Hugging Face's logo

a12354
/
ITFormer

Safetensors
Model card Files
xet
 Community
a12354 commited on
Commit
e36596a
·
verified ·
1 Parent(s): aabdb98

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/added_tokens.json +24 -0
  2. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/chat_template.jinja +54 -0
  3. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-400/vocab.json +0 -0
  4. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/added_tokens.json +24 -0
  5. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/chat_template.jinja +54 -0
  6. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/config.json +27 -0
  7. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/generation_config.json +4 -0
  8. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/merges.txt +0 -0
  9. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/special_tokens_map.json +31 -0
  10. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/tokenizer_config.json +207 -0
  11. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/trainer_state.json +3534 -0
  12. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/vocab.json +0 -0
  13. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/config.json +27 -0
  14. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/generation_config.json +4 -0
  15. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/merges.txt +0 -0
  16. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/skipped_overlength_train.jsonl +7 -0
  17. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/special_tokens_map.json +31 -0
  18. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/tokenizer_config.json +207 -0
  19. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/tsqa_run_config.json +69 -0
  20. runs/itformer_tsqa_full_20260615_164812/checkpoint_final/vocab.json +0 -0
  21. runs/itformer_tsqa_full_20260615_164812/eval/metrics.json +110 -0
  22. runs/itformer_tsqa_full_20260615_164812/eval/predictions.jsonl +0 -0
  23. runs/itformer_tsqa_full_20260615_164812/logs/eval.log +128 -0
  24. runs/itformer_tsqa_full_20260615_164812/logs/train.log +0 -0
  25. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/added_tokens.json +24 -0
  26. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/chat_template.jinja +54 -0
  27. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/added_tokens.json +24 -0
  28. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/chat_template.jinja +54 -0
  29. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/config.json +27 -0
  30. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/generation_config.json +4 -0
  31. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/merges.txt +0 -0
  32. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/special_tokens_map.json +31 -0
  33. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/tokenizer_config.json +207 -0
  34. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/trainer_state.json +0 -0
  35. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/vocab.json +0 -0
  36. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/added_tokens.json +24 -0
  37. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/chat_template.jinja +54 -0
  38. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/config.json +27 -0
  39. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/generation_config.json +4 -0
  40. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/merges.txt +0 -0
  41. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/special_tokens_map.json +31 -0
  42. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/tokenizer_config.json +207 -0
  43. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/trainer_state.json +0 -0
  44. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/vocab.json +0 -0
  45. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/config.json +27 -0
  46. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/generation_config.json +4 -0
  47. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/merges.txt +0 -0
  48. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/skipped_overlength_train.jsonl +7 -0
  49. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/special_tokens_map.json +31 -0
  50. runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/tokenizer_config.json +207 -0
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-400/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "TLM"
4
+ ],
5
+ "freeze_ts_model": true,
6
+ "gradient_checkpointing": true,
7
+ "llm_attn_implementation": null,
8
+ "llm_model_path": "/mnt/share01/sqk/models/qwen2.5-7b-instruct",
9
+ "llm_torch_dtype": "bfloat16",
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "lora_r": 16,
13
+ "lora_target_modules": [
14
+ "q_proj",
15
+ "k_proj",
16
+ "v_proj",
17
+ "o_proj",
18
+ "gate_proj",
19
+ "up_proj",
20
+ "down_proj"
21
+ ],
22
+ "model_type": "vlm_model",
23
+ "torch_dtype": "bfloat16",
24
+ "transformers_version": "4.54.1",
25
+ "ts_pad_num": 25,
26
+ "use_lora": true
27
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.54.1"
4
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "clean_up_tokenization_spaces": false,
199
+ "eos_token": "<|im_end|>",
200
+ "errors": "replace",
201
+ "extra_special_tokens": {},
202
+ "model_max_length": 131072,
203
+ "pad_token": "<|endoftext|>",
204
+ "split_special_tokens": false,
205
+ "tokenizer_class": "Qwen2Tokenizer",
206
+ "unk_token": null
207
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,3534 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.0,
6
+ "eval_steps": 999999999,
7
+ "global_step": 500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.004004004004004004,
14
+ "grad_norm": 552.2235107421875,
15
+ "learning_rate": 0.0,
16
+ "loss": 71.2359,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.008008008008008008,
21
+ "grad_norm": 468.91900634765625,
22
+ "learning_rate": 4.0000000000000003e-07,
23
+ "loss": 60.8415,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 0.012012012012012012,
28
+ "grad_norm": 515.1337890625,
29
+ "learning_rate": 8.000000000000001e-07,
30
+ "loss": 61.7703,
31
+ "step": 3
32
+ },
33
+ {
34
+ "epoch": 0.016016016016016016,
35
+ "grad_norm": 685.647216796875,
36
+ "learning_rate": 1.2000000000000002e-06,
37
+ "loss": 63.8354,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 0.02002002002002002,
42
+ "grad_norm": 581.6423950195312,
43
+ "learning_rate": 1.6000000000000001e-06,
44
+ "loss": 61.5803,
45
+ "step": 5
46
+ },
47
+ {
48
+ "epoch": 0.024024024024024024,
49
+ "grad_norm": 496.3015441894531,
50
+ "learning_rate": 2.0000000000000003e-06,
51
+ "loss": 56.5872,
52
+ "step": 6
53
+ },
54
+ {
55
+ "epoch": 0.028028028028028028,
56
+ "grad_norm": 902.9198608398438,
57
+ "learning_rate": 2.4000000000000003e-06,
58
+ "loss": 63.4979,
59
+ "step": 7
60
+ },
61
+ {
62
+ "epoch": 0.03203203203203203,
63
+ "grad_norm": 887.4134521484375,
64
+ "learning_rate": 2.8000000000000003e-06,
65
+ "loss": 62.5185,
66
+ "step": 8
67
+ },
68
+ {
69
+ "epoch": 0.036036036036036036,
70
+ "grad_norm": 511.5171203613281,
71
+ "learning_rate": 3.2000000000000003e-06,
72
+ "loss": 54.4745,
73
+ "step": 9
74
+ },
75
+ {
76
+ "epoch": 0.04004004004004004,
77
+ "grad_norm": 548.1619262695312,
78
+ "learning_rate": 3.6000000000000003e-06,
79
+ "loss": 51.5734,
80
+ "step": 10
81
+ },
82
+ {
83
+ "epoch": 0.044044044044044044,
84
+ "grad_norm": 547.2471313476562,
85
+ "learning_rate": 4.000000000000001e-06,
86
+ "loss": 46.0137,
87
+ "step": 11
88
+ },
89
+ {
90
+ "epoch": 0.04804804804804805,
91
+ "grad_norm": 485.0998229980469,
92
+ "learning_rate": 4.4e-06,
93
+ "loss": 46.7793,
94
+ "step": 12
95
+ },
96
+ {
97
+ "epoch": 0.05205205205205205,
98
+ "grad_norm": 611.9149169921875,
99
+ "learning_rate": 4.800000000000001e-06,
100
+ "loss": 46.6272,
101
+ "step": 13
102
+ },
103
+ {
104
+ "epoch": 0.056056056056056056,
105
+ "grad_norm": 510.6531066894531,
106
+ "learning_rate": 5.2e-06,
107
+ "loss": 46.6201,
108
+ "step": 14
109
+ },
110
+ {
111
+ "epoch": 0.06006006006006006,
112
+ "grad_norm": 363.2168884277344,
113
+ "learning_rate": 5.600000000000001e-06,
114
+ "loss": 37.0976,
115
+ "step": 15
116
+ },
117
+ {
118
+ "epoch": 0.06406406406406406,
119
+ "grad_norm": 500.39654541015625,
120
+ "learning_rate": 6e-06,
121
+ "loss": 34.2463,
122
+ "step": 16
123
+ },
124
+ {
125
+ "epoch": 0.06806806806806807,
126
+ "grad_norm": 589.0038452148438,
127
+ "learning_rate": 6.4000000000000006e-06,
128
+ "loss": 39.3048,
129
+ "step": 17
130
+ },
131
+ {
132
+ "epoch": 0.07207207207207207,
133
+ "grad_norm": 467.6886291503906,
134
+ "learning_rate": 6.800000000000001e-06,
135
+ "loss": 36.2025,
136
+ "step": 18
137
+ },
138
+ {
139
+ "epoch": 0.07607607607607608,
140
+ "grad_norm": 628.8820190429688,
141
+ "learning_rate": 7.2000000000000005e-06,
142
+ "loss": 24.7629,
143
+ "step": 19
144
+ },
145
+ {
146
+ "epoch": 0.08008008008008008,
147
+ "grad_norm": 475.3155212402344,
148
+ "learning_rate": 7.600000000000001e-06,
149
+ "loss": 26.2968,
150
+ "step": 20
151
+ },
152
+ {
153
+ "epoch": 0.08408408408408409,
154
+ "grad_norm": 261.9303894042969,
155
+ "learning_rate": 8.000000000000001e-06,
156
+ "loss": 27.2021,
157
+ "step": 21
158
+ },
159
+ {
160
+ "epoch": 0.08808808808808809,
161
+ "grad_norm": 366.16851806640625,
162
+ "learning_rate": 8.400000000000001e-06,
163
+ "loss": 27.2075,
164
+ "step": 22
165
+ },
166
+ {
167
+ "epoch": 0.0920920920920921,
168
+ "grad_norm": 269.6435546875,
169
+ "learning_rate": 8.8e-06,
170
+ "loss": 22.1728,
171
+ "step": 23
172
+ },
173
+ {
174
+ "epoch": 0.0960960960960961,
175
+ "grad_norm": 309.0314025878906,
176
+ "learning_rate": 9.200000000000002e-06,
177
+ "loss": 24.6544,
178
+ "step": 24
179
+ },
180
+ {
181
+ "epoch": 0.1001001001001001,
182
+ "grad_norm": 327.53106689453125,
183
+ "learning_rate": 9.600000000000001e-06,
184
+ "loss": 20.2396,
185
+ "step": 25
186
+ },
187
+ {
188
+ "epoch": 0.1041041041041041,
189
+ "grad_norm": 232.21571350097656,
190
+ "learning_rate": 1e-05,
191
+ "loss": 24.7279,
192
+ "step": 26
193
+ },
194
+ {
195
+ "epoch": 0.10810810810810811,
196
+ "grad_norm": 264.31024169921875,
197
+ "learning_rate": 1.04e-05,
198
+ "loss": 15.7913,
199
+ "step": 27
200
+ },
201
+ {
202
+ "epoch": 0.11211211211211211,
203
+ "grad_norm": 221.6844024658203,
204
+ "learning_rate": 1.0800000000000002e-05,
205
+ "loss": 17.3857,
206
+ "step": 28
207
+ },
208
+ {
209
+ "epoch": 0.11611611611611612,
210
+ "grad_norm": 204.84378051757812,
211
+ "learning_rate": 1.1200000000000001e-05,
212
+ "loss": 17.8263,
213
+ "step": 29
214
+ },
215
+ {
216
+ "epoch": 0.12012012012012012,
217
+ "grad_norm": 181.392333984375,
218
+ "learning_rate": 1.16e-05,
219
+ "loss": 12.1824,
220
+ "step": 30
221
+ },
222
+ {
223
+ "epoch": 0.12412412412412413,
224
+ "grad_norm": 305.5450134277344,
225
+ "learning_rate": 1.2e-05,
226
+ "loss": 15.9066,
227
+ "step": 31
228
+ },
229
+ {
230
+ "epoch": 0.12812812812812813,
231
+ "grad_norm": 203.30410766601562,
232
+ "learning_rate": 1.2400000000000002e-05,
233
+ "loss": 15.9298,
234
+ "step": 32
235
+ },
236
+ {
237
+ "epoch": 0.13213213213213212,
238
+ "grad_norm": 197.00323486328125,
239
+ "learning_rate": 1.2800000000000001e-05,
240
+ "loss": 18.7559,
241
+ "step": 33
242
+ },
243
+ {
244
+ "epoch": 0.13613613613613615,
245
+ "grad_norm": 150.67620849609375,
246
+ "learning_rate": 1.3200000000000002e-05,
247
+ "loss": 11.0149,
248
+ "step": 34
249
+ },
250
+ {
251
+ "epoch": 0.14014014014014015,
252
+ "grad_norm": 125.64725494384766,
253
+ "learning_rate": 1.3600000000000002e-05,
254
+ "loss": 14.8016,
255
+ "step": 35
256
+ },
257
+ {
258
+ "epoch": 0.14414414414414414,
259
+ "grad_norm": 153.85958862304688,
260
+ "learning_rate": 1.4e-05,
261
+ "loss": 22.8046,
262
+ "step": 36
263
+ },
264
+ {
265
+ "epoch": 0.14814814814814814,
266
+ "grad_norm": 130.13233947753906,
267
+ "learning_rate": 1.4400000000000001e-05,
268
+ "loss": 16.6846,
269
+ "step": 37
270
+ },
271
+ {
272
+ "epoch": 0.15215215215215216,
273
+ "grad_norm": 169.4436798095703,
274
+ "learning_rate": 1.48e-05,
275
+ "loss": 11.7838,
276
+ "step": 38
277
+ },
278
+ {
279
+ "epoch": 0.15615615615615616,
280
+ "grad_norm": 106.24337005615234,
281
+ "learning_rate": 1.5200000000000002e-05,
282
+ "loss": 11.4211,
283
+ "step": 39
284
+ },
285
+ {
286
+ "epoch": 0.16016016016016016,
287
+ "grad_norm": 139.6412811279297,
288
+ "learning_rate": 1.5600000000000003e-05,
289
+ "loss": 11.6741,
290
+ "step": 40
291
+ },
292
+ {
293
+ "epoch": 0.16416416416416416,
294
+ "grad_norm": 95.60179901123047,
295
+ "learning_rate": 1.6000000000000003e-05,
296
+ "loss": 15.9865,
297
+ "step": 41
298
+ },
299
+ {
300
+ "epoch": 0.16816816816816818,
301
+ "grad_norm": 85.54633331298828,
302
+ "learning_rate": 1.64e-05,
303
+ "loss": 12.6641,
304
+ "step": 42
305
+ },
306
+ {
307
+ "epoch": 0.17217217217217218,
308
+ "grad_norm": 75.1412582397461,
309
+ "learning_rate": 1.6800000000000002e-05,
310
+ "loss": 11.6642,
311
+ "step": 43
312
+ },
313
+ {
314
+ "epoch": 0.17617617617617617,
315
+ "grad_norm": 133.28509521484375,
316
+ "learning_rate": 1.72e-05,
317
+ "loss": 9.6996,
318
+ "step": 44
319
+ },
320
+ {
321
+ "epoch": 0.18018018018018017,
322
+ "grad_norm": 108.03025817871094,
323
+ "learning_rate": 1.76e-05,
324
+ "loss": 10.9586,
325
+ "step": 45
326
+ },
327
+ {
328
+ "epoch": 0.1841841841841842,
329
+ "grad_norm": 80.12256622314453,
330
+ "learning_rate": 1.8e-05,
331
+ "loss": 13.3997,
332
+ "step": 46
333
+ },
334
+ {
335
+ "epoch": 0.1881881881881882,
336
+ "grad_norm": 89.64915466308594,
337
+ "learning_rate": 1.8400000000000003e-05,
338
+ "loss": 12.9766,
339
+ "step": 47
340
+ },
341
+ {
342
+ "epoch": 0.1921921921921922,
343
+ "grad_norm": 57.482357025146484,
344
+ "learning_rate": 1.88e-05,
345
+ "loss": 12.1332,
346
+ "step": 48
347
+ },
348
+ {
349
+ "epoch": 0.1961961961961962,
350
+ "grad_norm": 322.14215087890625,
351
+ "learning_rate": 1.9200000000000003e-05,
352
+ "loss": 11.2119,
353
+ "step": 49
354
+ },
355
+ {
356
+ "epoch": 0.2002002002002002,
357
+ "grad_norm": 72.71914672851562,
358
+ "learning_rate": 1.9600000000000002e-05,
359
+ "loss": 13.4729,
360
+ "step": 50
361
+ },
362
+ {
363
+ "epoch": 0.2042042042042042,
364
+ "grad_norm": 62.99766540527344,
365
+ "learning_rate": 2e-05,
366
+ "loss": 8.6551,
367
+ "step": 51
368
+ },
369
+ {
370
+ "epoch": 0.2082082082082082,
371
+ "grad_norm": 57.57048416137695,
372
+ "learning_rate": 1.9999756307053947e-05,
373
+ "loss": 13.909,
374
+ "step": 52
