Hugging Face's logo

furproxy
/
9b-dci

Image-Text-to-Text
Transformers
Safetensors
qwen3_5
llama-factory
full
Generated from Trainer
conversational
Model card Files
xet
 Community
furproxy commited on
Commit
ddefa79
·
verified ·
1 Parent(s): 8cf5165

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 +376 -0
  14. trainer_state.json +2668 -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
+ 5k DCI dataset, 5k docci dataset, 2k regularization
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", 7.5e-6), language_lr),
40
+ "vision": _fallback(getattr(training_args, "vision_muon_lr", 3e-6), vision_lr),
41
+ "merger": _fallback(getattr(training_args, "merger_muon_lr", 1.5e-5), merger_lr),
42
+ }
43
+
44
+ family_to_adamw_lr = {
45
+ "language": _fallback(getattr(training_args, "language_adamw_lr", 7.5e-6), language_lr),
46
+ "vision": _fallback(getattr(training_args, "vision_adamw_lr", 3e-6), vision_lr),
47
+ "merger": _fallback(getattr(training_args, "merger_adamw_lr", 1.5e-5), merger_lr),
48
+ }
49
+ - train_batch_size: 1
50
+ - eval_batch_size: 8
51
+ - seed: 42
52
+ - distributed_type: multi-GPU
53
+ - gradient_accumulation_steps: 32
54
+ - total_train_batch_size: 32
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: 0.1
58
+ - num_epochs: 2
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
+ 5k DCI dataset, 5k docci dataset, 2k regularization
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", 7.5e-6), language_lr),
40
+ "vision": _fallback(getattr(training_args, "vision_muon_lr", 3e-6), vision_lr),
41
+ "merger": _fallback(getattr(training_args, "merger_muon_lr", 1.5e-5), merger_lr),
42
+ }
43
+
44
+ family_to_adamw_lr = {
45
+ "language": _fallback(getattr(training_args, "language_adamw_lr", 7.5e-6), language_lr),
46
+ "vision": _fallback(getattr(training_args, "vision_adamw_lr", 3e-6), vision_lr),
47
+ "merger": _fallback(getattr(training_args, "merger_adamw_lr", 1.5e-5), merger_lr),
48
+ }
49
+ - train_batch_size: 1
50
+ - eval_batch_size: 8
51
+ - seed: 42
52
+ - distributed_type: multi-GPU
53
+ - gradient_accumulation_steps: 32
54
+ - total_train_batch_size: 32
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: 0.1
58
+ - num_epochs: 2
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": 4848.179638638243,
3
+ "epoch": 2.0,
4
+ "total_flos": 2.55147477658973e+18,
5
+ "train_loss": 1.3643777534166972,
6
+ "train_runtime": 10437.131,
7
+ "train_samples_per_second": 2.299,
8
+ "train_steps_per_second": 0.072
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 sha256:9e1633fa4962e6cdf3ab84c90b3073bb9c80144adf52e9834f2c121e9ab673d3
3
+ size 18819727384
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": 4848.179638638243,
3
+ "epoch": 2.0,
4
+ "total_flos": 2.55147477658973e+18,
5
+ "train_loss": 1.3643777534166972,
6
+ "train_runtime": 10437.131,
7
+ "train_samples_per_second": 2.299,
8
+ "train_steps_per_second": 0.072
9
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,376 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 2, "total_steps": 750, "loss": 1.3918027877807617, "lr": 1.0000000000000001e-07, "epoch": 0.005333333333333333, "percentage": 0.27, "elapsed_time": "0:05:26", "remaining_time": "1 day, 9:55:40"}
2
+ {"current_steps": 4, "total_steps": 750, "loss": 2.0975496768951416, "lr": 3.0000000000000004e-07, "epoch": 0.010666666666666666, "percentage": 0.53, "elapsed_time": "0:05:50", "remaining_time": "18:09:11"}
3
+ {"current_steps": 6, "total_steps": 750, "loss": 2.0533759593963623, "lr": 5e-07, "epoch": 0.016, "percentage": 0.8, "elapsed_time": "0:06:15", "remaining_time": "12:55:05"}
4
+ {"current_steps": 8, "total_steps": 750, "loss": 1.4977960586547852, "lr": 7.000000000000001e-07, "epoch": 0.021333333333333333, "percentage": 1.07, "elapsed_time": "0:06:47", "remaining_time": "10:29:16"}
5
+ {"current_steps": 10, "total_steps": 750, "loss": 1.8092024326324463, "lr": 9e-07, "epoch": 0.02666666666666667, "percentage": 1.33, "elapsed_time": "0:07:36", "remaining_time": "9:23:08"}
6
+ {"current_steps": 12, "total_steps": 750, "loss": 2.0490243434906006, "lr": 1.1e-06, "epoch": 0.032, "percentage": 1.6, "elapsed_time": "0:08:03", "remaining_time": "8:15:36"}
7
+ {"current_steps": 14, "total_steps": 750, "loss": 2.2412607669830322, "lr": 1.3e-06, "epoch": 0.037333333333333336, "percentage": 1.87, "elapsed_time": "0:08:28", "remaining_time": "7:25:25"}
8
+ {"current_steps": 16, "total_steps": 750, "loss": 1.9627114534378052, "lr": 1.5e-06, "epoch": 0.042666666666666665, "percentage": 2.13, "elapsed_time": "0:08:54", "remaining_time": "6:49:02"}
9
+ {"current_steps": 18, "total_steps": 750, "loss": 1.8048596382141113, "lr": 1.7e-06, "epoch": 0.048, "percentage": 2.4, "elapsed_time": "0:09:25", "remaining_time": "6:23:15"}
10
+ {"current_steps": 20, "total_steps": 750, "loss": 1.9537056684494019, "lr": 1.9000000000000002e-06, "epoch": 0.05333333333333334, "percentage": 2.67, "elapsed_time": "0:09:48", "remaining_time": "5:57:59"}
11
+ {"current_steps": 22, "total_steps": 750, "loss": 1.7351176738739014, "lr": 2.1000000000000002e-06, "epoch": 0.058666666666666666, "percentage": 2.93, "elapsed_time": "0:10:16", "remaining_time": "5:40:13"}
12
+ {"current_steps": 24, "total_steps": 750, "loss": 2.1719343662261963, "lr": 2.3e-06, "epoch": 0.064, "percentage": 3.2, "elapsed_time": "0:10:39", "remaining_time": "5:22:37"}
13
+ {"current_steps": 26, "total_steps": 750, "loss": 2.0036661624908447, "lr": 2.4999999999999998e-06, "epoch": 0.06933333333333333, "percentage": 3.47, "elapsed_time": "0:11:06", "remaining_time": "5:09:25"}
14
+ {"current_steps": 28, "total_steps": 750, "loss": 1.4876424074172974, "lr": 2.7e-06, "epoch": 0.07466666666666667, "percentage": 3.73, "elapsed_time": "0:11:36", "remaining_time": "4:59:20"}
15
+ {"current_steps": 30, "total_steps": 750, "loss": 1.7595216035842896, "lr": 2.9e-06, "epoch": 0.08, "percentage": 4.0, "elapsed_time": "0:12:02", "remaining_time": "4:49:03"}
16
+ {"current_steps": 32, "total_steps": 750, "loss": 1.7543615102767944, "lr": 3.1e-06, "epoch": 0.08533333333333333, "percentage": 4.27, "elapsed_time": "0:12:30", "remaining_time": "4:40:32"}
17
+ {"current_steps": 34, "total_steps": 750, "loss": 1.6767919063568115, "lr": 3.3e-06, "epoch": 0.09066666666666667, "percentage": 4.53, "elapsed_time": "0:12:59", "remaining_time": "4:33:32"}
18
+ {"current_steps": 36, "total_steps": 750, "loss": 1.9044243097305298, "lr": 3.5e-06, "epoch": 0.096, "percentage": 4.8, "elapsed_time": "0:13:26", "remaining_time": "4:26:38"}
19
+ {"current_steps": 38, "total_steps": 750, "loss": 1.9992799758911133, "lr": 3.7e-06, "epoch": 0.10133333333333333, "percentage": 5.07, "elapsed_time": "0:13:52", "remaining_time": "4:20:03"}
20
+ {"current_steps": 40, "total_steps": 750, "loss": 1.4217664003372192, "lr": 3.9e-06, "epoch": 0.10666666666666667, "percentage": 5.33, "elapsed_time": "0:14:26", "remaining_time": "4:16:21"}
21
+ {"current_steps": 42, "total_steps": 750, "loss": 2.2651872634887695, "lr": 4.1e-06, "epoch": 0.112, "percentage": 5.6, "elapsed_time": "0:14:51", "remaining_time": "4:10:30"}
22
+ {"current_steps": 44, "total_steps": 750, "loss": 1.8020230531692505, "lr": 4.3e-06, "epoch": 0.11733333333333333, "percentage": 5.87, "elapsed_time": "0:15:16", "remaining_time": "4:05:12"}
23
+ {"current_steps": 46, "total_steps": 750, "loss": 2.057128667831421, "lr": 4.5e-06, "epoch": 0.12266666666666666, "percentage": 6.13, "elapsed_time": "0:15:42", "remaining_time": "4:00:21"}
24
+ {"current_steps": 48, "total_steps": 750, "loss": 1.7972444295883179, "lr": 4.700000000000001e-06, "epoch": 0.128, "percentage": 6.4, "elapsed_time": "0:16:11", "remaining_time": "3:56:41"}
25
+ {"current_steps": 50, "total_steps": 750, "loss": 1.585660696029663, "lr": 4.9e-06, "epoch": 0.13333333333333333, "percentage": 6.67, "elapsed_time": "0:16:35", "remaining_time": "3:52:19"}
26
+ {"current_steps": 52, "total_steps": 750, "loss": 1.5591168403625488, "lr": 5.1e-06, "epoch": 0.13866666666666666, "percentage": 6.93, "elapsed_time": "0:17:03", "remaining_time": "3:49:03"}
27
+ {"current_steps": 54, "total_steps": 750, "loss": 1.2894660234451294, "lr": 5.3e-06, "epoch": 0.144, "percentage": 7.2, "elapsed_time": "0:17:35", "remaining_time": "3:46:46"}
28
+ {"current_steps": 56, "total_steps": 750, "loss": 1.8333218097686768, "lr": 5.5e-06, "epoch": 0.14933333333333335, "percentage": 7.47, "elapsed_time": "0:18:01", "remaining_time": "3:43:27"}
29
+ {"current_steps": 58, "total_steps": 750, "loss": 1.5420780181884766, "lr": 5.7000000000000005e-06, "epoch": 0.15466666666666667, "percentage": 7.73, "elapsed_time": "0:18:22", "remaining_time": "3:39:14"}
30
+ {"current_steps": 60, "total_steps": 750, "loss": 1.8294118642807007, "lr": 5.899999999999999e-06, "epoch": 0.16, "percentage": 8.0, "elapsed_time": "0:18:49", "remaining_time": "3:36:26"}
31
+ {"current_steps": 62, "total_steps": 750, "loss": 1.5082216262817383, "lr": 6.1e-06, "epoch": 0.16533333333333333, "percentage": 8.27, "elapsed_time": "0:19:16", "remaining_time": "3:33:53"}
32
+ {"current_steps": 64, "total_steps": 750, "loss": 1.7252726554870605, "lr": 6.3e-06, "epoch": 0.17066666666666666, "percentage": 8.53, "elapsed_time": "0:19:44", "remaining_time": "3:31:33"}
33
+ {"current_steps": 66, "total_steps": 750, "loss": 1.56448233127594, "lr": 6.5000000000000004e-06, "epoch": 0.176, "percentage": 8.8, "elapsed_time": "0:20:13", "remaining_time": "3:29:34"}
34
+ {"current_steps": 68, "total_steps": 750, "loss": 1.87124502658844, "lr": 6.7e-06, "epoch": 0.18133333333333335, "percentage": 9.07, "elapsed_time": "0:20:40", "remaining_time": "3:27:19"}
35
+ {"current_steps": 70, "total_steps": 750, "loss": 1.4997594356536865, "lr": 6.900000000000001e-06, "epoch": 0.18666666666666668, "percentage": 9.33, "elapsed_time": "0:21:03", "remaining_time": "3:24:33"}
36
+ {"current_steps": 72, "total_steps": 750, "loss": 1.9693466424942017, "lr": 7.1e-06, "epoch": 0.192, "percentage": 9.6, "elapsed_time": "0:21:28", "remaining_time": "3:22:11"}
37
+ {"current_steps": 74, "total_steps": 750, "loss": 1.5585707426071167, "lr": 7.3e-06, "epoch": 0.19733333333333333, "percentage": 9.87, "elapsed_time": "0:21:53", "remaining_time": "3:19:59"}
38
+ {"current_steps": 76, "total_steps": 750, "loss": 1.7230855226516724, "lr": 7.5e-06, "epoch": 0.20266666666666666, "percentage": 10.13, "elapsed_time": "0:22:17", "remaining_time": "3:17:42"}
39
+ {"current_steps": 78, "total_steps": 750, "loss": 1.7421209812164307, "lr": 7.4998619077668045e-06, "epoch": 0.208, "percentage": 10.4, "elapsed_time": "0:22:42", "remaining_time": "3:15:42"}
40
+ {"current_steps": 80, "total_steps": 750, "loss": 1.8539875745773315, "lr": 7.499447643032373e-06, "epoch": 0.21333333333333335, "percentage": 10.67, "elapsed_time": "0:23:06", "remaining_time": "3:13:35"}
41
+ {"current_steps": 82, "total_steps": 750, "loss": 1.410808801651001, "lr": 7.498757241691129e-06, "epoch": 0.21866666666666668, "percentage": 10.93, "elapsed_time": "0:23:37", "remaining_time": "3:12:28"}
42
+ {"current_steps": 84, "total_steps": 750, "loss": 1.5474530458450317, "lr": 7.497790763563665e-06, "epoch": 0.224, "percentage": 11.2, "elapsed_time": "0:24:06", "remaining_time": "3:11:07"}
43
+ {"current_steps": 86, "total_steps": 750, "loss": 1.8224376440048218, "lr": 7.496548292391548e-06, "epoch": 0.22933333333333333, "percentage": 11.47, "elapsed_time": "0:24:28", "remaining_time": "3:08:59"}
44
+ {"current_steps": 88, "total_steps": 750, "loss": 1.5387835502624512, "lr": 7.4950299358300644e-06, "epoch": 0.23466666666666666, "percentage": 11.73, "elapsed_time": "0:24:52", "remaining_time": "3:07:08"}
45
+ {"current_steps": 90, "total_steps": 750, "loss": 1.4174927473068237, "lr": 7.493235825438903e-06, "epoch": 0.24, "percentage": 12.0, "elapsed_time": "0:25:24", "remaining_time": "3:06:17"}
46
+ {"current_steps": 92, "total_steps": 750, "loss": 1.3185020685195923, "lr": 7.491166116670744e-06, "epoch": 0.24533333333333332, "percentage": 12.27, "elapsed_time": "0:25:56", "remaining_time": "3:05:31"}
47
+ {"current_steps": 94, "total_steps": 750, "loss": 1.3037068843841553, "lr": 7.488820988857795e-06, "epoch": 0.25066666666666665, "percentage": 12.53, "elapsed_time": "0:26:15", "remaining_time": "3:03:13"}
48
+ {"current_steps": 96, "total_steps": 750, "loss": 1.7521201372146606, "lr": 7.486200645196252e-06, "epoch": 0.256, "percentage": 12.8, "elapsed_time": "0:26:40", "remaining_time": "3:01:41"}
49
+ {"current_steps": 98, "total_steps": 750, "loss": 1.2269163131713867, "lr": 7.483305312728692e-06, "epoch": 0.2613333333333333, "percentage": 13.07, "elapsed_time": "0:27:05", "remaining_time": "3:00:11"}
50
+ {"current_steps": 100, "total_steps": 750, "loss": 1.4020113945007324, "lr": 7.480135242324401e-06, "epoch": 0.26666666666666666, "percentage": 13.33, "elapsed_time": "0:27:32", "remaining_time": "2:58:59"}
51
+ {"current_steps": 102, "total_steps": 750, "loss": 1.236190915107727, "lr": 7.476690708657641e-06, "epoch": 0.272, "percentage": 13.6, "elapsed_time": "0:28:04", "remaining_time": "2:58:21"}
52
+ {"current_steps": 104, "total_steps": 750, "loss": 1.3339173793792725, "lr": 7.472972010183844e-06, "epoch": 0.2773333333333333, "percentage": 13.87, "elapsed_time": "0:28:32", "remaining_time": "2:57:16"}
53
+ {"current_steps": 106, "total_steps": 750, "loss": 1.5956217050552368, "lr": 7.468979469113756e-06, "epoch": 0.2826666666666667, "percentage": 14.13, "elapsed_time": "0:29:00", "remaining_time": "2:56:13"}
54
+ {"current_steps": 108, "total_steps": 750, "loss": 1.2588270902633667, "lr": 7.4647134313855186e-06, "epoch": 0.288, "percentage": 14.4, "elapsed_time": "0:29:31", "remaining_time": "2:55:30"}
55
+ {"current_steps": 110, "total_steps": 750, "loss": 1.6625038385391235, "lr": 7.4601742666346926e-06, "epoch": 0.29333333333333333, "percentage": 14.67, "elapsed_time": "0:29:58", "remaining_time": "2:54:23"}
56
+ {"current_steps": 112, "total_steps": 750, "loss": 1.3837028741836548, "lr": 7.455362368162235e-06, "epoch": 0.2986666666666667, "percentage": 14.93, "elapsed_time": "0:30:21", "remaining_time": "2:52:58"}
57
+ {"current_steps": 114, "total_steps": 750, "loss": 1.5355887413024902, "lr": 7.450278152900414e-06, "epoch": 0.304, "percentage": 15.2, "elapsed_time": "0:30:49", "remaining_time": "2:51:57"}
58
+ {"current_steps": 116, "total_steps": 750, "loss": 1.6411131620407104, "lr": 7.4449220613766924e-06, "epoch": 0.30933333333333335, "percentage": 15.47, "elapsed_time": "0:31:15", "remaining_time": "2:50:52"}
59
+ {"current_steps": 118, "total_steps": 750, "loss": 1.5240530967712402, "lr": 7.439294557675548e-06, "epoch": 0.31466666666666665, "percentage": 15.73, "elapsed_time": "0:31:41", "remaining_time": "2:49:46"}
60
+ {"current_steps": 120, "total_steps": 750, "loss": 1.1345856189727783, "lr": 7.43339612939827e-06, "epoch": 0.32, "percentage": 16.0, "elapsed_time": "0:32:09", "remaining_time": "2:48:52"}
61
+ {"current_steps": 122, "total_steps": 750, "loss": 1.1475586891174316, "lr": 7.427227287620707e-06, "epoch": 0.3253333333333333, "percentage": 16.27, "elapsed_time": "0:32:35", "remaining_time": "2:47:44"}
62
+ {"current_steps": 124, "total_steps": 750, "loss": 1.5467785596847534, "lr": 7.420788566848982e-06, "epoch": 0.33066666666666666, "percentage": 16.53, "elapsed_time": "0:33:00", "remaining_time": "2:46:40"}
63
+ {"current_steps": 126, "total_steps": 750, "loss": 1.6301590204238892, "lr": 7.414080524973188e-06, "epoch": 0.336, "percentage": 16.8, "elapsed_time": "0:33:27", "remaining_time": "2:45:42"}
64
+ {"current_steps": 128, "total_steps": 750, "loss": 1.4552572965621948, "lr": 7.407103743219038e-06, "epoch": 0.3413333333333333, "percentage": 17.07, "elapsed_time": "0:33:53", "remaining_time": "2:44:40"}
65
+ {"current_steps": 130, "total_steps": 750, "loss": 1.7707996368408203, "lr": 7.399858826097512e-06, "epoch": 0.3466666666666667, "percentage": 17.33, "elapsed_time": "0:34:16", "remaining_time": "2:43:27"}
66
+ {"current_steps": 132, "total_steps": 750, "loss": 1.4176101684570312, "lr": 7.392346401352474e-06, "epoch": 0.352, "percentage": 17.6, "elapsed_time": "0:34:41", "remaining_time": "2:42:25"}
67
+ {"current_steps": 134, "total_steps": 750, "loss": 1.4162875413894653, "lr": 7.3845671199062815e-06, "epoch": 0.35733333333333334, "percentage": 17.87, "elapsed_time": "0:35:04", "remaining_time": "2:41:16"}
68
+ {"current_steps": 136, "total_steps": 750, "loss": 1.4608713388442993, "lr": 7.376521655803392e-06, "epoch": 0.3626666666666667, "percentage": 18.13, "elapsed_time": "0:35:33", "remaining_time": "2:40:32"}
69
+ {"current_steps": 138, "total_steps": 750, "loss": 1.4052457809448242, "lr": 7.368210706151946e-06, "epoch": 0.368, "percentage": 18.4, "elapsed_time": "0:36:00", "remaining_time": "2:39:39"}
70
+ {"current_steps": 140, "total_steps": 750, "loss": 1.137465476989746, "lr": 7.359634991063379e-06, "epoch": 0.37333333333333335, "percentage": 18.67, "elapsed_time": "0:36:27", "remaining_time": "2:38:49"}
71
+ {"current_steps": 142, "total_steps": 750, "loss": 1.4206374883651733, "lr": 7.350795253590018e-06, "epoch": 0.37866666666666665, "percentage": 18.93, "elapsed_time": "0:36:52", "remaining_time": "2:37:53"}
72
+ {"current_steps": 144, "total_steps": 750, "loss": 1.020829439163208, "lr": 7.3416922596607055e-06, "epoch": 0.384, "percentage": 19.2, "elapsed_time": "0:37:19", "remaining_time": "2:37:05"}
73
+ {"current_steps": 146, "total_steps": 750, "loss": 1.558069109916687, "lr": 7.332326798014425e-06, "epoch": 0.3893333333333333, "percentage": 19.47, "elapsed_time": "0:37:46", "remaining_time": "2:36:18"}
74
+ {"current_steps": 148, "total_steps": 750, "loss": 1.5617731809616089, "lr": 7.322699680131971e-06, "epoch": 0.39466666666666667, "percentage": 19.73, "elapsed_time": "0:38:13", "remaining_time": "2:35:30"}
75
+ {"current_steps": 150, "total_steps": 750, "loss": 1.3473962545394897, "lr": 7.312811740165632e-06, "epoch": 0.4, "percentage": 20.0, "elapsed_time": "0:38:39", "remaining_time": "2:34:36"}
76
+ {"current_steps": 152, "total_steps": 750, "loss": 1.4550005197525024, "lr": 7.302663834866909e-06, "epoch": 0.4053333333333333, "percentage": 20.27, "elapsed_time": "0:39:04", "remaining_time": "2:33:42"}
77
+ {"current_steps": 154, "total_steps": 750, "loss": 1.2182159423828125, "lr": 7.2922568435122956e-06, "epoch": 0.4106666666666667, "percentage": 20.53, "elapsed_time": "0:39:34", "remaining_time": "2:33:11"}
78
+ {"current_steps": 156, "total_steps": 750, "loss": 1.1797568798065186, "lr": 7.281591667827079e-06, "epoch": 0.416, "percentage": 20.8, "elapsed_time": "0:39:57", "remaining_time": "2:32:09"}
