm": 0.37990084290504456,
1072
+ "learning_rate": 5.6512762262824986e-06,
1073
+ "loss": 1.2481307983398438,
1074
+ "step": 304
1075
+ },
1076
+ {
1077
+ "epoch": 1.3189655172413794,
1078
+ "grad_norm": 0.3042522072792053,
1079
+ "learning_rate": 5.873886133523725e-06,
1080
+ "loss": 1.1840277910232544,
1081
+ "step": 306
1082
+ },
1083
+ {
1084
+ "epoch": 1.3275862068965516,
1085
+ "grad_norm": 0.6558886170387268,
1086
+ "learning_rate": 6.101238586604963e-06,
1087
+ "loss": 0.6794692277908325,
1088
+ "step": 308
1089
+ },
1090
+ {
1091
+ "epoch": 1.3362068965517242,
1092
+ "grad_norm": 0.18041777610778809,
1093
+ "learning_rate": 6.333123245011208e-06,
1094
+ "loss": 0.9396043419837952,
1095
+ "step": 310
1096
+ },
1097
+ {
1098
+ "epoch": 1.3448275862068966,
1099
+ "grad_norm": 0.5412018895149231,
1100
+ "learning_rate": 6.569325575150216e-06,
1101
+ "loss": 1.292694091796875,
1102
+ "step": 312
1103
+ },
1104
+ {
1105
+ "epoch": 1.353448275862069,
1106
+ "grad_norm": 0.48298704624176025,
1107
+ "learning_rate": 6.809627048833306e-06,
1108
+ "loss": 1.1025623083114624,
1109
+ "step": 314
1110
+ },
1111
+ {
1112
+ "epoch": 1.3620689655172413,
1113
+ "grad_norm": 0.40131139755249023,
1114
+ "learning_rate": 7.053805345451903e-06,
1115
+ "loss": 0.7764808535575867,
1116
+ "step": 316
1117
+ },
1118
+ {
1119
+ "epoch": 1.3706896551724137,
1120
+ "grad_norm": 0.5116880536079407,
1121
+ "learning_rate": 7.3016345576627605e-06,
1122
+ "loss": 1.2299896478652954,
1123
+ "step": 318
1124
+ },
1125
+ {
1126
+ "epoch": 1.3793103448275863,
1127
+ "grad_norm": 0.11877727508544922,
1128
+ "learning_rate": 7.552885400391432e-06,
1129
+ "loss": 0.8323980569839478,
1130
+ "step": 320
1131
+ },
1132
+ {
1133
+ "epoch": 1.3879310344827587,
1134
+ "grad_norm": 1.5366413593292236,
1135
+ "learning_rate": 7.80732542296087e-06,
1136
+ "loss": 1.1044297218322754,
1137
+ "step": 322
1138
+ },
1139
+ {
1140
+ "epoch": 1.396551724137931,
1141
+ "grad_norm": 0.4911424517631531,
1142
+ "learning_rate": 8.064719224148607e-06,
1143
+ "loss": 1.518593430519104,
1144
+ "step": 324
1145
+ },
1146
+ {
1147
+ "epoch": 1.4051724137931034,
1148
+ "grad_norm": 1.790497899055481,
1149
+ "learning_rate": 8.324828669973863e-06,
1150
+ "loss": 1.134735345840454,
1151
+ "step": 326
1152
+ },
1153
+ {
1154
+ "epoch": 1.4137931034482758,
1155
+ "grad_norm": 0.5793009996414185,
1156
+ "learning_rate": 8.587413114012767e-06,
1157
+ "loss": 0.9491145610809326,
1158
+ "step": 328
1159
+ },
1160
+ {
1161
+ "epoch": 1.4224137931034484,
1162
+ "grad_norm": 0.4418241083621979,
1163
+ "learning_rate": 8.852229620038169e-06,
1164
+ "loss": 0.7093353271484375,
1165
+ "step": 330
1166
+ },
1167
+ {
1168
+ "epoch": 1.4310344827586206,
1169
+ "grad_norm": 0.29193559288978577,
1170
+ "learning_rate": 9.119033186777797e-06,
1171
+ "loss": 1.178290605545044,
1172
+ "step": 332
1173
+ },
1174
+ {
1175
+ "epoch": 1.4396551724137931,
1176
+ "grad_norm": 0.7738173007965088,
1177
+ "learning_rate": 9.387576974583059e-06,
1178
+ "loss": 1.2559309005737305,
1179
+ "step": 334
1180
+ },
1181
+ {
1182
+ "epoch": 1.4482758620689655,
1183
+ "grad_norm": 0.4955233931541443,
1184
+ "learning_rate": 9.657612533798527e-06,
1185
+ "loss": 1.1395691633224487,
1186
+ "step": 336
1187
+ },
1188
+ {
1189
+ "epoch": 1.456896551724138,
1190
+ "grad_norm": 0.33920466899871826,
1191
+ "learning_rate": 9.928890034621103e-06,
1192
+ "loss": 0.824805498123169,
1193
+ "step": 338
1194
+ },
1195
+ {
1196
+ "epoch": 1.4655172413793103,
1197
+ "grad_norm": 0.43428322672843933,
1198
+ "learning_rate": 1.020115849823592e-05,
1199
+ "loss": 1.2302451133728027,
1200
+ "step": 340
1201
+ },
1202
+ {
1203
+ "epoch": 1.4741379310344827,
1204
+ "grad_norm": 0.2865593731403351,
1205
+ "learning_rate": 1.04741660290154e-05,
1206
+ "loss": 1.2292989492416382,
1207
+ "step": 342
1208
+ },
1209
+ {
1210
+ "epoch": 1.4827586206896552,
1211
+ "grad_norm": 0.32383060455322266,
1212
+ "learning_rate": 1.0747660047566425e-05,
1213
+ "loss": 0.9929612278938293,
1214
+ "step": 344
1215
+ },
1216
+ {
1217
+ "epoch": 1.4913793103448276,
1218
+ "grad_norm": 0.33553263545036316,
1219
+ "learning_rate": 1.1021387524410204e-05,
1220
+ "loss": 0.8662385940551758,
1221
+ "step": 346
1222
+ },
1223
+ {
1224
+ "epoch": 1.5,
1225
+ "grad_norm": 0.5123760104179382,
1226
+ "learning_rate": 1.129509521407843e-05,
1227
+ "loss": 0.1633024513721466,
1228
+ "step": 348
1229
+ },
1230
+ {
1231
+ "epoch": 1.5086206896551724,
1232
+ "grad_norm": 0.23127630352973938,
1233
+ "learning_rate": 1.1568529889409417e-05,
1234
+ "loss": 0.7720933556556702,
1235
+ "step": 350
1236
+ },
1237
+ {
1238
+ "epoch": 1.5172413793103448,
1239
+ "grad_norm": 0.4621216654777527,
1240
+ "learning_rate": 1.1841438575827132e-05,
1241
+ "loss": 0.43289288878440857,
1242
+ "step": 352
1243
+ },
1244
+ {
1245
+ "epoch": 1.5258620689655173,
1246
+ "grad_norm": 0.24998484551906586,
1247