375
+ },
376
+ {
377
+ "epoch": 0.2122122122122122,
378
+ "grad_norm": 56.78343963623047,
379
+ "learning_rate": 1.9999025240093045e-05,
380
+ "loss": 11.8662,
381
+ "step": 53
382
+ },
383
+ {
384
+ "epoch": 0.21621621621621623,
385
+ "grad_norm": 35.05379867553711,
386
+ "learning_rate": 1.9997806834748455e-05,
387
+ "loss": 8.6797,
388
+ "step": 54
389
+ },
390
+ {
391
+ "epoch": 0.22022022022022023,
392
+ "grad_norm": 91.13362884521484,
393
+ "learning_rate": 1.9996101150403543e-05,
394
+ "loss": 10.3888,
395
+ "step": 55
396
+ },
397
+ {
398
+ "epoch": 0.22422422422422422,
399
+ "grad_norm": 45.59146499633789,
400
+ "learning_rate": 1.999390827019096e-05,
401
+ "loss": 9.2633,
402
+ "step": 56
403
+ },
404
+ {
405
+ "epoch": 0.22822822822822822,
406
+ "grad_norm": 52.165740966796875,
407
+ "learning_rate": 1.9991228300988586e-05,
408
+ "loss": 11.1376,
409
+ "step": 57
410
+ },
411
+ {
412
+ "epoch": 0.23223223223223224,
413
+ "grad_norm": 45.246849060058594,
414
+ "learning_rate": 1.9988061373414342e-05,
415
+ "loss": 8.02,
416
+ "step": 58
417
+ },
418
+ {
419
+ "epoch": 0.23623623623623624,
420
+ "grad_norm": 95.1810073852539,
421
+ "learning_rate": 1.9984407641819812e-05,
422
+ "loss": 9.3933,
423
+ "step": 59
424
+ },
425
+ {
426
+ "epoch": 0.24024024024024024,
427
+ "grad_norm": 44.28993225097656,
428
+ "learning_rate": 1.9980267284282718e-05,
429
+ "loss": 9.2506,
430
+ "step": 60
431
+ },
432
+ {
433
+ "epoch": 0.24424424424424424,
434
+ "grad_norm": 50.0533447265625,
435
+ "learning_rate": 1.9975640502598243e-05,
436
+ "loss": 14.73,
437
+ "step": 61
438
+ },
439
+ {
440
+ "epoch": 0.24824824824824826,
441
+ "grad_norm": 38.94038391113281,
442
+ "learning_rate": 1.9970527522269204e-05,
443
+ "loss": 13.3703,
444
+ "step": 62
445
+ },
446
+ {
447
+ "epoch": 0.25225225225225223,
448
+ "grad_norm": 29.84514617919922,
449
+ "learning_rate": 1.9964928592495046e-05,
450
+ "loss": 9.4598,
451
+ "step": 63
452
+ },
453
+ {
454
+ "epoch": 0.25625625625625625,
455
+ "grad_norm": 31.93252182006836,
456
+ "learning_rate": 1.9958843986159705e-05,
457
+ "loss": 9.6933,
458
+ "step": 64
459
+ },
460
+ {
461
+ "epoch": 0.2602602602602603,
462
+ "grad_norm": 50.74762725830078,
463
+ "learning_rate": 1.9952273999818312e-05,
464
+ "loss": 7.6425,
465
+ "step": 65
466
+ },
467
+ {
468
+ "epoch": 0.26426426426426425,
469
+ "grad_norm": 31.24817657470703,
470
+ "learning_rate": 1.9945218953682736e-05,
471
+ "loss": 8.1771,
472
+ "step": 66
473
+ },
474
+ {
475
+ "epoch": 0.2682682682682683,
476
+ "grad_norm": 39.49109649658203,
477
+ "learning_rate": 1.9937679191605964e-05,
478
+ "loss": 11.6527,
479
+ "step": 67
480
+ },
481
+ {
482
+ "epoch": 0.2722722722722723,
483
+ "grad_norm": 36.089115142822266,
484
+ "learning_rate": 1.992965508106537e-05,
485
+ "loss": 11.0348,
486
+ "step": 68
487
+ },
488
+ {
489
+ "epoch": 0.27627627627627627,
490
+ "grad_norm": 39.42906951904297,
491
+ "learning_rate": 1.9921147013144782e-05,
492
+ "loss": 9.9453,
493
+ "step": 69
494
+ },
495
+ {
496
+ "epoch": 0.2802802802802803,
497
+ "grad_norm": 29.402158737182617,
498
+ "learning_rate": 1.991215540251542e-05,
499
+ "loss": 11.6107,
500
+ "step": 70
501
+ },
502
+ {
503
+ "epoch": 0.28428428428428426,
504
+ "grad_norm": 34.03204345703125,
505
+ "learning_rate": 1.9902680687415704e-05,
506
+ "loss": 11.755,
507
+ "step": 71
508
+ },
509
+ {
510
+ "epoch": 0.2882882882882883,
511
+ "grad_norm": 51.315208435058594,
512
+ "learning_rate": 1.9892723329629885e-05,
513
+ "loss": 10.8995,
514
+ "step": 72
515
+ },
516
+ {
517
+ "epoch": 0.2922922922922923,
518
+ "grad_norm": 29.3174991607666,
519
+ "learning_rate": 1.988228381446553e-05,
520
+ "loss": 9.8221,
521
+ "step": 73
522
+ },
523
+ {
524
+ "epoch": 0.2962962962962963,
525
+ "grad_norm": 26.321531295776367,
526
+ "learning_rate": 1.987136265072988e-05,
527
+ "loss": 6.513,
528
+ "step": 74
529
+ },
530
+ {
531
+ "epoch": 0.3003003003003003,
532
+ "grad_norm": 30.6387882232666,
533
+ "learning_rate": 1.985996037070505e-05,
534
+ "loss": 12.0313,
535
+ "step": 75
536
+ },
537
+ {
538
+ "epoch": 0.30430430430430433,
539
+ "grad_norm": 43.83311462402344,
540
+ "learning_rate": 1.9848077530122083e-05,
541
+ "loss": 8.8236,
542
+ "step": 76
543
+ },
544
+ {
545
+ "epoch": 0.3083083083083083,
546
+ "grad_norm": 44.98320770263672,
547
+ "learning_rate": 1.983571470813386e-05,
548
+ "loss": 9.3465,
549
+ "step": 77
550
+ },
551
+ {
552
+ "epoch": 0.3123123123123123,
553
+ "grad_norm": 34.703189849853516,
554
+ "learning_rate": 1.982287250728689e-05,
555
+ "loss": 12.3766,
556
+ "step": 78
557
+ },
558
+ {
559
+ "epoch": 0.3163163163163163,
560
+ "grad_norm": 45.290496826171875,
561
+ "learning_rate": 1.9809551553491918e-05,
562
+ "loss": 9.9243,
563
+ "step": 79
564
+ },
565
+ {
566
+ "epoch": 0.3203203203203203,
567
+ "grad_norm": 21.45256996154785,
568
+ "learning_rate": 1.979575249599344e-05,
569
+ "loss": 9.9581,
570
+ "step": 80
571
+ },
572
+ {
573
+ "epoch": 0.32432432432432434,
574
+ "grad_norm": 27.00803565979004,
575
+ "learning_rate": 1.9781476007338058e-05,
576
+ "loss": 13.5561,
577
+ "step": 81
578
+ },
579
+ {
580
+ "epoch": 0.3283283283283283,
581
+ "grad_norm": 24.8415470123291,
582
+ "learning_rate": 1.9766722783341682e-05,
583
+ "loss": 13.3993,
584
+ "step": 82
585
+ },
586
+ {
587
+ "epoch": 0.33233233233233234,
588
+ "grad_norm": 20.847021102905273,
589
+ "learning_rate": 1.9751493543055634e-05,
590
+ "loss": 9.1176,
591
+ "step": 83
592
+ },
593
+ {
594
+ "epoch": 0.33633633633633636,
595
+ "grad_norm": 43.43299865722656,
596
+ "learning_rate": 1.9735789028731603e-05,
597
+ "loss": 14.0172,
598
+ "step": 84
599
+ },
600
+ {
601
+ "epoch": 0.34034034034034033,
602
+ "grad_norm": 36.034950256347656,
603
+ "learning_rate": 1.9719610005785466e-05,
604
+ "loss": 10.4853,
605
+ "step": 85
606
+ },
607
+ {
608
+ "epoch": 0.34434434434434436,
609
+ "grad_norm": 42.38825607299805,
610
+ "learning_rate": 1.9702957262759964e-05,
611
+ "loss": 13.2634,
612
+ "step": 86
613
+ },
614
+ {
615
+ "epoch": 0.3483483483483483,
616
+ "grad_norm": 19.491371154785156,
617
+ "learning_rate": 1.9685831611286312e-05,
618
+ "loss": 11.0397,
619
+ "step": 87
620
+ },
621
+ {
622
+ "epoch": 0.35235235235235235,
623
+ "grad_norm": 20.302053451538086,
624
+ "learning_rate": 1.9668233886044597e-05,
625
+ "loss": 8.1921,
626
+ "step": 88
627
+ },
628
+ {
629
+ "epoch": 0.3563563563563564,
630
+ "grad_norm": 19.364652633666992,
631
+ "learning_rate": 1.9650164944723116e-05,
632
+ "loss": 7.3972,
633
+ "step": 89
634
+ },
635
+ {
636
+ "epoch": 0.36036036036036034,
637
+ "grad_norm": 20.764934539794922,
638
+ "learning_rate": 1.9631625667976584e-05,
639
+ "loss": 9.6309,
640
+ "step": 90
641
+ },
642
+ {
643
+ "epoch": 0.36436436436436437,
644
+ "grad_norm": 26.724882125854492,
645
+ "learning_rate": 1.961261695938319e-05,
646
+ "loss": 10.7433,
647
+ "step": 91
648
+ },
649
+ {
650
+ "epoch": 0.3683683683683684,
651
+ "grad_norm": 18.47719955444336,
652
+ "learning_rate": 1.9593139745400575e-05,
653
+ "loss": 10.475,
654
+ "step": 92
655
+ },
656
+ {
657
+ "epoch": 0.37237237237237236,
658
+ "grad_norm": 18.546852111816406,
659
+ "learning_rate": 1.9573194975320672e-05,
660
+ "loss": 8.6371,
661
+ "step": 93
662
+ },
663
+ {
664
+ "epoch": 0.3763763763763764,
665
+ "grad_norm": 19.576723098754883,
666
+ "learning_rate": 1.9552783621223437e-05,
667
+ "loss": 14.267,
668
+ "step": 94
669
+ },
670
+ {
671
+ "epoch": 0.38038038038038036,
672
+ "grad_norm": 32.930912017822266,
673
+ "learning_rate": 1.9531906677929472e-05,
674
+ "loss": 9.0235,
675
+ "step": 95
676
+ },
677
+ {
678
+ "epoch": 0.3843843843843844,
679
+ "grad_norm": 15.93973159790039,
680
+ "learning_rate": 1.9510565162951538e-05,
681
+ "loss": 11.111,
682
+ "step": 96
683
+ },
684
+ {
685
+ "epoch": 0.3883883883883884,
686
+ "grad_norm": 15.77849006652832,
687
+ "learning_rate": 1.9488760116444966e-05,
688
+ "loss": 10.8974,
689
+ "step": 97
690
+ },
691
+ {
692
+ "epoch": 0.3923923923923924,
693
+ "grad_norm": 129.76841735839844,
694
+ "learning_rate": 1.9466492601156964e-05,
695
+ "loss": 8.9554,
696
+ "step": 98
697
+ },
698
+ {
699
+ "epoch": 0.3963963963963964,
700
+ "grad_norm": 15.08523178100586,
701
+ "learning_rate": 1.944376370237481e-05,
702
+ "loss": 10.8524,
703
+ "step": 99
704
+ },
705
+ {
706
+ "epoch": 0.4004004004004004,
707
+ "grad_norm": 48.81966018676758,
708
+ "learning_rate": 1.942057452787297e-05,
709
+ "loss": 12.435,
710
+ "step": 100
711
+ },
712
+ {
713
+ "epoch": 0.4044044044044044,
714
+ "grad_norm": 12.905803680419922,
715
+ "learning_rate": 1.9396926207859085e-05,
716
+ "loss": 8.8466,
717
+ "step": 101
718
+ },
719
+ {
720
+ "epoch": 0.4084084084084084,
721
+ "grad_norm": 12.142631530761719,
722
+ "learning_rate": 1.937281989491892e-05,
723
+ "loss": 8.6941,
724
+ "step": 102
725
+ },
726
+ {
727
+ "epoch": 0.4124124124124124,
728
+ "grad_norm": 17.675872802734375,
729
+ "learning_rate": 1.9348256763960146e-05,
730
+ "loss": 8.9944,
731
+ "step": 103
732
+ },
733
+ {
734
+ "epoch": 0.4164164164164164,
735
+ "grad_norm": 16.91659164428711,
736
+ "learning_rate": 1.9323238012155125e-05,
737
+ "loss": 10.5381,
738
+ "step": 104
739
+ },
740
+ {
741
+ "epoch": 0.42042042042042044,
742
+ "grad_norm": 15.56108570098877,
743
+ "learning_rate": 1.9297764858882516e-05,
744
+ "loss": 10.0021,
745
+ "step": 105
746
+ },
747
+ {
748
+ "epoch": 0.4244244244244244,
749
+ "grad_norm": 17.15951156616211,
750
+ "learning_rate": 1.9271838545667876e-05,
751
+ "loss": 11.5249,
752
+ "step": 106
753
+ },
754
+ {
755
+ "epoch": 0.42842842842842843,
756
+ "grad_norm": 18.636600494384766,
757
+ "learning_rate": 1.9245460336123136e-05,
758
+ "loss": 8.6094,
759
+ "step": 107
760
+ },
761
+ {
762
+ "epoch": 0.43243243243243246,
763
+ "grad_norm": 16.616872787475586,
764
+ "learning_rate": 1.9218631515885007e-05,
765
+ "loss": 12.1505,
766
+ "step": 108
767
+ },
768
+ {
769
+ "epoch": 0.4364364364364364,
770
+ "grad_norm": 22.013092041015625,
771
+ "learning_rate": 1.9191353392552346e-05,
772
+ "loss": 13.9514,
773
+ "step": 109
774
+ },
775
+ {
776
+ "epoch": 0.44044044044044045,
777
+ "grad_norm": 20.6289005279541,
778
+ "learning_rate": 1.9163627295622397e-05,
779
+ "loss": 11.5033,
780
+ "step": 110
781
+ },
782
+ {
783
+ "epoch": 0.4444444444444444,
784
+ "grad_norm": 31.792627334594727,
785
+ "learning_rate": 1.913545457642601e-05,
786
+ "loss": 10.1545,
787
+ "step": 111
788
+ },
789
+ {
790
+ "epoch": 0.44844844844844844,
791
+ "grad_norm": 13.40929889678955,
792
+ "learning_rate": 1.910683660806177e-05,
793
+ "loss": 13.5651,
794
+ "step": 112
795
+ },
796
+ {
797
+ "epoch": 0.45245245245245247,
798
+ "grad_norm": 21.241134643554688,
799
+ "learning_rate": 1.907777478532909e-05,
800
+ "loss": 12.6508,
801
+ "step": 113
802
+ },
803
+ {
804
+ "epoch": 0.45645645645645644,
805
+ "grad_norm": 12.805201530456543,
806
+ "learning_rate": 1.9048270524660197e-05,
807
+ "loss": 9.422,
808
+ "step": 114
809
+ },
810
+ {
811
+ "epoch": 0.46046046046046046,
812
+ "grad_norm": 10.168803215026855,
813
+ "learning_rate": 1.901832526405114e-05,
814
+ "loss": 6.8204,
815
+ "step": 115
816
+ },
817
+ {
818
+ "epoch": 0.4644644644644645,
819
+ "grad_norm": 12.934833526611328,
820
+ "learning_rate": 1.8987940462991673e-05,
821
+ "loss": 7.2831,
822
+ "step": 116
823
+ },
824
+ {
825
+ "epoch": 0.46846846846846846,
826
+ "grad_norm": 12.582077980041504,
827
+ "learning_rate": 1.895711760239413e-05,
828
+ "loss": 12.9766,
829
+ "step": 117
830
+ },
831
+ {
832
+ "epoch": 0.4724724724724725,
833
+ "grad_norm": 11.12944221496582,
834
+ "learning_rate": 1.892585818452126e-05,
835
+ "loss": 8.7003,
836
+ "step": 118
837
+ },
838
+ {
839
+ "epoch": 0.47647647647647645,
840
+ "grad_norm": 19.274728775024414,
841
+ "learning_rate": 1.889416373291298e-05,
842
+ "loss": 8.0606,
843
+ "step": 119
844
+ },
845
+ {
846
+ "epoch": 0.4804804804804805,
847
+ "grad_norm": 13.808324813842773,
848
+ "learning_rate": 1.8862035792312148e-05,
849
+ "loss": 8.4136,
850
+ "step": 120
851
+ },
852
+ {
853
+ "epoch": 0.4844844844844845,
854
+ "grad_norm": 10.761307716369629,
855
+ "learning_rate": 1.8829475928589272e-05,
856
+ "loss": 10.2967,
857
+ "step": 121
858
+ },
859
+ {
860
+ "epoch": 0.48848848848848847,
861
+ "grad_norm": 24.919099807739258,
862
+ "learning_rate": 1.879648572866617e-05,
863
+ "loss": 8.8193,
864
+ "step": 122
865
+ },
866
+ {
867
+ "epoch": 0.4924924924924925,
868
+ "grad_norm": 13.044708251953125,
869
+ "learning_rate": 1.8763066800438638e-05,
870
+ "loss": 11.359,
871
+ "step": 123
872
+ },
873
+ {
874
+ "epoch": 0.4964964964964965,
875
+ "grad_norm": 16.658878326416016,
876
+ "learning_rate": 1.8729220772698096e-05,
877
+ "loss": 11.893,
878
+ "step": 124
879
+ },
880
+ {
881
+ "epoch": 0.5005005005005005,
882
+ "grad_norm": 12.654986381530762,
883
+ "learning_rate": 1.869494929505219e-05,
884
+ "loss": 10.1042,
885
+ "step": 125
886
+ },
887
+ {
888
+ "epoch": 0.5045045045045045,
889
+ "grad_norm": 10.13923454284668,
890
+ "learning_rate": 1.866025403784439e-05,
891
+ "loss": 8.2048,
892
+ "step": 126
893
+ },
894
+ {
895
+ "epoch": 0.5085085085085085,
896
+ "grad_norm": 12.3381986618042,
897
+ "learning_rate": 1.8625136692072577e-05,
898
+ "loss": 12.2311,
899
+ "step": 127
900
+ },
901
+ {
902
+ "epoch": 0.5125125125125125,
903
+ "grad_norm": 13.630654335021973,
904
+ "learning_rate": 1.8589598969306646e-05,
905
+ "loss": 11.5138,
906
+ "step": 128
907
+ },
908
+ {
909
+ "epoch": 0.5165165165165165,
910
+ "grad_norm": 14.09514045715332,
911
+ "learning_rate": 1.855364260160507e-05,
912
+ "loss": 13.0295,
913
+ "step": 129
914
+ },
915
+ {
916
+ "epoch": 0.5205205205205206,
917
+ "grad_norm": 13.090926170349121,
918
+ "learning_rate": 1.851726934143048e-05,
919
+ "loss": 11.4308,
920
+ "step": 130
921
+ },
922
+ {
923
+ "epoch": 0.5245245245245245,
924
+ "grad_norm": 9.631702423095703,
925
+ "learning_rate": 1.848048096156426e-05,
926
+ "loss": 7.2076,
927
+ "step": 131
928
+ },
929
+ {
930
+ "epoch": 0.5285285285285285,
931
+ "grad_norm": 11.130585670471191,
932
+ "learning_rate": 1.8443279255020153e-05,
933
+ "loss": 10.7137,
934
+ "step": 132
935
+ },
936
+ {
937
+ "epoch": 0.5325325325325325,
938
+ "grad_norm": 10.14908218383789,
939
+ "learning_rate": 1.8405666034956842e-05,
940
+ "loss": 10.6793,
941
+ "step": 133
942
+ },
943
+ {
944
+ "epoch": 0.5365365365365365,
945
+ "grad_norm": 10.643115997314453,
946
+ "learning_rate": 1.836764313458962e-05,
947
+ "loss": 9.6055,
948