79
+ {"current_steps": 158, "total_steps": 750, "loss": 1.3810535669326782, "lr": 7.270669231907212e-06, "epoch": 0.42133333333333334, "percentage": 21.07, "elapsed_time": "0:40:18", "remaining_time": "2:31:03"}
80
+ {"current_steps": 160, "total_steps": 750, "loss": 1.286556363105774, "lr": 7.2594904821392485e-06, "epoch": 0.4266666666666667, "percentage": 21.33, "elapsed_time": "0:40:40", "remaining_time": "2:30:00"}
81
+ {"current_steps": 162, "total_steps": 750, "loss": 1.4289522171020508, "lr": 7.248056387118338e-06, "epoch": 0.432, "percentage": 21.6, "elapsed_time": "0:41:07", "remaining_time": "2:29:17"}
82
+ {"current_steps": 164, "total_steps": 750, "loss": 1.243101716041565, "lr": 7.236367937564302e-06, "epoch": 0.43733333333333335, "percentage": 21.87, "elapsed_time": "0:41:39", "remaining_time": "2:28:52"}
83
+ {"current_steps": 166, "total_steps": 750, "loss": 1.0999314785003662, "lr": 7.224426146235791e-06, "epoch": 0.44266666666666665, "percentage": 22.13, "elapsed_time": "0:42:15", "remaining_time": "2:28:38"}
84
+ {"current_steps": 168, "total_steps": 750, "loss": 1.4978009462356567, "lr": 7.212232047842537e-06, "epoch": 0.448, "percentage": 22.4, "elapsed_time": "0:42:43", "remaining_time": "2:27:59"}
85
+ {"current_steps": 170, "total_steps": 750, "loss": 1.4373788833618164, "lr": 7.199786698955697e-06, "epoch": 0.4533333333333333, "percentage": 22.67, "elapsed_time": "0:43:09", "remaining_time": "2:27:14"}
86
+ {"current_steps": 172, "total_steps": 750, "loss": 1.4840056896209717, "lr": 7.187091177916301e-06, "epoch": 0.45866666666666667, "percentage": 22.93, "elapsed_time": "0:43:34", "remaining_time": "2:26:27"}
87
+ {"current_steps": 174, "total_steps": 750, "loss": 1.2911632061004639, "lr": 7.174146584741825e-06, "epoch": 0.464, "percentage": 23.2, "elapsed_time": "0:44:06", "remaining_time": "2:26:01"}
88
+ {"current_steps": 176, "total_steps": 750, "loss": 1.3464269638061523, "lr": 7.160954041030877e-06, "epoch": 0.4693333333333333, "percentage": 23.47, "elapsed_time": "0:44:36", "remaining_time": "2:25:27"}
89
+ {"current_steps": 178, "total_steps": 750, "loss": 1.4491277933120728, "lr": 7.1475146898660116e-06, "epoch": 0.4746666666666667, "percentage": 23.73, "elapsed_time": "0:45:02", "remaining_time": "2:24:43"}
90
+ {"current_steps": 180, "total_steps": 750, "loss": 1.318329930305481, "lr": 7.133829695714688e-06, "epoch": 0.48, "percentage": 24.0, "elapsed_time": "0:45:31", "remaining_time": "2:24:08"}
91
+ {"current_steps": 182, "total_steps": 750, "loss": 1.089525580406189, "lr": 7.1199002443283776e-06, "epoch": 0.48533333333333334, "percentage": 24.27, "elapsed_time": "0:45:57", "remaining_time": "2:23:24"}
92
+ {"current_steps": 184, "total_steps": 750, "loss": 1.356623888015747, "lr": 7.1057275426398146e-06, "epoch": 0.49066666666666664, "percentage": 24.53, "elapsed_time": "0:46:24", "remaining_time": "2:22:45"}
93
+ {"current_steps": 186, "total_steps": 750, "loss": 1.5700266361236572, "lr": 7.09131281865843e-06, "epoch": 0.496, "percentage": 24.8, "elapsed_time": "0:46:49", "remaining_time": "2:21:57"}
94
+ {"current_steps": 188, "total_steps": 750, "loss": 1.12882661819458, "lr": 7.0766573213639404e-06, "epoch": 0.5013333333333333, "percentage": 25.07, "elapsed_time": "0:47:11", "remaining_time": "2:21:04"}
95
+ {"current_steps": 190, "total_steps": 750, "loss": 1.3883382081985474, "lr": 7.061762320598134e-06, "epoch": 0.5066666666666667, "percentage": 25.33, "elapsed_time": "0:47:36", "remaining_time": "2:20:19"}
96
+ {"current_steps": 192, "total_steps": 750, "loss": 1.1575868129730225, "lr": 7.0466291069548415e-06, "epoch": 0.512, "percentage": 25.6, "elapsed_time": "0:48:02", "remaining_time": "2:19:38"}
97
+ {"current_steps": 194, "total_steps": 750, "loss": 1.458404302597046, "lr": 7.031258991668114e-06, "epoch": 0.5173333333333333, "percentage": 25.87, "elapsed_time": "0:48:31", "remaining_time": "2:19:04"}
98
+ {"current_steps": 196, "total_steps": 750, "loss": 1.7165433168411255, "lr": 7.015653306498608e-06, "epoch": 0.5226666666666666, "percentage": 26.13, "elapsed_time": "0:48:58", "remaining_time": "2:18:26"}
99
+ {"current_steps": 198, "total_steps": 750, "loss": 1.5549272298812866, "lr": 6.99981340361819e-06, "epoch": 0.528, "percentage": 26.4, "elapsed_time": "0:49:25", "remaining_time": "2:17:48"}
100
+ {"current_steps": 200, "total_steps": 750, "loss": 1.6522574424743652, "lr": 6.983740655492783e-06, "epoch": 0.5333333333333333, "percentage": 26.67, "elapsed_time": "0:49:51", "remaining_time": "2:17:07"}
101
+ {"current_steps": 202, "total_steps": 750, "loss": 1.5417842864990234, "lr": 6.967436454763442e-06, "epoch": 0.5386666666666666, "percentage": 26.93, "elapsed_time": "0:50:18", "remaining_time": "2:16:29"}
102
+ {"current_steps": 204, "total_steps": 750, "loss": 1.279434323310852, "lr": 6.950902214125689e-06, "epoch": 0.544, "percentage": 27.2, "elapsed_time": "0:50:42", "remaining_time": "2:15:42"}
103
+ {"current_steps": 206, "total_steps": 750, "loss": 1.5310899019241333, "lr": 6.9341393662071115e-06, "epoch": 0.5493333333333333, "percentage": 27.47, "elapsed_time": "0:51:08", "remaining_time": "2:15:03"}
104
+ {"current_steps": 208, "total_steps": 750, "loss": 1.1565041542053223, "lr": 6.9171493634432225e-06, "epoch": 0.5546666666666666, "percentage": 27.73, "elapsed_time": "0:51:34", "remaining_time": "2:14:24"}
105
+ {"current_steps": 210, "total_steps": 750, "loss": 1.3370246887207031, "lr": 6.899933677951622e-06, "epoch": 0.56, "percentage": 28.0, "elapsed_time": "0:52:04", "remaining_time": "2:13:53"}
106
+ {"current_steps": 212, "total_steps": 750, "loss": 1.330797791481018, "lr": 6.882493801404442e-06, "epoch": 0.5653333333333334, "percentage": 28.27, "elapsed_time": "0:52:31", "remaining_time": "2:13:18"}
107
+ {"current_steps": 214, "total_steps": 750, "loss": 1.4331892728805542, "lr": 6.8648312448990875e-06, "epoch": 0.5706666666666667, "percentage": 28.53, "elapsed_time": "0:52:59", "remaining_time": "2:12:43"}
108
+ {"current_steps": 216, "total_steps": 750, "loss": 1.5132179260253906, "lr": 6.846947538827327e-06, "epoch": 0.576, "percentage": 28.8, "elapsed_time": "0:53:24", "remaining_time": "2:12:03"}
109
+ {"current_steps": 218, "total_steps": 750, "loss": 1.2577152252197266, "lr": 6.828844232742671e-06, "epoch": 0.5813333333333334, "percentage": 29.07, "elapsed_time": "0:53:50", "remaining_time": "2:11:24"}
110
+ {"current_steps": 220, "total_steps": 750, "loss": 1.45955491065979, "lr": 6.810522895226115e-06, "epoch": 0.5866666666666667, "percentage": 29.33, "elapsed_time": "0:54:18", "remaining_time": "2:10:51"}
111
+ {"current_steps": 222, "total_steps": 750, "loss": 1.2781492471694946, "lr": 6.791985113750234e-06, "epoch": 0.592, "percentage": 29.6, "elapsed_time": "0:54:47", "remaining_time": "2:10:19"}
112
+ {"current_steps": 224, "total_steps": 750, "loss": 1.3114062547683716, "lr": 6.773232494541625e-06, "epoch": 0.5973333333333334, "percentage": 29.87, "elapsed_time": "0:55:16", "remaining_time": "2:09:47"}
113
+ {"current_steps": 226, "total_steps": 750, "loss": 1.4861233234405518, "lr": 6.754266662441742e-06, "epoch": 0.6026666666666667, "percentage": 30.13, "elapsed_time": "0:55:42", "remaining_time": "2:09:09"}
114
+ {"current_steps": 228, "total_steps": 750, "loss": 1.6322368383407593, "lr": 6.735089260766105e-06, "epoch": 0.608, "percentage": 30.4, "elapsed_time": "0:56:06", "remaining_time": "2:08:26"}
115
+ {"current_steps": 230, "total_steps": 750, "loss": 1.5255112648010254, "lr": 6.7157019511619145e-06, "epoch": 0.6133333333333333, "percentage": 30.67, "elapsed_time": "0:56:30", "remaining_time": "2:07:45"}
116
+ {"current_steps": 232, "total_steps": 750, "loss": 1.2472528219223022, "lr": 6.696106413464071e-06, "epoch": 0.6186666666666667, "percentage": 30.93, "elapsed_time": "0:56:57", "remaining_time": "2:07:11"}
117
+ {"current_steps": 234, "total_steps": 750, "loss": 1.5083709955215454, "lr": 6.676304345549635e-06, "epoch": 0.624, "percentage": 31.2, "elapsed_time": "0:57:22", "remaining_time": "2:06:32"}
118
+ {"current_steps": 236, "total_steps": 750, "loss": 1.216172218322754, "lr": 6.656297463190704e-06, "epoch": 0.6293333333333333, "percentage": 31.47, "elapsed_time": "0:57:50", "remaining_time": "2:05:58"}
119
+ {"current_steps": 238, "total_steps": 750, "loss": 1.2348906993865967, "lr": 6.63608749990575e-06, "epoch": 0.6346666666666667, "percentage": 31.73, "elapsed_time": "0:58:22", "remaining_time": "2:05:34"}
120
+ {"current_steps": 240, "total_steps": 750, "loss": 1.3027633428573608, "lr": 6.615676206809415e-06, "epoch": 0.64, "percentage": 32.0, "elapsed_time": "0:58:50", "remaining_time": "2:05:02"}
121
+ {"current_steps": 242, "total_steps": 750, "loss": 1.1347825527191162, "lr": 6.595065352460786e-06, "epoch": 0.6453333333333333, "percentage": 32.27, "elapsed_time": "0:59:21", "remaining_time": "2:04:35"}
122
+ {"current_steps": 244, "total_steps": 750, "loss": 1.271606206893921, "lr": 6.574256722710161e-06, "epoch": 0.6506666666666666, "percentage": 32.53, "elapsed_time": "0:59:52", "remaining_time": "2:04:09"}
123
+ {"current_steps": 246, "total_steps": 750, "loss": 1.297676682472229, "lr": 6.553252120544297e-06, "epoch": 0.656, "percentage": 32.8, "elapsed_time": "1:00:20", "remaining_time": "2:03:36"}
124
+ {"current_steps": 248, "total_steps": 750, "loss": 1.3999580144882202, "lr": 6.5320533659302055e-06, "epoch": 0.6613333333333333, "percentage": 33.07, "elapsed_time": "1:00:47", "remaining_time": "2:03:04"}
125
+ {"current_steps": 250, "total_steps": 750, "loss": 1.4784775972366333, "lr": 6.510662295657448e-06, "epoch": 0.6666666666666666, "percentage": 33.33, "elapsed_time": "1:01:14", "remaining_time": "2:02:28"}
126
+ {"current_steps": 252, "total_steps": 750, "loss": 1.3089486360549927, "lr": 6.489080763178991e-06, "epoch": 0.672, "percentage": 33.6, "elapsed_time": "1:01:43", "remaining_time": "2:01:57"}
127
+ {"current_steps": 254, "total_steps": 750, "loss": 1.5869042873382568, "lr": 6.467310638450608e-06, "epoch": 0.6773333333333333, "percentage": 33.87, "elapsed_time": "1:02:08", "remaining_time": "2:01:20"}
128
+ {"current_steps": 256, "total_steps": 750, "loss": 1.491261601448059, "lr": 6.44535380776886e-06, "epoch": 0.6826666666666666, "percentage": 34.13, "elapsed_time": "1:02:33", "remaining_time": "2:00:44"}
129
+ {"current_steps": 258, "total_steps": 750, "loss": 1.3518444299697876, "lr": 6.42321217360765e-06, "epoch": 0.688, "percentage": 34.4, "elapsed_time": "1:02:56", "remaining_time": "2:00:00"}
130
+ {"current_steps": 260, "total_steps": 750, "loss": 1.1858634948730469, "lr": 6.40088765445338e-06, "epoch": 0.6933333333333334, "percentage": 34.67, "elapsed_time": "1:03:23", "remaining_time": "1:59:28"}
131
+ {"current_steps": 262, "total_steps": 750, "loss": 1.3287537097930908, "lr": 6.37838218463873e-06, "epoch": 0.6986666666666667, "percentage": 34.93, "elapsed_time": "1:03:53", "remaining_time": "1:58:59"}
132
+ {"current_steps": 264, "total_steps": 750, "loss": 1.0604113340377808, "lr": 6.355697714175047e-06, "epoch": 0.704, "percentage": 35.2, "elapsed_time": "1:04:20", "remaining_time": "1:58:26"}
133
+ {"current_steps": 266, "total_steps": 750, "loss": 1.5299183130264282, "lr": 6.332836208583389e-06, "epoch": 0.7093333333333334, "percentage": 35.47, "elapsed_time": "1:04:47", "remaining_time": "1:57:53"}
134
+ {"current_steps": 268, "total_steps": 750, "loss": 1.2309736013412476, "lr": 6.309799648724218e-06, "epoch": 0.7146666666666667, "percentage": 35.73, "elapsed_time": "1:05:16", "remaining_time": "1:57:24"}
135
+ {"current_steps": 270, "total_steps": 750, "loss": 1.0823864936828613, "lr": 6.2865900306257665e-06, "epoch": 0.72, "percentage": 36.0, "elapsed_time": "1:05:44", "remaining_time": "1:56:51"}
136
+ {"current_steps": 272, "total_steps": 750, "loss": 1.2900185585021973, "lr": 6.263209365311089e-06, "epoch": 0.7253333333333334, "percentage": 36.27, "elapsed_time": "1:06:13", "remaining_time": "1:56:23"}
137
+ {"current_steps": 274, "total_steps": 750, "loss": 1.2881044149398804, "lr": 6.2396596786238205e-06, "epoch": 0.7306666666666667, "percentage": 36.53, "elapsed_time": "1:06:38", "remaining_time": "1:55:46"}
138
+ {"current_steps": 276, "total_steps": 750, "loss": 1.3503191471099854, "lr": 6.215943011052632e-06, "epoch": 0.736, "percentage": 36.8, "elapsed_time": "1:07:08", "remaining_time": "1:55:19"}
139
+ {"current_steps": 278, "total_steps": 750, "loss": 1.33072030544281, "lr": 6.192061417554441e-06, "epoch": 0.7413333333333333, "percentage": 37.07, "elapsed_time": "1:07:33", "remaining_time": "1:54:41"}
140
+ {"current_steps": 280, "total_steps": 750, "loss": 1.2632883787155151, "lr": 6.16801696737636e-06, "epoch": 0.7466666666666667, "percentage": 37.33, "elapsed_time": "1:08:01", "remaining_time": "1:54:10"}
141
+ {"current_steps": 282, "total_steps": 750, "loss": 1.285319447517395, "lr": 6.143811743876392e-06, "epoch": 0.752, "percentage": 37.6, "elapsed_time": "1:08:27", "remaining_time": "1:53:36"}
142
+ {"current_steps": 284, "total_steps": 750, "loss": 1.0543291568756104, "lr": 6.119447844342924e-06, "epoch": 0.7573333333333333, "percentage": 37.87, "elapsed_time": "1:08:57", "remaining_time": "1:53:09"}
143
+ {"current_steps": 286, "total_steps": 750, "loss": 1.5204623937606812, "lr": 6.094927379813005e-06, "epoch": 0.7626666666666667, "percentage": 38.13, "elapsed_time": "1:09:22", "remaining_time": "1:52:32"}
144
+ {"current_steps": 288, "total_steps": 750, "loss": 0.9127747416496277, "lr": 6.070252474889433e-06, "epoch": 0.768, "percentage": 38.4, "elapsed_time": "1:09:44", "remaining_time": "1:51:52"}
145
+ {"current_steps": 290, "total_steps": 750, "loss": 1.2986137866973877, "lr": 6.045425267556659e-06, "epoch": 0.7733333333333333, "percentage": 38.67, "elapsed_time": "1:10:10", "remaining_time": "1:51:18"}
146
+ {"current_steps": 292, "total_steps": 750, "loss": 1.482648253440857, "lr": 6.020447908995551e-06, "epoch": 0.7786666666666666, "percentage": 38.93, "elapsed_time": "1:10:36", "remaining_time": "1:50:45"}
147
+ {"current_steps": 294, "total_steps": 750, "loss": 1.1602120399475098, "lr": 5.995322563396996e-06, "epoch": 0.784, "percentage": 39.2, "elapsed_time": "1:11:01", "remaining_time": "1:50:09"}
148
+ {"current_steps": 296, "total_steps": 750, "loss": 1.2999836206436157, "lr": 5.970051407774374e-06, "epoch": 0.7893333333333333, "percentage": 39.47, "elapsed_time": "1:11:25", "remaining_time": "1:49:33"}
149
+ {"current_steps": 298, "total_steps": 750, "loss": 1.464931607246399, "lr": 5.9446366317749464e-06, "epoch": 0.7946666666666666, "percentage": 39.73, "elapsed_time": "1:11:52", "remaining_time": "1:49:01"}
150
+ {"current_steps": 300, "total_steps": 750, "loss": 1.494107961654663, "lr": 5.919080437490115e-06, "epoch": 0.8, "percentage": 40.0, "elapsed_time": "1:12:17", "remaining_time": "1:48:25"}
151
+ {"current_steps": 302, "total_steps": 750, "loss": 1.3205267190933228, "lr": 5.893385039264628e-06, "epoch": 0.8053333333333333, "percentage": 40.27, "elapsed_time": "1:12:45", "remaining_time": "1:47:55"}
152
+ {"current_steps": 304, "total_steps": 750, "loss": 1.392236351966858, "lr": 5.867552663504711e-06, "epoch": 0.8106666666666666, "percentage": 40.53, "elapsed_time": "1:13:13", "remaining_time": "1:47:25"}
153
+ {"current_steps": 306, "total_steps": 750, "loss": 1.054479718208313, "lr": 5.841585548485161e-06, "epoch": 0.816, "percentage": 40.8, "elapsed_time": "1:13:35", "remaining_time": "1:46:46"}
154
+ {"current_steps": 308, "total_steps": 750, "loss": 1.4559931755065918, "lr": 5.815485944155406e-06, "epoch": 0.8213333333333334, "percentage": 41.07, "elapsed_time": "1:14:01", "remaining_time": "1:46:14"}
155
+ {"current_steps": 310, "total_steps": 750, "loss": 1.2822375297546387, "lr": 5.78925611194456e-06, "epoch": 0.8266666666666667, "percentage": 41.33, "elapsed_time": "1:14:30", "remaining_time": "1:45:45"}
156
+ {"current_steps": 312, "total_steps": 750, "loss": 1.1534578800201416, "lr": 5.762898324565472e-06, "epoch": 0.832, "percentage": 41.6, "elapsed_time": "1:15:05", "remaining_time": "1:45:24"}
157
+ {"current_steps": 314, "total_steps": 750, "loss": 1.4816879034042358, "lr": 5.7364148658178136e-06, "epoch": 0.8373333333333334, "percentage": 41.87, "elapsed_time": "1:15:33", "remaining_time": "1:44:54"}
158
+ {"current_steps": 316, "total_steps": 750, "loss": 1.3202126026153564, "lr": 5.7098080303901845e-06, "epoch": 0.8426666666666667, "percentage": 42.13, "elapsed_time": "1:16:00", "remaining_time": "1:44:23"}
159
+ {"current_steps": 318, "total_steps": 750, "loss": 1.548636794090271, "lr": 5.683080123661299e-06, "epoch": 0.848, "percentage": 42.4, "elapsed_time": "1:16:24", "remaining_time": "1:43:48"}
160
+ {"current_steps": 320, "total_steps": 750, "loss": 1.1134190559387207, "lr": 5.656233461500226e-06, "epoch": 0.8533333333333334, "percentage": 42.67, "elapsed_time": "1:16:58", "remaining_time": "1:43:25"}
161
+ {"current_steps": 322, "total_steps": 750, "loss": 1.427626132965088, "lr": 5.6292703700657276e-06, "epoch": 0.8586666666666667, "percentage": 42.93, "elapsed_time": "1:17:21", "remaining_time": "1:42:49"}
162
+ {"current_steps": 324, "total_steps": 750, "loss": 1.2644096612930298, "lr": 5.602193185604715e-06, "epoch": 0.864, "percentage": 43.2, "elapsed_time": "1:17:46", "remaining_time": "1:42:15"}
163
+ {"current_steps": 326, "total_steps": 750, "loss": 1.0416737794876099, "lr": 5.575004254249812e-06, "epoch": 0.8693333333333333, "percentage": 43.47, "elapsed_time": "1:18:11", "remaining_time": "1:41:42"}
164
+ {"current_steps": 328, "total_steps": 750, "loss": 1.113074779510498, "lr": 5.547705931816078e-06, "epoch": 0.8746666666666667, "percentage": 43.73, "elapsed_time": "1:18:40", "remaining_time": "1:41:13"}
165
+ {"current_steps": 330, "total_steps": 750, "loss": 1.0335323810577393, "lr": 5.520300583596882e-06, "epoch": 0.88, "percentage": 44.0, "elapsed_time": "1:19:04", "remaining_time": "1:40:38"}
166
+ {"current_steps": 332, "total_steps": 750, "loss": 1.1414536237716675, "lr": 5.49279058415896e-06, "epoch": 0.8853333333333333, "percentage": 44.27, "elapsed_time": "1:19:31", "remaining_time": "1:40:07"}
167
+ {"current_steps": 334, "total_steps": 750, "loss": 0.8871914148330688, "lr": 5.465178317136669e-06, "epoch": 0.8906666666666667, "percentage": 44.53, "elapsed_time": "1:19:57", "remaining_time": "1:39:35"}
168
+ {"current_steps": 336, "total_steps": 750, "loss": 1.3287149667739868, "lr": 5.437466175025457e-06, "epoch": 0.896, "percentage": 44.8, "elapsed_time": "1:20:26", "remaining_time": "1:39:07"}