+ "learning_rate": 1.2113568785386663e-05,
1248
+ "loss": 0.7231047749519348,
1249
+ "step": 354
1250
+ },
1251
+ {
1252
+ "epoch": 1.5344827586206895,
1253
+ "grad_norm": 1.216570496559143,
1254
+ "learning_rate": 1.2384668750369462e-05,
1255
+ "loss": 0.5929489731788635,
1256
+ "step": 356
1257
+ },
1258
+ {
1259
+ "epoch": 1.543103448275862,
1260
+ "grad_norm": 0.45301932096481323,
1261
+ "learning_rate": 1.2654487656212154e-05,
1262
+ "loss": 1.3106164932250977,
1263
+ "step": 358
1264
+ },
1265
+ {
1266
+ "epoch": 1.5517241379310345,
1267
+ "grad_norm": 0.26960089802742004,
1268
+ "learning_rate": 1.2922775873553643e-05,
1269
+ "loss": 0.6820691823959351,
1270
+ "step": 360
1271
+ },
1272
+ {
1273
+ "epoch": 1.5603448275862069,
1274
+ "grad_norm": 0.5970767736434937,
1275
+ "learning_rate": 1.3189285189185581e-05,
1276
+ "loss": 1.2352055311203003,
1277
+ "step": 362
1278
+ },
1279
+ {
1280
+ "epoch": 1.5689655172413794,
1281
+ "grad_norm": 0.5552269816398621,
1282
+ "learning_rate": 1.3453769035692767e-05,
1283
+ "loss": 1.2412972450256348,
1284
+ "step": 364
1285
+ },
1286
+ {
1287
+ "epoch": 1.5775862068965516,
1288
+ "grad_norm": 0.373104453086853,
1289
+ "learning_rate": 1.3715982719570773e-05,
1290
+ "loss": 1.192505955696106,
1291
+ "step": 366
1292
+ },
1293
+ {
1294
+ "epoch": 1.5862068965517242,
1295
+ "grad_norm": 0.36400294303894043,
1296
+ "learning_rate": 1.397568364761001e-05,
1297
+ "loss": 1.2814549207687378,
1298
+ "step": 368
1299
+ },
1300
+ {
1301
+ "epoch": 1.5948275862068966,
1302
+ "grad_norm": 0.39620983600616455,
1303
+ "learning_rate": 1.4232631551336535e-05,
1304
+ "loss": 1.2581911087036133,
1305
+ "step": 370
1306
+ },
1307
+ {
1308
+ "epoch": 1.603448275862069,
1309
+ "grad_norm": 1.385265588760376,
1310
+ "learning_rate": 1.4486588709302146e-05,
1311
+ "loss": 1.0574073791503906,
1312
+ "step": 372
1313
+ },
1314
+ {
1315
+ "epoch": 1.6120689655172413,
1316
+ "grad_norm": 0.7239859104156494,
1317
+ "learning_rate": 1.4737320167018063e-05,
1318
+ "loss": 0.6223809123039246,
1319
+ "step": 374
1320
+ },
1321
+ {
1322
+ "epoch": 1.6206896551724137,
1323
+ "grad_norm": 2.1304306983947754,
1324
+ "learning_rate": 1.4984593954328579e-05,
1325
+ "loss": 0.8683441281318665,
1326
+ "step": 376
1327
+ },
1328
+ {
1329
+ "epoch": 1.6293103448275863,
1330
+ "grad_norm": 0.36936265230178833,
1331
+ "learning_rate": 1.5228181300023775e-05,
1332
+ "loss": 1.1772154569625854,
1333
+ "step": 378
1334
+ },
1335
+ {
1336
+ "epoch": 1.6379310344827587,
1337
+ "grad_norm": 0.45922592282295227,
1338
+ "learning_rate": 1.5467856843492662e-05,
1339
+ "loss": 1.0002669095993042,
1340
+ "step": 380
1341
+ },
1342
+ {
1343
+ "epoch": 1.646551724137931,
1344
+ "grad_norm": 1.180850625038147,
1345
+ "learning_rate": 1.5703398843220852e-05,
1346
+ "loss": 0.7394900321960449,
1347
+ "step": 382
1348
+ },
1349
+ {
1350
+ "epoch": 1.6551724137931034,
1351
+ "grad_norm": 0.7287831902503967,
1352
+ "learning_rate": 1.5934589381940026e-05,
1353
+ "loss": 0.7356396913528442,
1354
+ "step": 384
1355
+ },
1356
+ {
1357
+ "epoch": 1.6637931034482758,
1358
+ "grad_norm": 0.45975226163864136,
1359
+ "learning_rate": 1.6161214568239207e-05,
1360
+ "loss": 1.3296085596084595,
1361
+ "step": 386
1362
+ },
1363
+ {
1364
+ "epoch": 1.6724137931034484,
1365
+ "grad_norm": 0.8266739845275879,
1366
+ "learning_rate": 1.638306473445153e-05,
1367
+ "loss": 1.1994720697402954,
1368
+ "step": 388
1369
+ },
1370
+ {
1371
+ "epoch": 1.6810344827586206,
1372
+ "grad_norm": 0.2567940950393677,
1373
+ "learning_rate": 1.6599934630633217e-05,
1374
+ "loss": 1.0203474760055542,
1375
+ "step": 390
1376
+ },
1377
+ {
1378
+ "epoch": 1.6896551724137931,
1379
+ "grad_norm": 0.32722774147987366,
1380
+ "learning_rate": 1.681162361445544e-05,
1381
+ "loss": 0.8025456666946411,
1382
+ "step": 392
1383
+ },
1384
+ {
1385
+ "epoch": 1.6982758620689655,
1386
+ "grad_norm": 0.3686321973800659,
1387
+ "learning_rate": 1.7017935836833353e-05,
1388
+ "loss": 1.2664248943328857,
1389
+ "step": 394
1390
+ },
1391
+ {
1392
+ "epoch": 1.706896551724138,
1393
+ "grad_norm": 0.2905973792076111,
1394
+ "learning_rate": 1.7218680423120506e-05,
1395
+ "loss": 0.8020282983779907,
1396
+ "step": 396
1397
+ },
1398
+ {
1399
+ "epoch": 1.7155172413793105,
1400
+ "grad_norm": 0.2557852268218994,
1401
+ "learning_rate": 1.7413671649701026e-05,
1402
+ "loss": 0.8401910066604614,
1403
+ "step": 398
1404
+ },
1405
+ {
1406
+ "epoch": 1.7241379310344827,
1407
+ "grad_norm": 0.1601221263408661,
1408
+ "learning_rate": 1.7602729115816268e-05,
1409
+ "loss": 0.9220103621482849,
1410
+ "step": 400
1411
+ },
1412
+ {
1413
+ "epoch": 1.7327586206896552,
1414
+ "grad_norm": 0.7182515859603882,
1415
+ "learning_rate": 1.77856779104668e-05,
1416
+ "loss": 0.999250054359436,
1417
+ "step": 402
1418
+ },
1419
+ {
1420
+ "epoch": 1.7413793103448276,
1421
+ "grad_norm": 0.8001770973205566,
1422
+ "learning_rate": 1.7962348774235565e-05,