+ "step": 134
949
+ },
950
+ {
951
+ "epoch": 0.5405405405405406,
952
+ "grad_norm": 11.732181549072266,
953
+ "learning_rate": 1.8329212407100996e-05,
954
+ "loss": 10.2945,
955
+ "step": 135
956
+ },
957
+ {
958
+ "epoch": 0.5445445445445446,
959
+ "grad_norm": 10.43485164642334,
960
+ "learning_rate": 1.8290375725550417e-05,
961
+ "loss": 9.606,
962
+ "step": 136
963
+ },
964
+ {
965
+ "epoch": 0.5485485485485485,
966
+ "grad_norm": 10.665045738220215,
967
+ "learning_rate": 1.8251134982782952e-05,
968
+ "loss": 10.7148,
969
+ "step": 137
970
+ },
971
+ {
972
+ "epoch": 0.5525525525525525,
973
+ "grad_norm": 12.763725280761719,
974
+ "learning_rate": 1.821149209133704e-05,
975
+ "loss": 9.804,
976
+ "step": 138
977
+ },
978
+ {
979
+ "epoch": 0.5565565565565566,
980
+ "grad_norm": 8.759904861450195,
981
+ "learning_rate": 1.8171448983351284e-05,
982
+ "loss": 9.072,
983
+ "step": 139
984
+ },
985
+ {
986
+ "epoch": 0.5605605605605606,
987
+ "grad_norm": 12.307586669921875,
988
+ "learning_rate": 1.8131007610470278e-05,
989
+ "loss": 10.848,
990
+ "step": 140
991
+ },
992
+ {
993
+ "epoch": 0.5645645645645646,
994
+ "grad_norm": 9.156224250793457,
995
+ "learning_rate": 1.8090169943749477e-05,
996
+ "loss": 8.1876,
997
+ "step": 141
998
+ },
999
+ {
1000
+ "epoch": 0.5685685685685685,
1001
+ "grad_norm": 11.873414039611816,
1002
+ "learning_rate": 1.804893797355914e-05,
1003
+ "loss": 7.681,
1004
+ "step": 142
1005
+ },
1006
+ {
1007
+ "epoch": 0.5725725725725725,
1008
+ "grad_norm": 8.32168197631836,
1009
+ "learning_rate": 1.8007313709487334e-05,
1010
+ "loss": 8.085,
1011
+ "step": 143
1012
+ },
1013
+ {
1014
+ "epoch": 0.5765765765765766,
1015
+ "grad_norm": 11.67728328704834,
1016
+ "learning_rate": 1.7965299180241963e-05,
1017
+ "loss": 13.952,
1018
+ "step": 144
1019
+ },
1020
+ {
1021
+ "epoch": 0.5805805805805806,
1022
+ "grad_norm": 10.53978157043457,
1023
+ "learning_rate": 1.792289643355191e-05,
1024
+ "loss": 8.5771,
1025
+ "step": 145
1026
+ },
1027
+ {
1028
+ "epoch": 0.5845845845845846,
1029
+ "grad_norm": 13.081896781921387,
1030
+ "learning_rate": 1.788010753606722e-05,
1031
+ "loss": 7.9493,
1032
+ "step": 146
1033
+ },
1034
+ {
1035
+ "epoch": 0.5885885885885885,
1036
+ "grad_norm": 14.055414199829102,
1037
+ "learning_rate": 1.78369345732584e-05,
1038
+ "loss": 12.0875,
1039
+ "step": 147
1040
+ },
1041
+ {
1042
+ "epoch": 0.5925925925925926,
1043
+ "grad_norm": 9.662280082702637,
1044
+ "learning_rate": 1.7793379649314743e-05,
1045
+ "loss": 9.7244,
1046
+ "step": 148
1047
+ },
1048
+ {
1049
+ "epoch": 0.5965965965965966,
1050
+ "grad_norm": 9.558379173278809,
1051
+ "learning_rate": 1.7749444887041797e-05,
1052
+ "loss": 8.237,
1053
+ "step": 149
1054
+ },
1055
+ {
1056
+ "epoch": 0.6006006006006006,
1057
+ "grad_norm": 11.434876441955566,
1058
+ "learning_rate": 1.7705132427757895e-05,
1059
+ "loss": 8.1637,
1060
+ "step": 150
1061
+ },
1062
+ {
1063
+ "epoch": 0.6046046046046046,
1064
+ "grad_norm": 13.445454597473145,
1065
+ "learning_rate": 1.766044443118978e-05,
1066
+ "loss": 10.9758,
1067
+ "step": 151
1068
+ },
1069
+ {
1070
+ "epoch": 0.6086086086086087,
1071
+ "grad_norm": 24.660717010498047,
1072
+ "learning_rate": 1.761538307536737e-05,
1073
+ "loss": 8.8184,
1074
+ "step": 152
1075
+ },
1076
+ {
1077
+ "epoch": 0.6126126126126126,
1078
+ "grad_norm": 9.408225059509277,
1079
+ "learning_rate": 1.7569950556517566e-05,
1080
+ "loss": 9.8908,
1081
+ "step": 153
1082
+ },
1083
+ {
1084
+ "epoch": 0.6166166166166166,
1085
+ "grad_norm": 11.434803009033203,
1086
+ "learning_rate": 1.7524149088957244e-05,
1087
+ "loss": 10.0439,
1088
+ "step": 154
1089
+ },
1090
+ {
1091
+ "epoch": 0.6206206206206206,
1092
+ "grad_norm": 7.227975368499756,
1093
+ "learning_rate": 1.747798090498532e-05,
1094
+ "loss": 6.7221,
1095
+ "step": 155
1096
+ },
1097
+ {
1098
+ "epoch": 0.6246246246246246,
1099
+ "grad_norm": 6.4641194343566895,
1100
+ "learning_rate": 1.7431448254773943e-05,
1101
+ "loss": 5.1012,
1102
+ "step": 156
1103
+ },
1104
+ {
1105
+ "epoch": 0.6286286286286287,
1106
+ "grad_norm": 10.321112632751465,
1107
+ "learning_rate": 1.7384553406258842e-05,
1108
+ "loss": 8.812,
1109
+ "step": 157
1110
+ },
1111
+ {
1112
+ "epoch": 0.6326326326326326,
1113
+ "grad_norm": 12.71799373626709,
1114
+ "learning_rate": 1.7337298645028764e-05,
1115
+ "loss": 10.6263,
1116
+ "step": 158
1117
+ },
1118
+ {
1119
+ "epoch": 0.6366366366366366,
1120
+ "grad_norm": 10.158978462219238,
1121
+ "learning_rate": 1.7289686274214116e-05,
1122
+ "loss": 10.0739,
1123
+ "step": 159
1124
+ },
1125
+ {
1126
+ "epoch": 0.6406406406406406,
1127
+ "grad_norm": 9.351489067077637,
1128
+ "learning_rate": 1.7241718614374678e-05,
1129
+ "loss": 7.3389,
1130
+ "step": 160
1131
+ },
1132
+ {
1133
+ "epoch": 0.6446446446446447,
1134
+ "grad_norm": 13.251178741455078,
1135
+ "learning_rate": 1.7193398003386514e-05,
1136
+ "loss": 11.0626,
1137
+ "step": 161
1138
+ },
1139
+ {
1140
+ "epoch": 0.6486486486486487,
1141
+ "grad_norm": 8.338618278503418,
1142
+ "learning_rate": 1.7144726796328034e-05,
1143
+ "loss": 8.3801,
1144
+ "step": 162
1145
+ },
1146
+ {
1147
+ "epoch": 0.6526526526526526,
1148
+ "grad_norm": 12.511034965515137,
1149
+ "learning_rate": 1.709570736536521e-05,
1150
+ "loss": 10.9749,
1151
+ "step": 163
1152
+ },
1153
+ {
1154
+ "epoch": 0.6566566566566566,
1155
+ "grad_norm": 8.098262786865234,
1156
+ "learning_rate": 1.7046342099635948e-05,
1157
+ "loss": 8.4176,
1158
+ "step": 164
1159
+ },
1160
+ {
1161
+ "epoch": 0.6606606606606606,
1162
+ "grad_norm": 12.58228874206543,
1163
+ "learning_rate": 1.6996633405133656e-05,
1164
+ "loss": 10.8064,
1165
+ "step": 165
1166
+ },
1167
+ {
1168
+ "epoch": 0.6646646646646647,
1169
+ "grad_norm": 11.529173851013184,
1170
+ "learning_rate": 1.6946583704589973e-05,
1171
+ "loss": 10.1078,
1172
+ "step": 166
1173
+ },
1174
+ {
1175
+ "epoch": 0.6686686686686687,
1176
+ "grad_norm": 12.31688117980957,
1177
+ "learning_rate": 1.68961954373567e-05,
1178
+ "loss": 9.094,
1179
+ "step": 167
1180
+ },
1181
+ {
1182
+ "epoch": 0.6726726726726727,
1183
+ "grad_norm": 9.920634269714355,
1184
+ "learning_rate": 1.684547105928689e-05,
1185
+ "loss": 10.8316,
1186
+ "step": 168
1187
+ },
1188
+ {
1189
+ "epoch": 0.6766766766766766,
1190
+ "grad_norm": 11.913458824157715,
1191
+ "learning_rate": 1.6794413042615168e-05,
1192
+ "loss": 6.7716,
1193
+ "step": 169
1194
+ },
1195
+ {
1196
+ "epoch": 0.6806806806806807,
1197
+ "grad_norm": 9.994155883789062,
1198
+ "learning_rate": 1.6743023875837233e-05,
1199
+ "loss": 9.5964,
1200
+ "step": 170
1201
+ },
1202
+ {
1203
+ "epoch": 0.6846846846846847,
1204
+ "grad_norm": 7.584497451782227,
1205
+ "learning_rate": 1.6691306063588583e-05,
1206
+ "loss": 8.6275,
1207
+ "step": 171
1208
+ },
1209
+ {
1210
+ "epoch": 0.6886886886886887,
1211
+ "grad_norm": 11.261972427368164,
1212
+ "learning_rate": 1.6639262126522417e-05,
1213
+ "loss": 7.6235,
1214
+ "step": 172
1215
+ },
1216
+ {
1217
+ "epoch": 0.6926926926926927,
1218
+ "grad_norm": 8.638460159301758,
1219
+ "learning_rate": 1.6586894601186804e-05,
1220
+ "loss": 9.0315,
1221
+ "step": 173
1222
+ },
1223
+ {
1224
+ "epoch": 0.6966966966966966,
1225
+ "grad_norm": 10.671465873718262,
1226
+ "learning_rate": 1.6534206039901057e-05,
1227
+ "loss": 9.5828,
1228
+ "step": 174
1229
+ },
1230
+ {
1231
+ "epoch": 0.7007007007007007,
1232
+ "grad_norm": 10.782361030578613,
1233
+ "learning_rate": 1.6481199010631312e-05,
1234
+ "loss": 10.148,
1235
+ "step": 175
1236
+ },
1237
+ {
1238
+ "epoch": 0.7047047047047047,
1239
+ "grad_norm": 8.802865982055664,
1240
+ "learning_rate": 1.6427876096865394e-05,
1241
+ "loss": 7.8103,
1242
+ "step": 176
1243
+ },
1244
+ {
1245
+ "epoch": 0.7087087087087087,
1246
+ "grad_norm": 9.869688034057617,
1247
+ "learning_rate": 1.63742398974869e-05,
1248
+ "loss": 11.8441,
1249
+ "step": 177
1250
+ },
1251
+ {
1252
+ "epoch": 0.7127127127127127,
1253
+ "grad_norm": 9.582351684570312,
1254
+ "learning_rate": 1.632029302664851e-05,
1255
+ "loss": 12.8614,
1256
+ "step": 178
1257
+ },
1258
+ {
1259
+ "epoch": 0.7167167167167167,
1260
+ "grad_norm": 8.890612602233887,
1261
+ "learning_rate": 1.6266038113644605e-05,
1262
+ "loss": 10.5914,
1263
+ "step": 179
1264
+ },
1265
+ {
1266
+ "epoch": 0.7207207207207207,
1267
+ "grad_norm": 8.70327091217041,
1268
+ "learning_rate": 1.6211477802783105e-05,
1269
+ "loss": 7.0284,
1270
+ "step": 180
1271
+ },
1272
+ {
1273
+ "epoch": 0.7247247247247247,
1274
+ "grad_norm": 10.544742584228516,
1275
+ "learning_rate": 1.6156614753256583e-05,
1276
+ "loss": 10.9661,
1277
+ "step": 181
1278
+ },
1279
+ {
1280
+ "epoch": 0.7287287287287287,
1281
+ "grad_norm": 9.516739845275879,
1282
+ "learning_rate": 1.610145163901268e-05,
1283
+ "loss": 10.4819,
1284
+ "step": 182
1285
+ },
1286
+ {
1287
+ "epoch": 0.7327327327327328,
1288
+ "grad_norm": 9.150429725646973,
1289
+ "learning_rate": 1.6045991148623752e-05,
1290
+ "loss": 7.1882,
1291
+ "step": 183
1292
+ },
1293
+ {
1294
+ "epoch": 0.7367367367367368,
1295
+ "grad_norm": 7.603723049163818,
1296
+ "learning_rate": 1.599023598515586e-05,
1297
+ "loss": 6.8023,
1298
+ "step": 184
1299
+ },
1300
+ {
1301
+ "epoch": 0.7407407407407407,
1302
+ "grad_norm": 7.485808849334717,
1303
+ "learning_rate": 1.5934188866037017e-05,
1304
+ "loss": 7.0446,
1305
+ "step": 185
1306
+ },
1307
+ {
1308
+ "epoch": 0.7447447447447447,
1309
+ "grad_norm": 12.322443962097168,
1310
+ "learning_rate": 1.5877852522924733e-05,
1311
+ "loss": 9.629,
1312
+ "step": 186
1313
+ },
1314
+ {
1315
+ "epoch": 0.7487487487487487,
1316
+ "grad_norm": 10.206731796264648,
1317
+ "learning_rate": 1.5821229701572897e-05,
1318
+ "loss": 8.2926,
1319
+ "step": 187
1320
+ },
1321
+ {
1322
+ "epoch": 0.7527527527527528,
1323
+ "grad_norm": 9.019510269165039,
1324
+ "learning_rate": 1.5764323161697933e-05,
1325
+ "loss": 8.6706,
1326
+ "step": 188
1327
+ },
1328
+ {
1329
+ "epoch": 0.7567567567567568,
1330
+ "grad_norm": 11.982542037963867,
1331
+ "learning_rate": 1.570713567684432e-05,
1332
+ "loss": 9.6295,
1333
+ "step": 189
1334
+ },
1335
+ {
1336
+ "epoch": 0.7607607607607607,
1337
+ "grad_norm": 8.122821807861328,
1338
+ "learning_rate": 1.564967003424938e-05,
1339
+ "loss": 9.8489,
1340
+ "step": 190
1341
+ },
1342
+ {
1343
+ "epoch": 0.7647647647647647,
1344
+ "grad_norm": 9.51779556274414,
1345
+ "learning_rate": 1.5591929034707468e-05,
1346
+ "loss": 11.341,
1347
+ "step": 191
1348
+ },
1349
+ {
1350
+ "epoch": 0.7687687687687688,
1351
+ "grad_norm": 8.324822425842285,
1352
+ "learning_rate": 1.553391549243344e-05,
1353
+ "loss": 8.3743,
1354
+ "step": 192
1355
+ },
1356
+ {
1357
+ "epoch": 0.7727727727727728,
1358
+ "grad_norm": 11.880289077758789,
1359
+ "learning_rate": 1.5475632234925505e-05,
1360
+ "loss": 11.0088,
1361
+ "step": 193
1362
+ },
1363
+ {
1364
+ "epoch": 0.7767767767767768,
1365
+ "grad_norm": 9.262480735778809,
1366
+ "learning_rate": 1.54170821028274e-05,
1367
+ "loss": 9.9509,
1368
+ "step": 194
1369
+ },
1370
+ {
1371
+ "epoch": 0.7807807807807807,
1372
+ "grad_norm": 7.56397819519043,
1373
+ "learning_rate": 1.5358267949789968e-05,
1374
+ "loss": 9.4004,
1375
+ "step": 195
1376
+ },
1377
+ {
1378
+ "epoch": 0.7847847847847848,
1379
+ "grad_norm": 12.214609146118164,
1380
+ "learning_rate": 1.529919264233205e-05,
1381
+ "loss": 9.4932,
1382
+ "step": 196
1383
+ },
1384
+ {
1385
+ "epoch": 0.7887887887887888,
1386
+ "grad_norm": 10.519573211669922,
1387
+ "learning_rate": 1.5239859059700794e-05,
1388
+ "loss": 8.6798,
1389
+ "step": 197
1390
+ },
1391
+ {
1392
+ "epoch": 0.7927927927927928,
1393
+ "grad_norm": 8.395581245422363,
1394
+ "learning_rate": 1.5180270093731305e-05,
1395
+ "loss": 11.1879,
1396
+ "step": 198
1397
+ },
1398
+ {
1399
+ "epoch": 0.7967967967967968,
1400
+ "grad_norm": 8.176620483398438,
1401
+ "learning_rate": 1.5120428648705716e-05,
1402
+ "loss": 11.1611,
1403
+ "step": 199
1404
+ },
1405
+ {
1406
+ "epoch": 0.8008008008008008,
1407
+ "grad_norm": 8.647604942321777,
1408
+ "learning_rate": 1.5060337641211637e-05,
1409
+ "loss": 8.4677,
1410
+ "step": 200
1411
+ },
1412
+ {
1413
+ "epoch": 0.8048048048048048,
1414
+ "grad_norm": 8.022457122802734,
1415
+ "learning_rate": 1.5000000000000002e-05,
1416
+ "loss": 7.4944,
1417
+ "step": 201
1418
+ },
1419
+ {
1420
+ "epoch": 0.8088088088088088,
1421
+ "grad_norm": 6.951892375946045,
1422
+ "learning_rate": 1.493941866584231e-05,
1423
+ "loss": 8.132,
1424
+ "step": 202
1425
+ },
1426
+ {
1427
+ "epoch": 0.8128128128128128,
1428
+ "grad_norm": 10.608201026916504,
1429
+ "learning_rate": 1.4878596591387329e-05,
1430
+ "loss": 10.2332,
1431
+ "step": 203
1432
+ },
1433
+ {
1434
+ "epoch": 0.8168168168168168,
1435
+ "grad_norm": 11.19121265411377,
1436
+ "learning_rate": 1.4817536741017153e-05,
1437
+ "loss": 12.4648,
1438
+ "step": 204
1439
+ },
1440
+ {
1441
+ "epoch": 0.8208208208208209,
1442
+ "grad_norm": 5.869455337524414,
1443
+ "learning_rate": 1.4756242090702756e-05,
1444
+ "loss": 7.7229,
1445
+ "step": 205
1446
+ },
1447
+ {
1448
+ "epoch": 0.8248248248248248,
1449
+ "grad_norm": 10.275627136230469,
1450
+ "learning_rate": 1.469471562785891e-05,
1451
+ "loss": 8.7337,
1452
+ "step": 206
1453
+ },
1454
+ {
1455
+ "epoch": 0.8288288288288288,
1456
+ "grad_norm": 9.787700653076172,
1457
+ "learning_rate": 1.463296035119862e-05,
1458
+ "loss": 11.5797,
1459
+ "step": 207
1460
+ },
1461
+ {
1462
+ "epoch": 0.8328328328328328,
1463
+ "grad_norm": 7.975783824920654,
1464
+ "learning_rate": 1.4570979270586944e-05,
1465
+ "loss": 6.6192,
1466
+ "step": 208
1467
+ },
1468
+ {
1469
+ "epoch": 0.8368368368368369,
1470
+ "grad_norm": 10.365467071533203,
1471
+ "learning_rate": 1.4508775406894308e-05,
1472
+ "loss": 9.8661,
1473
+ "step": 209
1474
+ },
1475
+ {
1476
+ "epoch": 0.8408408408408409,
1477
+ "grad_norm": 6.009556293487549,
1478
+ "learning_rate": 1.4446351791849276e-05,
1479
+ "loss": 7.116,
1480
+ "step": 210
1481
+ },
1482
+ {
1483
+ "epoch": 0.8448448448448449,
1484
+ "grad_norm": 4.778204917907715,
1485
+ "learning_rate": 1.4383711467890776e-05,
1486
+ "loss": 7.6266,
1487
+ "step": 211
1488
+ },
1489
+ {
1490
+ "epoch": 0.8488488488488488,
1491
+ "grad_norm": 6.596901893615723,
1492
+ "learning_rate": 1.4320857488019826e-05,
1493
+ "loss": 5.9151,
1494
+ "step": 212
1495
+ },
1496
+ {
1497
+ "epoch": 0.8528528528528528,
1498
+ "grad_norm": 10.031322479248047,
1499
+ "learning_rate": 1.4257792915650728e-05,
1500
+ "loss": 9.5717,
1501
+ "step": 213
1502
+ },
1503
+ {
1504