169
+ {"current_steps": 338, "total_steps": 750, "loss": 1.3277924060821533, "lr": 5.4096565589745565e-06, "epoch": 0.9013333333333333, "percentage": 45.07, "elapsed_time": "1:20:56", "remaining_time": "1:38:40"}
170
+ {"current_steps": 340, "total_steps": 750, "loss": 1.6219979524612427, "lr": 5.3817518785789386e-06, "epoch": 0.9066666666666666, "percentage": 45.33, "elapsed_time": "1:21:21", "remaining_time": "1:38:06"}
171
+ {"current_steps": 342, "total_steps": 750, "loss": 1.1041191816329956, "lr": 5.3537545516705285e-06, "epoch": 0.912, "percentage": 45.6, "elapsed_time": "1:21:48", "remaining_time": "1:37:35"}
172
+ {"current_steps": 344, "total_steps": 750, "loss": 1.5842766761779785, "lr": 5.325667004108713e-06, "epoch": 0.9173333333333333, "percentage": 45.87, "elapsed_time": "1:22:13", "remaining_time": "1:37:02"}
173
+ {"current_steps": 346, "total_steps": 750, "loss": 1.5403611660003662, "lr": 5.297491669570145e-06, "epoch": 0.9226666666666666, "percentage": 46.13, "elapsed_time": "1:22:40", "remaining_time": "1:36:31"}
174
+ {"current_steps": 348, "total_steps": 750, "loss": 1.387043833732605, "lr": 5.2692309893378806e-06, "epoch": 0.928, "percentage": 46.4, "elapsed_time": "1:23:04", "remaining_time": "1:35:58"}
175
+ {"current_steps": 350, "total_steps": 750, "loss": 1.1895140409469604, "lr": 5.240887412089846e-06, "epoch": 0.9333333333333333, "percentage": 46.67, "elapsed_time": "1:23:35", "remaining_time": "1:35:32"}
176
+ {"current_steps": 352, "total_steps": 750, "loss": 1.45750892162323, "lr": 5.2124633936866694e-06, "epoch": 0.9386666666666666, "percentage": 46.93, "elapsed_time": "1:24:01", "remaining_time": "1:35:00"}
177
+ {"current_steps": 354, "total_steps": 750, "loss": 1.2903810739517212, "lr": 5.183961396958894e-06, "epoch": 0.944, "percentage": 47.2, "elapsed_time": "1:24:31", "remaining_time": "1:34:32"}
178
+ {"current_steps": 356, "total_steps": 750, "loss": 1.0638062953948975, "lr": 5.155383891493582e-06, "epoch": 0.9493333333333334, "percentage": 47.47, "elapsed_time": "1:24:55", "remaining_time": "1:33:58"}
179
+ {"current_steps": 358, "total_steps": 750, "loss": 1.1334384679794312, "lr": 5.126733353420332e-06, "epoch": 0.9546666666666667, "percentage": 47.73, "elapsed_time": "1:25:24", "remaining_time": "1:33:30"}
180
+ {"current_steps": 360, "total_steps": 750, "loss": 1.616370677947998, "lr": 5.098012265196735e-06, "epoch": 0.96, "percentage": 48.0, "elapsed_time": "1:25:47", "remaining_time": "1:32:56"}
181
+ {"current_steps": 362, "total_steps": 750, "loss": 1.2166802883148193, "lr": 5.069223115393278e-06, "epoch": 0.9653333333333334, "percentage": 48.27, "elapsed_time": "1:26:17", "remaining_time": "1:32:29"}
182
+ {"current_steps": 364, "total_steps": 750, "loss": 1.460686206817627, "lr": 5.0403683984777156e-06, "epoch": 0.9706666666666667, "percentage": 48.53, "elapsed_time": "1:26:46", "remaining_time": "1:32:00"}
183
+ {"current_steps": 366, "total_steps": 750, "loss": 1.5133981704711914, "lr": 5.011450614598942e-06, "epoch": 0.976, "percentage": 48.8, "elapsed_time": "1:27:12", "remaining_time": "1:31:29"}
184
+ {"current_steps": 368, "total_steps": 750, "loss": 1.3987483978271484, "lr": 4.982472269370356e-06, "epoch": 0.9813333333333333, "percentage": 49.07, "elapsed_time": "1:27:40", "remaining_time": "1:31:00"}
185
+ {"current_steps": 370, "total_steps": 750, "loss": 1.187402606010437, "lr": 4.9534358736527624e-06, "epoch": 0.9866666666666667, "percentage": 49.33, "elapsed_time": "1:28:02", "remaining_time": "1:30:25"}
186
+ {"current_steps": 372, "total_steps": 750, "loss": 1.3969718217849731, "lr": 4.924343943336814e-06, "epoch": 0.992, "percentage": 49.6, "elapsed_time": "1:28:32", "remaining_time": "1:29:58"}
187
+ {"current_steps": 374, "total_steps": 750, "loss": 1.1329926252365112, "lr": 4.895198999125022e-06, "epoch": 0.9973333333333333, "percentage": 49.87, "elapsed_time": "1:29:03", "remaining_time": "1:29:32"}
188
+ {"current_steps": 376, "total_steps": 750, "loss": 1.426386833190918, "lr": 4.866003566313341e-06, "epoch": 1.0026666666666666, "percentage": 50.13, "elapsed_time": "1:29:29", "remaining_time": "1:29:01"}
189
+ {"current_steps": 378, "total_steps": 750, "loss": 1.2209362983703613, "lr": 4.83676017457237e-06, "epoch": 1.008, "percentage": 50.4, "elapsed_time": "1:29:53", "remaining_time": "1:28:27"}
190
+ {"current_steps": 380, "total_steps": 750, "loss": 1.1464167833328247, "lr": 4.807471357728167e-06, "epoch": 1.0133333333333334, "percentage": 50.67, "elapsed_time": "1:30:20", "remaining_time": "1:27:57"}
191
+ {"current_steps": 382, "total_steps": 750, "loss": 1.229437232017517, "lr": 4.778139653542688e-06, "epoch": 1.0186666666666666, "percentage": 50.93, "elapsed_time": "1:30:48", "remaining_time": "1:27:28"}
192
+ {"current_steps": 384, "total_steps": 750, "loss": 0.9976918697357178, "lr": 4.7487676034939156e-06, "epoch": 1.024, "percentage": 51.2, "elapsed_time": "1:31:13", "remaining_time": "1:26:56"}
193
+ {"current_steps": 386, "total_steps": 750, "loss": 1.025425672531128, "lr": 4.719357752555646e-06, "epoch": 1.0293333333333334, "percentage": 51.47, "elapsed_time": "1:31:38", "remaining_time": "1:26:25"}
194
+ {"current_steps": 388, "total_steps": 750, "loss": 1.0078941583633423, "lr": 4.689912648976973e-06, "epoch": 1.0346666666666666, "percentage": 51.73, "elapsed_time": "1:32:05", "remaining_time": "1:25:55"}
195
+ {"current_steps": 390, "total_steps": 750, "loss": 1.1683357954025269, "lr": 4.660434844061493e-06, "epoch": 1.04, "percentage": 52.0, "elapsed_time": "1:32:38", "remaining_time": "1:25:30"}
196
+ {"current_steps": 392, "total_steps": 750, "loss": 1.126926302909851, "lr": 4.630926891946251e-06, "epoch": 1.0453333333333332, "percentage": 52.27, "elapsed_time": "1:33:13", "remaining_time": "1:25:08"}
197
+ {"current_steps": 394, "total_steps": 750, "loss": 1.2865811586380005, "lr": 4.601391349380425e-06, "epoch": 1.0506666666666666, "percentage": 52.53, "elapsed_time": "1:33:41", "remaining_time": "1:24:38"}
198
+ {"current_steps": 396, "total_steps": 750, "loss": 1.5480324029922485, "lr": 4.571830775503799e-06, "epoch": 1.056, "percentage": 52.8, "elapsed_time": "1:34:05", "remaining_time": "1:24:06"}
199
+ {"current_steps": 398, "total_steps": 750, "loss": 1.2431285381317139, "lr": 4.542247731625022e-06, "epoch": 1.0613333333333332, "percentage": 53.07, "elapsed_time": "1:34:37", "remaining_time": "1:23:41"}
200
+ {"current_steps": 400, "total_steps": 750, "loss": 1.4781147241592407, "lr": 4.512644780999687e-06, "epoch": 1.0666666666666667, "percentage": 53.33, "elapsed_time": "1:35:02", "remaining_time": "1:23:09"}
201
+ {"current_steps": 402, "total_steps": 750, "loss": 1.0834038257598877, "lr": 4.48302448860822e-06, "epoch": 1.072, "percentage": 53.6, "elapsed_time": "1:35:25", "remaining_time": "1:22:35"}
202
+ {"current_steps": 404, "total_steps": 750, "loss": 1.366401195526123, "lr": 4.453389420933649e-06, "epoch": 1.0773333333333333, "percentage": 53.87, "elapsed_time": "1:35:51", "remaining_time": "1:22:05"}
203
+ {"current_steps": 406, "total_steps": 750, "loss": 1.0470842123031616, "lr": 4.423742145739222e-06, "epoch": 1.0826666666666667, "percentage": 54.13, "elapsed_time": "1:36:18", "remaining_time": "1:21:36"}
204
+ {"current_steps": 408, "total_steps": 750, "loss": 1.0181280374526978, "lr": 4.394085231845921e-06, "epoch": 1.088, "percentage": 54.4, "elapsed_time": "1:36:46", "remaining_time": "1:21:06"}
205
+ {"current_steps": 410, "total_steps": 750, "loss": 1.034490704536438, "lr": 4.364421248909884e-06, "epoch": 1.0933333333333333, "percentage": 54.67, "elapsed_time": "1:37:13", "remaining_time": "1:20:37"}
206
+ {"current_steps": 412, "total_steps": 750, "loss": 1.3856372833251953, "lr": 4.334752767199752e-06, "epoch": 1.0986666666666667, "percentage": 54.93, "elapsed_time": "1:37:36", "remaining_time": "1:20:04"}
207
+ {"current_steps": 414, "total_steps": 750, "loss": 1.139554500579834, "lr": 4.305082357373971e-06, "epoch": 1.104, "percentage": 55.2, "elapsed_time": "1:38:00", "remaining_time": "1:19:32"}
208
+ {"current_steps": 416, "total_steps": 750, "loss": 1.3391931056976318, "lr": 4.275412590258048e-06, "epoch": 1.1093333333333333, "percentage": 55.47, "elapsed_time": "1:38:28", "remaining_time": "1:19:04"}
209
+ {"current_steps": 418, "total_steps": 750, "loss": 1.5631487369537354, "lr": 4.2457460366218e-06, "epoch": 1.1146666666666667, "percentage": 55.73, "elapsed_time": "1:38:52", "remaining_time": "1:18:32"}
210
+ {"current_steps": 420, "total_steps": 750, "loss": 1.3829050064086914, "lr": 4.2160852669566135e-06, "epoch": 1.12, "percentage": 56.0, "elapsed_time": "1:39:20", "remaining_time": "1:18:02"}
211
+ {"current_steps": 422, "total_steps": 750, "loss": 1.0705004930496216, "lr": 4.1864328512527115e-06, "epoch": 1.1253333333333333, "percentage": 56.27, "elapsed_time": "1:39:42", "remaining_time": "1:17:30"}
212
+ {"current_steps": 424, "total_steps": 750, "loss": 1.4946973323822021, "lr": 4.156791358776483e-06, "epoch": 1.1306666666666667, "percentage": 56.53, "elapsed_time": "1:40:09", "remaining_time": "1:17:00"}
213
+ {"current_steps": 426, "total_steps": 750, "loss": 1.3657976388931274, "lr": 4.127163357847858e-06, "epoch": 1.1360000000000001, "percentage": 56.8, "elapsed_time": "1:40:38", "remaining_time": "1:16:33"}
214
+ {"current_steps": 428, "total_steps": 750, "loss": 1.211698055267334, "lr": 4.09755141561778e-06, "epoch": 1.1413333333333333, "percentage": 57.07, "elapsed_time": "1:41:05", "remaining_time": "1:16:03"}
215
+ {"current_steps": 430, "total_steps": 750, "loss": 1.5211821794509888, "lr": 4.0679580978457656e-06, "epoch": 1.1466666666666667, "percentage": 57.33, "elapsed_time": "1:41:30", "remaining_time": "1:15:32"}
216
+ {"current_steps": 432, "total_steps": 750, "loss": 1.0360568761825562, "lr": 4.038385968677596e-06, "epoch": 1.152, "percentage": 57.6, "elapsed_time": "1:42:03", "remaining_time": "1:15:07"}
217
+ {"current_steps": 434, "total_steps": 750, "loss": 1.6035184860229492, "lr": 4.008837590423144e-06, "epoch": 1.1573333333333333, "percentage": 57.87, "elapsed_time": "1:42:28", "remaining_time": "1:14:37"}
218
+ {"current_steps": 436, "total_steps": 750, "loss": 1.157009243965149, "lr": 3.979315523334355e-06, "epoch": 1.1626666666666667, "percentage": 58.13, "elapsed_time": "1:43:01", "remaining_time": "1:14:11"}
219
+ {"current_steps": 438, "total_steps": 750, "loss": 1.4968481063842773, "lr": 3.9498223253834165e-06, "epoch": 1.168, "percentage": 58.4, "elapsed_time": "1:43:27", "remaining_time": "1:13:41"}
220
+ {"current_steps": 440, "total_steps": 750, "loss": 1.229710340499878, "lr": 3.9203605520411165e-06, "epoch": 1.1733333333333333, "percentage": 58.67, "elapsed_time": "1:43:56", "remaining_time": "1:13:14"}
221
+ {"current_steps": 442, "total_steps": 750, "loss": 1.5977742671966553, "lr": 3.890932756055422e-06, "epoch": 1.1786666666666668, "percentage": 58.93, "elapsed_time": "1:44:21", "remaining_time": "1:12:43"}
222
+ {"current_steps": 444, "total_steps": 750, "loss": 1.2765097618103027, "lr": 3.861541487230299e-06, "epoch": 1.184, "percentage": 59.2, "elapsed_time": "1:44:42", "remaining_time": "1:12:09"}
223
+ {"current_steps": 446, "total_steps": 750, "loss": 1.3562519550323486, "lr": 3.832189292204773e-06, "epoch": 1.1893333333333334, "percentage": 59.47, "elapsed_time": "1:45:07", "remaining_time": "1:11:39"}
224
+ {"current_steps": 448, "total_steps": 750, "loss": 1.6301568746566772, "lr": 3.802878714232277e-06, "epoch": 1.1946666666666665, "percentage": 59.73, "elapsed_time": "1:45:32", "remaining_time": "1:11:08"}
225
+ {"current_steps": 450, "total_steps": 750, "loss": 1.264549732208252, "lr": 3.7736122929602887e-06, "epoch": 1.2, "percentage": 60.0, "elapsed_time": "1:46:01", "remaining_time": "1:10:41"}
226
+ {"current_steps": 452, "total_steps": 750, "loss": 1.4258700609207153, "lr": 3.744392564210283e-06, "epoch": 1.2053333333333334, "percentage": 60.27, "elapsed_time": "1:46:29", "remaining_time": "1:10:12"}
227
+ {"current_steps": 454, "total_steps": 750, "loss": 1.4381071329116821, "lr": 3.7152220597580023e-06, "epoch": 1.2106666666666666, "percentage": 60.53, "elapsed_time": "1:46:56", "remaining_time": "1:09:43"}
228
+ {"current_steps": 456, "total_steps": 750, "loss": 1.4425290822982788, "lr": 3.686103307114104e-06, "epoch": 1.216, "percentage": 60.8, "elapsed_time": "1:47:24", "remaining_time": "1:09:15"}
229
+ {"current_steps": 458, "total_steps": 750, "loss": 1.2543679475784302, "lr": 3.6570388293051458e-06, "epoch": 1.2213333333333334, "percentage": 61.07, "elapsed_time": "1:47:55", "remaining_time": "1:08:48"}
230
+ {"current_steps": 460, "total_steps": 750, "loss": 1.3114312887191772, "lr": 3.628031144654986e-06, "epoch": 1.2266666666666666, "percentage": 61.33, "elapsed_time": "1:48:19", "remaining_time": "1:08:17"}
231
+ {"current_steps": 462, "total_steps": 750, "loss": 1.4582160711288452, "lr": 3.5990827665665756e-06, "epoch": 1.232, "percentage": 61.6, "elapsed_time": "1:48:48", "remaining_time": "1:07:49"}
232
+ {"current_steps": 464, "total_steps": 750, "loss": 0.9986866116523743, "lr": 3.570196203304182e-06, "epoch": 1.2373333333333334, "percentage": 61.87, "elapsed_time": "1:49:17", "remaining_time": "1:07:21"}
233
+ {"current_steps": 466, "total_steps": 750, "loss": 1.124342918395996, "lr": 3.541373957776059e-06, "epoch": 1.2426666666666666, "percentage": 62.13, "elapsed_time": "1:49:41", "remaining_time": "1:06:51"}
234
+ {"current_steps": 468, "total_steps": 750, "loss": 1.3204665184020996, "lr": 3.512618527317581e-06, "epoch": 1.248, "percentage": 62.4, "elapsed_time": "1:50:07", "remaining_time": "1:06:21"}
235
+ {"current_steps": 470, "total_steps": 750, "loss": 1.1635738611221313, "lr": 3.483932403474856e-06, "epoch": 1.2533333333333334, "percentage": 62.67, "elapsed_time": "1:50:35", "remaining_time": "1:05:53"}
236
+ {"current_steps": 472, "total_steps": 750, "loss": 1.0901280641555786, "lr": 3.4553180717888407e-06, "epoch": 1.2586666666666666, "percentage": 62.93, "elapsed_time": "1:51:04", "remaining_time": "1:05:25"}
237
+ {"current_steps": 474, "total_steps": 750, "loss": 1.0830848217010498, "lr": 3.4267780115799846e-06, "epoch": 1.264, "percentage": 63.2, "elapsed_time": "1:51:25", "remaining_time": "1:04:52"}
238
+ {"current_steps": 476, "total_steps": 750, "loss": 1.3727480173110962, "lr": 3.398314695733401e-06, "epoch": 1.2693333333333334, "percentage": 63.47, "elapsed_time": "1:51:52", "remaining_time": "1:04:24"}
239
+ {"current_steps": 478, "total_steps": 750, "loss": 1.6773924827575684, "lr": 3.3699305904846027e-06, "epoch": 1.2746666666666666, "percentage": 63.73, "elapsed_time": "1:52:16", "remaining_time": "1:03:53"}
240
+ {"current_steps": 480, "total_steps": 750, "loss": 1.2972683906555176, "lr": 3.341628155205811e-06, "epoch": 1.28, "percentage": 64.0, "elapsed_time": "1:52:44", "remaining_time": "1:03:25"}
241
+ {"current_steps": 482, "total_steps": 750, "loss": 1.2166887521743774, "lr": 3.3134098421928692e-06, "epoch": 1.2853333333333334, "percentage": 64.27, "elapsed_time": "1:53:11", "remaining_time": "1:02:56"}
242
+ {"current_steps": 484, "total_steps": 750, "loss": 1.3262923955917358, "lr": 3.2852780964527466e-06, "epoch": 1.2906666666666666, "percentage": 64.53, "elapsed_time": "1:53:36", "remaining_time": "1:02:26"}
243
+ {"current_steps": 486, "total_steps": 750, "loss": 1.2588746547698975, "lr": 3.2572353554916966e-06, "epoch": 1.296, "percentage": 64.8, "elapsed_time": "1:54:03", "remaining_time": "1:01:57"}
244
+ {"current_steps": 488, "total_steps": 750, "loss": 0.8871427774429321, "lr": 3.229284049104057e-06, "epoch": 1.3013333333333335, "percentage": 65.07, "elapsed_time": "1:54:28", "remaining_time": "1:01:27"}
245
+ {"current_steps": 490, "total_steps": 750, "loss": 1.3596696853637695, "lr": 3.2014265991617147e-06, "epoch": 1.3066666666666666, "percentage": 65.33, "elapsed_time": "1:54:56", "remaining_time": "1:00:59"}
246
+ {"current_steps": 492, "total_steps": 750, "loss": 1.5204687118530273, "lr": 3.17366541940426e-06, "epoch": 1.312, "percentage": 65.6, "elapsed_time": "1:55:21", "remaining_time": "1:00:29"}
247
+ {"current_steps": 494, "total_steps": 750, "loss": 1.2328211069107056, "lr": 3.1460029152298484e-06, "epoch": 1.3173333333333335, "percentage": 65.87, "elapsed_time": "1:55:49", "remaining_time": "1:00:01"}
248
+ {"current_steps": 496, "total_steps": 750, "loss": 1.2612500190734863, "lr": 3.1184414834867804e-06, "epoch": 1.3226666666666667, "percentage": 66.13, "elapsed_time": "1:56:12", "remaining_time": "0:59:30"}
249
+ {"current_steps": 498, "total_steps": 750, "loss": 1.230236291885376, "lr": 3.0909835122658214e-06, "epoch": 1.328, "percentage": 66.4, "elapsed_time": "1:56:37", "remaining_time": "0:59:01"}
250
+ {"current_steps": 500, "total_steps": 750, "loss": 1.016282558441162, "lr": 3.063631380693286e-06, "epoch": 1.3333333333333333, "percentage": 66.67, "elapsed_time": "1:56:59", "remaining_time": "0:58:29"}
251
+ {"current_steps": 502, "total_steps": 750, "loss": 1.2729389667510986, "lr": 3.036387458724898e-06, "epoch": 1.3386666666666667, "percentage": 66.93, "elapsed_time": "1:57:29", "remaining_time": "0:58:02"}
252
+ {"current_steps": 504, "total_steps": 750, "loss": 1.2068243026733398, "lr": 3.009254106940435e-06, "epoch": 1.3439999999999999, "percentage": 67.2, "elapsed_time": "1:57:53", "remaining_time": "0:57:32"}
253
+ {"current_steps": 506, "total_steps": 750, "loss": 1.28195321559906, "lr": 2.982233676339206e-06, "epoch": 1.3493333333333333, "percentage": 67.47, "elapsed_time": "1:58:23", "remaining_time": "0:57:05"}
254
+ {"current_steps": 508, "total_steps": 750, "loss": 1.1186037063598633, "lr": 2.9553285081363308e-06, "epoch": 1.3546666666666667, "percentage": 67.73, "elapsed_time": "1:58:47", "remaining_time": "0:56:35"}
255
+ {"current_steps": 510, "total_steps": 750, "loss": 1.1410784721374512, "lr": 2.9285409335598982e-06, "epoch": 1.3599999999999999, "percentage": 68.0, "elapsed_time": "1:59:18", "remaining_time": "0:56:08"}
256
+ {"current_steps": 512, "total_steps": 750, "loss": 1.3254414796829224, "lr": 2.9018732736489615e-06, "epoch": 1.3653333333333333, "percentage": 68.27, "elapsed_time": "1:59:46", "remaining_time": "0:55:40"}
257
+ {"current_steps": 514, "total_steps": 750, "loss": 1.5107591152191162, "lr": 2.87532783905244e-06, "epoch": 1.3706666666666667, "percentage": 68.53, "elapsed_time": "2:00:14", "remaining_time": "0:55:12"}
258