1423
+ "loss": 1.7149617671966553,
1424
+ "step": 404
1425
+ },
1426
+ {
1427
+ "epoch": 1.75,
1428
+ "grad_norm": 1.6669609546661377,
1429
+ "learning_rate": 1.8132578255882124e-05,
1430
+ "loss": 0.9877924919128418,
1431
+ "step": 406
1432
+ },
1433
+ {
1434
+ "epoch": 1.7586206896551724,
1435
+ "grad_norm": 0.735676109790802,
1436
+ "learning_rate": 1.8296208863563515e-05,
1437
+ "loss": 1.2993067502975464,
1438
+ "step": 408
1439
+ },
1440
+ {
1441
+ "epoch": 1.7672413793103448,
1442
+ "grad_norm": 6.548970699310303,
1443
+ "learning_rate": 1.8453089210541577e-05,
1444
+ "loss": 0.8246227502822876,
1445
+ "step": 410
1446
+ },
1447
+ {
1448
+ "epoch": 1.7758620689655173,
1449
+ "grad_norm": 0.4565357565879822,
1450
+ "learning_rate": 1.8603074155241946e-05,
1451
+ "loss": 1.2052366733551025,
1452
+ "step": 412
1453
+ },
1454
+ {
1455
+ "epoch": 1.7844827586206895,
1456
+ "grad_norm": 0.3367665112018585,
1457
+ "learning_rate": 1.8746024935535238e-05,
1458
+ "loss": 1.217560887336731,
1459
+ "step": 414
1460
+ },
1461
+ {
1462
+ "epoch": 1.793103448275862,
1463
+ "grad_norm": 0.7779958844184875,
1464
+ "learning_rate": 1.8881809297116126e-05,
1465
+ "loss": 0.3758542239665985,
1466
+ "step": 416
1467
+ },
1468
+ {
1469
+ "epoch": 1.8017241379310345,
1470
+ "grad_norm": 0.5207919478416443,
1471
+ "learning_rate": 1.901030161586157e-05,
1472
+ "loss": 1.4365136623382568,
1473
+ "step": 418
1474
+ },
1475
+ {
1476
+ "epoch": 1.8103448275862069,
1477
+ "grad_norm": 6.575118541717529,
1478
+ "learning_rate": 1.9131383014054904e-05,
1479
+ "loss": 0.7461058497428894,
1480
+ "step": 420
1481
+ },
1482
+ {
1483
+ "epoch": 1.8189655172413794,
1484
+ "grad_norm": 0.35711878538131714,
1485
+ "learning_rate": 1.924494147036844e-05,
1486
+ "loss": 1.2452733516693115,
1487
+ "step": 422
1488
+ },
1489
+ {
1490
+ "epoch": 1.8275862068965516,
1491
+ "grad_norm": 0.410709410905838,
1492
+ "learning_rate": 1.9350871923502582e-05,
1493
+ "loss": 1.1930017471313477,
1494
+ "step": 424
1495
+ },
1496
+ {
1497
+ "epoch": 1.8362068965517242,
1498
+ "grad_norm": 0.6211258769035339,
1499
+ "learning_rate": 1.944907636938585e-05,
1500
+ "loss": 1.0968483686447144,
1501
+ "step": 426
1502
+ },
1503
+ {
1504
+ "epoch": 1.8448275862068966,
1505
+ "grad_norm": 7.162904739379883,
1506
+ "learning_rate": 1.953946395184563e-05,
1507
+ "loss": 0.3261430561542511,
1508
+ "step": 428
1509
+ },
1510
+ {
1511
+ "epoch": 1.853448275862069,
1512
+ "grad_norm": 7.713586807250977,
1513
+ "learning_rate": 1.962195104666592e-05,
1514
+ "loss": 0.6268512010574341,
1515
+ "step": 430
1516
+ },
1517
+ {
1518
+ "epoch": 1.8620689655172413,
1519
+ "grad_norm": 0.4543277323246002,
1520
+ "learning_rate": 1.9696461338954278e-05,
1521
+ "loss": 0.832156240940094,
1522
+ "step": 432
1523
+ },
1524
+ {
1525
+ "epoch": 1.8706896551724137,
1526
+ "grad_norm": 2.5041592121124268,
1527
+ "learning_rate": 1.976292589374634e-05,
1528
+ "loss": 0.5735204815864563,
1529
+ "step": 434
1530
+ },
1531
+ {
1532
+ "epoch": 1.8793103448275863,
1533
+ "grad_norm": 0.7300209403038025,
1534
+ "learning_rate": 1.9821283219782627e-05,
1535
+ "loss": 0.5290496945381165,
1536
+ "step": 436
1537
+ },
1538
+ {
1539
+ "epoch": 1.8879310344827587,
1540
+ "grad_norm": 0.34921202063560486,
1541
+ "learning_rate": 1.9871479326398668e-05,
1542
+ "loss": 1.232295036315918,
1543
+ "step": 438
1544
+ },
1545
+ {
1546
+ "epoch": 1.896551724137931,
1547
+ "grad_norm": 0.3079286217689514,
1548
+ "learning_rate": 1.9913467773475727e-05,
1549
+ "loss": 1.189243197441101,
1550
+ "step": 440
1551
+ },
1552
+ {
1553
+ "epoch": 1.9051724137931034,
1554
+ "grad_norm": 0.23536956310272217,
1555
+ "learning_rate": 1.9947209714405984e-05,
1556
+ "loss": 0.7931770086288452,
1557
+ "step": 442
1558
+ },
1559
+ {
1560
+ "epoch": 1.9137931034482758,
1561
+ "grad_norm": 1.4744566679000854,
1562
+ "learning_rate": 1.9972673932032394e-05,
1563
+ "loss": 0.7982081174850464,
1564
+ "step": 444
1565
+ },
1566
+ {
1567
+ "epoch": 1.9224137931034484,
1568
+ "grad_norm": 0.3188006281852722,
1569
+ "learning_rate": 1.9989836867529987e-05,
1570
+ "loss": 0.661638081073761,
1571
+ "step": 446
1572
+ },
1573
+ {
1574
+ "epoch": 1.9310344827586206,
1575
+ "grad_norm": 0.5140587687492371,
1576
+ "learning_rate": 1.999868264220186e-05,
1577
+ "loss": 1.3983592987060547,
1578
+ "step": 448
1579
+ },
1580
+ {
1581
+ "epoch": 1.9396551724137931,
1582
+ "grad_norm": 0.6525475382804871,
1583
+ "learning_rate": 1.999920307216975e-05,
1584
+ "loss": 1.5112402439117432,
1585
+ "step": 450
1586
+ },
1587
+ {
1588
+ "epoch": 1.9482758620689655,
1589
+ "grad_norm": 0.236190065741539,
1590
+ "learning_rate": 1.9991397675945524e-05,
1591
+ "loss": 1.2273776531219482,
1592
+ "step": 452
1593
+ },
1594
+ {
1595
+ "epoch": 1.956896551724138,
1596
+ "grad_norm": 0.2825621962547302,
1597
+ "learning_rate": 1.9975273674876653e-05,
1598
+ "loss": 0.7660670876502991,