+ "epoch": 0.8568568568568569,
1505
+ "grad_norm": 26.560138702392578,
1506
+ "learning_rate": 1.4194520824461773e-05,
1507
+ "loss": 8.7155,
1508
+ "step": 214
1509
+ },
1510
+ {
1511
+ "epoch": 0.8608608608608609,
1512
+ "grad_norm": 9.228643417358398,
1513
+ "learning_rate": 1.413104429824542e-05,
1514
+ "loss": 9.5688,
1515
+ "step": 215
1516
+ },
1517
+ {
1518
+ "epoch": 0.8648648648648649,
1519
+ "grad_norm": 8.710359573364258,
1520
+ "learning_rate": 1.4067366430758004e-05,
1521
+ "loss": 10.1115,
1522
+ "step": 216
1523
+ },
1524
+ {
1525
+ "epoch": 0.8688688688688688,
1526
+ "grad_norm": 9.233997344970703,
1527
+ "learning_rate": 1.4003490325568953e-05,
1528
+ "loss": 12.7164,
1529
+ "step": 217
1530
+ },
1531
+ {
1532
+ "epoch": 0.8728728728728729,
1533
+ "grad_norm": 9.377655029296875,
1534
+ "learning_rate": 1.3939419095909513e-05,
1535
+ "loss": 9.037,
1536
+ "step": 218
1537
+ },
1538
+ {
1539
+ "epoch": 0.8768768768768769,
1540
+ "grad_norm": 9.314506530761719,
1541
+ "learning_rate": 1.3875155864521031e-05,
1542
+ "loss": 10.3881,
1543
+ "step": 219
1544
+ },
1545
+ {
1546
+ "epoch": 0.8808808808808809,
1547
+ "grad_norm": 9.435461044311523,
1548
+ "learning_rate": 1.3810703763502744e-05,
1549
+ "loss": 10.1386,
1550
+ "step": 220
1551
+ },
1552
+ {
1553
+ "epoch": 0.8848848848848849,
1554
+ "grad_norm": 7.5539164543151855,
1555
+ "learning_rate": 1.3746065934159123e-05,
1556
+ "loss": 8.3585,
1557
+ "step": 221
1558
+ },
1559
+ {
1560
+ "epoch": 0.8888888888888888,
1561
+ "grad_norm": 7.970242977142334,
1562
+ "learning_rate": 1.3681245526846782e-05,
1563
+ "loss": 10.2223,
1564
+ "step": 222
1565
+ },
1566
+ {
1567
+ "epoch": 0.8928928928928929,
1568
+ "grad_norm": 11.161116600036621,
1569
+ "learning_rate": 1.3616245700820922e-05,
1570
+ "loss": 11.4815,
1571
+ "step": 223
1572
+ },
1573
+ {
1574
+ "epoch": 0.8968968968968969,
1575
+ "grad_norm": 7.511813640594482,
1576
+ "learning_rate": 1.3551069624081372e-05,
1577
+ "loss": 7.3804,
1578
+ "step": 224
1579
+ },
1580
+ {
1581
+ "epoch": 0.9009009009009009,
1582
+ "grad_norm": 9.041431427001953,
1583
+ "learning_rate": 1.3485720473218153e-05,
1584
+ "loss": 9.3529,
1585
+ "step": 225
1586
+ },
1587
+ {
1588
+ "epoch": 0.9049049049049049,
1589
+ "grad_norm": 7.143396854400635,
1590
+ "learning_rate": 1.342020143325669e-05,
1591
+ "loss": 8.7255,
1592
+ "step": 226
1593
+ },
1594
+ {
1595
+ "epoch": 0.908908908908909,
1596
+ "grad_norm": 8.11451244354248,
1597
+ "learning_rate": 1.3354515697502552e-05,
1598
+ "loss": 8.8726,
1599
+ "step": 227
1600
+ },
1601
+ {
1602
+ "epoch": 0.9129129129129129,
1603
+ "grad_norm": 9.570343971252441,
1604
+ "learning_rate": 1.3288666467385834e-05,
1605
+ "loss": 11.0756,
1606
+ "step": 228
1607
+ },
1608
+ {
1609
+ "epoch": 0.9169169169169169,
1610
+ "grad_norm": 7.341606616973877,
1611
+ "learning_rate": 1.3222656952305113e-05,
1612
+ "loss": 8.4637,
1613
+ "step": 229
1614
+ },
1615
+ {
1616
+ "epoch": 0.9209209209209209,
1617
+ "grad_norm": 9.357011795043945,
1618
+ "learning_rate": 1.3156490369471026e-05,
1619
+ "loss": 9.7954,
1620
+ "step": 230
1621
+ },
1622
+ {
1623
+ "epoch": 0.924924924924925,
1624
+ "grad_norm": 8.390585899353027,
1625
+ "learning_rate": 1.3090169943749475e-05,
1626
+ "loss": 7.4562,
1627
+ "step": 231
1628
+ },
1629
+ {
1630
+ "epoch": 0.928928928928929,
1631
+ "grad_norm": 9.880354881286621,
1632
+ "learning_rate": 1.3023698907504447e-05,
1633
+ "loss": 6.7831,
1634
+ "step": 232
1635
+ },
1636
+ {
1637
+ "epoch": 0.9329329329329329,
1638
+ "grad_norm": 11.886743545532227,
1639
+ "learning_rate": 1.2957080500440469e-05,
1640
+ "loss": 11.7537,
1641
+ "step": 233
1642
+ },
1643
+ {
1644
+ "epoch": 0.9369369369369369,
1645
+ "grad_norm": 8.389444351196289,
1646
+ "learning_rate": 1.2890317969444716e-05,
1647
+ "loss": 10.4195,
1648
+ "step": 234
1649
+ },
1650
+ {
1651
+ "epoch": 0.9409409409409409,
1652
+ "grad_norm": 8.356512069702148,
1653
+ "learning_rate": 1.2823414568428767e-05,
1654
+ "loss": 8.7304,
1655
+ "step": 235
1656
+ },
1657
+ {
1658
+ "epoch": 0.944944944944945,
1659
+ "grad_norm": 10.684683799743652,
1660
+ "learning_rate": 1.2756373558169992e-05,
1661
+ "loss": 10.0164,
1662
+ "step": 236
1663
+ },
1664
+ {
1665
+ "epoch": 0.948948948948949,
1666
+ "grad_norm": 8.662108421325684,
1667
+ "learning_rate": 1.2689198206152657e-05,
1668
+ "loss": 5.9544,
1669
+ "step": 237
1670
+ },
1671
+ {
1672
+ "epoch": 0.9529529529529529,
1673
+ "grad_norm": 10.215584754943848,
1674
+ "learning_rate": 1.2621891786408648e-05,
1675
+ "loss": 11.3221,
1676
+ "step": 238
1677
+ },
1678
+ {
1679
+ "epoch": 0.9569569569569569,
1680
+ "grad_norm": 10.895888328552246,
1681
+ "learning_rate": 1.2554457579357906e-05,
1682
+ "loss": 11.3697,
1683
+ "step": 239
1684
+ },
1685
+ {
1686
+ "epoch": 0.960960960960961,
1687
+ "grad_norm": 9.184041023254395,
1688
+ "learning_rate": 1.2486898871648552e-05,
1689
+ "loss": 8.0836,
1690
+ "step": 240
1691
+ },
1692
+ {
1693
+ "epoch": 0.964964964964965,
1694
+ "grad_norm": 11.528921127319336,
1695
+ "learning_rate": 1.2419218955996677e-05,
1696
+ "loss": 11.6967,
1697
+ "step": 241
1698
+ },
1699
+ {
1700
+ "epoch": 0.968968968968969,
1701
+ "grad_norm": 7.614370822906494,
1702
+ "learning_rate": 1.23514211310259e-05,
1703
+ "loss": 7.4309,
1704
+ "step": 242
1705
+ },
1706
+ {
1707
+ "epoch": 0.972972972972973,
1708
+ "grad_norm": 9.231164932250977,
1709
+ "learning_rate": 1.2283508701106559e-05,
1710
+ "loss": 8.9294,
1711
+ "step": 243
1712
+ },
1713
+ {
1714
+ "epoch": 0.9769769769769769,
1715
+ "grad_norm": 8.950560569763184,
1716
+ "learning_rate": 1.2215484976194675e-05,
1717
+ "loss": 8.4373,
1718
+ "step": 244
1719
+ },
1720
+ {
1721
+ "epoch": 0.980980980980981,
1722
+ "grad_norm": 8.459944725036621,
1723
+ "learning_rate": 1.2147353271670634e-05,
1724
+ "loss": 8.201,
1725
+ "step": 245
1726
+ },
1727
+ {
1728
+ "epoch": 0.984984984984985,
1729
+ "grad_norm": 7.455278396606445,
1730
+ "learning_rate": 1.2079116908177592e-05,
1731
+ "loss": 7.269,
1732
+ "step": 246
1733
+ },
1734
+ {
1735
+ "epoch": 0.988988988988989,
1736
+ "grad_norm": 16.996952056884766,
1737
+ "learning_rate": 1.2010779211459649e-05,
1738
+ "loss": 11.7109,
1739
+ "step": 247
1740
+ },
1741
+ {
1742
+ "epoch": 0.992992992992993,
1743
+ "grad_norm": 13.302130699157715,
1744
+ "learning_rate": 1.194234351219972e-05,
1745
+ "loss": 11.5493,
1746
+ "step": 248
1747
+ },
1748
+ {
1749
+ "epoch": 0.996996996996997,
1750
+ "grad_norm": 7.6457743644714355,
1751
+ "learning_rate": 1.187381314585725e-05,
1752
+ "loss": 8.1011,
1753
+ "step": 249
1754
+ },
1755
+ {
1756
+ "epoch": 1.0,
1757
+ "grad_norm": 7.505258083343506,
1758
+ "learning_rate": 1.1805191452505602e-05,
1759
+ "loss": 9.8145,
1760
+ "step": 250
1761
+ },
1762
+ {
1763
+ "epoch": 1.004004004004004,
1764
+ "grad_norm": 6.933170795440674,
1765
+ "learning_rate": 1.1736481776669307e-05,
1766
+ "loss": 7.6755,
1767
+ "step": 251
1768
+ },
1769
+ {
1770
+ "epoch": 1.008008008008008,
1771
+ "grad_norm": 6.536367893218994,
1772
+ "learning_rate": 1.1667687467161025e-05,
1773
+ "loss": 7.946,
1774
+ "step": 252
1775
+ },
1776
+ {
1777
+ "epoch": 1.012012012012012,
1778
+ "grad_norm": 8.585467338562012,
1779
+ "learning_rate": 1.159881187691835e-05,
1780
+ "loss": 9.8278,
1781
+ "step": 253
1782
+ },
1783
+ {
1784
+ "epoch": 1.016016016016016,
1785
+ "grad_norm": 5.9852776527404785,
1786
+ "learning_rate": 1.1529858362840383e-05,
1787
+ "loss": 5.818,
1788
+ "step": 254
1789
+ },
1790
+ {
1791
+ "epoch": 1.02002002002002,
1792
+ "grad_norm": 7.870911598205566,
1793
+ "learning_rate": 1.1460830285624119e-05,
1794
+ "loss": 9.4468,
1795
+ "step": 255
1796
+ },
1797
+ {
1798
+ "epoch": 1.024024024024024,
1799
+ "grad_norm": 10.146224975585938,
1800
+ "learning_rate": 1.1391731009600655e-05,
1801
+ "loss": 10.7049,
1802
+ "step": 256
1803
+ },
1804
+ {
1805
+ "epoch": 1.028028028028028,
1806
+ "grad_norm": 8.45648193359375,
1807
+ "learning_rate": 1.1322563902571227e-05,
1808
+ "loss": 9.0442,
1809
+ "step": 257
1810
+ },
1811
+ {
1812
+ "epoch": 1.032032032032032,
1813
+ "grad_norm": 6.834626197814941,
1814
+ "learning_rate": 1.1253332335643043e-05,
1815
+ "loss": 8.695,
1816
+ "step": 258
1817
+ },
1818
+ {
1819
+ "epoch": 1.0360360360360361,
1820
+ "grad_norm": 10.299759864807129,
1821
+ "learning_rate": 1.1184039683065014e-05,
1822
+ "loss": 10.9775,
1823
+ "step": 259
1824
+ },
1825
+ {
1826
+ "epoch": 1.04004004004004,
1827
+ "grad_norm": 9.172685623168945,
1828
+ "learning_rate": 1.1114689322063255e-05,
1829
+ "loss": 12.1397,
1830
+ "step": 260
1831
+ },
1832
+ {
1833
+ "epoch": 1.044044044044044,
1834
+ "grad_norm": 7.131484508514404,
1835
+ "learning_rate": 1.1045284632676535e-05,
1836
+ "loss": 5.0461,
1837
+ "step": 261
1838
+ },
1839
+ {
1840
+ "epoch": 1.048048048048048,
1841
+ "grad_norm": 9.110465049743652,
1842
+ "learning_rate": 1.0975828997591496e-05,
1843
+ "loss": 10.0956,
1844
+ "step": 262
1845
+ },
1846
+ {
1847
+ "epoch": 1.052052052052052,
1848
+ "grad_norm": 6.718713760375977,
1849
+ "learning_rate": 1.0906325801977804e-05,
1850
+ "loss": 7.3981,
1851
+ "step": 263
1852
+ },
1853
+ {
1854
+ "epoch": 1.0560560560560561,
1855
+ "grad_norm": 8.344298362731934,
1856
+ "learning_rate": 1.083677843332316e-05,
1857
+ "loss": 9.5594,
1858
+ "step": 264
1859
+ },
1860
+ {
1861
+ "epoch": 1.06006006006006,
1862
+ "grad_norm": 9.895957946777344,
1863
+ "learning_rate": 1.0767190281268187e-05,
1864
+ "loss": 11.5868,
1865
+ "step": 265
1866
+ },
1867
+ {
1868
+ "epoch": 1.064064064064064,
1869
+ "grad_norm": 8.519937515258789,
1870
+ "learning_rate": 1.0697564737441254e-05,
1871
+ "loss": 9.1346,
1872
+ "step": 266
1873
+ },
1874
+ {
1875
+ "epoch": 1.068068068068068,
1876
+ "grad_norm": 9.378730773925781,
1877
+ "learning_rate": 1.0627905195293135e-05,
1878
+ "loss": 9.8642,
1879
+ "step": 267
1880
+ },
1881
+ {
1882
+ "epoch": 1.072072072072072,
1883
+ "grad_norm": 9.651863098144531,
1884
+ "learning_rate": 1.055821504993164e-05,
1885
+ "loss": 8.3585,
1886
+ "step": 268
1887
+ },
1888
+ {
1889
+ "epoch": 1.0760760760760761,
1890
+ "grad_norm": 9.857842445373535,
1891
+ "learning_rate": 1.0488497697956134e-05,
1892
+ "loss": 11.3395,
1893
+ "step": 269
1894
+ },
1895
+ {
1896
+ "epoch": 1.08008008008008,
1897
+ "grad_norm": 8.212699890136719,
1898
+ "learning_rate": 1.0418756537291996e-05,
1899
+ "loss": 8.4237,
1900
+ "step": 270
1901
+ },
1902
+ {
1903
+ "epoch": 1.0840840840840842,
1904
+ "grad_norm": 7.172947883605957,
1905
+ "learning_rate": 1.0348994967025012e-05,
1906
+ "loss": 9.7379,
1907
+ "step": 271
1908
+ },
1909
+ {
1910
+ "epoch": 1.088088088088088,
1911
+ "grad_norm": 6.861180305480957,
1912
+ "learning_rate": 1.0279216387235691e-05,
1913
+ "loss": 8.7004,
1914
+ "step": 272
1915
+ },
1916
+ {
1917
+ "epoch": 1.092092092092092,
1918
+ "grad_norm": 9.049416542053223,
1919
+ "learning_rate": 1.0209424198833571e-05,
1920
+ "loss": 7.6999,
1921
+ "step": 273
1922
+ },
1923
+ {
1924
+ "epoch": 1.0960960960960962,
1925
+ "grad_norm": 9.505548477172852,
1926
+ "learning_rate": 1.0139621803391454e-05,
1927
+ "loss": 11.3466,
1928
+ "step": 274
1929
+ },
1930
+ {
1931
+ "epoch": 1.1001001001001,
1932
+ "grad_norm": 7.87801456451416,
1933
+ "learning_rate": 1.0069812602979617e-05,
1934
+ "loss": 8.511,
1935
+ "step": 275
1936
+ },
1937
+ {
1938
+ "epoch": 1.1041041041041042,
1939
+ "grad_norm": 8.974830627441406,
1940
+ "learning_rate": 1e-05,
1941
+ "loss": 11.6239,
1942
+ "step": 276
1943
+ },
1944
+ {
1945
+ "epoch": 1.1081081081081081,
1946
+ "grad_norm": 9.153380393981934,
1947
+ "learning_rate": 9.930187397020385e-06,
1948
+ "loss": 10.3482,
1949
+ "step": 277
1950
+ },
1951
+ {
1952
+ "epoch": 1.112112112112112,
1953
+ "grad_norm": 10.576739311218262,
1954
+ "learning_rate": 9.860378196608549e-06,
1955
+ "loss": 11.1896,
1956
+ "step": 278
1957
+ },
1958
+ {
1959
+ "epoch": 1.1161161161161162,
1960
+ "grad_norm": 9.567973136901855,
1961
+ "learning_rate": 9.790575801166432e-06,
1962
+ "loss": 9.2047,
1963
+ "step": 279
1964
+ },
1965
+ {
1966
+ "epoch": 1.12012012012012,
1967
+ "grad_norm": 8.562458992004395,
1968
+ "learning_rate": 9.720783612764314e-06,
1969
+ "loss": 9.7766,
1970
+ "step": 280
1971
+ },
1972
+ {
1973
+ "epoch": 1.1241241241241242,
1974
+ "grad_norm": 6.866599082946777,
1975
+ "learning_rate": 9.651005032974994e-06,
1976
+ "loss": 6.4877,
1977
+ "step": 281
1978
+ },
1979
+ {
1980
+ "epoch": 1.1281281281281281,
1981
+ "grad_norm": 11.195268630981445,
1982
+ "learning_rate": 9.581243462708007e-06,
1983
+ "loss": 12.8461,
1984
+ "step": 282
1985
+ },
1986
+ {
1987
+ "epoch": 1.132132132132132,
1988
+ "grad_norm": 9.039972305297852,
1989
+ "learning_rate": 9.511502302043867e-06,
1990
+ "loss": 9.8945,
1991
+ "step": 283
1992
+ },
1993
+ {
1994
+ "epoch": 1.1361361361361362,
1995
+ "grad_norm": 7.815824508666992,
1996
+ "learning_rate": 9.441784950068362e-06,
1997
+ "loss": 8.5909,
1998
+ "step": 284
1999
+ },
2000
+ {
2001
+ "epoch": 1.14014014014014,
2002
+ "grad_norm": 9.81783390045166,
2003
+ "learning_rate": 9.372094804706867e-06,
2004
+ "loss": 7.9823,
2005
+ "step": 285
2006
+ },
2007
+ {
2008
+ "epoch": 1.1441441441441442,
2009
+ "grad_norm": 11.14215087890625,
2010
+ "learning_rate": 9.302435262558748e-06,
2011
+ "loss": 9.8209,
2012
+ "step": 286
2013
+ },
2014
+ {
2015
+ "epoch": 1.1481481481481481,
2016
+ "grad_norm": 10.744006156921387,
2017
+ "learning_rate": 9.232809718731815e-06,
2018
+ "loss": 10.4147,
2019
+ "step": 287
2020
+ },
2021
+ {
2022
+ "epoch": 1.1521521521521523,
2023
+ "grad_norm": 10.524288177490234,
2024
+ "learning_rate": 9.163221566676847e-06,
2025
+ "loss": 10.7198,
2026
+ "step": 288
2027
+ },
2028
+ {
2029
+ "epoch": 1.1561561561561562,
2030
+ "grad_norm": 6.9448370933532715,
2031
+ "learning_rate": 9.093674198022201e-06,
2032
+ "loss": 7.5982,
2033
+ "step": 289
2034
+ },
2035
+ {
2036
+ "epoch": 1.16016016016016,
2037
+ "grad_norm": 10.453683853149414,
2038
+ "learning_rate": 9.024171002408507e-06,
2039
+ "loss": 12.5759,
2040
+ "step": 290
2041
+ },
2042
+ {
2043
+ "epoch": 1.1641641641641642,
2044
+ "grad_norm": 9.898215293884277,
2045
+ "learning_rate": 8.954715367323468e-06,
2046
+ "loss": 11.4182,
2047
+ "step": 291
2048
+ },
2049
+ {
2050
+ "epoch": 1.1681681681681682,
2051
+ "grad_norm": 6.9421772956848145,
2052
+ "learning_rate": 8.885310677936746e-06,
2053
+ "loss": 6.5644,
2054
+ "step": 292
2055
+ },
2056
+ {
2057
+ "epoch": 1.1721721721721723,
2058
+ "grad_norm": 10.001751899719238,