+ {"current_steps": 516, "total_steps": 750, "loss": 1.2509578466415405, "lr": 2.8489069298288992e-06, "epoch": 1.376, "percentage": 68.8, "elapsed_time": "2:00:40", "remaining_time": "0:54:43"}
259
+ {"current_steps": 518, "total_steps": 750, "loss": 1.3889421224594116, "lr": 2.8226128352472747e-06, "epoch": 1.3813333333333333, "percentage": 69.07, "elapsed_time": "2:01:08", "remaining_time": "0:54:15"}
260
+ {"current_steps": 520, "total_steps": 750, "loss": 1.4837677478790283, "lr": 2.796447833588497e-06, "epoch": 1.3866666666666667, "percentage": 69.33, "elapsed_time": "2:01:35", "remaining_time": "0:53:46"}
261
+ {"current_steps": 522, "total_steps": 750, "loss": 1.4461725950241089, "lr": 2.7704141919481057e-06, "epoch": 1.392, "percentage": 69.6, "elapsed_time": "2:02:02", "remaining_time": "0:53:18"}
262
+ {"current_steps": 524, "total_steps": 750, "loss": 1.299829363822937, "lr": 2.7445141660397996e-06, "epoch": 1.3973333333333333, "percentage": 69.87, "elapsed_time": "2:02:34", "remaining_time": "0:52:52"}
263
+ {"current_steps": 526, "total_steps": 750, "loss": 1.0224652290344238, "lr": 2.7187500000000004e-06, "epoch": 1.4026666666666667, "percentage": 70.13, "elapsed_time": "2:03:03", "remaining_time": "0:52:24"}
264
+ {"current_steps": 528, "total_steps": 750, "loss": 1.4109441041946411, "lr": 2.6931239261933957e-06, "epoch": 1.408, "percentage": 70.4, "elapsed_time": "2:03:29", "remaining_time": "0:51:55"}
265
+ {"current_steps": 530, "total_steps": 750, "loss": 1.3534106016159058, "lr": 2.6676381650195224e-06, "epoch": 1.4133333333333333, "percentage": 70.67, "elapsed_time": "2:03:55", "remaining_time": "0:51:26"}
266
+ {"current_steps": 532, "total_steps": 750, "loss": 1.5593756437301636, "lr": 2.6422949247203735e-06, "epoch": 1.4186666666666667, "percentage": 70.93, "elapsed_time": "2:04:20", "remaining_time": "0:50:57"}
267
+ {"current_steps": 534, "total_steps": 750, "loss": 1.290115475654602, "lr": 2.61709640118906e-06, "epoch": 1.424, "percentage": 71.2, "elapsed_time": "2:04:47", "remaining_time": "0:50:28"}
268
+ {"current_steps": 536, "total_steps": 750, "loss": 1.11957585811615, "lr": 2.5920447777795537e-06, "epoch": 1.4293333333333333, "percentage": 71.47, "elapsed_time": "2:05:19", "remaining_time": "0:50:02"}
269
+ {"current_steps": 538, "total_steps": 750, "loss": 1.0493898391723633, "lr": 2.5671422251174963e-06, "epoch": 1.4346666666666668, "percentage": 71.73, "elapsed_time": "2:05:52", "remaining_time": "0:49:36"}
270
+ {"current_steps": 540, "total_steps": 750, "loss": 1.2276326417922974, "lr": 2.5423909009121363e-06, "epoch": 1.44, "percentage": 72.0, "elapsed_time": "2:06:26", "remaining_time": "0:49:10"}
271
+ {"current_steps": 542, "total_steps": 750, "loss": 1.2422417402267456, "lr": 2.517792949769358e-06, "epoch": 1.4453333333333334, "percentage": 72.27, "elapsed_time": "2:06:56", "remaining_time": "0:48:42"}
272
+ {"current_steps": 544, "total_steps": 750, "loss": 1.4107698202133179, "lr": 2.4933505030058744e-06, "epoch": 1.4506666666666668, "percentage": 72.53, "elapsed_time": "2:07:25", "remaining_time": "0:48:14"}
273
+ {"current_steps": 546, "total_steps": 750, "loss": 1.0884898900985718, "lr": 2.4690656784645434e-06, "epoch": 1.456, "percentage": 72.8, "elapsed_time": "2:07:46", "remaining_time": "0:47:44"}
274
+ {"current_steps": 548, "total_steps": 750, "loss": 1.4010899066925049, "lr": 2.444940580330874e-06, "epoch": 1.4613333333333334, "percentage": 73.07, "elapsed_time": "2:08:13", "remaining_time": "0:47:15"}
275
+ {"current_steps": 550, "total_steps": 750, "loss": 1.2208497524261475, "lr": 2.420977298950702e-06, "epoch": 1.4666666666666668, "percentage": 73.33, "elapsed_time": "2:08:34", "remaining_time": "0:46:45"}
276
+ {"current_steps": 552, "total_steps": 750, "loss": 1.2561097145080566, "lr": 2.397177910649074e-06, "epoch": 1.472, "percentage": 73.6, "elapsed_time": "2:09:00", "remaining_time": "0:46:16"}
277
+ {"current_steps": 554, "total_steps": 750, "loss": 1.585281491279602, "lr": 2.3735444775503344e-06, "epoch": 1.4773333333333334, "percentage": 73.87, "elapsed_time": "2:09:25", "remaining_time": "0:45:47"}
278
+ {"current_steps": 556, "total_steps": 750, "loss": 1.4287792444229126, "lr": 2.350079047399464e-06, "epoch": 1.4826666666666668, "percentage": 74.13, "elapsed_time": "2:09:51", "remaining_time": "0:45:18"}
279
+ {"current_steps": 558, "total_steps": 750, "loss": 1.0933114290237427, "lr": 2.3267836533846328e-06, "epoch": 1.488, "percentage": 74.4, "elapsed_time": "2:10:19", "remaining_time": "0:44:50"}
280
+ {"current_steps": 560, "total_steps": 750, "loss": 1.1171455383300781, "lr": 2.3036603139610482e-06, "epoch": 1.4933333333333334, "percentage": 74.67, "elapsed_time": "2:10:47", "remaining_time": "0:44:22"}
281
+ {"current_steps": 562, "total_steps": 750, "loss": 1.2635574340820312, "lr": 2.2807110326760515e-06, "epoch": 1.4986666666666666, "percentage": 74.93, "elapsed_time": "2:11:14", "remaining_time": "0:43:54"}
282
+ {"current_steps": 564, "total_steps": 750, "loss": 1.4870187044143677, "lr": 2.25793779799553e-06, "epoch": 1.504, "percentage": 75.2, "elapsed_time": "2:11:38", "remaining_time": "0:43:24"}
283
+ {"current_steps": 566, "total_steps": 750, "loss": 1.1228148937225342, "lr": 2.2353425831316116e-06, "epoch": 1.5093333333333332, "percentage": 75.47, "elapsed_time": "2:12:11", "remaining_time": "0:42:58"}
284
+ {"current_steps": 568, "total_steps": 750, "loss": 1.393061876296997, "lr": 2.2129273458717058e-06, "epoch": 1.5146666666666668, "percentage": 75.73, "elapsed_time": "2:12:34", "remaining_time": "0:42:28"}
285
+ {"current_steps": 570, "total_steps": 750, "loss": 1.0268428325653076, "lr": 2.1906940284088615e-06, "epoch": 1.52, "percentage": 76.0, "elapsed_time": "2:13:06", "remaining_time": "0:42:01"}
286
+ {"current_steps": 572, "total_steps": 750, "loss": 1.3282322883605957, "lr": 2.1686445571734825e-06, "epoch": 1.5253333333333332, "percentage": 76.27, "elapsed_time": "2:13:29", "remaining_time": "0:41:32"}
287
+ {"current_steps": 574, "total_steps": 750, "loss": 1.2435400485992432, "lr": 2.146780842666417e-06, "epoch": 1.5306666666666666, "percentage": 76.53, "elapsed_time": "2:13:58", "remaining_time": "0:41:04"}
288
+ {"current_steps": 576, "total_steps": 750, "loss": 1.3963842391967773, "lr": 2.1251047792934116e-06, "epoch": 1.536, "percentage": 76.8, "elapsed_time": "2:14:22", "remaining_time": "0:40:35"}
289
+ {"current_steps": 578, "total_steps": 750, "loss": 1.2569538354873657, "lr": 2.1036182452009777e-06, "epoch": 1.5413333333333332, "percentage": 77.07, "elapsed_time": "2:14:49", "remaining_time": "0:40:07"}
290
+ {"current_steps": 580, "total_steps": 750, "loss": 1.3759268522262573, "lr": 2.082323102113647e-06, "epoch": 1.5466666666666666, "percentage": 77.33, "elapsed_time": "2:15:13", "remaining_time": "0:39:38"}
291
+ {"current_steps": 582, "total_steps": 750, "loss": 1.1200230121612549, "lr": 2.0612211951726712e-06, "epoch": 1.552, "percentage": 77.6, "elapsed_time": "2:15:42", "remaining_time": "0:39:10"}
292
+ {"current_steps": 584, "total_steps": 750, "loss": 1.6027384996414185, "lr": 2.040314352776137e-06, "epoch": 1.5573333333333332, "percentage": 77.87, "elapsed_time": "2:16:07", "remaining_time": "0:38:41"}
293
+ {"current_steps": 586, "total_steps": 750, "loss": 1.5927393436431885, "lr": 2.01960438642055e-06, "epoch": 1.5626666666666666, "percentage": 78.13, "elapsed_time": "2:16:33", "remaining_time": "0:38:13"}
294
+ {"current_steps": 588, "total_steps": 750, "loss": 1.0526584386825562, "lr": 1.9990930905438733e-06, "epoch": 1.568, "percentage": 78.4, "elapsed_time": "2:16:59", "remaining_time": "0:37:44"}
295
+ {"current_steps": 590, "total_steps": 750, "loss": 1.4229958057403564, "lr": 1.978782242370046e-06, "epoch": 1.5733333333333333, "percentage": 78.67, "elapsed_time": "2:17:23", "remaining_time": "0:37:15"}
296
+ {"current_steps": 592, "total_steps": 750, "loss": 1.2796956300735474, "lr": 1.958673601754995e-06, "epoch": 1.5786666666666667, "percentage": 78.93, "elapsed_time": "2:17:55", "remaining_time": "0:36:48"}
297
+ {"current_steps": 594, "total_steps": 750, "loss": 1.5249470472335815, "lr": 1.9387689110341494e-06, "epoch": 1.584, "percentage": 79.2, "elapsed_time": "2:18:22", "remaining_time": "0:36:20"}
298
+ {"current_steps": 596, "total_steps": 750, "loss": 1.0016456842422485, "lr": 1.9190698948714714e-06, "epoch": 1.5893333333333333, "percentage": 79.47, "elapsed_time": "2:18:49", "remaining_time": "0:35:52"}
299
+ {"current_steps": 598, "total_steps": 750, "loss": 1.4921225309371948, "lr": 1.899578260110027e-06, "epoch": 1.5946666666666667, "percentage": 79.73, "elapsed_time": "2:19:18", "remaining_time": "0:35:24"}
300
+ {"current_steps": 600, "total_steps": 750, "loss": 1.3857101202011108, "lr": 1.8802956956240875e-06, "epoch": 1.6, "percentage": 80.0, "elapsed_time": "2:19:43", "remaining_time": "0:34:55"}
301
+ {"current_steps": 602, "total_steps": 750, "loss": 1.0772466659545898, "lr": 1.8612238721728026e-06, "epoch": 1.6053333333333333, "percentage": 80.27, "elapsed_time": "2:20:09", "remaining_time": "0:34:27"}
302
+ {"current_steps": 604, "total_steps": 750, "loss": 1.374086618423462, "lr": 1.8423644422554278e-06, "epoch": 1.6106666666666667, "percentage": 80.53, "elapsed_time": "2:20:38", "remaining_time": "0:33:59"}
303
+ {"current_steps": 606, "total_steps": 750, "loss": 1.1497209072113037, "lr": 1.8237190399681486e-06, "epoch": 1.616, "percentage": 80.8, "elapsed_time": "2:21:12", "remaining_time": "0:33:33"}
304
+ {"current_steps": 608, "total_steps": 750, "loss": 1.1666005849838257, "lr": 1.8052892808624892e-06, "epoch": 1.6213333333333333, "percentage": 81.07, "elapsed_time": "2:21:35", "remaining_time": "0:33:04"}
305
+ {"current_steps": 610, "total_steps": 750, "loss": 1.4539345502853394, "lr": 1.78707676180533e-06, "epoch": 1.6266666666666667, "percentage": 81.33, "elapsed_time": "2:21:59", "remaining_time": "0:32:35"}
306
+ {"current_steps": 612, "total_steps": 750, "loss": 1.2192360162734985, "lr": 1.7690830608405475e-06, "epoch": 1.6320000000000001, "percentage": 81.6, "elapsed_time": "2:22:29", "remaining_time": "0:32:07"}
307
+ {"current_steps": 614, "total_steps": 750, "loss": 1.2787443399429321, "lr": 1.7513097370522843e-06, "epoch": 1.6373333333333333, "percentage": 81.87, "elapsed_time": "2:22:57", "remaining_time": "0:31:39"}
308
+ {"current_steps": 616, "total_steps": 750, "loss": 1.4955127239227295, "lr": 1.7337583304298553e-06, "epoch": 1.6426666666666667, "percentage": 82.13, "elapsed_time": "2:23:22", "remaining_time": "0:31:11"}
309
+ {"current_steps": 618, "total_steps": 750, "loss": 1.5240033864974976, "lr": 1.7164303617343188e-06, "epoch": 1.6480000000000001, "percentage": 82.4, "elapsed_time": "2:23:48", "remaining_time": "0:30:42"}
310
+ {"current_steps": 620, "total_steps": 750, "loss": 1.186129093170166, "lr": 1.6993273323667077e-06, "epoch": 1.6533333333333333, "percentage": 82.67, "elapsed_time": "2:24:17", "remaining_time": "0:30:15"}
311
+ {"current_steps": 622, "total_steps": 750, "loss": 1.2840181589126587, "lr": 1.6824507242379345e-06, "epoch": 1.6586666666666665, "percentage": 82.93, "elapsed_time": "2:24:42", "remaining_time": "0:29:46"}
312
+ {"current_steps": 624, "total_steps": 750, "loss": 1.5434945821762085, "lr": 1.6658019996403958e-06, "epoch": 1.6640000000000001, "percentage": 83.2, "elapsed_time": "2:25:06", "remaining_time": "0:29:18"}
313
+ {"current_steps": 626, "total_steps": 750, "loss": 1.5949230194091797, "lr": 1.6493826011212656e-06, "epoch": 1.6693333333333333, "percentage": 83.47, "elapsed_time": "2:25:31", "remaining_time": "0:28:49"}
314
+ {"current_steps": 628, "total_steps": 750, "loss": 1.1981688737869263, "lr": 1.6331939513575066e-06, "epoch": 1.6746666666666665, "percentage": 83.73, "elapsed_time": "2:25:59", "remaining_time": "0:28:21"}
315
+ {"current_steps": 630, "total_steps": 750, "loss": 1.3873530626296997, "lr": 1.6172374530326001e-06, "epoch": 1.6800000000000002, "percentage": 84.0, "elapsed_time": "2:26:22", "remaining_time": "0:27:52"}
316
+ {"current_steps": 632, "total_steps": 750, "loss": 1.1224573850631714, "lr": 1.60151448871501e-06, "epoch": 1.6853333333333333, "percentage": 84.27, "elapsed_time": "2:26:45", "remaining_time": "0:27:24"}
317
+ {"current_steps": 634, "total_steps": 750, "loss": 1.0328835248947144, "lr": 1.5860264207383851e-06, "epoch": 1.6906666666666665, "percentage": 84.53, "elapsed_time": "2:27:07", "remaining_time": "0:26:55"}
318
+ {"current_steps": 636, "total_steps": 750, "loss": 1.5136065483093262, "lr": 1.5707745910835233e-06, "epoch": 1.696, "percentage": 84.8, "elapsed_time": "2:27:33", "remaining_time": "0:26:26"}
319
+ {"current_steps": 638, "total_steps": 750, "loss": 1.5170190334320068, "lr": 1.5557603212620889e-06, "epoch": 1.7013333333333334, "percentage": 85.07, "elapsed_time": "2:28:00", "remaining_time": "0:25:58"}
320
+ {"current_steps": 640, "total_steps": 750, "loss": 1.4020171165466309, "lr": 1.5409849122021144e-06, "epoch": 1.7066666666666666, "percentage": 85.33, "elapsed_time": "2:28:23", "remaining_time": "0:25:30"}
321
+ {"current_steps": 642, "total_steps": 750, "loss": 1.5551862716674805, "lr": 1.5264496441352748e-06, "epoch": 1.712, "percentage": 85.6, "elapsed_time": "2:28:49", "remaining_time": "0:25:02"}
322
+ {"current_steps": 644, "total_steps": 750, "loss": 1.3862305879592896, "lr": 1.5121557764859648e-06, "epoch": 1.7173333333333334, "percentage": 85.87, "elapsed_time": "2:29:15", "remaining_time": "0:24:33"}
323
+ {"current_steps": 646, "total_steps": 750, "loss": 0.922488272190094, "lr": 1.4981045477621702e-06, "epoch": 1.7226666666666666, "percentage": 86.13, "elapsed_time": "2:29:37", "remaining_time": "0:24:05"}
324
+ {"current_steps": 648, "total_steps": 750, "loss": 1.240954041481018, "lr": 1.4842971754481615e-06, "epoch": 1.728, "percentage": 86.4, "elapsed_time": "2:30:02", "remaining_time": "0:23:37"}
325
+ {"current_steps": 650, "total_steps": 750, "loss": 1.0813591480255127, "lr": 1.4707348558989974e-06, "epoch": 1.7333333333333334, "percentage": 86.67, "elapsed_time": "2:30:30", "remaining_time": "0:23:09"}
326
+ {"current_steps": 652, "total_steps": 750, "loss": 1.4775949716567993, "lr": 1.4574187642368714e-06, "epoch": 1.7386666666666666, "percentage": 86.93, "elapsed_time": "2:30:57", "remaining_time": "0:22:41"}
327
+ {"current_steps": 654, "total_steps": 750, "loss": 1.2168411016464233, "lr": 1.444350054249288e-06, "epoch": 1.744, "percentage": 87.2, "elapsed_time": "2:31:31", "remaining_time": "0:22:14"}
328
+ {"current_steps": 656, "total_steps": 750, "loss": 1.4125171899795532, "lr": 1.431529858289091e-06, "epoch": 1.7493333333333334, "percentage": 87.47, "elapsed_time": "2:31:58", "remaining_time": "0:21:46"}
329
+ {"current_steps": 658, "total_steps": 750, "loss": 1.414473056793213, "lr": 1.418959287176354e-06, "epoch": 1.7546666666666666, "percentage": 87.73, "elapsed_time": "2:32:22", "remaining_time": "0:21:18"}
330
+ {"current_steps": 660, "total_steps": 750, "loss": 1.220368504524231, "lr": 1.406639430102127e-06, "epoch": 1.76, "percentage": 88.0, "elapsed_time": "2:32:49", "remaining_time": "0:20:50"}
331
+ {"current_steps": 662, "total_steps": 750, "loss": 1.2297767400741577, "lr": 1.3945713545340668e-06, "epoch": 1.7653333333333334, "percentage": 88.27, "elapsed_time": "2:33:18", "remaining_time": "0:20:22"}
332
+ {"current_steps": 664, "total_steps": 750, "loss": 1.3291184902191162, "lr": 1.3827561061239403e-06, "epoch": 1.7706666666666666, "percentage": 88.53, "elapsed_time": "2:33:47", "remaining_time": "0:19:55"}
333
+ {"current_steps": 666, "total_steps": 750, "loss": 1.0460172891616821, "lr": 1.3711947086170277e-06, "epoch": 1.776, "percentage": 88.8, "elapsed_time": "2:34:14", "remaining_time": "0:19:27"}
334
+ {"current_steps": 668, "total_steps": 750, "loss": 1.3426827192306519, "lr": 1.359888163763415e-06, "epoch": 1.7813333333333334, "percentage": 89.07, "elapsed_time": "2:34:40", "remaining_time": "0:18:59"}
335
+ {"current_steps": 670, "total_steps": 750, "loss": 1.1144654750823975, "lr": 1.3488374512311988e-06, "epoch": 1.7866666666666666, "percentage": 89.33, "elapsed_time": "2:35:08", "remaining_time": "0:18:31"}
336
+ {"current_steps": 672, "total_steps": 750, "loss": 1.2571556568145752, "lr": 1.3380435285216006e-06, "epoch": 1.792, "percentage": 89.6, "elapsed_time": "2:35:36", "remaining_time": "0:18:03"}
337
+ {"current_steps": 674, "total_steps": 750, "loss": 1.454203486442566, "lr": 1.327507330886004e-06, "epoch": 1.7973333333333334, "percentage": 89.87, "elapsed_time": "2:36:04", "remaining_time": "0:17:35"}
338
+ {"current_steps": 676, "total_steps": 750, "loss": 1.4048188924789429, "lr": 1.3172297712449172e-06, "epoch": 1.8026666666666666, "percentage": 90.13, "elapsed_time": "2:36:30", "remaining_time": "0:17:07"}
339
+ {"current_steps": 678, "total_steps": 750, "loss": 1.0630518198013306, "lr": 1.3072117401088736e-06, "epoch": 1.808, "percentage": 90.4, "elapsed_time": "2:37:03", "remaining_time": "0:16:40"}
340
+ {"current_steps": 680, "total_steps": 750, "loss": 1.1879171133041382, "lr": 1.2974541055012725e-06, "epoch": 1.8133333333333335, "percentage": 90.67, "elapsed_time": "2:37:27", "remaining_time": "0:16:12"}
341
+ {"current_steps": 682, "total_steps": 750, "loss": 1.380265235900879, "lr": 1.2879577128831673e-06, "epoch": 1.8186666666666667, "percentage": 90.93, "elapsed_time": "2:37:54", "remaining_time": "0:15:44"}
342
+ {"current_steps": 684, "total_steps": 750, "loss": 0.971758246421814, "lr": 1.278723385080009e-06, "epoch": 1.8239999999999998, "percentage": 91.2, "elapsed_time": "2:38:18", "remaining_time": "0:15:16"}
343
+ {"current_steps": 686, "total_steps": 750, "loss": 1.512132167816162, "lr": 1.2697519222103545e-06, "epoch": 1.8293333333333335, "percentage": 91.47, "elapsed_time": "2:38:44", "remaining_time": "0:14:48"}
344
+ {"current_steps": 688, "total_steps": 750, "loss": 1.005048155784607, "lr": 1.2610441016165355e-06, "epoch": 1.8346666666666667, "percentage": 91.73, "elapsed_time": "2:39:10", "remaining_time": "0:14:20"}
345
+ {"current_steps": 690, "total_steps": 750, "loss": 1.371572732925415, "lr": 1.2526006777973084e-06, "epoch": 1.8399999999999999, "percentage": 92.0, "elapsed_time": "2:39:37", "remaining_time": "0:13:52"}
346
+ {"current_steps": 692, "total_steps": 750, "loss": 1.332929253578186, "lr": 1.2444223823424778e-06, "epoch": 1.8453333333333335, "percentage": 92.27, "elapsed_time": "2:40:04", "remaining_time": "0:13:24"}
347
+ {"current_steps": 694, "total_steps": 750, "loss": 1.0624377727508545, "lr": 1.2365099238695073e-06, "epoch": 1.8506666666666667, "percentage": 92.53, "elapsed_time": "2:40:39", "remaining_time": "0:12:57"}