1599
+ "step": 454
1600
+ },
1601
+ {
1602
+ "epoch": 1.9655172413793105,
1603
+ "grad_norm": 0.21918132901191711,
1604
+ "learning_rate": 1.99508459864652e-05,
1605
+ "loss": 0.9497285485267639,
1606
+ "step": 456
1607
+ },
1608
+ {
1609
+ "epoch": 1.9741379310344827,
1610
+ "grad_norm": 0.38524022698402405,
1611
+ "learning_rate": 1.991813721056655e-05,
1612
+ "loss": 1.0104745626449585,
1613
+ "step": 458
1614
+ },
1615
+ {
1616
+ "epoch": 1.9827586206896552,
1617
+ "grad_norm": 3.191633701324463,
1618
+ "learning_rate": 1.987717760848061e-05,
1619
+ "loss": 0.919417679309845,
1620
+ "step": 460
1621
+ },
1622
+ {
1623
+ "epoch": 1.9913793103448276,
1624
+ "grad_norm": 0.2649468183517456,
1625
+ "learning_rate": 1.982800507495485e-05,
1626
+ "loss": 1.400168538093567,
1627
+ "step": 462
1628
+ },
1629
+ {
1630
+ "epoch": 2.0,
1631
+ "grad_norm": 0.6692625880241394,
1632
+ "learning_rate": 1.977066510312508e-05,
1633
+ "loss": 1.335584282875061,
1634
+ "step": 464
1635
+ },
1636
+ {
1637
+ "epoch": 2.0086206896551726,
1638
+ "grad_norm": 0.93243408203125,
1639
+ "learning_rate": 1.9705210742426404e-05,
1640
+ "loss": 1.1145689487457275,
1641
+ "step": 466
1642
+ },
1643
+ {
1644
+ "epoch": 2.0172413793103448,
1645
+ "grad_norm": 0.8251880407333374,
1646
+ "learning_rate": 1.9631702549513264e-05,
1647
+ "loss": 1.1242371797561646,
1648
+ "step": 468
1649
+ },
1650
+ {
1651
+ "epoch": 2.0258620689655173,
1652
+ "grad_norm": 1.484380841255188,
1653
+ "learning_rate": 1.9550208532234013e-05,
1654
+ "loss": 1.022644281387329,
1655
+ "step": 470
1656
+ },
1657
+ {
1658
+ "epoch": 2.0344827586206895,
1659
+ "grad_norm": 0.8342429995536804,
1660
+ "learning_rate": 1.9460804086711857e-05,
1661
+ "loss": 0.6067448258399963,
1662
+ "step": 472
1663
+ },
1664
+ {
1665
+ "epoch": 2.043103448275862,
1666
+ "grad_norm": 6.31473445892334,
1667
+ "learning_rate": 1.936357192759033e-05,
1668
+ "loss": 1.1570570468902588,
1669
+ "step": 474
1670
+ },
1671
+ {
1672
+ "epoch": 2.0517241379310347,
1673
+ "grad_norm": 0.2635251581668854,
1674
+ "learning_rate": 1.9258602011507896e-05,
1675
+ "loss": 1.0996088981628418,
1676
+ "step": 476
1677
+ },
1678
+ {
1679
+ "epoch": 2.060344827586207,
1680
+ "grad_norm": 0.5212433338165283,
1681
+ "learning_rate": 1.9145991453872398e-05,
1682
+ "loss": 0.6901102066040039,
1683
+ "step": 478
1684
+ },
1685
+ {
1686
+ "epoch": 2.0689655172413794,
1687
+ "grad_norm": 0.4480215311050415,
1688
+ "learning_rate": 1.902584443901246e-05,
1689
+ "loss": 1.4176607131958008,
1690
+ "step": 480
1691
+ },
1692
+ {
1693
+ "epoch": 2.0775862068965516,
1694
+ "grad_norm": 2.65861439704895,
1695
+ "learning_rate": 1.8898272123788897e-05,
1696
+ "loss": 1.0099068880081177,
1697
+ "step": 482
1698
+ },
1699
+ {
1700
+ "epoch": 2.086206896551724,
1701
+ "grad_norm": 0.35105887055397034,
1702
+ "learning_rate": 1.8763392534755276e-05,
1703
+ "loss": 0.9407973289489746,
1704
+ "step": 484
1705
+ },
1706
+ {
1707
+ "epoch": 2.0948275862068964,
1708
+ "grad_norm": 0.3949888348579407,
1709
+ "learning_rate": 1.862133045896286e-05,
1710
+ "loss": 1.2682446241378784,
1711
+ "step": 486
1712
+ },
1713
+ {
1714
+ "epoch": 2.103448275862069,
1715
+ "grad_norm": 0.16779205203056335,
1716
+ "learning_rate": 1.847221732851092e-05,
1717
+ "loss": 0.2819199562072754,
1718
+ "step": 488
1719
+ },
1720
+ {
1721
+ "epoch": 2.1120689655172415,
1722
+ "grad_norm": 1.363291621208191,
1723
+ "learning_rate": 1.831619109894921e-05,
1724
+ "loss": 0.4681054651737213,
1725
+ "step": 490
1726
+ },
1727
+ {
1728
+ "epoch": 2.1206896551724137,
1729
+ "grad_norm": 0.33102285861968994,
1730
+ "learning_rate": 1.81533961216451e-05,
1731
+ "loss": 0.9479153156280518,
1732
+ "step": 492
1733
+ },
1734
+ {
1735
+ "epoch": 2.1293103448275863,
1736
+ "grad_norm": 0.6415525674819946,
1737
+ "learning_rate": 1.79839830102335e-05,
1738
+ "loss": 1.2515679597854614,
1739
+ "step": 494
1740
+ },
1741
+ {
1742
+ "epoch": 2.1379310344827585,
1743
+ "grad_norm": 0.2858109474182129,
1744
+ "learning_rate": 1.780810850127307e-05,
1745
+ "loss": 1.110595464706421,
1746
+ "step": 496
1747
+ },
1748
+ {
1749
+ "epoch": 2.146551724137931,
1750
+ "grad_norm": 0.7833629846572876,
1751
+ "learning_rate": 1.7625935309237705e-05,
1752
+ "loss": 0.6906008720397949,
1753
+ "step": 498
1754
+ },
1755
+ {
1756
+ "epoch": 2.1551724137931036,
1757
+ "grad_norm": 0.3774907886981964,
1758
+ "learning_rate": 1.7437631975977327e-05,
1759
+ "loss": 0.8901933431625366,
1760
+ "step": 500
1761
+ },
1762
+ {
1763
+ "epoch": 2.163793103448276,
1764
+ "grad_norm": 1.0206127166748047,
1765
+ "learning_rate": 1.7243372714787447e-05,
1766
+ "loss": 1.1632163524627686,
1767
+ "step": 502
1768
+ },
1769
+ {
1770
+ "epoch": 2.1724137931034484,
1771
+ "grad_norm": 0.2904803454875946,
1772
+ "learning_rate": 1.704333724923158e-05,
1773
+ "loss": 1.0225820541381836,
1774
+ "step": 504
1775
+ },
1776
+ {
1777