2059
+ "learning_rate": 8.815960316934991e-06,
2060
+ "loss": 9.7358,
2061
+ "step": 293
2062
+ },
2063
+ {
2064
+ "epoch": 1.1761761761761762,
2065
+ "grad_norm": 10.748393058776855,
2066
+ "learning_rate": 8.746667664356957e-06,
2067
+ "loss": 11.6022,
2068
+ "step": 294
2069
+ },
2070
+ {
2071
+ "epoch": 1.1801801801801801,
2072
+ "grad_norm": 6.848690509796143,
2073
+ "learning_rate": 8.677436097428775e-06,
2074
+ "loss": 9.815,
2075
+ "step": 295
2076
+ },
2077
+ {
2078
+ "epoch": 1.1841841841841843,
2079
+ "grad_norm": 9.581942558288574,
2080
+ "learning_rate": 8.60826899039935e-06,
2081
+ "loss": 11.3877,
2082
+ "step": 296
2083
+ },
2084
+ {
2085
+ "epoch": 1.1881881881881882,
2086
+ "grad_norm": 7.306396007537842,
2087
+ "learning_rate": 8.539169714375885e-06,
2088
+ "loss": 9.4341,
2089
+ "step": 297
2090
+ },
2091
+ {
2092
+ "epoch": 1.1921921921921923,
2093
+ "grad_norm": 5.874427318572998,
2094
+ "learning_rate": 8.47014163715962e-06,
2095
+ "loss": 8.0625,
2096
+ "step": 298
2097
+ },
2098
+ {
2099
+ "epoch": 1.1961961961961962,
2100
+ "grad_norm": 11.7793550491333,
2101
+ "learning_rate": 8.401188123081653e-06,
2102
+ "loss": 12.9606,
2103
+ "step": 299
2104
+ },
2105
+ {
2106
+ "epoch": 1.2002002002002001,
2107
+ "grad_norm": 8.709701538085938,
2108
+ "learning_rate": 8.332312532838978e-06,
2109
+ "loss": 10.2525,
2110
+ "step": 300
2111
+ },
2112
+ {
2113
+ "epoch": 1.2042042042042043,
2114
+ "grad_norm": 5.573022365570068,
2115
+ "learning_rate": 8.263518223330698e-06,
2116
+ "loss": 6.523,
2117
+ "step": 301
2118
+ },
2119
+ {
2120
+ "epoch": 1.2082082082082082,
2121
+ "grad_norm": 13.79970932006836,
2122
+ "learning_rate": 8.194808547494401e-06,
2123
+ "loss": 8.5468,
2124
+ "step": 302
2125
+ },
2126
+ {
2127
+ "epoch": 1.2122122122122123,
2128
+ "grad_norm": 9.682428359985352,
2129
+ "learning_rate": 8.126186854142752e-06,
2130
+ "loss": 10.4565,
2131
+ "step": 303
2132
+ },
2133
+ {
2134
+ "epoch": 1.2162162162162162,
2135
+ "grad_norm": 8.019155502319336,
2136
+ "learning_rate": 8.057656487800283e-06,
2137
+ "loss": 7.2856,
2138
+ "step": 304
2139
+ },
2140
+ {
2141
+ "epoch": 1.2202202202202201,
2142
+ "grad_norm": 6.491253852844238,
2143
+ "learning_rate": 7.989220788540356e-06,
2144
+ "loss": 7.5846,
2145
+ "step": 305
2146
+ },
2147
+ {
2148
+ "epoch": 1.2242242242242243,
2149
+ "grad_norm": 11.169169425964355,
2150
+ "learning_rate": 7.92088309182241e-06,
2151
+ "loss": 11.8709,
2152
+ "step": 306
2153
+ },
2154
+ {
2155
+ "epoch": 1.2282282282282282,
2156
+ "grad_norm": 10.235918045043945,
2157
+ "learning_rate": 7.852646728329368e-06,
2158
+ "loss": 11.7238,
2159
+ "step": 307
2160
+ },
2161
+ {
2162
+ "epoch": 1.2322322322322323,
2163
+ "grad_norm": 9.817302703857422,
2164
+ "learning_rate": 7.784515023805328e-06,
2165
+ "loss": 10.9515,
2166
+ "step": 308
2167
+ },
2168
+ {
2169
+ "epoch": 1.2362362362362362,
2170
+ "grad_norm": 6.883071422576904,
2171
+ "learning_rate": 7.716491298893443e-06,
2172
+ "loss": 6.5977,
2173
+ "step": 309
2174
+ },
2175
+ {
2176
+ "epoch": 1.2402402402402402,
2177
+ "grad_norm": 5.05192756652832,
2178
+ "learning_rate": 7.6485788689741e-06,
2179
+ "loss": 5.8135,
2180
+ "step": 310
2181
+ },
2182
+ {
2183
+ "epoch": 1.2442442442442443,
2184
+ "grad_norm": 6.485847473144531,
2185
+ "learning_rate": 7.580781044003324e-06,
2186
+ "loss": 6.6049,
2187
+ "step": 311
2188
+ },
2189
+ {
2190
+ "epoch": 1.2482482482482482,
2191
+ "grad_norm": 13.060445785522461,
2192
+ "learning_rate": 7.513101128351454e-06,
2193
+ "loss": 13.6744,
2194
+ "step": 312
2195
+ },
2196
+ {
2197
+ "epoch": 1.2522522522522523,
2198
+ "grad_norm": 9.57452392578125,
2199
+ "learning_rate": 7.445542420642097e-06,
2200
+ "loss": 9.607,
2201
+ "step": 313
2202
+ },
2203
+ {
2204
+ "epoch": 1.2562562562562563,
2205
+ "grad_norm": 7.936678886413574,
2206
+ "learning_rate": 7.378108213591355e-06,
2207
+ "loss": 7.526,
2208
+ "step": 314
2209
+ },
2210
+ {
2211
+ "epoch": 1.2602602602602602,
2212
+ "grad_norm": 8.851930618286133,
2213
+ "learning_rate": 7.310801793847344e-06,
2214
+ "loss": 8.7499,
2215
+ "step": 315
2216
+ },
2217
+ {
2218
+ "epoch": 1.2642642642642643,
2219
+ "grad_norm": 7.699076175689697,
2220
+ "learning_rate": 7.243626441830009e-06,
2221
+ "loss": 6.2849,
2222
+ "step": 316
2223
+ },
2224
+ {
2225
+ "epoch": 1.2682682682682682,
2226
+ "grad_norm": 10.221113204956055,
2227
+ "learning_rate": 7.176585431571235e-06,
2228
+ "loss": 9.9155,
2229
+ "step": 317
2230
+ },
2231
+ {
2232
+ "epoch": 1.2722722722722724,
2233
+ "grad_norm": 9.064275741577148,
2234
+ "learning_rate": 7.109682030555283e-06,
2235
+ "loss": 11.269,
2236
+ "step": 318
2237
+ },
2238
+ {
2239
+ "epoch": 1.2762762762762763,
2240
+ "grad_norm": 8.74006175994873,
2241
+ "learning_rate": 7.042919499559538e-06,
2242
+ "loss": 8.0676,
2243
+ "step": 319
2244
+ },
2245
+ {
2246
+ "epoch": 1.2802802802802802,
2247
+ "grad_norm": 6.647096633911133,
2248
+ "learning_rate": 6.976301092495556e-06,
2249
+ "loss": 6.7589,
2250
+ "step": 320
2251
+ },
2252
+ {
2253
+ "epoch": 1.2842842842842843,
2254
+ "grad_norm": 10.923158645629883,
2255
+ "learning_rate": 6.909830056250527e-06,
2256
+ "loss": 11.9449,
2257
+ "step": 321
2258
+ },
2259
+ {
2260
+ "epoch": 1.2882882882882882,
2261
+ "grad_norm": 12.557632446289062,
2262
+ "learning_rate": 6.843509630528977e-06,
2263
+ "loss": 11.5468,
2264
+ "step": 322
2265
+ },
2266
+ {
2267
+ "epoch": 1.2922922922922924,
2268
+ "grad_norm": 8.553438186645508,
2269
+ "learning_rate": 6.777343047694891e-06,
2270
+ "loss": 11.6,
2271
+ "step": 323
2272
+ },
2273
+ {
2274
+ "epoch": 1.2962962962962963,
2275
+ "grad_norm": 9.020846366882324,
2276
+ "learning_rate": 6.711333532614168e-06,
2277
+ "loss": 9.4851,
2278
+ "step": 324
2279
+ },
2280
+ {
2281
+ "epoch": 1.3003003003003002,
2282
+ "grad_norm": 9.486271858215332,
2283
+ "learning_rate": 6.645484302497452e-06,
2284
+ "loss": 7.9016,
2285
+ "step": 325
2286
+ },
2287
+ {
2288
+ "epoch": 1.3043043043043043,
2289
+ "grad_norm": 10.151366233825684,
2290
+ "learning_rate": 6.579798566743314e-06,
2291
+ "loss": 9.7636,
2292
+ "step": 326
2293
+ },
2294
+ {
2295
+ "epoch": 1.3083083083083082,
2296
+ "grad_norm": 7.098061561584473,
2297
+ "learning_rate": 6.5142795267818505e-06,
2298
+ "loss": 8.5963,
2299
+ "step": 327
2300
+ },
2301
+ {
2302
+ "epoch": 1.3123123123123124,
2303
+ "grad_norm": 10.057488441467285,
2304
+ "learning_rate": 6.448930375918632e-06,
2305
+ "loss": 10.6946,
2306
+ "step": 328
2307
+ },
2308
+ {
2309
+ "epoch": 1.3163163163163163,
2310
+ "grad_norm": 11.017156600952148,
2311
+ "learning_rate": 6.383754299179079e-06,
2312
+ "loss": 10.1259,
2313
+ "step": 329
2314
+ },
2315
+ {
2316
+ "epoch": 1.3203203203203202,
2317
+ "grad_norm": 9.51500415802002,
2318
+ "learning_rate": 6.318754473153221e-06,
2319
+ "loss": 10.976,
2320
+ "step": 330
2321
+ },
2322
+ {
2323
+ "epoch": 1.3243243243243243,
2324
+ "grad_norm": 10.10836410522461,
2325
+ "learning_rate": 6.25393406584088e-06,
2326
+ "loss": 11.5025,
2327
+ "step": 331
2328
+ },
2329
+ {
2330
+ "epoch": 1.3283283283283283,
2331
+ "grad_norm": 9.623117446899414,
2332
+ "learning_rate": 6.18929623649726e-06,
2333
+ "loss": 11.3624,
2334
+ "step": 332
2335
+ },
2336
+ {
2337
+ "epoch": 1.3323323323323324,
2338
+ "grad_norm": 9.546333312988281,
2339
+ "learning_rate": 6.124844135478971e-06,
2340
+ "loss": 11.5228,
2341
+ "step": 333
2342
+ },
2343
+ {
2344
+ "epoch": 1.3363363363363363,
2345
+ "grad_norm": 9.087567329406738,
2346
+ "learning_rate": 6.06058090409049e-06,
2347
+ "loss": 11.568,
2348
+ "step": 334
2349
+ },
2350
+ {
2351
+ "epoch": 1.3403403403403402,
2352
+ "grad_norm": 7.236462593078613,
2353
+ "learning_rate": 5.996509674431053e-06,
2354
+ "loss": 5.3294,
2355
+ "step": 335
2356
+ },
2357
+ {
2358
+ "epoch": 1.3443443443443444,
2359
+ "grad_norm": 8.476216316223145,
2360
+ "learning_rate": 5.932633569242e-06,
2361
+ "loss": 7.8108,
2362
+ "step": 336
2363
+ },
2364
+ {
2365
+ "epoch": 1.3483483483483483,
2366
+ "grad_norm": 8.417974472045898,
2367
+ "learning_rate": 5.868955701754584e-06,
2368
+ "loss": 6.7529,
2369
+ "step": 337
2370
+ },
2371
+ {
2372
+ "epoch": 1.3523523523523524,
2373
+ "grad_norm": 6.996325969696045,
2374
+ "learning_rate": 5.8054791755382286e-06,
2375
+ "loss": 8.3493,
2376
+ "step": 338
2377
+ },
2378
+ {
2379
+ "epoch": 1.3563563563563563,
2380
+ "grad_norm": 9.619551658630371,
2381
+ "learning_rate": 5.742207084349274e-06,
2382
+ "loss": 10.4045,
2383
+ "step": 339
2384
+ },
2385
+ {
2386
+ "epoch": 1.3603603603603602,
2387
+ "grad_norm": 10.322511672973633,
2388
+ "learning_rate": 5.679142511980176e-06,
2389
+ "loss": 13.1554,
2390
+ "step": 340
2391
+ },
2392
+ {
2393
+ "epoch": 1.3643643643643644,
2394
+ "grad_norm": 5.681710720062256,
2395
+ "learning_rate": 5.616288532109225e-06,
2396
+ "loss": 3.6891,
2397
+ "step": 341
2398
+ },
2399
+ {
2400
+ "epoch": 1.3683683683683685,
2401
+ "grad_norm": 10.557039260864258,
2402
+ "learning_rate": 5.553648208150728e-06,
2403
+ "loss": 10.9024,
2404
+ "step": 342
2405
+ },
2406
+ {
2407
+ "epoch": 1.3723723723723724,
2408
+ "grad_norm": 8.481843948364258,
2409
+ "learning_rate": 5.491224593105695e-06,
2410
+ "loss": 7.9044,
2411
+ "step": 343
2412
+ },
2413
+ {
2414
+ "epoch": 1.3763763763763763,
2415
+ "grad_norm": 6.216835975646973,
2416
+ "learning_rate": 5.429020729413062e-06,
2417
+ "loss": 8.3118,
2418
+ "step": 344
2419
+ },
2420
+ {
2421
+ "epoch": 1.3803803803803802,
2422
+ "grad_norm": 9.639850616455078,
2423
+ "learning_rate": 5.367039648801386e-06,
2424
+ "loss": 9.459,
2425
+ "step": 345
2426
+ },
2427
+ {
2428
+ "epoch": 1.3843843843843844,
2429
+ "grad_norm": 8.770609855651855,
2430
+ "learning_rate": 5.305284372141095e-06,
2431
+ "loss": 5.8499,
2432
+ "step": 346
2433
+ },
2434
+ {
2435
+ "epoch": 1.3883883883883885,
2436
+ "grad_norm": 10.659659385681152,
2437
+ "learning_rate": 5.243757909297247e-06,
2438
+ "loss": 11.1739,
2439
+ "step": 347
2440
+ },
2441
+ {
2442
+ "epoch": 1.3923923923923924,
2443
+ "grad_norm": 12.134528160095215,
2444
+ "learning_rate": 5.1824632589828465e-06,
2445
+ "loss": 12.6911,
2446
+ "step": 348
2447
+ },
2448
+ {
2449
+ "epoch": 1.3963963963963963,
2450
+ "grad_norm": 7.723568439483643,
2451
+ "learning_rate": 5.121403408612672e-06,
2452
+ "loss": 8.274,
2453
+ "step": 349
2454
+ },
2455
+ {
2456
+ "epoch": 1.4004004004004005,
2457
+ "grad_norm": 8.672338485717773,
2458
+ "learning_rate": 5.060581334157693e-06,
2459
+ "loss": 8.4151,
2460
+ "step": 350
2461
+ },
2462
+ {
2463
+ "epoch": 1.4044044044044044,
2464
+ "grad_norm": 7.8654303550720215,
2465
+ "learning_rate": 5.000000000000003e-06,
2466
+ "loss": 7.2695,
2467
+ "step": 351
2468
+ },
2469
+ {
2470
+ "epoch": 1.4084084084084085,
2471
+ "grad_norm": 9.311978340148926,
2472
+ "learning_rate": 4.939662358788364e-06,
2473
+ "loss": 8.9974,
2474
+ "step": 352
2475
+ },
2476
+ {
2477
+ "epoch": 1.4124124124124124,
2478
+ "grad_norm": 8.940287590026855,
2479
+ "learning_rate": 4.879571351294287e-06,
2480
+ "loss": 9.7042,
2481
+ "step": 353
2482
+ },
2483
+ {
2484
+ "epoch": 1.4164164164164164,
2485
+ "grad_norm": 7.047165870666504,
2486
+ "learning_rate": 4.8197299062687e-06,
2487
+ "loss": 8.4618,
2488
+ "step": 354
2489
+ },
2490
+ {
2491
+ "epoch": 1.4204204204204205,
2492
+ "grad_norm": 6.649487495422363,
2493
+ "learning_rate": 4.76014094029921e-06,
2494
+ "loss": 8.1405,
2495
+ "step": 355
2496
+ },
2497
+ {
2498
+ "epoch": 1.4244244244244244,
2499
+ "grad_norm": 6.8228840827941895,
2500
+ "learning_rate": 4.700807357667953e-06,
2501
+ "loss": 7.5051,
2502
+ "step": 356
2503
+ },
2504
+ {
2505
+ "epoch": 1.4284284284284285,
2506
+ "grad_norm": 6.64699649810791,
2507
+ "learning_rate": 4.641732050210032e-06,
2508
+ "loss": 7.9086,
2509
+ "step": 357
2510
+ },
2511
+ {
2512
+ "epoch": 1.4324324324324325,
2513
+ "grad_norm": 8.171652793884277,
2514
+ "learning_rate": 4.582917897172603e-06,
2515
+ "loss": 8.5123,
2516
+ "step": 358
2517
+ },
2518
+ {
2519
+ "epoch": 1.4364364364364364,
2520
+ "grad_norm": 8.410653114318848,
2521
+ "learning_rate": 4.524367765074499e-06,
2522
+ "loss": 7.1971,
2523
+ "step": 359
2524
+ },
2525
+ {
2526
+ "epoch": 1.4404404404404405,
2527
+ "grad_norm": 6.568076133728027,
2528
+ "learning_rate": 4.46608450756656e-06,
2529
+ "loss": 9.118,
2530
+ "step": 360
2531
+ },
2532
+ {
2533
+ "epoch": 1.4444444444444444,
2534
+ "grad_norm": 8.250543594360352,
2535
+ "learning_rate": 4.408070965292534e-06,
2536
+ "loss": 6.788,
2537
+ "step": 361
2538
+ },
2539
+ {
2540
+ "epoch": 1.4484484484484486,
2541
+ "grad_norm": 9.262479782104492,
2542
+ "learning_rate": 4.350329965750622e-06,
2543
+ "loss": 9.7014,
2544
+ "step": 362
2545
+ },
2546
+ {
2547
+ "epoch": 1.4524524524524525,
2548
+ "grad_norm": 8.645161628723145,
2549
+ "learning_rate": 4.292864323155684e-06,
2550
+ "loss": 6.8988,
2551
+ "step": 363
2552
+ },
2553
+ {
2554
+ "epoch": 1.4564564564564564,
2555
+ "grad_norm": 9.62594223022461,
2556
+ "learning_rate": 4.235676838302069e-06,
2557
+ "loss": 6.8996,
2558
+ "step": 364
2559
+ },
2560
+ {
2561
+ "epoch": 1.4604604604604605,
2562
+ "grad_norm": 9.494356155395508,
2563
+ "learning_rate": 4.178770298427107e-06,
2564
+ "loss": 11.2865,
2565
+ "step": 365
2566
+ },
2567
+ {
2568
+ "epoch": 1.4644644644644644,
2569
+ "grad_norm": 8.535139083862305,
2570
+ "learning_rate": 4.12214747707527e-06,
2571
+ "loss": 8.7372,
2572
+ "step": 366
2573
+ },
2574
+ {
2575
+ "epoch": 1.4684684684684686,
2576
+ "grad_norm": 8.055177688598633,
2577
+ "learning_rate": 4.065811133962987e-06,
2578
+ "loss": 7.4024,
2579
+ "step": 367
2580
+ },
2581
+ {
2582
+ "epoch": 1.4724724724724725,
2583
+ "grad_norm": 8.173779487609863,
2584
+ "learning_rate": 4.009764014844143e-06,
2585
+ "loss": 8.9576,
2586
+ "step": 368
2587
+ },
2588
+ {
2589
+ "epoch": 1.4764764764764764,
2590
+ "grad_norm": 9.618171691894531,
2591
+ "learning_rate": 3.954008851376252e-06,
2592
+ "loss": 10.6677,
2593
+ "step": 369
2594
+ },
2595
+ {
2596
+ "epoch": 1.4804804804804805,
2597
+ "grad_norm": 8.679189682006836,
2598
+ "learning_rate": 3.898548360987325e-06,
2599
+ "loss": 7.8925,
2600
+ "step": 370
2601
+ },
2602
+ {
2603
+ "epoch": 1.4844844844844844,
2604
+ "grad_norm": 7.810166835784912,
2605
+ "learning_rate": 3.8433852467434175e-06,
2606
+ "loss": 10.0909,
2607
+ "step": 371
2608
+ },
2609
+ {
2610
+ "epoch": 1.4884884884884886,
2611
+ "grad_norm": 8.003090858459473,
2612
+ "learning_rate": 3.7885221972168974e-06,
2613
+ "loss": 8.3157,