348
+ {"current_steps": 696, "total_steps": 750, "loss": 1.4434648752212524, "lr": 1.2288639879621216e-06, "epoch": 1.8559999999999999, "percentage": 92.8, "elapsed_time": "2:41:05", "remaining_time": "0:12:29"}
349
+ {"current_steps": 698, "total_steps": 750, "loss": 1.115839958190918, "lr": 1.2214852371109014e-06, "epoch": 1.8613333333333333, "percentage": 93.07, "elapsed_time": "2:41:32", "remaining_time": "0:12:02"}
350
+ {"current_steps": 700, "total_steps": 750, "loss": 1.3425531387329102, "lr": 1.214374310655884e-06, "epoch": 1.8666666666666667, "percentage": 93.33, "elapsed_time": "2:42:00", "remaining_time": "0:11:34"}
351
+ {"current_steps": 702, "total_steps": 750, "loss": 1.2544267177581787, "lr": 1.207531824731164e-06, "epoch": 1.8719999999999999, "percentage": 93.6, "elapsed_time": "2:42:26", "remaining_time": "0:11:06"}
352
+ {"current_steps": 704, "total_steps": 750, "loss": 1.2201001644134521, "lr": 1.2009583722115107e-06, "epoch": 1.8773333333333333, "percentage": 93.87, "elapsed_time": "2:42:50", "remaining_time": "0:10:38"}
353
+ {"current_steps": 706, "total_steps": 750, "loss": 1.1109111309051514, "lr": 1.1946545226609944e-06, "epoch": 1.8826666666666667, "percentage": 94.13, "elapsed_time": "2:43:20", "remaining_time": "0:10:10"}
354
+ {"current_steps": 708, "total_steps": 750, "loss": 1.2960035800933838, "lr": 1.1886208222836388e-06, "epoch": 1.888, "percentage": 94.4, "elapsed_time": "2:43:46", "remaining_time": "0:09:42"}
355
+ {"current_steps": 710, "total_steps": 750, "loss": 1.3226895332336426, "lr": 1.1828577938760926e-06, "epoch": 1.8933333333333333, "percentage": 94.67, "elapsed_time": "2:44:15", "remaining_time": "0:09:15"}
356
+ {"current_steps": 712, "total_steps": 750, "loss": 1.459092140197754, "lr": 1.1773659367823326e-06, "epoch": 1.8986666666666667, "percentage": 94.93, "elapsed_time": "2:44:41", "remaining_time": "0:08:47"}
357
+ {"current_steps": 714, "total_steps": 750, "loss": 0.9945974349975586, "lr": 1.1721457268503963e-06, "epoch": 1.904, "percentage": 95.2, "elapsed_time": "2:45:11", "remaining_time": "0:08:19"}
358
+ {"current_steps": 716, "total_steps": 750, "loss": 1.3499224185943604, "lr": 1.1671976163911537e-06, "epoch": 1.9093333333333333, "percentage": 95.47, "elapsed_time": "2:45:39", "remaining_time": "0:07:51"}
359
+ {"current_steps": 718, "total_steps": 750, "loss": 1.203792929649353, "lr": 1.162522034139113e-06, "epoch": 1.9146666666666667, "percentage": 95.73, "elapsed_time": "2:46:01", "remaining_time": "0:07:23"}
360
+ {"current_steps": 720, "total_steps": 750, "loss": 1.432945728302002, "lr": 1.1581193852152746e-06, "epoch": 1.92, "percentage": 96.0, "elapsed_time": "2:46:28", "remaining_time": "0:06:56"}
361
+ {"current_steps": 722, "total_steps": 750, "loss": 1.4233765602111816, "lr": 1.1539900510920285e-06, "epoch": 1.9253333333333333, "percentage": 96.27, "elapsed_time": "2:46:53", "remaining_time": "0:06:28"}
362
+ {"current_steps": 724, "total_steps": 750, "loss": 1.5078463554382324, "lr": 1.1501343895601022e-06, "epoch": 1.9306666666666668, "percentage": 96.53, "elapsed_time": "2:47:19", "remaining_time": "0:06:00"}
363
+ {"current_steps": 726, "total_steps": 750, "loss": 1.041426658630371, "lr": 1.1465527346975566e-06, "epoch": 1.936, "percentage": 96.8, "elapsed_time": "2:47:43", "remaining_time": "0:05:32"}
364
+ {"current_steps": 728, "total_steps": 750, "loss": 1.1554670333862305, "lr": 1.1432453968408437e-06, "epoch": 1.9413333333333334, "percentage": 97.07, "elapsed_time": "2:48:12", "remaining_time": "0:05:04"}
365
+ {"current_steps": 730, "total_steps": 750, "loss": 1.238165020942688, "lr": 1.1402126625579132e-06, "epoch": 1.9466666666666668, "percentage": 97.33, "elapsed_time": "2:48:39", "remaining_time": "0:04:37"}
366
+ {"current_steps": 732, "total_steps": 750, "loss": 1.2469568252563477, "lr": 1.1374547946233843e-06, "epoch": 1.952, "percentage": 97.6, "elapsed_time": "2:49:07", "remaining_time": "0:04:09"}
367
+ {"current_steps": 734, "total_steps": 750, "loss": 1.100600004196167, "lr": 1.1349720319957783e-06, "epoch": 1.9573333333333334, "percentage": 97.87, "elapsed_time": "2:49:35", "remaining_time": "0:03:41"}
368
+ {"current_steps": 736, "total_steps": 750, "loss": 1.256953239440918, "lr": 1.1327645897968104e-06, "epoch": 1.9626666666666668, "percentage": 98.13, "elapsed_time": "2:50:06", "remaining_time": "0:03:14"}
369
+ {"current_steps": 738, "total_steps": 750, "loss": 1.0144098997116089, "lr": 1.1308326592927547e-06, "epoch": 1.968, "percentage": 98.4, "elapsed_time": "2:50:31", "remaining_time": "0:02:46"}
370
+ {"current_steps": 740, "total_steps": 750, "loss": 1.1728276014328003, "lr": 1.1291764078778685e-06, "epoch": 1.9733333333333334, "percentage": 98.67, "elapsed_time": "2:50:57", "remaining_time": "0:02:18"}
371
+ {"current_steps": 742, "total_steps": 750, "loss": 1.084470272064209, "lr": 1.127795979059889e-06, "epoch": 1.9786666666666668, "percentage": 98.93, "elapsed_time": "2:51:25", "remaining_time": "0:01:50"}
372
+ {"current_steps": 744, "total_steps": 750, "loss": 1.2899092435836792, "lr": 1.1266914924475997e-06, "epoch": 1.984, "percentage": 99.2, "elapsed_time": "2:51:52", "remaining_time": "0:01:23"}
373
+ {"current_steps": 746, "total_steps": 750, "loss": 1.3732484579086304, "lr": 1.1258630437404658e-06, "epoch": 1.9893333333333332, "percentage": 99.47, "elapsed_time": "2:52:20", "remaining_time": "0:00:55"}
374
+ {"current_steps": 748, "total_steps": 750, "loss": 1.2392138242721558, "lr": 1.1253107047203428e-06, "epoch": 1.9946666666666668, "percentage": 99.73, "elapsed_time": "2:52:52", "remaining_time": "0:00:27"}
375
+ {"current_steps": 750, "total_steps": 750, "loss": 1.4283043146133423, "lr": 1.1250345232452565e-06, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "2:53:18", "remaining_time": "0:00:00"}
376
+ {"current_steps": 750, "total_steps": 750, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "2:53:57", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,2668 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 750,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.005333333333333333,
14
+ "grad_norm": 0.5234375,
15
+ "learning_rate": 1.0000000000000001e-07,
16
+ "loss": 1.3918027877807617,
17
+ "step": 2
18
+ },
19
+ {
20
+ "epoch": 0.010666666666666666,
21
+ "grad_norm": 1.3203125,
22
+ "learning_rate": 3.0000000000000004e-07,
23
+ "loss": 2.0975496768951416,
24
+ "step": 4
25
+ },
26
+ {
27
+ "epoch": 0.016,
28
+ "grad_norm": 1.671875,
29
+ "learning_rate": 5e-07,
30
+ "loss": 2.0533759593963623,
31
+ "step": 6
32
+ },
33
+ {
34
+ "epoch": 0.021333333333333333,
35
+ "grad_norm": 0.60546875,
36
+ "learning_rate": 7.000000000000001e-07,
37
+ "loss": 1.4977960586547852,
38
+ "step": 8
39
+ },
40
+ {
41
+ "epoch": 0.02666666666666667,
42
+ "grad_norm": 1.28125,
43
+ "learning_rate": 9e-07,
44
+ "loss": 1.8092024326324463,
45
+ "step": 10
46
+ },
47
+ {
48
+ "epoch": 0.032,
49
+ "grad_norm": 0.82421875,
50
+ "learning_rate": 1.1e-06,
51
+ "loss": 2.0490243434906006,
52
+ "step": 12
53
+ },
54
+ {
55
+ "epoch": 0.037333333333333336,
56
+ "grad_norm": 1.671875,
57
+ "learning_rate": 1.3e-06,
58
+ "loss": 2.2412607669830322,
59
+ "step": 14
60
+ },
61
+ {
62
+ "epoch": 0.042666666666666665,
63
+ "grad_norm": 0.8828125,
64
+ "learning_rate": 1.5e-06,
65
+ "loss": 1.9627114534378052,
66
+ "step": 16
67
+ },
68
+ {
69
+ "epoch": 0.048,
70
+ "grad_norm": 2.390625,
71
+ "learning_rate": 1.7e-06,
72
+ "loss": 1.8048596382141113,
73
+ "step": 18
74
+ },
75
+ {
76
+ "epoch": 0.05333333333333334,
77
+ "grad_norm": 1.3046875,
78
+ "learning_rate": 1.9000000000000002e-06,
79
+ "loss": 1.9537056684494019,
80
+ "step": 20
81
+ },
82
+ {
83
+ "epoch": 0.058666666666666666,
84
+ "grad_norm": 1.09375,
85
+ "learning_rate": 2.1000000000000002e-06,
86
+ "loss": 1.7351176738739014,
87
+ "step": 22
88
+ },
89
+ {
90
+ "epoch": 0.064,
91
+ "grad_norm": 0.93359375,
92
+ "learning_rate": 2.3e-06,
93
+ "loss": 2.1719343662261963,
94
+ "step": 24
95
+ },
96
+ {
97
+ "epoch": 0.06933333333333333,
98
+ "grad_norm": 0.73828125,
99
+ "learning_rate": 2.4999999999999998e-06,
100
+ "loss": 2.0036661624908447,
101
+ "step": 26
102
+ },
103
+ {
104
+ "epoch": 0.07466666666666667,
105
+ "grad_norm": 0.6328125,
106
+ "learning_rate": 2.7e-06,
107
+ "loss": 1.4876424074172974,
108
+ "step": 28
109
+ },
110
+ {
111
+ "epoch": 0.08,
112
+ "grad_norm": 0.93359375,
113
+ "learning_rate": 2.9e-06,
114
+ "loss": 1.7595216035842896,
115
+ "step": 30
116
+ },
117
+ {
118
+ "epoch": 0.08533333333333333,
119
+ "grad_norm": 0.92578125,
120
+ "learning_rate": 3.1e-06,
121
+ "loss": 1.7543615102767944,
122
+ "step": 32
123
+ },
124
+ {
125
+ "epoch": 0.09066666666666667,
126
+ "grad_norm": 0.703125,
127
+ "learning_rate": 3.3e-06,
128
+ "loss": 1.6767919063568115,
129
+ "step": 34
130
+ },
131
+ {
132
+ "epoch": 0.096,
133
+ "grad_norm": 0.7265625,
134
+ "learning_rate": 3.5e-06,
135
+ "loss": 1.9044243097305298,
136
+ "step": 36
137
+ },
138
+ {
139
+ "epoch": 0.10133333333333333,
140
+ "grad_norm": 0.78125,
141
+ "learning_rate": 3.7e-06,
142
+ "loss": 1.9992799758911133,
143
+ "step": 38
144
+ },
145
+ {
146
+ "epoch": 0.10666666666666667,
147
+ "grad_norm": 0.2021484375,
148
+ "learning_rate": 3.9e-06,
149
+ "loss": 1.4217664003372192,
150
+ "step": 40
151
+ },
152
+ {
153
+ "epoch": 0.112,
154
+ "grad_norm": 1.0859375,
155
+ "learning_rate": 4.1e-06,
156
+ "loss": 2.2651872634887695,
157
+ "step": 42
158
+ },
159
+ {
160
+ "epoch": 0.11733333333333333,
161
+ "grad_norm": 1.015625,
162
+ "learning_rate": 4.3e-06,
163
+ "loss": 1.8020230531692505,
164
+ "step": 44
165
+ },
166
+ {
167
+ "epoch": 0.12266666666666666,
168
+ "grad_norm": 0.6328125,
169
+ "learning_rate": 4.5e-06,
170
+ "loss": 2.057128667831421,
171
+ "step": 46
172
+ },
173
+ {
174
+ "epoch": 0.128,
175
+ "grad_norm": 0.287109375,
176
+ "learning_rate": 4.700000000000001e-06,
177
+ "loss": 1.7972444295883179,
178
+ "step": 48
179
+ },
180
+ {
181
+ "epoch": 0.13333333333333333,
182
+ "grad_norm": 1.1328125,
183
+ "learning_rate": 4.9e-06,
184
+ "loss": 1.585660696029663,
185
+ "step": 50
186
+ },
187
+ {
188
+ "epoch": 0.13866666666666666,
189
+ "grad_norm": 1.4140625,
190
+ "learning_rate": 5.1e-06,
191
+ "loss": 1.5591168403625488,
192
+ "step": 52
193
+ },
194
+ {
195
+ "epoch": 0.144,
196
+ "grad_norm": 0.2119140625,
197
+ "learning_rate": 5.3e-06,
198
+ "loss": 1.2894660234451294,
199
+ "step": 54
200
+ },
201
+ {
202
+ "epoch": 0.14933333333333335,
203
+ "grad_norm": 0.73046875,
204
+ "learning_rate": 5.5e-06,
205
+ "loss": 1.8333218097686768,
206
+ "step": 56
207
+ },
208
+ {
209
+ "epoch": 0.15466666666666667,
210
+ "grad_norm": 1.1015625,
211
+ "learning_rate": 5.7000000000000005e-06,
212
+ "loss": 1.5420780181884766,
213
+ "step": 58
214
+ },
215
+ {
216
+ "epoch": 0.16,
217
+ "grad_norm": 0.828125,
218
+ "learning_rate": 5.899999999999999e-06,
219
+ "loss": 1.8294118642807007,
220
+ "step": 60
221
+ },
222
+ {
223
+ "epoch": 0.16533333333333333,
224
+ "grad_norm": 0.3515625,
225
+ "learning_rate": 6.1e-06,
226
+ "loss": 1.5082216262817383,
227
+ "step": 62
228
+ },
229
+ {
230
+ "epoch": 0.17066666666666666,
231
+ "grad_norm": 0.314453125,
232
+ "learning_rate": 6.3e-06,
233
+ "loss": 1.7252726554870605,
234
+ "step": 64
235
+ },
236
+ {
237
+ "epoch": 0.176,
238
+ "grad_norm": 0.435546875,
239
+ "learning_rate": 6.5000000000000004e-06,
240
+ "loss": 1.56448233127594,
241
+ "step": 66
242
+ },
243
+ {
244
+ "epoch": 0.18133333333333335,
245
+ "grad_norm": 1.1015625,
246
+ "learning_rate": 6.7e-06,
247
+ "loss": 1.87124502658844,
248
+ "step": 68
249
+ },
250
+ {
251
+ "epoch": 0.18666666666666668,
252
+ "grad_norm": 0.5859375,
253
+ "learning_rate": 6.900000000000001e-06,
254
+ "loss": 1.4997594356536865,
255
+ "step": 70
256
+ },
257
+ {
258
+ "epoch": 0.192,
259
+ "grad_norm": 0.609375,
260
+ "learning_rate": 7.1e-06,
261
+ "loss": 1.9693466424942017,
262
+ "step": 72
263
+ },
264
+ {
265
+ "epoch": 0.19733333333333333,
266
+ "grad_norm": 0.95703125,
267
+ "learning_rate": 7.3e-06,
268
+ "loss": 1.5585707426071167,
269
+ "step": 74
270
+ },
271
+ {
272
+ "epoch": 0.20266666666666666,
273
+ "grad_norm": 1.859375,
274
+ "learning_rate": 7.5e-06,
275
+ "loss": 1.7230855226516724,
276
+ "step": 76
277
+ },
278
+ {
279
+ "epoch": 0.208,
280
+ "grad_norm": 0.38671875,
281
+ "learning_rate": 7.4998619077668045e-06,
282
+ "loss": 1.7421209812164307,
283
+ "step": 78
284
+ },
285
+ {
286
+ "epoch": 0.21333333333333335,
287
+ "grad_norm": 0.466796875,
288
+ "learning_rate": 7.499447643032373e-06,
289
+ "loss": 1.8539875745773315,
290
+ "step": 80
291
+ },
292
+ {
293
+ "epoch": 0.21866666666666668,
294
+ "grad_norm": 0.2734375,
295
+ "learning_rate": 7.498757241691129e-06,
296
+ "loss": 1.410808801651001,
297
+ "step": 82
298
+ },
299
+ {
300
+ "epoch": 0.224,
301
+ "grad_norm": 0.39453125,
302
+ "learning_rate": 7.497790763563665e-06,
303
+ "loss": 1.5474530458450317,
304
+ "step": 84
305
+ },
306
+ {
307
+ "epoch": 0.22933333333333333,
308
+ "grad_norm": 0.69921875,
309
+ "learning_rate": 7.496548292391548e-06,
310
+ "loss": 1.8224376440048218,
311
+ "step": 86
312
+ },
313
+ {
314
+ "epoch": 0.23466666666666666,
315
+ "grad_norm": 1.265625,
316
+ "learning_rate": 7.4950299358300644e-06,
317
+ "loss": 1.5387835502624512,
318
+ "step": 88
319
+ },
320
+ {
321
+ "epoch": 0.24,
322
+ "grad_norm": 0.1806640625,
323
+ "learning_rate": 7.493235825438903e-06,
324
+ "loss": 1.4174927473068237,
325
+ "step": 90
326
+ },
327
+ {
328
+ "epoch": 0.24533333333333332,
329
+ "grad_norm": 0.53125,
330
+ "learning_rate": 7.491166116670744e-06,
331
+ "loss": 1.3185020685195923,
332
+ "step": 92
333
+ },
334
+ {
335
+ "epoch": 0.25066666666666665,
336
+ "grad_norm": 2.078125,
337
+ "learning_rate": 7.488820988857795e-06,
338
+ "loss": 1.3037068843841553,
339
+ "step": 94
340
+ },
341
+ {
342
+ "epoch": 0.256,
343
+ "grad_norm": 0.330078125,
344
+ "learning_rate": 7.486200645196252e-06,
345
+ "loss": 1.7521201372146606,
346
+ "step": 96
347
+ },
348
+ {
349
+ "epoch": 0.2613333333333333,
350
+ "grad_norm": 0.71875,
351
+ "learning_rate": 7.483305312728692e-06,
352
+ "loss": 1.2269163131713867,
353
+ "step": 98
354
+ },
355
+ {
356
+ "epoch": 0.26666666666666666,
357
+ "grad_norm": 0.41796875,
358
+ "learning_rate": 7.480135242324401e-06,
359
+ "loss": 1.4020113945007324,
360
+ "step": 100
361
+ },
362
+ {
363
+ "epoch": 0.272,
364
+ "grad_norm": 0.1875,
365
+ "learning_rate": 7.476690708657641e-06,
366
+ "loss": 1.236190915107727,
367
+ "step": 102
368
+ },
369
+ {
370
+ "epoch": 0.2773333333333333,
371
+ "grad_norm": 0.765625,
372
+ "learning_rate": 7.472972010183844e-06,
373
+ "loss": 1.3339173793792725,
374
+ "step": 104
375
+ },
376
+ {
377
+ "epoch": 0.2826666666666667,
378
+ "grad_norm": 1.6640625,
379
+ "learning_rate": 7.468979469113756e-06,
380
+ "loss": 1.5956217050552368,
381
+ "step": 106
382
+ },
383
+ {
384
+ "epoch": 0.288,
385
+ "grad_norm": 0.2119140625,
386
+ "learning_rate": 7.4647134313855186e-06,
387
+ "loss": 1.2588270902633667,
388
+ "step": 108
389
+ },
390
+ {
391
+ "epoch": 0.29333333333333333,
392
+ "grad_norm": 0.52734375,
393
+ "learning_rate": 7.4601742666346926e-06,
394
+ "loss": 1.6625038385391235,
395
+ "step": 110
396
+ },
397
+ {
398
+ "epoch": 0.2986666666666667,
399
+ "grad_norm": 0.7109375,
400
+ "learning_rate": 7.455362368162235e-06,
401
+ "loss": 1.3837028741836548,
402
+ "step": 112
403
+ },
404
+ {
405
+ "epoch": 0.304,
406
+ "grad_norm": 0.283203125,
407
+ "learning_rate": 7.450278152900414e-06,
408
+ "loss": 1.5355887413024902,
409
+ "step": 114
410
+ },
411
+ {
412
+ "epoch": 0.30933333333333335,
413
+ "grad_norm": 0.65625,
414
+ "learning_rate": 7.4449220613766924e-06,
415
+ "loss": 1.6411131620407104,
416
+ "step": 116
417
+ },
418
+ {
419
+ "epoch": 0.31466666666666665,
420
+ "grad_norm": 1.6484375,
421
+ "learning_rate": 7.439294557675548e-06,
422
+ "loss": 1.5240530967712402,
423
+ "step": 118
424
+ },
425
+ {
426
+ "epoch": 0.32,
427
+ "grad_norm": 0.609375,
428
+ "learning_rate": 7.43339612939827e-06,
429
+ "loss": 1.1345856189727783,
430
+ "step": 120
431
+ },
432
+ {
433
+ "epoch": 0.3253333333333333,
434
+ "grad_norm": 0.373046875,
435
+ "learning_rate": 7.427227287620707e-06,
436
+ "loss": 1.1475586891174316,
437
+ "step": 122
438
+ },
439
+ {
440
+ "epoch": 0.33066666666666666,
441
+ "grad_norm": 0.52734375,
442
+ "learning_rate": 7.420788566848982e-06,
443
+ "loss": 1.5467785596847534,
444
+ "step": 124
445
+ },
446
+ {
447
+ "epoch": 0.336,
448
+ "grad_norm": 0.287109375,
449
+ "learning_rate": 7.414080524973188e-06,
450
+ "loss": 1.6301590204238892,
451
+ "step": 126
452
+ },
453
+ {
454
+ "epoch": 0.3413333333333333,
455
+ "grad_norm": 0.57421875,
456
+ "learning_rate": 7.407103743219038e-06,
457
+ "loss": 1.4552572965621948,
458
+ "step": 128
459
+ },
460
+ {
461
+ "epoch": 0.3466666666666667,
462
+ "grad_norm": 0.73046875,
463
+ "learning_rate": 7.399858826097512e-06,
464
+ "loss": 1.7707996368408203,
465
+ "step": 130
466
+ },
467
+ {
468
+ "epoch": 0.352,
469
+ "grad_norm": 0.64453125,
470
+ "learning_rate": 7.392346401352474e-06,
471
+ "loss": 1.4176101684570312,
472
+ "step": 132
473
+ },
474
+ {
475
+ "epoch": 0.35733333333333334,
476
+ "grad_norm": 0.61328125,
477
+ "learning_rate": 7.3845671199062815e-06,
478
+ "loss": 1.4162875413894653,
479
+ "step": 134
480
+ },
481
+ {
482
+ "epoch": 0.3626666666666667,
483
+ "grad_norm": 0.55859375,
484
+ "learning_rate": 7.376521655803392e-06,
485
+ "loss": 1.4608713388442993,
486
+ "step": 136
487
+ },
488
+ {
489
+ "epoch": 0.368,
490
+ "grad_norm": 0.392578125,
491
+ "learning_rate": 7.368210706151946e-06,
492
+ "loss": 1.4052457809448242,
493
+ "step": 138
494
+ },
495
+ {
496
+ "epoch": 0.37333333333333335,
497
+ "grad_norm": 0.427734375,
498
+ "learning_rate": 7.359634991063379e-06,
499
+ "loss": 1.137465476989746,
500
+ "step": 140
501
+ },
502
+ {
503
+ "epoch": 0.37866666666666665,
504
+ "grad_norm": 0.8359375,
505
+ "learning_rate": 7.350795253590018e-06,
506
+ "loss": 1.4206374883651733,
507
+ "step": 142
508
+ },
509
+ {
510
+ "epoch": 0.384,
511
+ "grad_norm": 0.11474609375,
512
+ "learning_rate": 7.3416922596607055e-06,
513
+ "loss": 1.020829439163208,
514
+ "step": 144
515
+ },
516
+ {
517
+ "epoch": 0.3893333333333333,
518
+ "grad_norm": 0.2421875,
519
+ "learning_rate": 7.332326798014425e-06,
520
+ "loss": 1.558069109916687,
521
+ "step": 146
522
+ },
523
+ {
524
+ "epoch": 0.39466666666666667,
525
+ "grad_norm": 0.26171875,
526