+ "epoch": 2.1810344827586206,
1778
+ "grad_norm": 1.0189746618270874,
1779
+ "learning_rate": 1.683771064686572e-05,
1780
+ "loss": 0.596871018409729,
1781
+ "step": 506
1782
+ },
1783
+ {
1784
+ "epoch": 2.189655172413793,
1785
+ "grad_norm": 0.2902408540248871,
1786
+ "learning_rate": 1.6626683148018704e-05,
1787
+ "loss": 1.1479698419570923,
1788
+ "step": 508
1789
+ },
1790
+ {
1791
+ "epoch": 2.1982758620689653,
1792
+ "grad_norm": 0.32355165481567383,
1793
+ "learning_rate": 1.6410449989786864e-05,
1794
+ "loss": 1.113614559173584,
1795
+ "step": 510
1796
+ },
1797
+ {
1798
+ "epoch": 2.206896551724138,
1799
+ "grad_norm": 0.34692347049713135,
1800
+ "learning_rate": 1.6189211225405815e-05,
1801
+ "loss": 0.7449208498001099,
1802
+ "step": 512
1803
+ },
1804
+ {
1805
+ "epoch": 2.2155172413793105,
1806
+ "grad_norm": 0.30047309398651123,
1807
+ "learning_rate": 1.596317153916643e-05,
1808
+ "loss": 1.1159427165985107,
1809
+ "step": 514
1810
+ },
1811
+ {
1812
+ "epoch": 2.2241379310344827,
1813
+ "grad_norm": 0.6437045335769653,
1814
+ "learning_rate": 1.5732540057046285e-05,
1815
+ "loss": 0.7390024662017822,
1816
+ "step": 516
1817
+ },
1818
+ {
1819
+ "epoch": 2.2327586206896552,
1820
+ "grad_norm": 0.46292436122894287,
1821
+ "learning_rate": 1.5497530153231886e-05,
1822
+ "loss": 0.7803560495376587,
1823
+ "step": 518
1824
+ },
1825
+ {
1826
+ "epoch": 2.2413793103448274,
1827
+ "grad_norm": 0.5340967774391174,
1828
+ "learning_rate": 1.5258359252710415e-05,
1829
+ "loss": 1.3723326921463013,
1830
+ "step": 520
1831
+ },
1832
+ {
1833
+ "epoch": 2.25,
1834
+ "grad_norm": 0.31911370158195496,
1835
+ "learning_rate": 1.5015248630113865e-05,
1836
+ "loss": 0.7922369241714478,
1837
+ "step": 522
1838
+ },
1839
+ {
1840
+ "epoch": 2.2586206896551726,
1841
+ "grad_norm": 0.4369973838329315,
1842
+ "learning_rate": 1.476842320500153e-05,
1843
+ "loss": 0.7873719334602356,
1844
+ "step": 524
1845
+ },
1846
+ {
1847
+ "epoch": 2.2672413793103448,
1848
+ "grad_norm": 5.220350742340088,
1849
+ "learning_rate": 1.4518111333770401e-05,
1850
+ "loss": 0.7667290568351746,
1851
+ "step": 526
1852
+ },
1853
+ {
1854
+ "epoch": 2.2758620689655173,
1855
+ "grad_norm": 0.7911840081214905,
1856
+ "learning_rate": 1.4264544598385801e-05,
1857
+ "loss": 1.106222152709961,
1858
+ "step": 528
1859
+ },
1860
+ {
1861
+ "epoch": 2.2844827586206895,
1862
+ "grad_norm": 0.27636000514030457,
1863
+ "learning_rate": 1.4007957592127807e-05,
1864
+ "loss": 1.0902438163757324,
1865
+ "step": 530
1866
+ },
1867
+ {
1868
+ "epoch": 2.293103448275862,
1869
+ "grad_norm": 0.4640883505344391,
1870
+ "learning_rate": 1.374858770255177e-05,
1871
+ "loss": 0.8156579732894897,
1872
+ "step": 532
1873
+ },
1874
+ {
1875
+ "epoch": 2.3017241379310347,
1876
+ "grad_norm": 0.8998560905456543,
1877
+ "learning_rate": 1.3486674891863518e-05,
1878
+ "loss": 1.139012098312378,
1879
+ "step": 534
1880
+ },
1881
+ {
1882
+ "epoch": 2.310344827586207,
1883
+ "grad_norm": 0.4182785451412201,
1884
+ "learning_rate": 1.322246147491271e-05,
1885
+ "loss": 1.191633701324463,
1886
+ "step": 536
1887
+ },
1888
+ {
1889
+ "epoch": 2.3189655172413794,
1890
+ "grad_norm": 0.39193910360336304,
1891
+ "learning_rate": 1.2956191895009415e-05,
1892
+ "loss": 0.7655143141746521,
1893
+ "step": 538
1894
+ },
1895
+ {
1896
+ "epoch": 2.3275862068965516,
1897
+ "grad_norm": 0.4902004301548004,
1898
+ "learning_rate": 1.268811249777167e-05,
1899
+ "loss": 0.9563652276992798,
1900
+ "step": 540
1901
+ },
1902
+ {
1903
+ "epoch": 2.336206896551724,
1904
+ "grad_norm": 0.6500739455223083,
1905
+ "learning_rate": 1.2418471303212933e-05,
1906
+ "loss": 0.7211173176765442,
1907
+ "step": 542
1908
+ },
1909
+ {
1910
+ "epoch": 2.344827586206897,
1911
+ "grad_norm": 0.32692751288414,
1912
+ "learning_rate": 1.2147517776280517e-05,
1913
+ "loss": 1.0490473508834839,
1914
+ "step": 544
1915
+ },
1916
+ {
1917
+ "epoch": 2.353448275862069,
1918
+ "grad_norm": 0.8291403651237488,
1919
+ "learning_rate": 1.1875502596057116e-05,
1920
+ "loss": 0.8839560151100159,
1921
+ "step": 546
1922
+ },
1923
+ {
1924
+ "epoch": 2.3620689655172415,
1925
+ "grad_norm": 0.28376197814941406,
1926
+ "learning_rate": 1.1602677423839195e-05,
1927
+ "loss": 1.127523422241211,
1928
+ "step": 548
1929
+ },
1930
+ {
1931
+ "epoch": 2.3706896551724137,
1932
+ "grad_norm": 0.604331910610199,
1933
+ "learning_rate": 1.1329294670306513e-05,
1934
+ "loss": 0.6361532807350159,
1935
+ "step": 550
1936
+ },
1937
+ {
1938
+ "epoch": 2.3793103448275863,
1939
+ "grad_norm": 0.6158839464187622,
1940
+ "learning_rate": 1.1055607261998416e-05,
1941
+ "loss": 0.31083130836486816,
1942
+ "step": 552
1943
+ },
1944
+ {
1945
+ "epoch": 2.3879310344827585,
1946
+ "grad_norm": 0.3060431480407715,
1947
+ "learning_rate": 1.078186840731281e-05,
1948
+ "loss": 1.2638636827468872,
1949
+ "step": 554
1950
+ },
1951
+ {
1952
+ "epoch": 2.396551724137931,
1953