2614
+ "step": 372
2615
+ },
2616
+ {
2617
+ "epoch": 1.4924924924924925,
2618
+ "grad_norm": 8.543599128723145,
2619
+ "learning_rate": 3.7339618863553983e-06,
2620
+ "loss": 11.6619,
2621
+ "step": 373
2622
+ },
2623
+ {
2624
+ "epoch": 1.4964964964964964,
2625
+ "grad_norm": 8.370789527893066,
2626
+ "learning_rate": 3.679706973351491e-06,
2627
+ "loss": 9.0243,
2628
+ "step": 374
2629
+ },
2630
+ {
2631
+ "epoch": 1.5005005005005005,
2632
+ "grad_norm": 8.351577758789062,
2633
+ "learning_rate": 3.625760102513103e-06,
2634
+ "loss": 9.7003,
2635
+ "step": 375
2636
+ },
2637
+ {
2638
+ "epoch": 1.5045045045045045,
2639
+ "grad_norm": 7.890990734100342,
2640
+ "learning_rate": 3.5721239031346067e-06,
2641
+ "loss": 9.2858,
2642
+ "step": 376
2643
+ },
2644
+ {
2645
+ "epoch": 1.5085085085085086,
2646
+ "grad_norm": 9.778692245483398,
2647
+ "learning_rate": 3.5188009893686916e-06,
2648
+ "loss": 11.4346,
2649
+ "step": 377
2650
+ },
2651
+ {
2652
+ "epoch": 1.5125125125125125,
2653
+ "grad_norm": 9.475290298461914,
2654
+ "learning_rate": 3.4657939600989453e-06,
2655
+ "loss": 10.973,
2656
+ "step": 378
2657
+ },
2658
+ {
2659
+ "epoch": 1.5165165165165164,
2660
+ "grad_norm": 10.027473449707031,
2661
+ "learning_rate": 3.4131053988131947e-06,
2662
+ "loss": 12.3965,
2663
+ "step": 379
2664
+ },
2665
+ {
2666
+ "epoch": 1.5205205205205206,
2667
+ "grad_norm": 9.450143814086914,
2668
+ "learning_rate": 3.360737873477584e-06,
2669
+ "loss": 9.4957,
2670
+ "step": 380
2671
+ },
2672
+ {
2673
+ "epoch": 1.5245245245245245,
2674
+ "grad_norm": 10.000890731811523,
2675
+ "learning_rate": 3.308693936411421e-06,
2676
+ "loss": 11.4026,
2677
+ "step": 381
2678
+ },
2679
+ {
2680
+ "epoch": 1.5285285285285286,
2681
+ "grad_norm": 8.858360290527344,
2682
+ "learning_rate": 3.2569761241627694e-06,
2683
+ "loss": 7.8786,
2684
+ "step": 382
2685
+ },
2686
+ {
2687
+ "epoch": 1.5325325325325325,
2688
+ "grad_norm": 7.490108966827393,
2689
+ "learning_rate": 3.2055869573848374e-06,
2690
+ "loss": 9.3472,
2691
+ "step": 383
2692
+ },
2693
+ {
2694
+ "epoch": 1.5365365365365364,
2695
+ "grad_norm": 9.823856353759766,
2696
+ "learning_rate": 3.1545289407131128e-06,
2697
+ "loss": 11.4032,
2698
+ "step": 384
2699
+ },
2700
+ {
2701
+ "epoch": 1.5405405405405406,
2702
+ "grad_norm": 8.761564254760742,
2703
+ "learning_rate": 3.103804562643302e-06,
2704
+ "loss": 8.7117,
2705
+ "step": 385
2706
+ },
2707
+ {
2708
+ "epoch": 1.5445445445445447,
2709
+ "grad_norm": 5.970858573913574,
2710
+ "learning_rate": 3.0534162954100264e-06,
2711
+ "loss": 6.9834,
2712
+ "step": 386
2713
+ },
2714
+ {
2715
+ "epoch": 1.5485485485485486,
2716
+ "grad_norm": 7.699877738952637,
2717
+ "learning_rate": 3.003366594866345e-06,
2718
+ "loss": 7.6222,
2719
+ "step": 387
2720
+ },
2721
+ {
2722
+ "epoch": 1.5525525525525525,
2723
+ "grad_norm": 9.768478393554688,
2724
+ "learning_rate": 2.953657900364053e-06,
2725
+ "loss": 10.7794,
2726
+ "step": 388
2727
+ },
2728
+ {
2729
+ "epoch": 1.5565565565565564,
2730
+ "grad_norm": 7.5719380378723145,
2731
+ "learning_rate": 2.9042926346347932e-06,
2732
+ "loss": 6.1284,
2733
+ "step": 389
2734
+ },
2735
+ {
2736
+ "epoch": 1.5605605605605606,
2737
+ "grad_norm": 9.800040245056152,
2738
+ "learning_rate": 2.855273203671969e-06,
2739
+ "loss": 10.5813,
2740
+ "step": 390
2741
+ },
2742
+ {
2743
+ "epoch": 1.5645645645645647,
2744
+ "grad_norm": 8.183945655822754,
2745
+ "learning_rate": 2.8066019966134907e-06,
2746
+ "loss": 7.0301,
2747
+ "step": 391
2748
+ },
2749
+ {
2750
+ "epoch": 1.5685685685685686,
2751
+ "grad_norm": 8.762014389038086,
2752
+ "learning_rate": 2.7582813856253276e-06,
2753
+ "loss": 10.3296,
2754
+ "step": 392
2755
+ },
2756
+ {
2757
+ "epoch": 1.5725725725725725,
2758
+ "grad_norm": 10.513341903686523,
2759
+ "learning_rate": 2.7103137257858867e-06,
2760
+ "loss": 10.5659,
2761
+ "step": 393
2762
+ },
2763
+ {
2764
+ "epoch": 1.5765765765765765,
2765
+ "grad_norm": 9.811934471130371,
2766
+ "learning_rate": 2.6627013549712355e-06,
2767
+ "loss": 12.2094,
2768
+ "step": 394
2769
+ },
2770
+ {
2771
+ "epoch": 1.5805805805805806,
2772
+ "grad_norm": 9.031659126281738,
2773
+ "learning_rate": 2.615446593741161e-06,
2774
+ "loss": 8.0043,
2775
+ "step": 395
2776
+ },
2777
+ {
2778
+ "epoch": 1.5845845845845847,
2779
+ "grad_norm": 7.04182243347168,
2780
+ "learning_rate": 2.5685517452260566e-06,
2781
+ "loss": 8.0887,
2782
+ "step": 396
2783
+ },
2784
+ {
2785
+ "epoch": 1.5885885885885886,
2786
+ "grad_norm": 10.211063385009766,
2787
+ "learning_rate": 2.522019095014683e-06,
2788
+ "loss": 12.3084,
2789
+ "step": 397
2790
+ },
2791
+ {
2792
+ "epoch": 1.5925925925925926,
2793
+ "grad_norm": 7.216320037841797,
2794
+ "learning_rate": 2.4758509110427576e-06,
2795
+ "loss": 7.1561,
2796
+ "step": 398
2797
+ },
2798
+ {
2799
+ "epoch": 1.5965965965965965,
2800
+ "grad_norm": 9.875468254089355,
2801
+ "learning_rate": 2.4300494434824373e-06,
2802
+ "loss": 11.823,
2803
+ "step": 399
2804
+ },
2805
+ {
2806
+ "epoch": 1.6006006006006006,
2807
+ "grad_norm": 9.655021667480469,
2808
+ "learning_rate": 2.3846169246326345e-06,
2809
+ "loss": 9.9917,
2810
+ "step": 400
2811
+ },
2812
+ {
2813
+ "epoch": 1.6046046046046047,
2814
+ "grad_norm": 7.658693313598633,
2815
+ "learning_rate": 2.339555568810221e-06,
2816
+ "loss": 7.3462,
2817
+ "step": 401
2818
+ },
2819
+ {
2820
+ "epoch": 1.6086086086086087,
2821
+ "grad_norm": 8.90809440612793,
2822
+ "learning_rate": 2.2948675722421086e-06,
2823
+ "loss": 9.3165,
2824
+ "step": 402
2825
+ },
2826
+ {
2827
+ "epoch": 1.6126126126126126,
2828
+ "grad_norm": 9.23449420928955,
2829
+ "learning_rate": 2.2505551129582047e-06,
2830
+ "loss": 11.721,
2831
+ "step": 403
2832
+ },
2833
+ {
2834
+ "epoch": 1.6166166166166165,
2835
+ "grad_norm": 7.537600040435791,
2836
+ "learning_rate": 2.206620350685257e-06,
2837
+ "loss": 10.1761,
2838
+ "step": 404
2839
+ },
2840
+ {
2841
+ "epoch": 1.6206206206206206,
2842
+ "grad_norm": 7.198468208312988,
2843
+ "learning_rate": 2.163065426741603e-06,
2844
+ "loss": 6.6432,
2845
+ "step": 405
2846
+ },
2847
+ {
2848
+ "epoch": 1.6246246246246248,
2849
+ "grad_norm": 9.12747859954834,
2850
+ "learning_rate": 2.119892463932781e-06,
2851
+ "loss": 8.6486,
2852
+ "step": 406
2853
+ },
2854
+ {
2855
+ "epoch": 1.6286286286286287,
2856
+ "grad_norm": 11.575620651245117,
2857
+ "learning_rate": 2.0771035664480944e-06,
2858
+ "loss": 11.8213,
2859
+ "step": 407
2860
+ },
2861
+ {
2862
+ "epoch": 1.6326326326326326,
2863
+ "grad_norm": 9.51784896850586,
2864
+ "learning_rate": 2.0347008197580376e-06,
2865
+ "loss": 9.1633,
2866
+ "step": 408
2867
+ },
2868
+ {
2869
+ "epoch": 1.6366366366366365,
2870
+ "grad_norm": 8.156274795532227,
2871
+ "learning_rate": 1.9926862905126663e-06,
2872
+ "loss": 7.8501,
2873
+ "step": 409
2874
+ },
2875
+ {
2876
+ "epoch": 1.6406406406406406,
2877
+ "grad_norm": 8.703547477722168,
2878
+ "learning_rate": 1.95106202644086e-06,
2879
+ "loss": 7.7957,
2880
+ "step": 410
2881
+ },
2882
+ {
2883
+ "epoch": 1.6446446446446448,
2884
+ "grad_norm": 8.195082664489746,
2885
+ "learning_rate": 1.9098300562505266e-06,
2886
+ "loss": 7.3787,
2887
+ "step": 411
2888
+ },
2889
+ {
2890
+ "epoch": 1.6486486486486487,
2891
+ "grad_norm": 8.791090965270996,
2892
+ "learning_rate": 1.8689923895297247e-06,
2893
+ "loss": 8.48,
2894
+ "step": 412
2895
+ },
2896
+ {
2897
+ "epoch": 1.6526526526526526,
2898
+ "grad_norm": 7.4949870109558105,
2899
+ "learning_rate": 1.8285510166487154e-06,
2900
+ "loss": 6.1946,
2901
+ "step": 413
2902
+ },
2903
+ {
2904
+ "epoch": 1.6566566566566565,
2905
+ "grad_norm": 8.243803024291992,
2906
+ "learning_rate": 1.7885079086629598e-06,
2907
+ "loss": 5.4843,
2908
+ "step": 414
2909
+ },
2910
+ {
2911
+ "epoch": 1.6606606606606606,
2912
+ "grad_norm": 6.237216949462891,
2913
+ "learning_rate": 1.7488650172170496e-06,
2914
+ "loss": 8.3262,
2915
+ "step": 415
2916
+ },
2917
+ {
2918
+ "epoch": 1.6646646646646648,
2919
+ "grad_norm": 7.631178855895996,
2920
+ "learning_rate": 1.709624274449584e-06,
2921
+ "loss": 6.0657,
2922
+ "step": 416
2923
+ },
2924
+ {
2925
+ "epoch": 1.6686686686686687,
2926
+ "grad_norm": 10.616560935974121,
2927
+ "learning_rate": 1.6707875928990059e-06,
2928
+ "loss": 11.2167,
2929
+ "step": 417
2930
+ },
2931
+ {
2932
+ "epoch": 1.6726726726726726,
2933
+ "grad_norm": 7.889260292053223,
2934
+ "learning_rate": 1.6323568654103838e-06,
2935
+ "loss": 7.768,
2936
+ "step": 418
2937
+ },
2938
+ {
2939
+ "epoch": 1.6766766766766765,
2940
+ "grad_norm": 7.279766082763672,
2941
+ "learning_rate": 1.5943339650431578e-06,
2942
+ "loss": 9.2693,
2943
+ "step": 419
2944
+ },
2945
+ {
2946
+ "epoch": 1.6806806806806807,
2947
+ "grad_norm": 8.203940391540527,
2948
+ "learning_rate": 1.5567207449798517e-06,
2949
+ "loss": 8.1825,
2950
+ "step": 420
2951
+ },
2952
+ {
2953
+ "epoch": 1.6846846846846848,
2954
+ "grad_norm": 7.227617263793945,
2955
+ "learning_rate": 1.5195190384357405e-06,
2956
+ "loss": 8.1531,
2957
+ "step": 421
2958
+ },
2959
+ {
2960
+ "epoch": 1.6886886886886887,
2961
+ "grad_norm": 8.23820972442627,
2962
+ "learning_rate": 1.4827306585695234e-06,
2963
+ "loss": 7.8877,
2964
+ "step": 422
2965
+ },
2966
+ {
2967
+ "epoch": 1.6926926926926926,
2968
+ "grad_norm": 9.185068130493164,
2969
+ "learning_rate": 1.446357398394934e-06,
2970
+ "loss": 9.8151,
2971
+ "step": 423
2972
+ },
2973
+ {
2974
+ "epoch": 1.6966966966966965,
2975
+ "grad_norm": 7.578010082244873,
2976
+ "learning_rate": 1.4104010306933558e-06,
2977
+ "loss": 8.0134,
2978
+ "step": 424
2979
+ },
2980
+ {
2981
+ "epoch": 1.7007007007007007,
2982
+ "grad_norm": 9.514433860778809,
2983
+ "learning_rate": 1.3748633079274254e-06,
2984
+ "loss": 10.7415,
2985
+ "step": 425
2986
+ },
2987
+ {
2988
+ "epoch": 1.7047047047047048,
2989
+ "grad_norm": 8.607891082763672,
2990
+ "learning_rate": 1.339745962155613e-06,
2991
+ "loss": 9.117,
2992
+ "step": 426
2993
+ },
2994
+ {
2995
+ "epoch": 1.7087087087087087,
2996
+ "grad_norm": 9.616765975952148,
2997
+ "learning_rate": 1.30505070494781e-06,
2998
+ "loss": 9.6697,
2999
+ "step": 427
3000
+ },
3001
+ {
3002
+ "epoch": 1.7127127127127126,
3003
+ "grad_norm": 6.407479286193848,
3004
+ "learning_rate": 1.2707792273019049e-06,
3005
+ "loss": 5.5711,
3006
+ "step": 428
3007
+ },
3008
+ {
3009
+ "epoch": 1.7167167167167166,
3010
+ "grad_norm": 7.340726375579834,
3011
+ "learning_rate": 1.2369331995613664e-06,
3012
+ "loss": 8.495,
3013
+ "step": 429
3014
+ },
3015
+ {
3016
+ "epoch": 1.7207207207207207,
3017
+ "grad_norm": 8.43381404876709,
3018
+ "learning_rate": 1.2035142713338366e-06,
3019
+ "loss": 7.7537,
3020
+ "step": 430
3021
+ },
3022
+ {
3023
+ "epoch": 1.7247247247247248,
3024
+ "grad_norm": 7.116838455200195,
3025
+ "learning_rate": 1.1705240714107301e-06,
3026
+ "loss": 7.4401,
3027
+ "step": 431
3028
+ },
3029
+ {
3030
+ "epoch": 1.7287287287287287,
3031
+ "grad_norm": 5.795578956604004,
3032
+ "learning_rate": 1.1379642076878528e-06,
3033
+ "loss": 7.5497,
3034
+ "step": 432
3035
+ },
3036
+ {
3037
+ "epoch": 1.7327327327327327,
3038
+ "grad_norm": 10.752535820007324,
3039
+ "learning_rate": 1.1058362670870248e-06,
3040
+ "loss": 10.7059,
3041
+ "step": 433
3042
+ },
3043
+ {
3044
+ "epoch": 1.7367367367367368,
3045
+ "grad_norm": 9.655488967895508,
3046
+ "learning_rate": 1.0741418154787443e-06,
3047
+ "loss": 8.5591,
3048
+ "step": 434
3049
+ },
3050
+ {
3051
+ "epoch": 1.7407407407407407,
3052
+ "grad_norm": 7.492973804473877,
3053
+ "learning_rate": 1.042882397605871e-06,
3054
+ "loss": 8.6837,
3055
+ "step": 435
3056
+ },
3057
+ {
3058
+ "epoch": 1.7447447447447448,
3059
+ "grad_norm": 9.218620300292969,
3060
+ "learning_rate": 1.012059537008332e-06,
3061
+ "loss": 9.744,
3062
+ "step": 436
3063
+ },
3064
+ {
3065
+ "epoch": 1.7487487487487487,
3066
+ "grad_norm": 8.91235637664795,
3067
+ "learning_rate": 9.816747359488632e-07,
3068
+ "loss": 8.5032,
3069
+ "step": 437
3070
+ },
3071
+ {
3072
+ "epoch": 1.7527527527527527,
3073
+ "grad_norm": 8.027016639709473,
3074
+ "learning_rate": 9.517294753398066e-07,
3075
+ "loss": 7.9958,
3076
+ "step": 438
3077
+ },
3078
+ {
3079
+ "epoch": 1.7567567567567568,
3080
+ "grad_norm": 7.357422828674316,
3081
+ "learning_rate": 9.222252146709143e-07,
3082
+ "loss": 8.4258,
3083
+ "step": 439
3084
+ },
3085
+ {
3086
+ "epoch": 1.7607607607607607,
3087
+ "grad_norm": 10.137444496154785,
3088
+ "learning_rate": 8.931633919382299e-07,
3089
+ "loss": 10.453,
3090
+ "step": 440
3091
+ },
3092
+ {
3093
+ "epoch": 1.7647647647647648,
3094
+ "grad_norm": 9.656020164489746,
3095
+ "learning_rate": 8.645454235739903e-07,
3096
+ "loss": 8.4509,
3097
+ "step": 441
3098
+ },
3099
+ {
3100
+ "epoch": 1.7687687687687688,
3101
+ "grad_norm": 8.812692642211914,
3102
+ "learning_rate": 8.363727043776037e-07,
3103
+ "loss": 10.0055,
3104
+ "step": 442
3105
+ },
3106
+ {
3107
+ "epoch": 1.7727727727727727,
3108
+ "grad_norm": 8.220873832702637,
3109
+ "learning_rate": 8.086466074476562e-07,
3110
+ "loss": 8.6968,
3111
+ "step": 443
3112
+ },
3113
+ {
3114
+ "epoch": 1.7767767767767768,
3115
+ "grad_norm": 8.308377265930176,
3116
+ "learning_rate": 7.81368484114996e-07,
3117
+ "loss": 8.6779,
3118
+ "step": 444
3119
+ },
3120
+ {
3121
+ "epoch": 1.7807807807807807,
3122
+ "grad_norm": 6.530799865722656,
3123
+ "learning_rate": 7.545396638768698e-07,
3124
+ "loss": 6.6627,
3125
+ "step": 445
3126
+ },
3127
+ {
3128
+ "epoch": 1.7847847847847849,
3129
+ "grad_norm": 9.846541404724121,
3130
+ "learning_rate": 7.281614543321269e-07,
3131
+ "loss": 10.6289,
3132
+ "step": 446
3133
+ },
3134
+ {
3135
+ "epoch": 1.7887887887887888,
3136
+ "grad_norm": 10.847795486450195,
3137
+ "learning_rate": 7.022351411174866e-07,
3138
+ "loss": 10.4342,
3139
+ "step": 447
3140
+ },
3141
+ {
3142
+ "epoch": 1.7927927927927927,
3143
+ "grad_norm": 10.292729377746582,
3144
+ "learning_rate": 6.767619878448783e-07,
3145
+ "loss": 9.7337,
3146
+ "step": 448
3147
+ },
3148
+ {
3149
+ "epoch": 1.7967967967967968,
3150
+ "grad_norm": 6.869995594024658,
3151
+ "learning_rate": 6.517432360398556e-07,
3152
+ "loss": 7.9103,
3153
+ "step": 449
3154
+ },
3155
+ {
3156
+ "epoch": 1.800800800800801,
3157
+ "grad_norm": 10.077024459838867,
3158
+ "learning_rate": 6.271801050810856e-07,
3159
+ "loss": 9.3392,
3160
+ "step": 450
3161
+ },
3162
+ {
3163
+ "epoch": 1.8048048048048049,
3164
+ "grad_norm": 10.323750495910645,
3165
+ "learning_rate": 6.030737921409169e-07,
3166
+ "loss": 9.5517,
3167
+ "step": 451