+ "learning_rate": 7.322699680131971e-06,
527
+ "loss": 1.5617731809616089,
528
+ "step": 148
529
+ },
530
+ {
531
+ "epoch": 0.4,
532
+ "grad_norm": 0.40234375,
533
+ "learning_rate": 7.312811740165632e-06,
534
+ "loss": 1.3473962545394897,
535
+ "step": 150
536
+ },
537
+ {
538
+ "epoch": 0.4053333333333333,
539
+ "grad_norm": 0.71875,
540
+ "learning_rate": 7.302663834866909e-06,
541
+ "loss": 1.4550005197525024,
542
+ "step": 152
543
+ },
544
+ {
545
+ "epoch": 0.4106666666666667,
546
+ "grad_norm": 0.4140625,
547
+ "learning_rate": 7.2922568435122956e-06,
548
+ "loss": 1.2182159423828125,
549
+ "step": 154
550
+ },
551
+ {
552
+ "epoch": 0.416,
553
+ "grad_norm": 0.515625,
554
+ "learning_rate": 7.281591667827079e-06,
555
+ "loss": 1.1797568798065186,
556
+ "step": 156
557
+ },
558
+ {
559
+ "epoch": 0.42133333333333334,
560
+ "grad_norm": 0.50390625,
561
+ "learning_rate": 7.270669231907212e-06,
562
+ "loss": 1.3810535669326782,
563
+ "step": 158
564
+ },
565
+ {
566
+ "epoch": 0.4266666666666667,
567
+ "grad_norm": 0.412109375,
568
+ "learning_rate": 7.2594904821392485e-06,
569
+ "loss": 1.286556363105774,
570
+ "step": 160
571
+ },
572
+ {
573
+ "epoch": 0.432,
574
+ "grad_norm": 0.419921875,
575
+ "learning_rate": 7.248056387118338e-06,
576
+ "loss": 1.4289522171020508,
577
+ "step": 162
578
+ },
579
+ {
580
+ "epoch": 0.43733333333333335,
581
+ "grad_norm": 0.66015625,
582
+ "learning_rate": 7.236367937564302e-06,
583
+ "loss": 1.243101716041565,
584
+ "step": 164
585
+ },
586
+ {
587
+ "epoch": 0.44266666666666665,
588
+ "grad_norm": 0.283203125,
589
+ "learning_rate": 7.224426146235791e-06,
590
+ "loss": 1.0999314785003662,
591
+ "step": 166
592
+ },
593
+ {
594
+ "epoch": 0.448,
595
+ "grad_norm": 0.3046875,
596
+ "learning_rate": 7.212232047842537e-06,
597
+ "loss": 1.4978009462356567,
598
+ "step": 168
599
+ },
600
+ {
601
+ "epoch": 0.4533333333333333,
602
+ "grad_norm": 5.53125,
603
+ "learning_rate": 7.199786698955697e-06,
604
+ "loss": 1.4373788833618164,
605
+ "step": 170
606
+ },
607
+ {
608
+ "epoch": 0.45866666666666667,
609
+ "grad_norm": 0.69921875,
610
+ "learning_rate": 7.187091177916301e-06,
611
+ "loss": 1.4840056896209717,
612
+ "step": 172
613
+ },
614
+ {
615
+ "epoch": 0.464,
616
+ "grad_norm": 0.2021484375,
617
+ "learning_rate": 7.174146584741825e-06,
618
+ "loss": 1.2911632061004639,
619
+ "step": 174
620
+ },
621
+ {
622
+ "epoch": 0.4693333333333333,
623
+ "grad_norm": 0.279296875,
624
+ "learning_rate": 7.160954041030877e-06,
625
+ "loss": 1.3464269638061523,
626
+ "step": 176
627
+ },
628
+ {
629
+ "epoch": 0.4746666666666667,
630
+ "grad_norm": 0.357421875,
631
+ "learning_rate": 7.1475146898660116e-06,
632
+ "loss": 1.4491277933120728,
633
+ "step": 178
634
+ },
635
+ {
636
+ "epoch": 0.48,
637
+ "grad_norm": 0.349609375,
638
+ "learning_rate": 7.133829695714688e-06,
639
+ "loss": 1.318329930305481,
640
+ "step": 180
641
+ },
642
+ {
643
+ "epoch": 0.48533333333333334,
644
+ "grad_norm": 0.78515625,
645
+ "learning_rate": 7.1199002443283776e-06,
646
+ "loss": 1.089525580406189,
647
+ "step": 182
648
+ },
649
+ {
650
+ "epoch": 0.49066666666666664,
651
+ "grad_norm": 0.263671875,
652
+ "learning_rate": 7.1057275426398146e-06,
653
+ "loss": 1.356623888015747,
654
+ "step": 184
655
+ },
656
+ {
657
+ "epoch": 0.496,
658
+ "grad_norm": 0.51171875,
659
+ "learning_rate": 7.09131281865843e-06,
660
+ "loss": 1.5700266361236572,
661
+ "step": 186
662
+ },
663
+ {
664
+ "epoch": 0.5013333333333333,
665
+ "grad_norm": 0.45703125,
666
+ "learning_rate": 7.0766573213639404e-06,
667
+ "loss": 1.12882661819458,
668
+ "step": 188
669
+ },
670
+ {
671
+ "epoch": 0.5066666666666667,
672
+ "grad_norm": 0.53125,
673
+ "learning_rate": 7.061762320598134e-06,
674
+ "loss": 1.3883382081985474,
675
+ "step": 190
676
+ },
677
+ {
678
+ "epoch": 0.512,
679
+ "grad_norm": 0.263671875,
680
+ "learning_rate": 7.0466291069548415e-06,
681
+ "loss": 1.1575868129730225,
682
+ "step": 192
683
+ },
684
+ {
685
+ "epoch": 0.5173333333333333,
686
+ "grad_norm": 0.291015625,
687
+ "learning_rate": 7.031258991668114e-06,
688
+ "loss": 1.458404302597046,
689
+ "step": 194
690
+ },
691
+ {
692
+ "epoch": 0.5226666666666666,
693
+ "grad_norm": 0.5703125,
694
+ "learning_rate": 7.015653306498608e-06,
695
+ "loss": 1.7165433168411255,
696
+ "step": 196
697
+ },
698
+ {
699
+ "epoch": 0.528,
700
+ "grad_norm": 0.486328125,
701
+ "learning_rate": 6.99981340361819e-06,
702
+ "loss": 1.5549272298812866,
703
+ "step": 198
704
+ },
705
+ {
706
+ "epoch": 0.5333333333333333,
707
+ "grad_norm": 0.70703125,
708
+ "learning_rate": 6.983740655492783e-06,
709
+ "loss": 1.6522574424743652,
710
+ "step": 200
711
+ },
712
+ {
713
+ "epoch": 0.5386666666666666,
714
+ "grad_norm": 0.84375,
715
+ "learning_rate": 6.967436454763442e-06,
716
+ "loss": 1.5417842864990234,
717
+ "step": 202
718
+ },
719
+ {
720
+ "epoch": 0.544,
721
+ "grad_norm": 0.58203125,
722
+ "learning_rate": 6.950902214125689e-06,
723
+ "loss": 1.279434323310852,
724
+ "step": 204
725
+ },
726
+ {
727
+ "epoch": 0.5493333333333333,
728
+ "grad_norm": 0.474609375,
729
+ "learning_rate": 6.9341393662071115e-06,
730
+ "loss": 1.5310899019241333,
731
+ "step": 206
732
+ },
733
+ {
734
+ "epoch": 0.5546666666666666,
735
+ "grad_norm": 0.4375,
736
+ "learning_rate": 6.9171493634432225e-06,
737
+ "loss": 1.1565041542053223,
738
+ "step": 208
739
+ },
740
+ {
741
+ "epoch": 0.56,
742
+ "grad_norm": 0.3828125,
743
+ "learning_rate": 6.899933677951622e-06,
744
+ "loss": 1.3370246887207031,
745
+ "step": 210
746
+ },
747
+ {
748
+ "epoch": 0.5653333333333334,
749
+ "grad_norm": 0.37890625,
750
+ "learning_rate": 6.882493801404442e-06,
751
+ "loss": 1.330797791481018,
752
+ "step": 212
753
+ },
754
+ {
755
+ "epoch": 0.5706666666666667,
756
+ "grad_norm": 0.81640625,
757
+ "learning_rate": 6.8648312448990875e-06,
758
+ "loss": 1.4331892728805542,
759
+ "step": 214
760
+ },
761
+ {
762
+ "epoch": 0.576,
763
+ "grad_norm": 1.6875,
764
+ "learning_rate": 6.846947538827327e-06,
765
+ "loss": 1.5132179260253906,
766
+ "step": 216
767
+ },
768
+ {
769
+ "epoch": 0.5813333333333334,
770
+ "grad_norm": 1.46875,
771
+ "learning_rate": 6.828844232742671e-06,
772
+ "loss": 1.2577152252197266,
773
+ "step": 218
774
+ },
775
+ {
776
+ "epoch": 0.5866666666666667,
777
+ "grad_norm": 0.9453125,
778
+ "learning_rate": 6.810522895226115e-06,
779
+ "loss": 1.45955491065979,
780
+ "step": 220
781
+ },
782
+ {
783
+ "epoch": 0.592,
784
+ "grad_norm": 0.2470703125,
785
+ "learning_rate": 6.791985113750234e-06,
786
+ "loss": 1.2781492471694946,
787
+ "step": 222
788
+ },
789
+ {
790
+ "epoch": 0.5973333333333334,
791
+ "grad_norm": 0.40625,
792
+ "learning_rate": 6.773232494541625e-06,
793
+ "loss": 1.3114062547683716,
794
+ "step": 224
795
+ },
796
+ {
797
+ "epoch": 0.6026666666666667,
798
+ "grad_norm": 1.0546875,
799
+ "learning_rate": 6.754266662441742e-06,
800
+ "loss": 1.4861233234405518,
801
+ "step": 226
802
+ },
803
+ {
804
+ "epoch": 0.608,
805
+ "grad_norm": 0.671875,
806
+ "learning_rate": 6.735089260766105e-06,
807
+ "loss": 1.6322368383407593,
808
+ "step": 228
809
+ },
810
+ {
811
+ "epoch": 0.6133333333333333,
812
+ "grad_norm": 0.7421875,
813
+ "learning_rate": 6.7157019511619145e-06,
814
+ "loss": 1.5255112648010254,
815
+ "step": 230
816
+ },
817
+ {
818
+ "epoch": 0.6186666666666667,
819
+ "grad_norm": 0.56640625,
820
+ "learning_rate": 6.696106413464071e-06,
821
+ "loss": 1.2472528219223022,
822
+ "step": 232
823
+ },
824
+ {
825
+ "epoch": 0.624,
826
+ "grad_norm": 0.5625,
827
+ "learning_rate": 6.676304345549635e-06,
828
+ "loss": 1.5083709955215454,
829
+ "step": 234
830
+ },
831
+ {
832
+ "epoch": 0.6293333333333333,
833
+ "grad_norm": 0.62890625,
834
+ "learning_rate": 6.656297463190704e-06,
835
+ "loss": 1.216172218322754,
836
+ "step": 236
837
+ },
838
+ {
839
+ "epoch": 0.6346666666666667,
840
+ "grad_norm": 0.5859375,
841
+ "learning_rate": 6.63608749990575e-06,
842
+ "loss": 1.2348906993865967,
843
+ "step": 238
844
+ },
845
+ {
846
+ "epoch": 0.64,
847
+ "grad_norm": 0.345703125,
848
+ "learning_rate": 6.615676206809415e-06,
849
+ "loss": 1.3027633428573608,
850
+ "step": 240
851
+ },
852
+ {
853
+ "epoch": 0.6453333333333333,
854
+ "grad_norm": 0.38671875,
855
+ "learning_rate": 6.595065352460786e-06,
856
+ "loss": 1.1347825527191162,
857
+ "step": 242
858
+ },
859
+ {
860
+ "epoch": 0.6506666666666666,
861
+ "grad_norm": 0.5625,
862
+ "learning_rate": 6.574256722710161e-06,
863
+ "loss": 1.271606206893921,
864
+ "step": 244
865
+ },
866
+ {
867
+ "epoch": 0.656,
868
+ "grad_norm": 0.69921875,
869
+ "learning_rate": 6.553252120544297e-06,
870
+ "loss": 1.297676682472229,
871
+ "step": 246
872
+ },
873
+ {
874
+ "epoch": 0.6613333333333333,
875
+ "grad_norm": 0.357421875,
876
+ "learning_rate": 6.5320533659302055e-06,
877
+ "loss": 1.3999580144882202,
878
+ "step": 248
879
+ },
880
+ {
881
+ "epoch": 0.6666666666666666,
882
+ "grad_norm": 0.498046875,
883
+ "learning_rate": 6.510662295657448e-06,
884
+ "loss": 1.4784775972366333,
885
+ "step": 250
886
+ },
887
+ {
888
+ "epoch": 0.672,
889
+ "grad_norm": 0.33984375,
890
+ "learning_rate": 6.489080763178991e-06,
891
+ "loss": 1.3089486360549927,
892
+ "step": 252
893
+ },
894
+ {
895
+ "epoch": 0.6773333333333333,
896
+ "grad_norm": 0.451171875,
897
+ "learning_rate": 6.467310638450608e-06,
898
+ "loss": 1.5869042873382568,
899
+ "step": 254
900
+ },
901
+ {
902
+ "epoch": 0.6826666666666666,
903
+ "grad_norm": 0.5390625,
904
+ "learning_rate": 6.44535380776886e-06,
905
+ "loss": 1.491261601448059,
906
+ "step": 256
907
+ },
908
+ {
909
+ "epoch": 0.688,
910
+ "grad_norm": 0.53125,
911
+ "learning_rate": 6.42321217360765e-06,
912
+ "loss": 1.3518444299697876,
913
+ "step": 258
914
+ },
915
+ {
916
+ "epoch": 0.6933333333333334,
917
+ "grad_norm": 0.408203125,
918
+ "learning_rate": 6.40088765445338e-06,
919
+ "loss": 1.1858634948730469,
920
+ "step": 260
921
+ },
922
+ {
923
+ "epoch": 0.6986666666666667,
924
+ "grad_norm": 0.44921875,
925
+ "learning_rate": 6.37838218463873e-06,
926
+ "loss": 1.3287537097930908,
927
+ "step": 262
928
+ },
929
+ {
930
+ "epoch": 0.704,
931
+ "grad_norm": 0.478515625,
932
+ "learning_rate": 6.355697714175047e-06,
933
+ "loss": 1.0604113340377808,
934
+ "step": 264
935
+ },
936
+ {
937
+ "epoch": 0.7093333333333334,
938
+ "grad_norm": 0.9453125,
939
+ "learning_rate": 6.332836208583389e-06,
940
+ "loss": 1.5299183130264282,
941
+ "step": 266
942
+ },
943
+ {
944
+ "epoch": 0.7146666666666667,
945
+ "grad_norm": 0.421875,
946
+ "learning_rate": 6.309799648724218e-06,
947
+ "loss": 1.2309736013412476,
948
+ "step": 268
949
+ },
950
+ {
951
+ "epoch": 0.72,
952
+ "grad_norm": 0.7265625,
953
+ "learning_rate": 6.2865900306257665e-06,
954
+ "loss": 1.0823864936828613,
955
+ "step": 270
956
+ },
957
+ {
958
+ "epoch": 0.7253333333333334,
959
+ "grad_norm": 0.435546875,
960
+ "learning_rate": 6.263209365311089e-06,
961
+ "loss": 1.2900185585021973,
962
+ "step": 272
963
+ },
964
+ {
965
+ "epoch": 0.7306666666666667,
966
+ "grad_norm": 0.6875,
967
+ "learning_rate": 6.2396596786238205e-06,
968
+ "loss": 1.2881044149398804,
969
+ "step": 274
970
+ },
971
+ {
972
+ "epoch": 0.736,
973
+ "grad_norm": 1.03125,
974
+ "learning_rate": 6.215943011052632e-06,
975
+ "loss": 1.3503191471099854,
976
+ "step": 276
977
+ },
978
+ {
979
+ "epoch": 0.7413333333333333,
980
+ "grad_norm": 0.63671875,
981
+ "learning_rate": 6.192061417554441e-06,
982
+ "loss": 1.33072030544281,
983
+ "step": 278
984
+ },
985
+ {
986
+ "epoch": 0.7466666666666667,
987
+ "grad_norm": 0.42578125,
988
+ "learning_rate": 6.16801696737636e-06,
989
+ "loss": 1.2632883787155151,
990
+ "step": 280
991
+ },
992
+ {
993
+ "epoch": 0.752,
994
+ "grad_norm": 0.349609375,
995
+ "learning_rate": 6.143811743876392e-06,
996
+ "loss": 1.285319447517395,
997
+ "step": 282
998
+ },
999
+ {
1000
+ "epoch": 0.7573333333333333,
1001
+ "grad_norm": 0.287109375,
1002
+ "learning_rate": 6.119447844342924e-06,
1003
+ "loss": 1.0543291568756104,
1004
+ "step": 284
1005
+ },
1006
+ {
1007
+ "epoch": 0.7626666666666667,
1008
+ "grad_norm": 1.0,
1009
+ "learning_rate": 6.094927379813005e-06,
1010
+ "loss": 1.5204623937606812,
1011
+ "step": 286
1012
+ },
1013
+ {
1014
+ "epoch": 0.768,
1015
+ "grad_norm": 0.5390625,
1016
+ "learning_rate": 6.070252474889433e-06,
1017
+ "loss": 0.9127747416496277,
1018
+ "step": 288
1019
+ },
1020
+ {
1021
+ "epoch": 0.7733333333333333,
1022
+ "grad_norm": 0.279296875,
1023
+ "learning_rate": 6.045425267556659e-06,
1024
+ "loss": 1.2986137866973877,
1025
+ "step": 290
1026
+ },
1027
+ {
1028
+ "epoch": 0.7786666666666666,
1029
+ "grad_norm": 0.58984375,
1030
+ "learning_rate": 6.020447908995551e-06,
1031
+ "loss": 1.482648253440857,
1032
+ "step": 292
1033
+ },
1034
+ {
1035
+ "epoch": 0.784,
1036
+ "grad_norm": 0.39453125,
1037
+ "learning_rate": 5.995322563396996e-06,
1038
+ "loss": 1.1602120399475098,
1039
+ "step": 294
1040
+ },
1041
+ {
1042
+ "epoch": 0.7893333333333333,
1043
+ "grad_norm": 1.0859375,
1044
+ "learning_rate": 5.970051407774374e-06,
1045
+ "loss": 1.2999836206436157,
1046
+ "step": 296
1047
+ },
1048
+ {
1049
+ "epoch": 0.7946666666666666,
1050
+ "grad_norm": 0.31640625,
1051
+ "learning_rate": 5.9446366317749464e-06,
1052
+ "loss": 1.464931607246399,
1053
+ "step": 298
1054
+ },
1055
+ {
1056
+ "epoch": 0.8,
1057
+ "grad_norm": 3.78125,
1058
+ "learning_rate": 5.919080437490115e-06,
1059
+ "loss": 1.494107961654663,
1060
+ "step": 300
1061
+ },
1062
+ {
1063
+ "epoch": 0.8053333333333333,
1064
+ "grad_norm": 0.45703125,
1065
+ "learning_rate": 5.893385039264628e-06,
1066
+ "loss": 1.3205267190933228,
1067
+ "step": 302
1068
+ },
1069
+ {
1070
+ "epoch": 0.8106666666666666,
1071
+ "grad_norm": 0.306640625,
1072
+ "learning_rate": 5.867552663504711e-06,
1073
+ "loss": 1.392236351966858,
1074
+ "step": 304
1075
+ },
1076
+ {
1077
+ "epoch": 0.816,
1078
+ "grad_norm": 2.109375,
1079
+ "learning_rate": 5.841585548485161e-06,
1080
+ "loss": 1.054479718208313,
1081
+ "step": 306
1082
+ },
1083
+ {
1084
+ "epoch": 0.8213333333333334,
1085
+ "grad_norm": 0.474609375,
1086
+ "learning_rate": 5.815485944155406e-06,
1087
+ "loss": 1.4559931755065918,
1088
+ "step": 308
1089
+ },
1090
+ {
1091
+ "epoch": 0.8266666666666667,
1092
+ "grad_norm": 1.6484375,
1093
+ "learning_rate": 5.78925611194456e-06,
1094
+ "loss": 1.2822375297546387,
1095
+ "step": 310
1096
+ },
1097
+ {
1098
+ "epoch": 0.832,
1099
+ "grad_norm": 0.5234375,
1100
+ "learning_rate": 5.762898324565472e-06,
1101
+ "loss": 1.1534578800201416,
1102
+ "step": 312
1103
+ },
1104
+ {
1105
+ "epoch": 0.8373333333333334,
1106
+ "grad_norm": 0.40234375,
1107
+ "learning_rate": 5.7364148658178136e-06,
1108
+ "loss": 1.4816879034042358,
1109
+ "step": 314
1110
+ },
1111
+ {
1112
+ "epoch": 0.8426666666666667,
1113
+ "grad_norm": 0.48828125,
1114
+ "learning_rate": 5.7098080303901845e-06,
1115
+ "loss": 1.3202126026153564,
1116
+ "step": 316
1117
+ },
1118
+ {
1119
+ "epoch": 0.848,
1120
+ "grad_norm": 0.57421875,
1121
+ "learning_rate": 5.683080123661299e-06,
1122
+ "loss": 1.548636794090271,
1123
+ "step": 318
1124
+ },
1125
+ {
1126
+ "epoch": 0.8533333333333334,
1127
+ "grad_norm": 0.18359375,
1128
+ "learning_rate": 5.656233461500226e-06,
1129
+ "loss": 1.1134190559387207,
1130
+ "step": 320
1131
+ },
1132
+ {
1133
+ "epoch": 0.8586666666666667,
1134
+ "grad_norm": 0.369140625,
1135
+ "learning_rate": 5.6292703700657276e-06,
1136
+ "loss": 1.427626132965088,
1137
+ "step": 322
1138
+ },
1139
+ {
1140
+ "epoch": 0.864,
1141
+ "grad_norm": 0.66796875,
1142
+ "learning_rate": 5.602193185604715e-06,
1143
+ "loss": 1.2644096612930298,
1144
+ "step": 324
1145
+ },
1146
+ {
1147
+ "epoch": 0.8693333333333333,
1148
+ "grad_norm": 0.419921875,
1149
+ "learning_rate": 5.575004254249812e-06,
1150
+ "loss": 1.0416737794876099,
1151
+ "step": 326
1152
+ },
1153
+ {
1154
+ "epoch": 0.8746666666666667,
1155
+ "grad_norm": 0.32421875,
1156
+ "learning_rate": 5.547705931816078e-06,
1157
+ "loss": 1.113074779510498,
1158
+ "step": 328
1159
+ },
1160
+ {
1161
+ "epoch": 0.88,
1162
+ "grad_norm": 1.2578125,
1163
+ "learning_rate": 5.520300583596882e-06,
1164
+ "loss": 1.0335323810577393,
1165
+ "step": 330
1166
+ },
1167
+ {
1168
+ "epoch": 0.8853333333333333,
1169
+ "grad_norm": 0.3125,
1170
+ "learning_rate": 5.49279058415896e-06,
1171
+ "loss": 1.1414536237716675,
1172
+ "step": 332
1173
+ },
1174
+ {
1175
+ "epoch": 0.8906666666666667,
1176
+ "grad_norm": 0.37890625,
1177
+ "learning_rate": 5.465178317136669e-06,
1178
+ "loss": 0.8871914148330688,
1179
+ "step": 334
1180
+ },
1181
+ {
1182
+ "epoch": 0.896,
1183
+ "grad_norm": 0.578125,
1184
+ "learning_rate": 5.437466175025457e-06,
1185
+ "loss": 1.3287149667739868,
1186
+ "step": 336
1187
+ },
1188
+ {
1189
+ "epoch": 0.9013333333333333,
1190
+ "grad_norm": 0.59765625,
1191
+ "learning_rate": 5.4096565589745565e-06,
1192
+ "loss": 1.3277924060821533,
1193
+ "step": 338
1194
+ },
1195
+ {
1196
+ "epoch": 0.9066666666666666,
1197
+ "grad_norm": 0.55078125,
1198
+ "learning_rate": 5.3817518785789386e-06,
1199
+ "loss": 1.6219979524612427,
1200
+ "step": 340
1201
+ },
1202
+ {
1203
+ "epoch": 0.912,
1204
+ "grad_norm": 0.337890625,
1205
+ "learning_rate": 5.3537545516705285e-06,
1206
+ "loss": 1.1041191816329956,
1207
+ "step": 342
1208
+ },
1209
+ {
1210
+ "epoch": 0.9173333333333333,
1211
+ "grad_norm": 0.54296875,
1212
+ "learning_rate": 5.325667004108713e-06,
1213
+ "loss": 1.5842766761779785,
1214
+ "step": 344
1215
+ },
1216
+ {
1217
+ "epoch": 0.9226666666666666,
1218
+ "grad_norm": 0.77734375,
1219
+ "learning_rate": 5.297491669570145e-06,
1220
+ "loss": 1.5403611660003662,
1221
+ "step": 346
1222
+ },
1223
+ {
1224
+ "epoch": 0.928,
1225
+ "grad_norm": 0.39453125,
1226
+ "learning_rate": 5.2692309893378806e-06,
1227
+ "loss": 1.387043833732605,
1228
+ "step": 348
1229
+ },
1230
+ {
1231
+ "epoch": 0.9333333333333333,
1232
+ "grad_norm": 0.376953125,
1233
+ "learning_rate": 5.240887412089846e-06,
1234
+ "loss": 1.1895140409469604,
1235
+ "step": 350
1236
+ },
1237
+ {
1238
+ "epoch": 0.9386666666666666,
1239
+ "grad_norm": 0.7734375,
1240
+ "learning_rate": 5.2124633936866694e-06,
1241
+ "loss": 1.45750892162323,
1242
+ "step": 352
1243
+ },
1244
+ {
1245
+ "epoch": 0.944,
1246
+ "grad_norm": 0.48828125,
1247
+ "learning_rate": 5.183961396958894e-06,
1248
+ "loss": 1.2903810739517212,
1249
+ "step": 354
1250
+ },
1251
+ {
1252
+ "epoch": 0.9493333333333334,
1253
+ "grad_norm": 0.365234375,
1254
+ "learning_rate": 5.155383891493582e-06,
1255
+ "loss": 1.0638062953948975,
1256