+ "grad_norm": 0.25409573316574097,
1954
+ "learning_rate": 1.0508331362244472e-05,
1955
+ "loss": 1.079472303390503,
1956
+ "step": 556
1957
+ },
1958
+ {
1959
+ "epoch": 2.405172413793103,
1960
+ "grad_norm": 0.15330657362937927,
1961
+ "learning_rate": 1.0235249196079217e-05,
1962
+ "loss": 0.4111257493495941,
1963
+ "step": 558
1964
+ },
1965
+ {
1966
+ "epoch": 2.413793103448276,
1967
+ "grad_norm": 0.5213512778282166,
1968
+ "learning_rate": 9.96287455726086e-06,
1969
+ "loss": 0.9900328516960144,
1970
+ "step": 560
1971
+ },
1972
+ {
1973
+ "epoch": 2.4224137931034484,
1974
+ "grad_norm": 0.28381043672561646,
1975
+ "learning_rate": 9.691459439647469e-06,
1976
+ "loss": 0.8341249823570251,
1977
+ "step": 562
1978
+ },
1979
+ {
1980
+ "epoch": 2.4310344827586206,
1981
+ "grad_norm": 0.676983654499054,
1982
+ "learning_rate": 9.421254949373347e-06,
1983
+ "loss": 0.8743176460266113,
1984
+ "step": 564
1985
+ },
1986
+ {
1987
+ "epoch": 2.439655172413793,
1988
+ "grad_norm": 0.3603588938713074,
1989
+ "learning_rate": 9.152511072532206e-06,
1990
+ "loss": 1.056296467781067,
1991
+ "step": 566
1992
+ },
1993
+ {
1994
+ "epoch": 2.4482758620689653,
1995
+ "grad_norm": 1.0118433237075806,
1996
+ "learning_rate": 8.88547644389654e-06,
1997
+ "loss": 1.413284182548523,
1998
+ "step": 568
1999
+ },
2000
+ {
2001
+ "epoch": 2.456896551724138,
2002
+ "grad_norm": 1.1313719749450684,
2003
+ "learning_rate": 8.62039811688737e-06,
2004
+ "loss": 0.1686408817768097,
2005
+ "step": 570
2006
+ },
2007
+ {
2008
+ "epoch": 2.4655172413793105,
2009
+ "grad_norm": 0.2529284656047821,
2010
+ "learning_rate": 8.357521335006843e-06,
2011
+ "loss": 0.9291278123855591,
2012
+ "step": 572
2013
+ },
2014
+ {
2015
+ "epoch": 2.4741379310344827,
2016
+ "grad_norm": 0.26292645931243896,
2017
+ "learning_rate": 8.097089304945431e-06,
2018
+ "loss": 1.1657003164291382,
2019
+ "step": 574
2020
+ },
2021
+ {
2022
+ "epoch": 2.4827586206896552,
2023
+ "grad_norm": 0.46254923939704895,
2024
+ "learning_rate": 7.839342971573412e-06,
2025
+ "loss": 1.1208579540252686,
2026
+ "step": 576
2027
+ },
2028
+ {
2029
+ "epoch": 2.4913793103448274,
2030
+ "grad_norm": 0.5573145151138306,
2031
+ "learning_rate": 7.584520795025116e-06,
2032
+ "loss": 0.6583477258682251,
2033
+ "step": 578
2034
+ },
2035
+ {
2036
+ "epoch": 2.5,
2037
+ "grad_norm": 0.3368423283100128,
2038
+ "learning_rate": 7.332858530081812e-06,
2039
+ "loss": 1.1409282684326172,
2040
+ "step": 580
2041
+ },
2042
+ {
2043
+ "epoch": 2.5086206896551726,
2044
+ "grad_norm": 0.22401149570941925,
2045
+ "learning_rate": 7.084589008057583e-06,
2046
+ "loss": 0.9950229525566101,
2047
+ "step": 582
2048
+ },
2049
+ {
2050
+ "epoch": 2.5172413793103448,
2051
+ "grad_norm": 0.4168458580970764,
2052
+ "learning_rate": 6.839941921389941e-06,
2053
+ "loss": 0.677608847618103,
2054
+ "step": 584
2055
+ },
2056
+ {
2057
+ "epoch": 2.5258620689655173,
2058
+ "grad_norm": 0.3188498616218567,
2059
+ "learning_rate": 6.599143611134356e-06,
2060
+ "loss": 0.21669866144657135,
2061
+ "step": 586
2062
+ },
2063
+ {
2064
+ "epoch": 2.5344827586206895,
2065
+ "grad_norm": 0.41348910331726074,
2066
+ "learning_rate": 6.362416857559549e-06,
2067
+ "loss": 0.7037891745567322,
2068
+ "step": 588
2069
+ },
2070
+ {
2071
+ "epoch": 2.543103448275862,
2072
+ "grad_norm": 3.121192216873169,
2073
+ "learning_rate": 6.129980674036977e-06,
2074
+ "loss": 0.9864909648895264,
2075
+ "step": 590
2076
+ },
2077
+ {
2078
+ "epoch": 2.5517241379310347,
2079
+ "grad_norm": 0.6518086194992065,
2080
+ "learning_rate": 5.90205010441544e-06,
2081
+ "loss": 0.8599920272827148,
2082
+ "step": 592
2083
+ },
2084
+ {
2085
+ "epoch": 2.560344827586207,
2086
+ "grad_norm": 0.6577050685882568,
2087
+ "learning_rate": 5.678836024068174e-06,
2088
+ "loss": 0.7931281328201294,
2089
+ "step": 594
2090
+ },
2091
+ {
2092
+ "epoch": 2.5689655172413794,
2093
+ "grad_norm": 0.1971147507429123,
2094
+ "learning_rate": 5.4605449447964406e-06,
2095
+ "loss": 0.7438328862190247,
2096
+ "step": 596
2097
+ },
2098
+ {
2099
+ "epoch": 2.5775862068965516,
2100
+ "grad_norm": 0.687308132648468,
2101
+ "learning_rate": 5.24737882377032e-06,
2102
+ "loss": 0.7791666984558105,
2103
+ "step": 598
2104
+ },
2105
+ {
2106
+ "epoch": 2.586206896551724,
2107
+ "grad_norm": 0.24792003631591797,
2108
+ "learning_rate": 5.039534876683188e-06,
2109
+ "loss": 1.1297264099121094,
2110
+ "step": 600
2111
+ },
2112
+ {
2113
+ "epoch": 2.594827586206897,
2114
+ "grad_norm": 0.431932657957077,
2115
+ "learning_rate": 4.837205395292953e-06,
2116
+ "loss": 0.6758660674095154,
2117
+ "step": 602
2118
+ },
2119
+ {
2120
+ "epoch": 2.603448275862069,
2121
+ "grad_norm": 1.7007827758789062,
2122
+ "learning_rate": 4.640577569518789e-06,
2123
+ "loss": 1.114516019821167,
2124
+ "step": 604
2125
+ },
2126
+ {
2127
+ "epoch": 2.612068965517241,
2128
+ "grad_norm": 0.2791440188884735,
2129
+ "learning_rate": 4.449833314257892e-06,