3168
+ },
3169
+ {
3170
+ "epoch": 1.8088088088088088,
3171
+ "grad_norm": 7.406447410583496,
3172
+ "learning_rate": 5.794254721270331e-07,
3173
+ "loss": 6.6254,
3174
+ "step": 452
3175
+ },
3176
+ {
3177
+ "epoch": 1.8128128128128127,
3178
+ "grad_norm": 8.898077964782715,
3179
+ "learning_rate": 5.562362976251901e-07,
3180
+ "loss": 9.9607,
3181
+ "step": 453
3182
+ },
3183
+ {
3184
+ "epoch": 1.8168168168168168,
3185
+ "grad_norm": 7.905872821807861,
3186
+ "learning_rate": 5.335073988430373e-07,
3187
+ "loss": 8.5673,
3188
+ "step": 454
3189
+ },
3190
+ {
3191
+ "epoch": 1.820820820820821,
3192
+ "grad_norm": 11.556071281433105,
3193
+ "learning_rate": 5.112398835550348e-07,
3194
+ "loss": 10.4564,
3195
+ "step": 455
3196
+ },
3197
+ {
3198
+ "epoch": 1.8248248248248249,
3199
+ "grad_norm": 8.942086219787598,
3200
+ "learning_rate": 4.894348370484648e-07,
3201
+ "loss": 10.8804,
3202
+ "step": 456
3203
+ },
3204
+ {
3205
+ "epoch": 1.8288288288288288,
3206
+ "grad_norm": 9.902061462402344,
3207
+ "learning_rate": 4.6809332207053083e-07,
3208
+ "loss": 9.289,
3209
+ "step": 457
3210
+ },
3211
+ {
3212
+ "epoch": 1.8328328328328327,
3213
+ "grad_norm": 9.8366060256958,
3214
+ "learning_rate": 4.4721637877656377e-07,
3215
+ "loss": 9.5864,
3216
+ "step": 458
3217
+ },
3218
+ {
3219
+ "epoch": 1.8368368368368369,
3220
+ "grad_norm": 8.741854667663574,
3221
+ "learning_rate": 4.268050246793276e-07,
3222
+ "loss": 11.5834,
3223
+ "step": 459
3224
+ },
3225
+ {
3226
+ "epoch": 1.840840840840841,
3227
+ "grad_norm": 10.229823112487793,
3228
+ "learning_rate": 4.068602545994249e-07,
3229
+ "loss": 11.6159,
3230
+ "step": 460
3231
+ },
3232
+ {
3233
+ "epoch": 1.844844844844845,
3234
+ "grad_norm": 7.879222869873047,
3235
+ "learning_rate": 3.8738304061681107e-07,
3236
+ "loss": 8.8435,
3237
+ "step": 461
3238
+ },
3239
+ {
3240
+ "epoch": 1.8488488488488488,
3241
+ "grad_norm": 8.602821350097656,
3242
+ "learning_rate": 3.68374332023419e-07,
3243
+ "loss": 7.9848,
3244
+ "step": 462
3245
+ },
3246
+ {
3247
+ "epoch": 1.8528528528528527,
3248
+ "grad_norm": 9.40288257598877,
3249
+ "learning_rate": 3.498350552768859e-07,
3250
+ "loss": 7.7615,
3251
+ "step": 463
3252
+ },
3253
+ {
3254
+ "epoch": 1.8568568568568569,
3255
+ "grad_norm": 9.402653694152832,
3256
+ "learning_rate": 3.3176611395540625e-07,
3257
+ "loss": 8.1317,
3258
+ "step": 464
3259
+ },
3260
+ {
3261
+ "epoch": 1.860860860860861,
3262
+ "grad_norm": 7.863263130187988,
3263
+ "learning_rate": 3.1416838871368925e-07,
3264
+ "loss": 10.661,
3265
+ "step": 465
3266
+ },
3267
+ {
3268
+ "epoch": 1.864864864864865,
3269
+ "grad_norm": 11.480242729187012,
3270
+ "learning_rate": 2.970427372400353e-07,
3271
+ "loss": 8.2887,
3272
+ "step": 466
3273
+ },
3274
+ {
3275
+ "epoch": 1.8688688688688688,
3276
+ "grad_norm": 8.683096885681152,
3277
+ "learning_rate": 2.8038999421453827e-07,
3278
+ "loss": 9.6233,
3279
+ "step": 467
3280
+ },
3281
+ {
3282
+ "epoch": 1.8728728728728727,
3283
+ "grad_norm": 7.530503273010254,
3284
+ "learning_rate": 2.6421097126839714e-07,
3285
+ "loss": 7.0061,
3286
+ "step": 468
3287
+ },
3288
+ {
3289
+ "epoch": 1.8768768768768769,
3290
+ "grad_norm": 11.571393013000488,
3291
+ "learning_rate": 2.4850645694436736e-07,
3292
+ "loss": 10.4153,
3293
+ "step": 469
3294
+ },
3295
+ {
3296
+ "epoch": 1.880880880880881,
3297
+ "grad_norm": 9.118682861328125,
3298
+ "learning_rate": 2.332772166583208e-07,
3299
+ "loss": 8.9374,
3300
+ "step": 470
3301
+ },
3302
+ {
3303
+ "epoch": 1.884884884884885,
3304
+ "grad_norm": 10.62685489654541,
3305
+ "learning_rate": 2.1852399266194312e-07,
3306
+ "loss": 11.9423,
3307
+ "step": 471
3308
+ },
3309
+ {
3310
+ "epoch": 1.8888888888888888,
3311
+ "grad_norm": 10.59912395477295,
3312
+ "learning_rate": 2.0424750400655947e-07,
3313
+ "loss": 13.7211,
3314
+ "step": 472
3315
+ },
3316
+ {
3317
+ "epoch": 1.8928928928928928,
3318
+ "grad_norm": 9.825209617614746,
3319
+ "learning_rate": 1.9044844650808468e-07,
3320
+ "loss": 12.1137,
3321
+ "step": 473
3322
+ },
3323
+ {
3324
+ "epoch": 1.896896896896897,
3325
+ "grad_norm": 8.697061538696289,
3326
+ "learning_rate": 1.7712749271311392e-07,
3327
+ "loss": 10.0648,
3328
+ "step": 474
3329
+ },
3330
+ {
3331
+ "epoch": 1.900900900900901,
3332
+ "grad_norm": 11.070192337036133,
3333
+ "learning_rate": 1.6428529186614195e-07,
3334
+ "loss": 10.2718,
3335
+ "step": 475
3336
+ },
3337
+ {
3338
+ "epoch": 1.904904904904905,
3339
+ "grad_norm": 6.94769287109375,
3340
+ "learning_rate": 1.519224698779198e-07,
3341
+ "loss": 5.7507,
3342
+ "step": 476
3343
+ },
3344
+ {
3345
+ "epoch": 1.9089089089089089,
3346
+ "grad_norm": 9.053165435791016,
3347
+ "learning_rate": 1.400396292949513e-07,
3348
+ "loss": 8.6296,
3349
+ "step": 477
3350
+ },
3351
+ {
3352
+ "epoch": 1.9129129129129128,
3353
+ "grad_norm": 7.860747337341309,
3354
+ "learning_rate": 1.2863734927012094e-07,
3355
+ "loss": 7.3664,
3356
+ "step": 478
3357
+ },
3358
+ {
3359
+ "epoch": 1.916916916916917,
3360
+ "grad_norm": 10.029607772827148,
3361
+ "learning_rate": 1.1771618553447217e-07,
3362
+ "loss": 9.7694,
3363
+ "step": 479
3364
+ },
3365
+ {
3366
+ "epoch": 1.920920920920921,
3367
+ "grad_norm": 9.870777130126953,
3368
+ "learning_rate": 1.0727667037011668e-07,
3369
+ "loss": 8.9218,
3370
+ "step": 480
3371
+ },
3372
+ {
3373
+ "epoch": 1.924924924924925,
3374
+ "grad_norm": 9.642688751220703,
3375
+ "learning_rate": 9.731931258429638e-08,
3376
+ "loss": 11.7123,
3377
+ "step": 481
3378
+ },
3379
+ {
3380
+ "epoch": 1.9289289289289289,
3381
+ "grad_norm": 8.105547904968262,
3382
+ "learning_rate": 8.784459748458318e-08,
3383
+ "loss": 5.0593,
3384
+ "step": 482
3385
+ },
3386
+ {
3387
+ "epoch": 1.9329329329329328,
3388
+ "grad_norm": 10.861283302307129,
3389
+ "learning_rate": 7.885298685522235e-08,
3390
+ "loss": 9.342,
3391
+ "step": 483
3392
+ },
3393
+ {
3394
+ "epoch": 1.936936936936937,
3395
+ "grad_norm": 9.56164264678955,
3396
+ "learning_rate": 7.034491893463059e-08,
3397
+ "loss": 9.4425,
3398
+ "step": 484
3399
+ },
3400
+ {
3401
+ "epoch": 1.940940940940941,
3402
+ "grad_norm": 9.342721939086914,
3403
+ "learning_rate": 6.232080839403631e-08,
3404
+ "loss": 9.2435,
3405
+ "step": 485
3406
+ },
3407
+ {
3408
+ "epoch": 1.944944944944945,
3409
+ "grad_norm": 8.120580673217773,
3410
+ "learning_rate": 5.4781046317267103e-08,
3411
+ "loss": 8.4095,
3412
+ "step": 486
3413
+ },
3414
+ {
3415
+ "epoch": 1.9489489489489489,
3416
+ "grad_norm": 8.107100486755371,
3417
+ "learning_rate": 4.772600018168816e-08,
3418
+ "loss": 8.3244,
3419
+ "step": 487
3420
+ },
3421
+ {
3422
+ "epoch": 1.9529529529529528,
3423
+ "grad_norm": 11.232532501220703,
3424
+ "learning_rate": 4.115601384029666e-08,
3425
+ "loss": 9.7187,
3426
+ "step": 488
3427
+ },
3428
+ {
3429
+ "epoch": 1.956956956956957,
3430
+ "grad_norm": 10.383572578430176,
3431
+ "learning_rate": 3.50714075049563e-08,
3432
+ "loss": 11.3603,
3433
+ "step": 489
3434
+ },
3435
+ {
3436
+ "epoch": 1.960960960960961,
3437
+ "grad_norm": 6.417677402496338,
3438
+ "learning_rate": 2.947247773079753e-08,
3439
+ "loss": 7.2631,
3440
+ "step": 490
3441
+ },
3442
+ {
3443
+ "epoch": 1.964964964964965,
3444
+ "grad_norm": 9.289351463317871,
3445
+ "learning_rate": 2.4359497401758026e-08,
3446
+ "loss": 7.9686,
3447
+ "step": 491
3448
+ },
3449
+ {
3450
+ "epoch": 1.968968968968969,
3451
+ "grad_norm": 6.331518173217773,
3452
+ "learning_rate": 1.973271571728441e-08,
3453
+ "loss": 6.2156,
3454
+ "step": 492
3455
+ },
3456
+ {
3457
+ "epoch": 1.972972972972973,
3458
+ "grad_norm": 9.099044799804688,
3459
+ "learning_rate": 1.5592358180189782e-08,
3460
+ "loss": 11.0908,
3461
+ "step": 493
3462
+ },
3463
+ {
3464
+ "epoch": 1.976976976976977,
3465
+ "grad_norm": 7.2669243812561035,
3466
+ "learning_rate": 1.1938626585660252e-08,
3467
+ "loss": 7.5373,
3468
+ "step": 494
3469
+ },
3470
+ {
3471
+ "epoch": 1.980980980980981,
3472
+ "grad_norm": 8.350067138671875,
3473
+ "learning_rate": 8.771699011416169e-09,
3474
+ "loss": 8.9676,
3475
+ "step": 495
3476
+ },
3477
+ {
3478
+ "epoch": 1.984984984984985,
3479
+ "grad_norm": 8.300844192504883,
3480
+ "learning_rate": 6.091729809042379e-09,
3481
+ "loss": 9.1963,
3482
+ "step": 496
3483
+ },
3484
+ {
3485
+ "epoch": 1.988988988988989,
3486
+ "grad_norm": 8.056188583374023,
3487
+ "learning_rate": 3.898849596456477e-09,
3488
+ "loss": 8.1463,
3489
+ "step": 497
3490
+ },
3491
+ {
3492
+ "epoch": 1.992992992992993,
3493
+ "grad_norm": 11.183734893798828,
3494
+ "learning_rate": 2.193165251545004e-09,
3495
+ "loss": 12.0939,
3496
+ "step": 498
3497
+ },
3498
+ {
3499
+ "epoch": 1.996996996996997,
3500
+ "grad_norm": 8.850930213928223,
3501
+ "learning_rate": 9.74759906957612e-10,
3502
+ "loss": 8.3097,
3503
+ "step": 499
3504
+ },
3505
+ {
3506
+ "epoch": 2.0,
3507
+ "grad_norm": 6.024604320526123,
3508
+ "learning_rate": 2.436929460525317e-10,
3509
+ "loss": 7.6412,
3510
+ "step": 500
3511
+ }
3512
+ ],
3513
+ "logging_steps": 1,
3514
+ "max_steps": 500,
3515
+ "num_input_tokens_seen": 0,
3516
+ "num_train_epochs": 2,
3517
+ "save_steps": 100,
3518
+ "stateful_callbacks": {
3519
+ "TrainerControl": {
3520
+ "args": {
3521
+ "should_epoch_stop": false,
3522
+ "should_evaluate": false,
3523
+ "should_log": false,
3524
+ "should_save": true,
3525
+ "should_training_stop": true
3526
+ },
3527
+ "attributes": {}
3528
+ }
3529
+ },
3530
+ "total_flos": 1.4982206597575475e+17,
3531
+ "train_batch_size": 1,
3532
+ "trial_name": null,
3533
+ "trial_params": null
3534
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/checkpoint-500/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "TLM"
4
+ ],
5
+ "freeze_ts_model": true,
6
+ "gradient_checkpointing": true,
7
+ "llm_attn_implementation": null,
8
+ "llm_model_path": "/mnt/share01/sqk/models/qwen2.5-7b-instruct",
9
+ "llm_torch_dtype": "bfloat16",
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "lora_r": 16,
13
+ "lora_target_modules": [
14
+ "q_proj",
15
+ "k_proj",
16
+ "v_proj",
17
+ "o_proj",
18
+ "gate_proj",
19
+ "up_proj",
20
+ "down_proj"
21
+ ],
22
+ "model_type": "vlm_model",
23
+ "torch_dtype": "bfloat16",
24
+ "transformers_version": "4.54.1",
25
+ "ts_pad_num": 25,
26
+ "use_lora": true
27
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.54.1"
4
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/skipped_overlength_train.jsonl ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {"id": "forecasting_011619_56748ac9", "line_number": 4058, "source_type": "forecasting", "input_tokens": 4200, "instruction_tokens": 4153, "max_seq_length": 4096}
2
+ {"id": "forecasting_031782_0bf68658", "line_number": 4074, "source_type": "forecasting", "input_tokens": 6114, "instruction_tokens": 6067, "max_seq_length": 4096}
3
+ {"id": "forecasting_037187_4a95592f", "line_number": 4234, "source_type": "forecasting", "input_tokens": 4438, "instruction_tokens": 4390, "max_seq_length": 4096}
4
+ {"id": "forecasting_036903_44ae43fa", "line_number": 4325, "source_type": "forecasting", "input_tokens": 5016, "instruction_tokens": 4968, "max_seq_length": 4096}
5
+ {"id": "forecasting_004128_65a4fddf", "line_number": 4824, "source_type": "forecasting", "input_tokens": 5871, "instruction_tokens": 5824, "max_seq_length": 4096}
6
+ {"id": "forecasting_027194_66448c4b", "line_number": 5817, "source_type": "forecasting", "input_tokens": 4616, "instruction_tokens": 4568, "max_seq_length": 4096}
7
+ {"id": "forecasting_008995_b0cbd927", "line_number": 5886, "source_type": "forecasting", "input_tokens": 4431, "instruction_tokens": 4384, "max_seq_length": 4096}
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/tokenizer_config.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "clean_up_tokenization_spaces": false,
199
+ "eos_token": "<|im_end|>",
200
+ "errors": "replace",
201
+ "extra_special_tokens": {},
202
+ "model_max_length": 131072,
203
+ "pad_token": "<|endoftext|>",
204
+ "split_special_tokens": false,
205
+ "tokenizer_class": "Qwen2Tokenizer",
206
+ "unk_token": null
207
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/tsqa_run_config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train_path": "/mnt/share01/sqk/datasets/Time-MQA_TSQA/tmp/train.jsonl",
3
+ "eval_path": "/mnt/share01/sqk/datasets/Time-MQA_TSQA/tmp/eval.jsonl",
4
+ "llm_model_path": "/mnt/share01/sqk/models/qwen2.5-7b-instruct",
5
+ "load_ts_encoder": "/mnt/share01/sqk/ITFormer/save/pretrain/model.safetensors",
6
+ "output_dir": "/mnt/share01/sqk/ITFormer/runs/itformer_tsqa_full_20260615_164812/checkpoint_final",
7
+ "max_train_samples": 0,
8
+ "max_eval_samples": 0,
9
+ "max_seq_length": 4096,
10
+ "seed": 42,
11
+ "model": "TimeSeriesEncoder",
12
+ "d_model": 512,
13
+ "n_heads": 8,
14
+ "e_layers": 4,
15
+ "patch_len": 60,
16
+ "stride": 60,
17
+ "input_len": 600,
18
+ "dropout": 0.1,
19
+ "freeze_ts_model": true,
20
+ "pretrain": false,
21
+ "min_mask_ratio": 0.7,
22
+ "max_mask_ratio": 0.8,
23
+ "it_d_model": 3584,
24
+ "it_n_heads": 28,
25
+ "it_layers": 2,
26
+ "it_dropout": 0.1,
27
+ "prefix_num": 25,
28
+ "adapter_type": "itformer",
29
+ "llm_attn_implementation": null,
30
+ "llm_torch_dtype": "bfloat16",
31
+ "use_lora": true,
32
+ "lora_r": 16,
33
+ "lora_alpha": 32,
34
+ "lora_dropout": 0.05,
35
+ "lora_target_modules": [
36
+ "q_proj",
37
+ "k_proj",
38
+ "v_proj",
39
+ "o_proj",
40
+ "gate_proj",
41
+ "up_proj",
42
+ "down_proj"
43
+ ],
44
+ "gradient_checkpointing": true,
45
+ "do_train": true,
46
+ "per_device_train_batch_size": 1,
47
+ "per_device_eval_batch_size": 1,
48
+ "gradient_accumulation_steps": 16,
49
+ "learning_rate": 2e-05,
50
+ "num_train_epochs": 2.0,
51
+ "weight_decay": 1e-06,
52
+ "fp16": false,
53
+ "bf16": true,
54
+ "dataloader_pin_memory": true,
55
+ "dataloader_num_workers": 4,
56
+ "save_steps": 100,
57
+ "save_total_limit": 2,
58
+ "logging_steps": 1,
59
+ "eval_steps": 999999999,
60
+ "eval_stragy": "no",
61
+ "shuffle": true,
62
+ "report_to": "none",
63
+ "mode": "train",
64
+ "ts_pad_num": 25,
65
+ "llm_d_model": 3584,
66
+ "actual_train_samples": 7993,
67
+ "actual_eval_samples": 800,
68
+ "skipped_overlength_train_samples": 7
69
+ }
runs/itformer_tsqa_full_20260615_164812/checkpoint_final/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_164812/eval/metrics.json ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "by_group": {
3
+ "anomaly_detection": {
4
+ "count": 200,
5
+ "accuracy": 0.52,
6
+ "correct": 104,
7
+ "parsed": 200
8
+ },
9
+ "classification": {
10
+ "count": 200,