+ "step": 356
1257
+ },
1258
+ {
1259
+ "epoch": 0.9546666666666667,
1260
+ "grad_norm": 0.392578125,
1261
+ "learning_rate": 5.126733353420332e-06,
1262
+ "loss": 1.1334384679794312,
1263
+ "step": 358
1264
+ },
1265
+ {
1266
+ "epoch": 0.96,
1267
+ "grad_norm": 0.8671875,
1268
+ "learning_rate": 5.098012265196735e-06,
1269
+ "loss": 1.616370677947998,
1270
+ "step": 360
1271
+ },
1272
+ {
1273
+ "epoch": 0.9653333333333334,
1274
+ "grad_norm": 0.458984375,
1275
+ "learning_rate": 5.069223115393278e-06,
1276
+ "loss": 1.2166802883148193,
1277
+ "step": 362
1278
+ },
1279
+ {
1280
+ "epoch": 0.9706666666666667,
1281
+ "grad_norm": 0.439453125,
1282
+ "learning_rate": 5.0403683984777156e-06,
1283
+ "loss": 1.460686206817627,
1284
+ "step": 364
1285
+ },
1286
+ {
1287
+ "epoch": 0.976,
1288
+ "grad_norm": 0.412109375,
1289
+ "learning_rate": 5.011450614598942e-06,
1290
+ "loss": 1.5133981704711914,
1291
+ "step": 366
1292
+ },
1293
+ {
1294
+ "epoch": 0.9813333333333333,
1295
+ "grad_norm": 0.51953125,
1296
+ "learning_rate": 4.982472269370356e-06,
1297
+ "loss": 1.3987483978271484,
1298
+ "step": 368
1299
+ },
1300
+ {
1301
+ "epoch": 0.9866666666666667,
1302
+ "grad_norm": 0.5,
1303
+ "learning_rate": 4.9534358736527624e-06,
1304
+ "loss": 1.187402606010437,
1305
+ "step": 370
1306
+ },
1307
+ {
1308
+ "epoch": 0.992,
1309
+ "grad_norm": 0.388671875,
1310
+ "learning_rate": 4.924343943336814e-06,
1311
+ "loss": 1.3969718217849731,
1312
+ "step": 372
1313
+ },
1314
+ {
1315
+ "epoch": 0.9973333333333333,
1316
+ "grad_norm": 0.5078125,
1317
+ "learning_rate": 4.895198999125022e-06,
1318
+ "loss": 1.1329926252365112,
1319
+ "step": 374
1320
+ },
1321
+ {
1322
+ "epoch": 1.0026666666666666,
1323
+ "grad_norm": 0.76953125,
1324
+ "learning_rate": 4.866003566313341e-06,
1325
+ "loss": 1.426386833190918,
1326
+ "step": 376
1327
+ },
1328
+ {
1329
+ "epoch": 1.008,
1330
+ "grad_norm": 0.5859375,
1331
+ "learning_rate": 4.83676017457237e-06,
1332
+ "loss": 1.2209362983703613,
1333
+ "step": 378
1334
+ },
1335
+ {
1336
+ "epoch": 1.0133333333333334,
1337
+ "grad_norm": 0.59375,
1338
+ "learning_rate": 4.807471357728167e-06,
1339
+ "loss": 1.1464167833328247,
1340
+ "step": 380
1341
+ },
1342
+ {
1343
+ "epoch": 1.0186666666666666,
1344
+ "grad_norm": 1.0078125,
1345
+ "learning_rate": 4.778139653542688e-06,
1346
+ "loss": 1.229437232017517,
1347
+ "step": 382
1348
+ },
1349
+ {
1350
+ "epoch": 1.024,
1351
+ "grad_norm": 0.337890625,
1352
+ "learning_rate": 4.7487676034939156e-06,
1353
+ "loss": 0.9976918697357178,
1354
+ "step": 384
1355
+ },
1356
+ {
1357
+ "epoch": 1.0293333333333334,
1358
+ "grad_norm": 0.310546875,
1359
+ "learning_rate": 4.719357752555646e-06,
1360
+ "loss": 1.025425672531128,
1361
+ "step": 386
1362
+ },
1363
+ {
1364
+ "epoch": 1.0346666666666666,
1365
+ "grad_norm": 0.2216796875,
1366
+ "learning_rate": 4.689912648976973e-06,
1367
+ "loss": 1.0078941583633423,
1368
+ "step": 388
1369
+ },
1370
+ {
1371
+ "epoch": 1.04,
1372
+ "grad_norm": 0.4296875,
1373
+ "learning_rate": 4.660434844061493e-06,
1374
+ "loss": 1.1683357954025269,
1375
+ "step": 390
1376
+ },
1377
+ {
1378
+ "epoch": 1.0453333333333332,
1379
+ "grad_norm": 0.51953125,
1380
+ "learning_rate": 4.630926891946251e-06,
1381
+ "loss": 1.126926302909851,
1382
+ "step": 392
1383
+ },
1384
+ {
1385
+ "epoch": 1.0506666666666666,
1386
+ "grad_norm": 0.5,
1387
+ "learning_rate": 4.601391349380425e-06,
1388
+ "loss": 1.2865811586380005,
1389
+ "step": 394
1390
+ },
1391
+ {
1392
+ "epoch": 1.056,
1393
+ "grad_norm": 0.7734375,
1394
+ "learning_rate": 4.571830775503799e-06,
1395
+ "loss": 1.5480324029922485,
1396
+ "step": 396
1397
+ },
1398
+ {
1399
+ "epoch": 1.0613333333333332,
1400
+ "grad_norm": 0.55859375,
1401
+ "learning_rate": 4.542247731625022e-06,
1402
+ "loss": 1.2431285381317139,
1403
+ "step": 398
1404
+ },
1405
+ {
1406
+ "epoch": 1.0666666666666667,
1407
+ "grad_norm": 0.546875,
1408
+ "learning_rate": 4.512644780999687e-06,
1409
+ "loss": 1.4781147241592407,
1410
+ "step": 400
1411
+ },
1412
+ {
1413
+ "epoch": 1.072,
1414
+ "grad_norm": 2.515625,
1415
+ "learning_rate": 4.48302448860822e-06,
1416
+ "loss": 1.0834038257598877,
1417
+ "step": 402
1418
+ },
1419
+ {
1420
+ "epoch": 1.0773333333333333,
1421
+ "grad_norm": 0.6328125,
1422
+ "learning_rate": 4.453389420933649e-06,
1423
+ "loss": 1.366401195526123,
1424
+ "step": 404
1425
+ },
1426
+ {
1427
+ "epoch": 1.0826666666666667,
1428
+ "grad_norm": 0.271484375,
1429
+ "learning_rate": 4.423742145739222e-06,
1430
+ "loss": 1.0470842123031616,
1431
+ "step": 406
1432
+ },
1433
+ {
1434
+ "epoch": 1.088,
1435
+ "grad_norm": 0.431640625,
1436
+ "learning_rate": 4.394085231845921e-06,
1437
+ "loss": 1.0181280374526978,
1438
+ "step": 408
1439
+ },
1440
+ {
1441
+ "epoch": 1.0933333333333333,
1442
+ "grad_norm": 0.65625,
1443
+ "learning_rate": 4.364421248909884e-06,
1444
+ "loss": 1.034490704536438,
1445
+ "step": 410
1446
+ },
1447
+ {
1448
+ "epoch": 1.0986666666666667,
1449
+ "grad_norm": 0.70703125,
1450
+ "learning_rate": 4.334752767199752e-06,
1451
+ "loss": 1.3856372833251953,
1452
+ "step": 412
1453
+ },
1454
+ {
1455
+ "epoch": 1.104,
1456
+ "grad_norm": 0.373046875,
1457
+ "learning_rate": 4.305082357373971e-06,
1458
+ "loss": 1.139554500579834,
1459
+ "step": 414
1460
+ },
1461
+ {
1462
+ "epoch": 1.1093333333333333,
1463
+ "grad_norm": 0.51171875,
1464
+ "learning_rate": 4.275412590258048e-06,
1465
+ "loss": 1.3391931056976318,
1466
+ "step": 416
1467
+ },
1468
+ {
1469
+ "epoch": 1.1146666666666667,
1470
+ "grad_norm": 0.349609375,
1471
+ "learning_rate": 4.2457460366218e-06,
1472
+ "loss": 1.5631487369537354,
1473
+ "step": 418
1474
+ },
1475
+ {
1476
+ "epoch": 1.12,
1477
+ "grad_norm": 0.6875,
1478
+ "learning_rate": 4.2160852669566135e-06,
1479
+ "loss": 1.3829050064086914,
1480
+ "step": 420
1481
+ },
1482
+ {
1483
+ "epoch": 1.1253333333333333,
1484
+ "grad_norm": 0.97265625,
1485
+ "learning_rate": 4.1864328512527115e-06,
1486
+ "loss": 1.0705004930496216,
1487
+ "step": 422
1488
+ },
1489
+ {
1490
+ "epoch": 1.1306666666666667,
1491
+ "grad_norm": 0.68359375,
1492
+ "learning_rate": 4.156791358776483e-06,
1493
+ "loss": 1.4946973323822021,
1494
+ "step": 424
1495
+ },
1496
+ {
1497
+ "epoch": 1.1360000000000001,
1498
+ "grad_norm": 0.384765625,
1499
+ "learning_rate": 4.127163357847858e-06,
1500
+ "loss": 1.3657976388931274,
1501
+ "step": 426
1502
+ },
1503
+ {
1504
+ "epoch": 1.1413333333333333,
1505
+ "grad_norm": 0.447265625,
1506
+ "learning_rate": 4.09755141561778e-06,
1507
+ "loss": 1.211698055267334,
1508
+ "step": 428
1509
+ },
1510
+ {
1511
+ "epoch": 1.1466666666666667,
1512
+ "grad_norm": 12.8125,
1513
+ "learning_rate": 4.0679580978457656e-06,
1514
+ "loss": 1.5211821794509888,
1515
+ "step": 430
1516
+ },
1517
+ {
1518
+ "epoch": 1.152,
1519
+ "grad_norm": 0.46875,
1520
+ "learning_rate": 4.038385968677596e-06,
1521
+ "loss": 1.0360568761825562,
1522
+ "step": 432
1523
+ },
1524
+ {
1525
+ "epoch": 1.1573333333333333,
1526
+ "grad_norm": 1.3125,
1527
+ "learning_rate": 4.008837590423144e-06,
1528
+ "loss": 1.6035184860229492,
1529
+ "step": 434
1530
+ },
1531
+ {
1532
+ "epoch": 1.1626666666666667,
1533
+ "grad_norm": 0.48828125,
1534
+ "learning_rate": 3.979315523334355e-06,
1535
+ "loss": 1.157009243965149,
1536
+ "step": 436
1537
+ },
1538
+ {
1539
+ "epoch": 1.168,
1540
+ "grad_norm": 0.458984375,
1541
+ "learning_rate": 3.9498223253834165e-06,
1542
+ "loss": 1.4968481063842773,
1543
+ "step": 438
1544
+ },
1545
+ {
1546
+ "epoch": 1.1733333333333333,
1547
+ "grad_norm": 0.5234375,
1548
+ "learning_rate": 3.9203605520411165e-06,
1549
+ "loss": 1.229710340499878,
1550
+ "step": 440
1551
+ },
1552
+ {
1553
+ "epoch": 1.1786666666666668,
1554
+ "grad_norm": 0.734375,
1555
+ "learning_rate": 3.890932756055422e-06,
1556
+ "loss": 1.5977742671966553,
1557
+ "step": 442
1558
+ },
1559
+ {
1560
+ "epoch": 1.184,
1561
+ "grad_norm": 0.75390625,
1562
+ "learning_rate": 3.861541487230299e-06,
1563
+ "loss": 1.2765097618103027,
1564
+ "step": 444
1565
+ },
1566
+ {
1567
+ "epoch": 1.1893333333333334,
1568
+ "grad_norm": 0.59765625,
1569
+ "learning_rate": 3.832189292204773e-06,
1570
+ "loss": 1.3562519550323486,
1571
+ "step": 446
1572
+ },
1573
+ {
1574
+ "epoch": 1.1946666666666665,
1575
+ "grad_norm": 0.4453125,
1576
+ "learning_rate": 3.802878714232277e-06,
1577
+ "loss": 1.6301568746566772,
1578
+ "step": 448
1579
+ },
1580
+ {
1581
+ "epoch": 1.2,
1582
+ "grad_norm": 0.47265625,
1583
+ "learning_rate": 3.7736122929602887e-06,
1584
+ "loss": 1.264549732208252,
1585
+ "step": 450
1586
+ },
1587
+ {
1588
+ "epoch": 1.2053333333333334,
1589
+ "grad_norm": 0.443359375,
1590
+ "learning_rate": 3.744392564210283e-06,
1591
+ "loss": 1.4258700609207153,
1592
+ "step": 452
1593
+ },
1594
+ {
1595
+ "epoch": 1.2106666666666666,
1596
+ "grad_norm": 0.56640625,
1597
+ "learning_rate": 3.7152220597580023e-06,
1598
+ "loss": 1.4381071329116821,
1599
+ "step": 454
1600
+ },
1601
+ {
1602
+ "epoch": 1.216,
1603
+ "grad_norm": 0.42578125,
1604
+ "learning_rate": 3.686103307114104e-06,
1605
+ "loss": 1.4425290822982788,
1606
+ "step": 456
1607
+ },
1608
+ {
1609
+ "epoch": 1.2213333333333334,
1610
+ "grad_norm": 0.419921875,
1611
+ "learning_rate": 3.6570388293051458e-06,
1612
+ "loss": 1.2543679475784302,
1613
+ "step": 458
1614
+ },
1615
+ {
1616
+ "epoch": 1.2266666666666666,
1617
+ "grad_norm": 0.392578125,
1618
+ "learning_rate": 3.628031144654986e-06,
1619
+ "loss": 1.3114312887191772,
1620
+ "step": 460
1621
+ },
1622
+ {
1623
+ "epoch": 1.232,
1624
+ "grad_norm": 0.53125,
1625
+ "learning_rate": 3.5990827665665756e-06,
1626
+ "loss": 1.4582160711288452,
1627
+ "step": 462
1628
+ },
1629
+ {
1630
+ "epoch": 1.2373333333333334,
1631
+ "grad_norm": 0.53125,
1632
+ "learning_rate": 3.570196203304182e-06,
1633
+ "loss": 0.9986866116523743,
1634
+ "step": 464
1635
+ },
1636
+ {
1637
+ "epoch": 1.2426666666666666,
1638
+ "grad_norm": 0.8984375,
1639
+ "learning_rate": 3.541373957776059e-06,
1640
+ "loss": 1.124342918395996,
1641
+ "step": 466
1642
+ },
1643
+ {
1644
+ "epoch": 1.248,
1645
+ "grad_norm": 0.62109375,
1646
+ "learning_rate": 3.512618527317581e-06,
1647
+ "loss": 1.3204665184020996,
1648
+ "step": 468
1649
+ },
1650
+ {
1651
+ "epoch": 1.2533333333333334,
1652
+ "grad_norm": 0.349609375,
1653
+ "learning_rate": 3.483932403474856e-06,
1654
+ "loss": 1.1635738611221313,
1655
+ "step": 470
1656
+ },
1657
+ {
1658
+ "epoch": 1.2586666666666666,
1659
+ "grad_norm": 0.52734375,
1660
+ "learning_rate": 3.4553180717888407e-06,
1661
+ "loss": 1.0901280641555786,
1662
+ "step": 472
1663
+ },
1664
+ {
1665
+ "epoch": 1.264,
1666
+ "grad_norm": 0.41015625,
1667
+ "learning_rate": 3.4267780115799846e-06,
1668
+ "loss": 1.0830848217010498,
1669
+ "step": 474
1670
+ },
1671
+ {
1672
+ "epoch": 1.2693333333333334,
1673
+ "grad_norm": 0.310546875,
1674
+ "learning_rate": 3.398314695733401e-06,
1675
+ "loss": 1.3727480173110962,
1676
+ "step": 476
1677
+ },
1678
+ {
1679
+ "epoch": 1.2746666666666666,
1680
+ "grad_norm": 0.578125,
1681
+ "learning_rate": 3.3699305904846027e-06,
1682
+ "loss": 1.6773924827575684,
1683
+ "step": 478
1684
+ },
1685
+ {
1686
+ "epoch": 1.28,
1687
+ "grad_norm": 0.55078125,
1688
+ "learning_rate": 3.341628155205811e-06,
1689
+ "loss": 1.2972683906555176,
1690
+ "step": 480
1691
+ },
1692
+ {
1693
+ "epoch": 1.2853333333333334,
1694
+ "grad_norm": 0.71875,
1695
+ "learning_rate": 3.3134098421928692e-06,
1696
+ "loss": 1.2166887521743774,
1697
+ "step": 482
1698
+ },
1699
+ {
1700
+ "epoch": 1.2906666666666666,
1701
+ "grad_norm": 1.5078125,
1702
+ "learning_rate": 3.2852780964527466e-06,
1703
+ "loss": 1.3262923955917358,
1704
+ "step": 484
1705
+ },
1706
+ {
1707
+ "epoch": 1.296,
1708
+ "grad_norm": 0.2216796875,
1709
+ "learning_rate": 3.2572353554916966e-06,
1710
+ "loss": 1.2588746547698975,
1711
+ "step": 486
1712
+ },
1713
+ {
1714
+ "epoch": 1.3013333333333335,
1715
+ "grad_norm": 0.43359375,
1716
+ "learning_rate": 3.229284049104057e-06,
1717
+ "loss": 0.8871427774429321,
1718
+ "step": 488
1719
+ },
1720
+ {
1721
+ "epoch": 1.3066666666666666,
1722
+ "grad_norm": 0.259765625,
1723
+ "learning_rate": 3.2014265991617147e-06,
1724
+ "loss": 1.3596696853637695,
1725
+ "step": 490
1726
+ },
1727
+ {
1728
+ "epoch": 1.312,
1729
+ "grad_norm": 0.62890625,
1730
+ "learning_rate": 3.17366541940426e-06,
1731
+ "loss": 1.5204687118530273,
1732
+ "step": 492
1733
+ },
1734
+ {
1735
+ "epoch": 1.3173333333333335,
1736
+ "grad_norm": 0.32421875,
1737
+ "learning_rate": 3.1460029152298484e-06,
1738
+ "loss": 1.2328211069107056,
1739
+ "step": 494
1740
+ },
1741
+ {
1742
+ "epoch": 1.3226666666666667,
1743
+ "grad_norm": 0.43359375,
1744
+ "learning_rate": 3.1184414834867804e-06,
1745
+ "loss": 1.2612500190734863,
1746
+ "step": 496
1747
+ },
1748
+ {
1749
+ "epoch": 1.328,
1750
+ "grad_norm": 0.328125,
1751
+ "learning_rate": 3.0909835122658214e-06,
1752
+ "loss": 1.230236291885376,
1753
+ "step": 498
1754
+ },
1755
+ {
1756
+ "epoch": 1.3333333333333333,
1757
+ "grad_norm": 0.8046875,
1758
+ "learning_rate": 3.063631380693286e-06,
1759
+ "loss": 1.016282558441162,
1760
+ "step": 500
1761
+ },
1762
+ {
1763
+ "epoch": 1.3386666666666667,
1764
+ "grad_norm": 0.451171875,
1765
+ "learning_rate": 3.036387458724898e-06,
1766
+ "loss": 1.2729389667510986,
1767
+ "step": 502
1768
+ },
1769
+ {
1770
+ "epoch": 1.3439999999999999,
1771
+ "grad_norm": 0.396484375,
1772
+ "learning_rate": 3.009254106940435e-06,
1773
+ "loss": 1.2068243026733398,
1774
+ "step": 504
1775
+ },
1776
+ {
1777
+ "epoch": 1.3493333333333333,
1778
+ "grad_norm": 0.51953125,
1779
+ "learning_rate": 2.982233676339206e-06,
1780
+ "loss": 1.28195321559906,
1781
+ "step": 506
1782
+ },
1783
+ {
1784
+ "epoch": 1.3546666666666667,
1785
+ "grad_norm": 0.275390625,
1786
+ "learning_rate": 2.9553285081363308e-06,
1787
+ "loss": 1.1186037063598633,
1788
+ "step": 508
1789
+ },
1790
+ {
1791
+ "epoch": 1.3599999999999999,
1792
+ "grad_norm": 0.2255859375,
1793
+ "learning_rate": 2.9285409335598982e-06,
1794
+ "loss": 1.1410784721374512,
1795
+ "step": 510
1796
+ },
1797
+ {
1798
+ "epoch": 1.3653333333333333,
1799
+ "grad_norm": 0.59765625,
1800
+ "learning_rate": 2.9018732736489615e-06,
1801
+ "loss": 1.3254414796829224,
1802
+ "step": 512
1803
+ },
1804
+ {
1805
+ "epoch": 1.3706666666666667,
1806
+ "grad_norm": 0.515625,
1807
+ "learning_rate": 2.87532783905244e-06,
1808
+ "loss": 1.5107591152191162,
1809
+ "step": 514
1810
+ },
1811
+ {
1812
+ "epoch": 1.376,
1813
+ "grad_norm": 0.46484375,
1814
+ "learning_rate": 2.8489069298288992e-06,
1815
+ "loss": 1.2509578466415405,
1816
+ "step": 516
1817
+ },
1818
+ {
1819
+ "epoch": 1.3813333333333333,
1820
+ "grad_norm": 0.40234375,
1821
+ "learning_rate": 2.8226128352472747e-06,
1822
+ "loss": 1.3889421224594116,
1823
+ "step": 518
1824
+ },
1825
+ {
1826
+ "epoch": 1.3866666666666667,
1827
+ "grad_norm": 0.9140625,
1828
+ "learning_rate": 2.796447833588497e-06,
1829
+ "loss": 1.4837677478790283,
1830
+ "step": 520
1831
+ },
1832
+ {
1833
+ "epoch": 1.392,
1834
+ "grad_norm": 0.359375,
1835
+ "learning_rate": 2.7704141919481057e-06,
1836
+ "loss": 1.4461725950241089,
1837
+ "step": 522
1838
+ },
1839
+ {
1840
+ "epoch": 1.3973333333333333,
1841
+ "grad_norm": 0.625,
1842
+ "learning_rate": 2.7445141660397996e-06,
1843
+ "loss": 1.299829363822937,
1844
+ "step": 524
1845
+ },
1846
+ {
1847
+ "epoch": 1.4026666666666667,
1848
+ "grad_norm": 0.5234375,
1849
+ "learning_rate": 2.7187500000000004e-06,
1850
+ "loss": 1.0224652290344238,
1851
+ "step": 526
1852
+ },
1853
+ {
1854
+ "epoch": 1.408,
1855
+ "grad_norm": 0.53515625,
1856
+ "learning_rate": 2.6931239261933957e-06,
1857
+ "loss": 1.4109441041946411,
1858
+ "step": 528
1859
+ },
1860
+ {
1861
+ "epoch": 1.4133333333333333,
1862
+ "grad_norm": 0.40234375,
1863
+ "learning_rate": 2.6676381650195224e-06,
1864
+ "loss": 1.3534106016159058,
1865
+ "step": 530
1866
+ },
1867
+ {
1868
+ "epoch": 1.4186666666666667,
1869
+ "grad_norm": 0.89453125,
1870
+ "learning_rate": 2.6422949247203735e-06,
1871
+ "loss": 1.5593756437301636,
1872
+ "step": 532
1873
+ },
1874
+ {
1875
+ "epoch": 1.424,
1876
+ "grad_norm": 0.45703125,
1877
+ "learning_rate": 2.61709640118906e-06,
1878
+ "loss": 1.290115475654602,
1879
+ "step": 534
1880
+ },
1881
+ {
1882
+ "epoch": 1.4293333333333333,
1883
+ "grad_norm": 0.8125,
1884
+ "learning_rate": 2.5920447777795537e-06,
1885
+ "loss": 1.11957585811615,
1886
+ "step": 536
1887
+ },
1888
+ {
1889
+ "epoch": 1.4346666666666668,
1890
+ "grad_norm": 4.375,
1891
+ "learning_rate": 2.5671422251174963e-06,
1892
+ "loss": 1.0493898391723633,
1893
+ "step": 538
1894
+ },
1895
+ {
1896
+ "epoch": 1.44,
1897
+ "grad_norm": 0.78515625,
1898
+ "learning_rate": 2.5423909009121363e-06,
1899
+ "loss": 1.2276326417922974,
1900
+ "step": 540
1901
+ },
1902
+ {
1903
+ "epoch": 1.4453333333333334,
1904
+ "grad_norm": 0.28125,
1905
+ "learning_rate": 2.517792949769358e-06,
1906
+ "loss": 1.2422417402267456,
1907
+ "step": 542
1908
+ },
1909
+ {
1910
+ "epoch": 1.4506666666666668,
1911
+ "grad_norm": 0.8359375,
1912
+ "learning_rate": 2.4933505030058744e-06,
1913
+ "loss": 1.4107698202133179,
1914
+ "step": 544
1915
+ },
1916
+ {
1917
+ "epoch": 1.456,
1918
+ "grad_norm": 0.357421875,
1919
+ "learning_rate": 2.4690656784645434e-06,
1920
+ "loss": 1.0884898900985718,
1921
+ "step": 546
1922
+ },
1923
+ {
1924
+ "epoch": 1.4613333333333334,
1925
+ "grad_norm": 0.408203125,
1926
+ "learning_rate": 2.444940580330874e-06,
1927
+ "loss": 1.4010899066925049,
1928
+ "step": 548
1929
+ },
1930
+ {
1931
+ "epoch": 1.4666666666666668,
1932
+ "grad_norm": 1.03125,
1933
+ "learning_rate": 2.420977298950702e-06,
1934
+ "loss": 1.2208497524261475,
1935
+ "step": 550
1936
+ },
1937
+ {
1938
+ "epoch": 1.472,
1939
+ "grad_norm": 0.41796875,
1940
+ "learning_rate": 2.397177910649074e-06,
1941
+ "loss": 1.2561097145080566,
1942
+ "step": 552
1943
+ },
1944
+ {
1945
+ "epoch": 1.4773333333333334,
1946
+ "grad_norm": 0.52734375,
1947
+ "learning_rate": 2.3735444775503344e-06,
1948
+ "loss": 1.585281491279602,
1949
+ "step": 554
1950
+ },
1951
+ {
1952
+ "epoch": 1.4826666666666668,
1953
+ "grad_norm": 1.1875,
1954
+ "learning_rate": 2.350079047399464e-06,
1955
+ "loss": 1.4287792444229126,
1956
+ "step": 556
1957
+ },
1958
+ {
1959
+ "epoch": 1.488,
1960
+ "grad_norm": 0.87890625,
1961
+ "learning_rate": 2.3267836533846328e-06,
1962
+ "loss": 1.0933114290237427,
1963
+ "step": 558
1964
+ },
1965
+ {
1966
+ "epoch": 1.4933333333333334,
1967
+ "grad_norm": 0.8359375,
1968