2130
+ "loss": 0.6865538954734802,
2131
+ "step": 606
2132
+ },
2133
+ {
2134
+ "epoch": 2.6206896551724137,
2135
+ "grad_norm": 0.14818964898586273,
2136
+ "learning_rate": 4.265149101082668e-06,
2137
+ "loss": 0.667411744594574,
2138
+ "step": 608
2139
+ },
2140
+ {
2141
+ "epoch": 2.6293103448275863,
2142
+ "grad_norm": 0.47603490948677063,
2143
+ "learning_rate": 4.086695794973827e-06,
2144
+ "loss": 0.999786376953125,
2145
+ "step": 610
2146
+ },
2147
+ {
2148
+ "epoch": 2.637931034482759,
2149
+ "grad_norm": 0.26607102155685425,
2150
+ "learning_rate": 3.914638496240613e-06,
2151
+ "loss": 0.8935456275939941,
2152
+ "step": 612
2153
+ },
2154
+ {
2155
+ "epoch": 2.646551724137931,
2156
+ "grad_norm": 0.5287325382232666,
2157
+ "learning_rate": 3.7491363877742946e-06,
2158
+ "loss": 0.293642520904541,
2159
+ "step": 614
2160
+ },
2161
+ {
2162
+ "epoch": 2.655172413793103,
2163
+ "grad_norm": 0.7885716557502747,
2164
+ "learning_rate": 3.5903425877764024e-06,
2165
+ "loss": 0.860302209854126,
2166
+ "step": 616
2167
+ },
2168
+ {
2169
+ "epoch": 2.663793103448276,
2170
+ "grad_norm": 0.23788996040821075,
2171
+ "learning_rate": 3.4384040080977486e-06,
2172
+ "loss": 0.679624617099762,
2173
+ "step": 618
2174
+ },
2175
+ {
2176
+ "epoch": 2.6724137931034484,
2177
+ "grad_norm": 0.5915703773498535,
2178
+ "learning_rate": 3.2934612183194636e-06,
2179
+ "loss": 0.6118066310882568,
2180
+ "step": 620
2181
+ },
2182
+ {
2183
+ "epoch": 2.6810344827586206,
2184
+ "grad_norm": 0.347929984331131,
2185
+ "learning_rate": 3.1556483157016765e-06,
2186
+ "loss": 1.0104352235794067,
2187
+ "step": 622
2188
+ },
2189
+ {
2190
+ "epoch": 2.689655172413793,
2191
+ "grad_norm": 0.3776116967201233,
2192
+ "learning_rate": 3.0250928011203114e-06,
2193
+ "loss": 0.7297549843788147,
2194
+ "step": 624
2195
+ },
2196
+ {
2197
+ "epoch": 2.6982758620689653,
2198
+ "grad_norm": 0.33936139941215515,
2199
+ "learning_rate": 2.9019154611065938e-06,
2200
+ "loss": 1.0180128812789917,
2201
+ "step": 626
2202
+ },
2203
+ {
2204
+ "epoch": 2.706896551724138,
2205
+ "grad_norm": 1.2539973258972168,
2206
+ "learning_rate": 2.7862302560985105e-06,
2207
+ "loss": 0.729310154914856,
2208
+ "step": 628
2209
+ },
2210
+ {
2211
+ "epoch": 2.7155172413793105,
2212
+ "grad_norm": 0.11071612685918808,
2213
+ "learning_rate": 2.6781442150076485e-06,
2214
+ "loss": 0.5759393572807312,
2215
+ "step": 630
2216
+ },
2217
+ {
2218
+ "epoch": 2.7241379310344827,
2219
+ "grad_norm": 0.3599656820297241,
2220
+ "learning_rate": 2.577757336198829e-06,
2221
+ "loss": 0.8420278429985046,
2222
+ "step": 632
2223
+ },
2224
+ {
2225
+ "epoch": 2.7327586206896552,
2226
+ "grad_norm": 0.45908743143081665,
2227
+ "learning_rate": 2.4851624949742624e-06,
2228
+ "loss": 1.203715443611145,
2229
+ "step": 634
2230
+ },
2231
+ {
2232
+ "epoch": 2.7413793103448274,
2233
+ "grad_norm": 0.4836440682411194,
2234
+ "learning_rate": 2.4004453576477363e-06,
2235
+ "loss": 1.251405119895935,
2236
+ "step": 636
2237
+ },
2238
+ {
2239
+ "epoch": 2.75,
2240
+ "grad_norm": 0.406963974237442,
2241
+ "learning_rate": 2.323684302288429e-06,
2242
+ "loss": 0.5981881022453308,
2243
+ "step": 638
2244
+ },
2245
+ {
2246
+ "epoch": 2.7586206896551726,
2247
+ "grad_norm": 0.4914894700050354,
2248
+ "learning_rate": 2.2549503462075524e-06,
2249
+ "loss": 0.4858713448047638,
2250
+ "step": 640
2251
+ },
2252
+ {
2253
+ "epoch": 2.7672413793103448,
2254
+ "grad_norm": 0.49679532647132874,
2255
+ "learning_rate": 2.1943070802550147e-06,
2256
+ "loss": 1.0236581563949585,
2257
+ "step": 642
2258
+ },
2259
+ {
2260
+ "epoch": 2.7758620689655173,
2261
+ "grad_norm": 0.5968631505966187,
2262
+ "learning_rate": 2.141810609986821e-06,
2263
+ "loss": 0.7582760453224182,
2264
+ "step": 644
2265
+ },
2266
+ {
2267
+ "epoch": 2.7844827586206895,
2268
+ "grad_norm": 0.47396087646484375,
2269
+ "learning_rate": 2.0975095037577143e-06,
2270
+ "loss": 0.7772106528282166,
2271
+ "step": 646
2272
+ },
2273
+ {
2274
+ "epoch": 2.793103448275862,
2275
+ "grad_norm": 0.29245540499687195,
2276
+ "learning_rate": 2.0614447477869856e-06,
2277
+ "loss": 0.9372343420982361,
2278
+ "step": 648
2279
+ },
2280
+ {
2281
+ "epoch": 2.8017241379310347,
2282
+ "grad_norm": 0.3278985023498535,
2283
+ "learning_rate": 2.0336497082391155e-06,
2284
+ "loss": 1.1293443441390991,
2285
+ "step": 650
2286
+ },
2287
+ {
2288
+ "epoch": 2.810344827586207,
2289
+ "grad_norm": 0.2128603756427765,
2290
+ "learning_rate": 2.0141501003542617e-06,
2291
+ "loss": 0.21890364587306976,
2292
+ "step": 652
2293
+ },
2294
+ {
2295
+ "epoch": 2.8189655172413794,
2296
+ "grad_norm": 0.2553781569004059,
2297
+ "learning_rate": 2.002963964657213e-06,
2298
+ "loss": 0.9752658605575562,
2299
+ "step": 654
2300
+ },
2301
+ {
2302
+ "epoch": 2.8275862068965516,
2303
+ "grad_norm": 0.6789045333862305,
2304
+ "learning_rate": 2.0001016502667552e-06,
2305
+ "loss": 1.158425211906433,
2306
+ "step": 656
2307
+ },
2308
+ {
2309
+ "epoch": 2.836206896551724,
2310
+ "grad_norm": 0.2958366572856903,
2311
+ "learning_rate": 2.0055658053209428e-06,
2312
+ "loss": 0.09864208102226257,
2313
+ "step": 658
2314
+ },
2315
+ {
2316
+ "epoch": 2.844827586206897,
2317
+ "grad_norm": 0.5826700925827026,
2318
+ "learning_rate": 2.0193513745271104e-06,
2319
+ "loss": 1.1222366094589233,
2320
+ "step": 660
2321
+ },
2322
+ {
2323
+ "epoch": 2.853448275862069,
2324
+ "grad_norm": 0.2867080867290497,
2325
+ "learning_rate": 2.041445603838903e-06,
2326
+ "loss": 0.7978563904762268,
2327
+ "step": 662
2328
+ },
2329
+ {
2330
+ "epoch": 2.862068965517241,
2331
+ "grad_norm": 0.9346190690994263,
2332
+ "learning_rate": 2.071828052255974e-06,
2333
+ "loss": 1.0888785123825073,
2334
+ "step": 664
2335
+ },
2336
+ {
2337
+ "epoch": 2.8706896551724137,
2338
+ "grad_norm": 0.4730575382709503,
2339
+ "learning_rate": 2.110470610735471e-06,
2340
+ "loss": 0.8604229688644409,
2341
+ "step": 666
2342
+ },
2343
+ {
2344
+ "epoch": 2.8793103448275863,
2345
+ "grad_norm": 0.3757993280887604,
2346
+ "learning_rate": 2.157337528197775e-06,
2347
+ "loss": 0.6924110054969788,
2348
+ "step": 668
2349
+ },
2350
+ {
2351
+ "epoch": 2.887931034482759,
2352
+ "grad_norm": 0.7365322709083557,
2353
+ "learning_rate": 2.2123854446024656e-06,
2354
+ "loss": 1.1259899139404297,
2355
+ "step": 670
2356
+ },
2357
+ {
2358
+ "epoch": 2.896551724137931,
2359
+ "grad_norm": 0.20995132625102997,
2360
+ "learning_rate": 2.275563431063895e-06,
2361
+ "loss": 0.5768133997917175,
2362
+ "step": 672
2363
+ },
2364
+ {
2365
+ "epoch": 2.905172413793103,
2366
+ "grad_norm": 0.12330807000398636,
2367
+ "learning_rate": 2.346813036969242e-06,
2368
+ "loss": 0.655694842338562,
2369
+ "step": 674
2370
+ },
2371
+ {
2372
+ "epoch": 2.913793103448276,
2373
+ "grad_norm": 0.295114129781723,
2374
+ "learning_rate": 2.4260683440554954e-06,
2375
+ "loss": 0.6339759230613708,
2376
+ "step": 676
2377
+ },
2378
+ {
2379
+ "epoch": 2.9224137931034484,
2380
+ "grad_norm": 0.3862989544868469,
2381
+ "learning_rate": 2.5132560273952874e-06,
2382
+ "loss": 0.9285910129547119,
2383
+ "step": 678
2384
+ },
2385
+ {
2386
+ "epoch": 2.9310344827586206,
2387
+ "grad_norm": 0.33736833930015564,
2388
+ "learning_rate": 2.6082954232351966e-06,
2389
+ "loss": 0.8067626953125,
2390
+ "step": 680
2391
+ },
2392
+ {
2393
+ "epoch": 2.939655172413793,
2394
+ "grad_norm": 0.23088297247886658,
2395
+ "learning_rate": 2.711098603623742e-06,
2396
+ "loss": 1.2526490688323975,
2397
+ "step": 682
2398
+ },
2399
+ {
2400
+ "epoch": 2.9482758620689653,
2401
+ "grad_norm": 0.29236695170402527,
2402
+ "learning_rate": 2.821570457759993e-06,
2403
+ "loss": 1.1392483711242676,
2404
+ "step": 684
2405
+ },
2406
+ {
2407
+ "epoch": 2.956896551724138,
2408
+ "grad_norm": 2.1297762393951416,
2409
+ "learning_rate": 2.9396087799875866e-06,
2410
+ "loss": 0.7813910841941833,
2411
+ "step": 686
2412
+ },
2413
+ {
2414
+ "epoch": 2.9655172413793105,
2415
+ "grad_norm": 0.7009169459342957,
2416
+ "learning_rate": 3.065104364352743e-06,
2417
+ "loss": 0.9301526546478271,
2418
+ "step": 688
2419
+ },
2420
+ {
2421
+ "epoch": 2.9741379310344827,
2422
+ "grad_norm": 0.6672054529190063,
2423
+ "learning_rate": 3.197941105638702e-06,
2424
+ "loss": 1.025092601776123,
2425
+ "step": 690
2426
+ },
2427
+ {
2428
+ "epoch": 2.9827586206896552,
2429
+ "grad_norm": 0.5058484673500061,
2430
+ "learning_rate": 3.337996106783247e-06,
2431
+ "loss": 0.6867604851722717,
2432
+ "step": 692
2433
+ },
2434
+ {
2435
+ "epoch": 2.9913793103448274,
2436
+ "grad_norm": 0.34245193004608154,
2437
+ "learning_rate": 3.4851397925798167e-06,
2438
+ "loss": 1.19440495967865,
2439
+ "step": 694
2440
+ },
2441
+ {
2442
+ "epoch": 3.0,
2443
+ "grad_norm": 0.4390437602996826,
2444
+ "learning_rate": 3.6392360295571228e-06,
2445
+ "loss": 1.113608717918396,
2446
+ "step": 696
2447
+ },
2448
+ {
2449
+ "epoch": 3.0,
2450
+ "step": 696,
2451
+ "total_flos": 2.7548743231466373e+18,
2452
+ "train_loss": 1.0648414060130886,
2453
+ "train_runtime": 11016.7324,
2454
+ "train_samples_per_second": 3.791,
2455
+ "train_steps_per_second": 0.063
2456
+ }
2457
+ ],
2458
+ "logging_steps": 2,
2459
+ "max_steps": 696,
2460
+ "num_input_tokens_seen": 0,
2461
+ "num_train_epochs": 3,
2462
+ "save_steps": 99999,
2463
+ "stateful_callbacks": {
2464
+ "TrainerControl": {
2465
+ "args": {
2466
+ "should_epoch_stop": false,
2467
+ "should_evaluate": false,
2468
+ "should_log": false,
2469
+ "should_save": false,
2470
+ "should_training_stop": false
2471
+ },
2472
+ "attributes": {}
2473
+ }
2474
+ },
2475
+ "total_flos": 2.7548743231466373e+18,
2476
+ "train_batch_size": 2,
2477
+ "trial_name": null,
2478
+ "trial_params": null
2479
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47b7e242741438090b2388ed1d624ef69beb0f140286d6d60a52f84bdd74d361
3
+ size 5649
training_loss.png ADDED