+ "learning_rate": 2.3036603139610482e-06,
1969
+ "loss": 1.1171455383300781,
1970
+ "step": 560
1971
+ },
1972
+ {
1973
+ "epoch": 1.4986666666666666,
1974
+ "grad_norm": 0.64453125,
1975
+ "learning_rate": 2.2807110326760515e-06,
1976
+ "loss": 1.2635574340820312,
1977
+ "step": 562
1978
+ },
1979
+ {
1980
+ "epoch": 1.504,
1981
+ "grad_norm": 1.015625,
1982
+ "learning_rate": 2.25793779799553e-06,
1983
+ "loss": 1.4870187044143677,
1984
+ "step": 564
1985
+ },
1986
+ {
1987
+ "epoch": 1.5093333333333332,
1988
+ "grad_norm": 0.197265625,
1989
+ "learning_rate": 2.2353425831316116e-06,
1990
+ "loss": 1.1228148937225342,
1991
+ "step": 566
1992
+ },
1993
+ {
1994
+ "epoch": 1.5146666666666668,
1995
+ "grad_norm": 0.462890625,
1996
+ "learning_rate": 2.2129273458717058e-06,
1997
+ "loss": 1.393061876296997,
1998
+ "step": 568
1999
+ },
2000
+ {
2001
+ "epoch": 1.52,
2002
+ "grad_norm": 0.45703125,
2003
+ "learning_rate": 2.1906940284088615e-06,
2004
+ "loss": 1.0268428325653076,
2005
+ "step": 570
2006
+ },
2007
+ {
2008
+ "epoch": 1.5253333333333332,
2009
+ "grad_norm": 0.80859375,
2010
+ "learning_rate": 2.1686445571734825e-06,
2011
+ "loss": 1.3282322883605957,
2012
+ "step": 572
2013
+ },
2014
+ {
2015
+ "epoch": 1.5306666666666666,
2016
+ "grad_norm": 0.390625,
2017
+ "learning_rate": 2.146780842666417e-06,
2018
+ "loss": 1.2435400485992432,
2019
+ "step": 574
2020
+ },
2021
+ {
2022
+ "epoch": 1.536,
2023
+ "grad_norm": 0.61328125,
2024
+ "learning_rate": 2.1251047792934116e-06,
2025
+ "loss": 1.3963842391967773,
2026
+ "step": 576
2027
+ },
2028
+ {
2029
+ "epoch": 1.5413333333333332,
2030
+ "grad_norm": 0.40234375,
2031
+ "learning_rate": 2.1036182452009777e-06,
2032
+ "loss": 1.2569538354873657,
2033
+ "step": 578
2034
+ },
2035
+ {
2036
+ "epoch": 1.5466666666666666,
2037
+ "grad_norm": 0.58203125,
2038
+ "learning_rate": 2.082323102113647e-06,
2039
+ "loss": 1.3759268522262573,
2040
+ "step": 580
2041
+ },
2042
+ {
2043
+ "epoch": 1.552,
2044
+ "grad_norm": 0.4140625,
2045
+ "learning_rate": 2.0612211951726712e-06,
2046
+ "loss": 1.1200230121612549,
2047
+ "step": 582
2048
+ },
2049
+ {
2050
+ "epoch": 1.5573333333333332,
2051
+ "grad_norm": 0.51171875,
2052
+ "learning_rate": 2.040314352776137e-06,
2053
+ "loss": 1.6027384996414185,
2054
+ "step": 584
2055
+ },
2056
+ {
2057
+ "epoch": 1.5626666666666666,
2058
+ "grad_norm": 0.66015625,
2059
+ "learning_rate": 2.01960438642055e-06,
2060
+ "loss": 1.5927393436431885,
2061
+ "step": 586
2062
+ },
2063
+ {
2064
+ "epoch": 1.568,
2065
+ "grad_norm": 0.306640625,
2066
+ "learning_rate": 1.9990930905438733e-06,
2067
+ "loss": 1.0526584386825562,
2068
+ "step": 588
2069
+ },
2070
+ {
2071
+ "epoch": 1.5733333333333333,
2072
+ "grad_norm": 0.65625,
2073
+ "learning_rate": 1.978782242370046e-06,
2074
+ "loss": 1.4229958057403564,
2075
+ "step": 590
2076
+ },
2077
+ {
2078
+ "epoch": 1.5786666666666667,
2079
+ "grad_norm": 0.2392578125,
2080
+ "learning_rate": 1.958673601754995e-06,
2081
+ "loss": 1.2796956300735474,
2082
+ "step": 592
2083
+ },
2084
+ {
2085
+ "epoch": 1.584,
2086
+ "grad_norm": 0.57421875,
2087
+ "learning_rate": 1.9387689110341494e-06,
2088
+ "loss": 1.5249470472335815,
2089
+ "step": 594
2090
+ },
2091
+ {
2092
+ "epoch": 1.5893333333333333,
2093
+ "grad_norm": 0.37890625,
2094
+ "learning_rate": 1.9190698948714714e-06,
2095
+ "loss": 1.0016456842422485,
2096
+ "step": 596
2097
+ },
2098
+ {
2099
+ "epoch": 1.5946666666666667,
2100
+ "grad_norm": 0.29296875,
2101
+ "learning_rate": 1.899578260110027e-06,
2102
+ "loss": 1.4921225309371948,
2103
+ "step": 598
2104
+ },
2105
+ {
2106
+ "epoch": 1.6,
2107
+ "grad_norm": 0.5625,
2108
+ "learning_rate": 1.8802956956240875e-06,
2109
+ "loss": 1.3857101202011108,
2110
+ "step": 600
2111
+ },
2112
+ {
2113
+ "epoch": 1.6053333333333333,
2114
+ "grad_norm": 0.376953125,
2115
+ "learning_rate": 1.8612238721728026e-06,
2116
+ "loss": 1.0772466659545898,
2117
+ "step": 602
2118
+ },
2119
+ {
2120
+ "epoch": 1.6106666666666667,
2121
+ "grad_norm": 0.69921875,
2122
+ "learning_rate": 1.8423644422554278e-06,
2123
+ "loss": 1.374086618423462,
2124
+ "step": 604
2125
+ },
2126
+ {
2127
+ "epoch": 1.616,
2128
+ "grad_norm": 0.2392578125,
2129
+ "learning_rate": 1.8237190399681486e-06,
2130
+ "loss": 1.1497209072113037,
2131
+ "step": 606
2132
+ },
2133
+ {
2134
+ "epoch": 1.6213333333333333,
2135
+ "grad_norm": 0.31640625,
2136
+ "learning_rate": 1.8052892808624892e-06,
2137
+ "loss": 1.1666005849838257,
2138
+ "step": 608
2139
+ },
2140
+ {
2141
+ "epoch": 1.6266666666666667,
2142
+ "grad_norm": 0.5390625,
2143
+ "learning_rate": 1.78707676180533e-06,
2144
+ "loss": 1.4539345502853394,
2145
+ "step": 610
2146
+ },
2147
+ {
2148
+ "epoch": 1.6320000000000001,
2149
+ "grad_norm": 0.62890625,
2150
+ "learning_rate": 1.7690830608405475e-06,
2151
+ "loss": 1.2192360162734985,
2152
+ "step": 612
2153
+ },
2154
+ {
2155
+ "epoch": 1.6373333333333333,
2156
+ "grad_norm": 0.40234375,
2157
+ "learning_rate": 1.7513097370522843e-06,
2158
+ "loss": 1.2787443399429321,
2159
+ "step": 614
2160
+ },
2161
+ {
2162
+ "epoch": 1.6426666666666667,
2163
+ "grad_norm": 0.8203125,
2164
+ "learning_rate": 1.7337583304298553e-06,
2165
+ "loss": 1.4955127239227295,
2166
+ "step": 616
2167
+ },
2168
+ {
2169
+ "epoch": 1.6480000000000001,
2170
+ "grad_norm": 0.8359375,
2171
+ "learning_rate": 1.7164303617343188e-06,
2172
+ "loss": 1.5240033864974976,
2173
+ "step": 618
2174
+ },
2175
+ {
2176
+ "epoch": 1.6533333333333333,
2177
+ "grad_norm": 0.333984375,
2178
+ "learning_rate": 1.6993273323667077e-06,
2179
+ "loss": 1.186129093170166,
2180
+ "step": 620
2181
+ },
2182
+ {
2183
+ "epoch": 1.6586666666666665,
2184
+ "grad_norm": 0.65234375,
2185
+ "learning_rate": 1.6824507242379345e-06,
2186
+ "loss": 1.2840181589126587,
2187
+ "step": 622
2188
+ },
2189
+ {
2190
+ "epoch": 1.6640000000000001,
2191
+ "grad_norm": 0.5390625,
2192
+ "learning_rate": 1.6658019996403958e-06,
2193
+ "loss": 1.5434945821762085,
2194
+ "step": 624
2195
+ },
2196
+ {
2197
+ "epoch": 1.6693333333333333,
2198
+ "grad_norm": 0.68359375,
2199
+ "learning_rate": 1.6493826011212656e-06,
2200
+ "loss": 1.5949230194091797,
2201
+ "step": 626
2202
+ },
2203
+ {
2204
+ "epoch": 1.6746666666666665,
2205
+ "grad_norm": 0.5,
2206
+ "learning_rate": 1.6331939513575066e-06,
2207
+ "loss": 1.1981688737869263,
2208
+ "step": 628
2209
+ },
2210
+ {
2211
+ "epoch": 1.6800000000000002,
2212
+ "grad_norm": 0.6953125,
2213
+ "learning_rate": 1.6172374530326001e-06,
2214
+ "loss": 1.3873530626296997,
2215
+ "step": 630
2216
+ },
2217
+ {
2218
+ "epoch": 1.6853333333333333,
2219
+ "grad_norm": 0.5703125,
2220
+ "learning_rate": 1.60151448871501e-06,
2221
+ "loss": 1.1224573850631714,
2222
+ "step": 632
2223
+ },
2224
+ {
2225
+ "epoch": 1.6906666666666665,
2226
+ "grad_norm": 0.640625,
2227
+ "learning_rate": 1.5860264207383851e-06,
2228
+ "loss": 1.0328835248947144,
2229
+ "step": 634
2230
+ },
2231
+ {
2232
+ "epoch": 1.696,
2233
+ "grad_norm": 0.9375,
2234
+ "learning_rate": 1.5707745910835233e-06,
2235
+ "loss": 1.5136065483093262,
2236
+ "step": 636
2237
+ },
2238
+ {
2239
+ "epoch": 1.7013333333333334,
2240
+ "grad_norm": 1.2421875,
2241
+ "learning_rate": 1.5557603212620889e-06,
2242
+ "loss": 1.5170190334320068,
2243
+ "step": 638
2244
+ },
2245
+ {
2246
+ "epoch": 1.7066666666666666,
2247
+ "grad_norm": 0.41015625,
2248
+ "learning_rate": 1.5409849122021144e-06,
2249
+ "loss": 1.4020171165466309,
2250
+ "step": 640
2251
+ },
2252
+ {
2253
+ "epoch": 1.712,
2254
+ "grad_norm": 0.60546875,
2255
+ "learning_rate": 1.5264496441352748e-06,
2256
+ "loss": 1.5551862716674805,
2257
+ "step": 642
2258
+ },
2259
+ {
2260
+ "epoch": 1.7173333333333334,
2261
+ "grad_norm": 0.396484375,
2262
+ "learning_rate": 1.5121557764859648e-06,
2263
+ "loss": 1.3862305879592896,
2264
+ "step": 644
2265
+ },
2266
+ {
2267
+ "epoch": 1.7226666666666666,
2268
+ "grad_norm": 0.41015625,
2269
+ "learning_rate": 1.4981045477621702e-06,
2270
+ "loss": 0.922488272190094,
2271
+ "step": 646
2272
+ },
2273
+ {
2274
+ "epoch": 1.728,
2275
+ "grad_norm": 0.318359375,
2276
+ "learning_rate": 1.4842971754481615e-06,
2277
+ "loss": 1.240954041481018,
2278
+ "step": 648
2279
+ },
2280
+ {
2281
+ "epoch": 1.7333333333333334,
2282
+ "grad_norm": 0.24609375,
2283
+ "learning_rate": 1.4707348558989974e-06,
2284
+ "loss": 1.0813591480255127,
2285
+ "step": 650
2286
+ },
2287
+ {
2288
+ "epoch": 1.7386666666666666,
2289
+ "grad_norm": 0.63671875,
2290
+ "learning_rate": 1.4574187642368714e-06,
2291
+ "loss": 1.4775949716567993,
2292
+ "step": 652
2293
+ },
2294
+ {
2295
+ "epoch": 1.744,
2296
+ "grad_norm": 0.30859375,
2297
+ "learning_rate": 1.444350054249288e-06,
2298
+ "loss": 1.2168411016464233,
2299
+ "step": 654
2300
+ },
2301
+ {
2302
+ "epoch": 1.7493333333333334,
2303
+ "grad_norm": 0.5625,
2304
+ "learning_rate": 1.431529858289091e-06,
2305
+ "loss": 1.4125171899795532,
2306
+ "step": 656
2307
+ },
2308
+ {
2309
+ "epoch": 1.7546666666666666,
2310
+ "grad_norm": 0.53125,
2311
+ "learning_rate": 1.418959287176354e-06,
2312
+ "loss": 1.414473056793213,
2313
+ "step": 658
2314
+ },
2315
+ {
2316
+ "epoch": 1.76,
2317
+ "grad_norm": 0.486328125,
2318
+ "learning_rate": 1.406639430102127e-06,
2319
+ "loss": 1.220368504524231,
2320
+ "step": 660
2321
+ },
2322
+ {
2323
+ "epoch": 1.7653333333333334,
2324
+ "grad_norm": 0.4921875,
2325
+ "learning_rate": 1.3945713545340668e-06,
2326
+ "loss": 1.2297767400741577,
2327
+ "step": 662
2328
+ },
2329
+ {
2330
+ "epoch": 1.7706666666666666,
2331
+ "grad_norm": 0.2412109375,
2332
+ "learning_rate": 1.3827561061239403e-06,
2333
+ "loss": 1.3291184902191162,
2334
+ "step": 664
2335
+ },
2336
+ {
2337
+ "epoch": 1.776,
2338
+ "grad_norm": 0.65625,
2339
+ "learning_rate": 1.3711947086170277e-06,
2340
+ "loss": 1.0460172891616821,
2341
+ "step": 666
2342
+ },
2343
+ {
2344
+ "epoch": 1.7813333333333334,
2345
+ "grad_norm": 0.40234375,
2346
+ "learning_rate": 1.359888163763415e-06,
2347
+ "loss": 1.3426827192306519,
2348
+ "step": 668
2349
+ },
2350
+ {
2351
+ "epoch": 1.7866666666666666,
2352
+ "grad_norm": 0.3984375,
2353
+ "learning_rate": 1.3488374512311988e-06,
2354
+ "loss": 1.1144654750823975,
2355
+ "step": 670
2356
+ },
2357
+ {
2358
+ "epoch": 1.792,
2359
+ "grad_norm": 0.5546875,
2360
+ "learning_rate": 1.3380435285216006e-06,
2361
+ "loss": 1.2571556568145752,
2362
+ "step": 672
2363
+ },
2364
+ {
2365
+ "epoch": 1.7973333333333334,
2366
+ "grad_norm": 0.6953125,
2367
+ "learning_rate": 1.327507330886004e-06,
2368
+ "loss": 1.454203486442566,
2369
+ "step": 674
2370
+ },
2371
+ {
2372
+ "epoch": 1.8026666666666666,
2373
+ "grad_norm": 0.48046875,
2374
+ "learning_rate": 1.3172297712449172e-06,
2375
+ "loss": 1.4048188924789429,
2376
+ "step": 676
2377
+ },
2378
+ {
2379
+ "epoch": 1.808,
2380
+ "grad_norm": 0.58984375,
2381
+ "learning_rate": 1.3072117401088736e-06,
2382
+ "loss": 1.0630518198013306,
2383
+ "step": 678
2384
+ },
2385
+ {
2386
+ "epoch": 1.8133333333333335,
2387
+ "grad_norm": 0.9609375,
2388
+ "learning_rate": 1.2974541055012725e-06,
2389
+ "loss": 1.1879171133041382,
2390
+ "step": 680
2391
+ },
2392
+ {
2393
+ "epoch": 1.8186666666666667,
2394
+ "grad_norm": 0.734375,
2395
+ "learning_rate": 1.2879577128831673e-06,
2396
+ "loss": 1.380265235900879,
2397
+ "step": 682
2398
+ },
2399
+ {
2400
+ "epoch": 1.8239999999999998,
2401
+ "grad_norm": 0.36328125,
2402
+ "learning_rate": 1.278723385080009e-06,
2403
+ "loss": 0.971758246421814,
2404
+ "step": 684
2405
+ },
2406
+ {
2407
+ "epoch": 1.8293333333333335,
2408
+ "grad_norm": 0.5390625,
2409
+ "learning_rate": 1.2697519222103545e-06,
2410
+ "loss": 1.512132167816162,
2411
+ "step": 686
2412
+ },
2413
+ {
2414
+ "epoch": 1.8346666666666667,
2415
+ "grad_norm": 0.431640625,
2416
+ "learning_rate": 1.2610441016165355e-06,
2417
+ "loss": 1.005048155784607,
2418
+ "step": 688
2419
+ },
2420
+ {
2421
+ "epoch": 1.8399999999999999,
2422
+ "grad_norm": 0.294921875,
2423
+ "learning_rate": 1.2526006777973084e-06,
2424
+ "loss": 1.371572732925415,
2425
+ "step": 690
2426
+ },
2427
+ {
2428
+ "epoch": 1.8453333333333335,
2429
+ "grad_norm": 0.5703125,
2430
+ "learning_rate": 1.2444223823424778e-06,
2431
+ "loss": 1.332929253578186,
2432
+ "step": 692
2433
+ },
2434
+ {
2435
+ "epoch": 1.8506666666666667,
2436
+ "grad_norm": 0.30078125,
2437
+ "learning_rate": 1.2365099238695073e-06,
2438
+ "loss": 1.0624377727508545,
2439
+ "step": 694
2440
+ },
2441
+ {
2442
+ "epoch": 1.8559999999999999,
2443
+ "grad_norm": 0.60546875,
2444
+ "learning_rate": 1.2288639879621216e-06,
2445
+ "loss": 1.4434648752212524,
2446
+ "step": 696
2447
+ },
2448
+ {
2449
+ "epoch": 1.8613333333333333,
2450
+ "grad_norm": 0.59375,
2451
+ "learning_rate": 1.2214852371109014e-06,
2452
+ "loss": 1.115839958190918,
2453
+ "step": 698
2454
+ },
2455
+ {
2456
+ "epoch": 1.8666666666666667,
2457
+ "grad_norm": 0.294921875,
2458
+ "learning_rate": 1.214374310655884e-06,
2459
+ "loss": 1.3425531387329102,
2460
+ "step": 700
2461
+ },
2462
+ {
2463
+ "epoch": 1.8719999999999999,
2464
+ "grad_norm": 0.41015625,
2465
+ "learning_rate": 1.207531824731164e-06,
2466
+ "loss": 1.2544267177581787,
2467
+ "step": 702
2468
+ },
2469
+ {
2470
+ "epoch": 1.8773333333333333,
2471
+ "grad_norm": 0.48828125,
2472
+ "learning_rate": 1.2009583722115107e-06,
2473
+ "loss": 1.2201001644134521,
2474
+ "step": 704
2475
+ },
2476
+ {
2477
+ "epoch": 1.8826666666666667,
2478
+ "grad_norm": 0.5859375,
2479
+ "learning_rate": 1.1946545226609944e-06,
2480
+ "loss": 1.1109111309051514,
2481
+ "step": 706
2482
+ },
2483
+ {
2484
+ "epoch": 1.888,
2485
+ "grad_norm": 0.62109375,
2486
+ "learning_rate": 1.1886208222836388e-06,
2487
+ "loss": 1.2960035800933838,
2488
+ "step": 708
2489
+ },
2490
+ {
2491
+ "epoch": 1.8933333333333333,
2492
+ "grad_norm": 0.62890625,
2493
+ "learning_rate": 1.1828577938760926e-06,
2494
+ "loss": 1.3226895332336426,
2495
+ "step": 710
2496
+ },
2497
+ {
2498
+ "epoch": 1.8986666666666667,
2499
+ "grad_norm": 0.4296875,
2500
+ "learning_rate": 1.1773659367823326e-06,
2501
+ "loss": 1.459092140197754,
2502
+ "step": 712
2503
+ },
2504
+ {
2505
+ "epoch": 1.904,
2506
+ "grad_norm": 0.263671875,
2507
+ "learning_rate": 1.1721457268503963e-06,
2508
+ "loss": 0.9945974349975586,
2509
+ "step": 714
2510
+ },
2511
+ {
2512
+ "epoch": 1.9093333333333333,
2513
+ "grad_norm": 1.4921875,
2514
+ "learning_rate": 1.1671976163911537e-06,
2515
+ "loss": 1.3499224185943604,
2516
+ "step": 716
2517
+ },
2518
+ {
2519
+ "epoch": 1.9146666666666667,
2520
+ "grad_norm": 0.5,
2521
+ "learning_rate": 1.162522034139113e-06,
2522
+ "loss": 1.203792929649353,
2523
+ "step": 718
2524
+ },
2525
+ {
2526
+ "epoch": 1.92,
2527
+ "grad_norm": 0.357421875,
2528
+ "learning_rate": 1.1581193852152746e-06,
2529
+ "loss": 1.432945728302002,
2530
+ "step": 720
2531
+ },
2532
+ {
2533
+ "epoch": 1.9253333333333333,
2534
+ "grad_norm": 0.64453125,
2535
+ "learning_rate": 1.1539900510920285e-06,
2536
+ "loss": 1.4233765602111816,
2537
+ "step": 722
2538
+ },
2539
+ {
2540
+ "epoch": 1.9306666666666668,
2541
+ "grad_norm": 0.83984375,
2542
+ "learning_rate": 1.1501343895601022e-06,
2543
+ "loss": 1.5078463554382324,
2544
+ "step": 724
2545
+ },
2546
+ {
2547
+ "epoch": 1.936,
2548
+ "grad_norm": 0.298828125,
2549
+ "learning_rate": 1.1465527346975566e-06,
2550
+ "loss": 1.041426658630371,
2551
+ "step": 726
2552
+ },
2553
+ {
2554
+ "epoch": 1.9413333333333334,
2555
+ "grad_norm": 0.87890625,
2556
+ "learning_rate": 1.1432453968408437e-06,
2557
+ "loss": 1.1554670333862305,
2558
+ "step": 728
2559
+ },
2560
+ {
2561
+ "epoch": 1.9466666666666668,
2562
+ "grad_norm": 0.66015625,
2563
+ "learning_rate": 1.1402126625579132e-06,
2564
+ "loss": 1.238165020942688,
2565
+ "step": 730
2566
+ },
2567
+ {
2568
+ "epoch": 1.952,
2569
+ "grad_norm": 0.25,
2570
+ "learning_rate": 1.1374547946233843e-06,
2571
+ "loss": 1.2469568252563477,
2572
+ "step": 732
2573
+ },
2574
+ {
2575
+ "epoch": 1.9573333333333334,
2576
+ "grad_norm": 0.94921875,
2577
+ "learning_rate": 1.1349720319957783e-06,
2578
+ "loss": 1.100600004196167,
2579
+ "step": 734
2580
+ },
2581
+ {
2582
+ "epoch": 1.9626666666666668,
2583
+ "grad_norm": 0.287109375,
2584
+ "learning_rate": 1.1327645897968104e-06,
2585
+ "loss": 1.256953239440918,
2586
+ "step": 736
2587
+ },
2588
+ {
2589
+ "epoch": 1.968,
2590
+ "grad_norm": 0.408203125,
2591
+ "learning_rate": 1.1308326592927547e-06,
2592
+ "loss": 1.0144098997116089,
2593
+ "step": 738
2594
+ },
2595
+ {
2596
+ "epoch": 1.9733333333333334,
2597
+ "grad_norm": 0.5234375,
2598
+ "learning_rate": 1.1291764078778685e-06,
2599
+ "loss": 1.1728276014328003,
2600
+ "step": 740
2601
+ },
2602
+ {
2603
+ "epoch": 1.9786666666666668,
2604
+ "grad_norm": 0.70703125,
2605
+ "learning_rate": 1.127795979059889e-06,
2606
+ "loss": 1.084470272064209,
2607
+ "step": 742
2608
+ },
2609
+ {
2610
+ "epoch": 1.984,
2611
+ "grad_norm": 0.625,
2612
+ "learning_rate": 1.1266914924475997e-06,
2613
+ "loss": 1.2899092435836792,
2614
+ "step": 744
2615
+ },
2616
+ {
2617
+ "epoch": 1.9893333333333332,
2618
+ "grad_norm": 1.0390625,
2619
+ "learning_rate": 1.1258630437404658e-06,
2620
+ "loss": 1.3732484579086304,
2621
+ "step": 746
2622
+ },
2623
+ {
2624
+ "epoch": 1.9946666666666668,
2625
+ "grad_norm": 0.42578125,
2626
+ "learning_rate": 1.1253107047203428e-06,
2627
+ "loss": 1.2392138242721558,
2628
+ "step": 748
2629
+ },
2630
+ {
2631
+ "epoch": 2.0,
2632
+ "grad_norm": 0.375,
2633
+ "learning_rate": 1.1250345232452565e-06,
2634
+ "loss": 1.4283043146133423,
2635
+ "step": 750
2636
+ },
2637
+ {
2638
+ "epoch": 2.0,
2639
+ "step": 750,
2640
+ "total_flos": 2.55147477658973e+18,
2641
+ "train_loss": 1.3643777534166972,
2642
+ "train_runtime": 10437.131,
2643
+ "train_samples_per_second": 2.299,
2644
+ "train_steps_per_second": 0.072
2645
+ }
2646
+ ],
2647
+ "logging_steps": 2,
2648
+ "max_steps": 750,
2649
+ "num_input_tokens_seen": 0,
2650
+ "num_train_epochs": 2,
2651
+ "save_steps": 99999,
2652
+ "stateful_callbacks": {
2653
+ "TrainerControl": {
2654
+ "args": {
2655
+ "should_epoch_stop": false,
2656
+ "should_evaluate": false,
2657
+ "should_log": false,
2658
+ "should_save": true,
2659
+ "should_training_stop": true
2660
+ },
2661
+ "attributes": {}
2662
+ }
2663
+ },
2664
+ "total_flos": 2.55147477658973e+18,
2665
+ "train_batch_size": 1,
2666
+ "trial_name": null,
2667
+ "trial_params": null
2668
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c40f4704287310d254e687199863d05ca5f8314516c2a6e786ee901ba498fc68
3
+ size 5649
training_loss.png ADDED