11
+ "accuracy": 0.68,
12
+ "correct": 136,
13
+ "parsed": 200
14
+ },
15
+ "forecasting": {
16
+ "count": 200,
17
+ "valid_samples": 194,
18
+ "valid_points": 3815,
19
+ "mse": 11348803.821745949,
20
+ "mae": 711.87058641765
21
+ },
22
+ "open_ended": {
23
+ "count": 200,
24
+ "accuracy": 0.46,
25
+ "parsed_accuracy": 0.467005076142132,
26
+ "parse_rate": 0.985,
27
+ "correct": 92,
28
+ "parsed": 197,
29
+ "unparsed": 3,
30
+ "by_format": {
31
+ "multiple_choice": {
32
+ "count": 67,
33
+ "accuracy": 0.47761194029850745,
34
+ "parsed_accuracy": 0.48484848484848486,
35
+ "correct": 32,
36
+ "parsed": 66,
37
+ "unparsed": 1
38
+ },
39
+ "open_ended_question": {
40
+ "count": 67,
41
+ "accuracy": 0.26865671641791045,
42
+ "parsed_accuracy": 0.27692307692307694,
43
+ "correct": 18,
44
+ "parsed": 65,
45
+ "unparsed": 2
46
+ },
47
+ "true_false": {
48
+ "count": 66,
49
+ "accuracy": 0.6363636363636364,
50
+ "parsed_accuracy": 0.6363636363636364,
51
+ "correct": 42,
52
+ "parsed": 66,
53
+ "unparsed": 0
54
+ }
55
+ },
56
+ "by_method": {
57
+ "anomaly": {
58
+ "count": 2,
59
+ "accuracy": 1.0,
60
+ "correct": 2
61
+ },
62
+ "cyclical": {
63
+ "count": 3,
64
+ "accuracy": 0.6666666666666666,
65
+ "correct": 2
66
+ },
67
+ "multiple_choice": {
68
+ "count": 64,
69
+ "accuracy": 0.5,
70
+ "correct": 32
71
+ },
72
+ "numeric_scalar": {
73
+ "count": 30,
74
+ "accuracy": 0.0,
75
+ "correct": 0
76
+ },
77
+ "numeric_sequence": {
78
+ "count": 1,
79
+ "accuracy": 0.0,
80
+ "correct": 0
81
+ },
82
+ "seasonality": {
83
+ "count": 3,
84
+ "accuracy": 1.0,
85
+ "correct": 3
86
+ },
87
+ "structural_break": {
88
+ "count": 1,
89
+ "accuracy": 1.0,
90
+ "correct": 1
91
+ },
92
+ "trend": {
93
+ "count": 20,
94
+ "accuracy": 0.3,
95
+ "correct": 6
96
+ },
97
+ "true_false": {
98
+ "count": 65,
99
+ "accuracy": 0.6307692307692307,
100
+ "correct": 41
101
+ },
102
+ "volatility": {
103
+ "count": 8,
104
+ "accuracy": 0.625,
105
+ "correct": 5
106
+ }
107
+ }
108
+ }
109
+ }
110
+ }
runs/itformer_tsqa_full_20260615_164812/eval/predictions.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_164812/logs/eval.log ADDED
@@ -0,0 +1,128 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+
3
+ ✅ Loaded LLM model from: /mnt/share01/sqk/models/qwen2.5-7b-instruct
4
+ trainable params: 40,370,176 || all params: 7,655,986,688 || trainable%: 0.5273
5
+ 🔌 Using adapter: itformer
6
+ Loading model weights from: /mnt/share01/sqk/ITFormer/runs/itformer_tsqa_full_20260615_164812/checkpoint_final/model.safetensors
7
+ Found 0 frozen LLM weights (will be ignored)
8
+ Found 392 LoRA tensors
9
+ Found 150 non-LLM tensors
10
+ LLM missing keys (ignored): 339 keys
11
+
12
+ self.gen = func(*args, **kwds)
13
+ /home/suiqk/anaconda3/envs/scalerag-ts-v4/lib/python3.11/contextlib.py:105: FutureWarning: `torch.backends.cuda.sdp_kernel()` is deprecated. In the future, this context manager will be removed. Please see `torch.nn.attention.sdpa_kernel()` for the new context manager, with updated signature.
14
+ self.gen = func(*args, **kwds)
15
+
16
+ {
17
+ "by_group": {
18
+ "anomaly_detection": {
19
+ "count": 200,
20
+ "accuracy": 0.52,
21
+ "correct": 104,
22
+ "parsed": 200
23
+ },
24
+ "classification": {
25
+ "count": 200,
26
+ "accuracy": 0.68,
27
+ "correct": 136,
28
+ "parsed": 200
29
+ },
30
+ "forecasting": {
31
+ "count": 200,
32
+ "valid_samples": 194,
33
+ "valid_points": 3815,
34
+ "mse": 11348803.821745949,
35
+ "mae": 711.87058641765
36
+ },
37
+ "open_ended": {
38
+ "count": 200,
39
+ "accuracy": 0.46,
40
+ "parsed_accuracy": 0.467005076142132,
41
+ "parse_rate": 0.985,
42
+ "correct": 92,
43
+ "parsed": 197,
44
+ "unparsed": 3,
45
+ "by_format": {
46
+ "multiple_choice": {
47
+ "count": 67,
48
+ "accuracy": 0.47761194029850745,
49
+ "parsed_accuracy": 0.48484848484848486,
50
+ "correct": 32,
51
+ "parsed": 66,
52
+ "unparsed": 1
53
+ },
54
+ "open_ended_question": {
55
+ "count": 67,
56
+ "accuracy": 0.26865671641791045,
57
+ "parsed_accuracy": 0.27692307692307694,
58
+ "correct": 18,
59
+ "parsed": 65,
60
+ "unparsed": 2
61
+ },
62
+ "true_false": {
63
+ "count": 66,
64
+ "accuracy": 0.6363636363636364,
65
+ "parsed_accuracy": 0.6363636363636364,
66
+ "correct": 42,
67
+ "parsed": 66,
68
+ "unparsed": 0
69
+ }
70
+ },
71
+ "by_method": {
72
+ "anomaly": {
73
+ "count": 2,
74
+ "accuracy": 1.0,
75
+ "correct": 2
76
+ },
77
+ "cyclical": {
78
+ "count": 3,
79
+ "accuracy": 0.6666666666666666,
80
+ "correct": 2
81
+ },
82
+ "multiple_choice": {
83
+ "count": 64,
84
+ "accuracy": 0.5,
85
+ "correct": 32
86
+ },
87
+ "numeric_scalar": {
88
+ "count": 30,
89
+ "accuracy": 0.0,
90
+ "correct": 0
91
+ },
92
+ "numeric_sequence": {
93
+ "count": 1,
94
+ "accuracy": 0.0,
95
+ "correct": 0
96
+ },
97
+ "seasonality": {
98
+ "count": 3,
99
+ "accuracy": 1.0,
100
+ "correct": 3
101
+ },
102
+ "structural_break": {
103
+ "count": 1,
104
+ "accuracy": 1.0,
105
+ "correct": 1
106
+ },
107
+ "trend": {
108
+ "count": 20,
109
+ "accuracy": 0.3,
110
+ "correct": 6
111
+ },
112
+ "true_false": {
113
+ "count": 65,
114
+ "accuracy": 0.6307692307692307,
115
+ "correct": 41
116
+ },
117
+ "volatility": {
118
+ "count": 8,
119
+ "accuracy": 0.625,
120
+ "correct": 5
121
+ }
122
+ }
123
+ }
124
+ }
125
+ }
126
+ Predictions: /mnt/share01/sqk/ITFormer/runs/itformer_tsqa_full_20260615_164812/eval/predictions.jsonl
127
+ Metrics: /mnt/share01/sqk/ITFormer/runs/itformer_tsqa_full_20260615_164812/eval/metrics.json
128
+ [rank0]:[W615 18:05:20.343498175 ProcessGroupNCCL.cpp:1496] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
runs/itformer_tsqa_full_20260615_164812/logs/train.log ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "TLM"
4
+ ],
5
+ "freeze_ts_model": true,
6
+ "gradient_checkpointing": true,
7
+ "llm_attn_implementation": null,
8
+ "llm_model_path": "/mnt/share01/sqk/models/qwen2.5-7b-instruct",
9
+ "llm_torch_dtype": "bfloat16",
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "lora_r": 16,
13
+ "lora_target_modules": [
14
+ "q_proj",
15
+ "k_proj",
16
+ "v_proj",
17
+ "o_proj",
18
+ "gate_proj",
19
+ "up_proj",
20
+ "down_proj"
21
+ ],
22
+ "model_type": "vlm_model",
23
+ "torch_dtype": "bfloat16",
24
+ "transformers_version": "4.54.1",
25
+ "ts_pad_num": 25,
26
+ "use_lora": true
27
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.54.1"
4
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/tokenizer_config.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "clean_up_tokenization_spaces": false,
199
+ "eos_token": "<|im_end|>",
200
+ "errors": "replace",
201
+ "extra_special_tokens": {},
202
+ "model_max_length": 131072,
203
+ "pad_token": "<|endoftext|>",
204
+ "split_special_tokens": false,
205
+ "tokenizer_class": "Qwen2Tokenizer",
206
+ "unk_token": null
207
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3500/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/added_tokens.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</tool_call>": 151658,
3
+ "<tool_call>": 151657,
4
+ "<|box_end|>": 151649,
5
+ "<|box_start|>": 151648,
6
+ "<|endoftext|>": 151643,
7
+ "<|file_sep|>": 151664,
8
+ "<|fim_middle|>": 151660,
9
+ "<|fim_pad|>": 151662,
10
+ "<|fim_prefix|>": 151659,
11
+ "<|fim_suffix|>": 151661,
12
+ "<|im_end|>": 151645,
13
+ "<|im_start|>": 151644,
14
+ "<|image_pad|>": 151655,
15
+ "<|object_ref_end|>": 151647,
16
+ "<|object_ref_start|>": 151646,
17
+ "<|quad_end|>": 151651,
18
+ "<|quad_start|>": 151650,
19
+ "<|repo_name|>": 151663,
20
+ "<|video_pad|>": 151656,
21
+ "<|vision_end|>": 151653,
22
+ "<|vision_pad|>": 151654,
23
+ "<|vision_start|>": 151652
24
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "TLM"
4
+ ],
5
+ "freeze_ts_model": true,
6
+ "gradient_checkpointing": true,
7
+ "llm_attn_implementation": null,
8
+ "llm_model_path": "/mnt/share01/sqk/models/qwen2.5-7b-instruct",
9
+ "llm_torch_dtype": "bfloat16",
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "lora_r": 16,
13
+ "lora_target_modules": [
14
+ "q_proj",
15
+ "k_proj",
16
+ "v_proj",
17
+ "o_proj",
18
+ "gate_proj",
19
+ "up_proj",
20
+ "down_proj"
21
+ ],
22
+ "model_type": "vlm_model",
23
+ "torch_dtype": "bfloat16",
24
+ "transformers_version": "4.54.1",
25
+ "ts_pad_num": 25,
26
+ "use_lora": true
27
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.54.1"
4
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/tokenizer_config.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "clean_up_tokenization_spaces": false,
199
+ "eos_token": "<|im_end|>",
200
+ "errors": "replace",
201
+ "extra_special_tokens": {},
202
+ "model_max_length": 131072,
203
+ "pad_token": "<|endoftext|>",
204
+ "split_special_tokens": false,
205
+ "tokenizer_class": "Qwen2Tokenizer",
206
+ "unk_token": null
207
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/checkpoint-3996/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "TLM"
4
+ ],
5
+ "freeze_ts_model": true,
6
+ "gradient_checkpointing": true,
7
+ "llm_attn_implementation": null,
8
+ "llm_model_path": "/mnt/share01/sqk/models/qwen2.5-7b-instruct",
9
+ "llm_torch_dtype": "bfloat16",
10
+ "lora_alpha": 32,
11
+ "lora_dropout": 0.05,
12
+ "lora_r": 16,
13
+ "lora_target_modules": [
14
+ "q_proj",
15
+ "k_proj",
16
+ "v_proj",
17
+ "o_proj",
18
+ "gate_proj",
19
+ "up_proj",
20
+ "down_proj"
21
+ ],
22
+ "model_type": "vlm_model",
23
+ "torch_dtype": "bfloat16",
24
+ "transformers_version": "4.54.1",
25
+ "ts_pad_num": 25,
26
+ "use_lora": true
27
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.54.1"
4
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/skipped_overlength_train.jsonl ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {"id": "forecasting_011619_56748ac9", "line_number": 4058, "source_type": "forecasting", "input_tokens": 4301, "instruction_tokens": 4254, "max_seq_length": 4096}
2
+ {"id": "forecasting_031782_0bf68658", "line_number": 4074, "source_type": "forecasting", "input_tokens": 6214, "instruction_tokens": 6167, "max_seq_length": 4096}
3
+ {"id": "forecasting_037187_4a95592f", "line_number": 4234, "source_type": "forecasting", "input_tokens": 4539, "instruction_tokens": 4491, "max_seq_length": 4096}
4
+ {"id": "forecasting_036903_44ae43fa", "line_number": 4325, "source_type": "forecasting", "input_tokens": 5117, "instruction_tokens": 5069, "max_seq_length": 4096}
5
+ {"id": "forecasting_004128_65a4fddf", "line_number": 4824, "source_type": "forecasting", "input_tokens": 5974, "instruction_tokens": 5927, "max_seq_length": 4096}
6
+ {"id": "forecasting_027194_66448c4b", "line_number": 5817, "source_type": "forecasting", "input_tokens": 4717, "instruction_tokens": 4669, "max_seq_length": 4096}
7
+ {"id": "forecasting_008995_b0cbd927", "line_number": 5886, "source_type": "forecasting", "input_tokens": 4532, "instruction_tokens": 4485, "max_seq_length": 4096}
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/special_tokens_map.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|object_ref_start|>",
6
+ "<|object_ref_end|>",
7
+ "<|box_start|>",
8
+ "<|box_end|>",
9
+ "<|quad_start|>",
10
+ "<|quad_end|>",
11
+ "<|vision_start|>",
12
+ "<|vision_end|>",
13
+ "<|vision_pad|>",
14
+ "<|image_pad|>",
15
+ "<|video_pad|>"
16
+ ],
17
+ "eos_token": {
18
+ "content": "<|im_end|>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ },
24
+ "pad_token": {
25
+ "content": "<|endoftext|>",
26
+ "lstrip": false,
27
+ "normalized": false,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
runs/itformer_tsqa_full_20260615_182013_itformer/checkpoint_final/tokenizer_config.json ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "added_tokens_decoder": {
5
+ "151643": {
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "151644": {
14
+ "content": "<|im_start|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "151645": {
22
+ "content": "<|im_end|>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "151646": {
30
+ "content": "<|object_ref_start|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "151647": {
38
+ "content": "<|object_ref_end|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ },
45
+ "151648": {
46
+ "content": "<|box_start|>",
47
+ "lstrip": false,
48
+ "normalized": false,
49
+ "rstrip": false,
50
+ "single_word": false,
51
+ "special": true
52
+ },
53
+ "151649": {
54
+ "content": "<|box_end|>",
55
+ "lstrip": false,
56
+ "normalized": false,
57
+ "rstrip": false,
58
+ "single_word": false,
59
+ "special": true
60
+ },
61
+ "151650": {
62
+ "content": "<|quad_start|>",
63
+ "lstrip": false,
64
+ "normalized": false,
65
+ "rstrip": false,
66
+ "single_word": false,
67
+ "special": true
68
+ },
69
+ "151651": {
70
+ "content": "<|quad_end|>",
71
+ "lstrip": false,
72
+ "normalized": false,
73
+ "rstrip": false,
74
+ "single_word": false,
75
+ "special": true
76
+ },
77
+ "151652": {
78
+ "content": "<|vision_start|>",
79
+ "lstrip": false,
80
+ "normalized": false,
81
+ "rstrip": false,
82
+ "single_word": false,
83
+ "special": true
84
+ },
85
+ "151653": {
86
+ "content": "<|vision_end|>",
87
+ "lstrip": false,
88
+ "normalized": false,
89
+ "rstrip": false,
90
+ "single_word": false,
91
+ "special": true
92
+ },
93
+ "151654": {
94
+ "content": "<|vision_pad|>",
95
+ "lstrip": false,
96
+ "normalized": false,
97
+ "rstrip": false,
98
+ "single_word": false,
99
+ "special": true
100
+ },
101
+ "151655": {
102
+ "content": "<|image_pad|>",
103
+ "lstrip": false,
104
+ "normalized": false,
105
+ "rstrip": false,
106
+ "single_word": false,
107
+ "special": true
108
+ },
109
+ "151656": {
110
+ "content": "<|video_pad|>",
111
+ "lstrip": false,
112
+ "normalized": false,
113
+ "rstrip": false,
114
+ "single_word": false,
115
+ "special": true
116
+ },
117
+ "151657": {
118
+ "content": "<tool_call>",
119
+ "lstrip": false,
120
+ "normalized": false,
121
+ "rstrip": false,
122
+ "single_word": false,
123
+ "special": false
124
+ },
125
+ "151658": {
126
+ "content": "</tool_call>",
127
+ "lstrip": false,
128
+ "normalized": false,
129
+ "rstrip": false,
130
+ "single_word": false,
131
+ "special": false
132
+ },
133
+ "151659": {
134
+ "content": "<|fim_prefix|>",
135
+ "lstrip": false,
136
+ "normalized": false,
137
+ "rstrip": false,
138
+ "single_word": false,
139
+ "special": false
140
+ },
141
+ "151660": {
142
+ "content": "<|fim_middle|>",
143
+ "lstrip": false,
144
+ "normalized": false,
145
+ "rstrip": false,
146
+ "single_word": false,
147
+ "special": false
148
+ },
149
+ "151661": {
150
+ "content": "<|fim_suffix|>",
151
+ "lstrip": false,
152
+ "normalized": false,
153
+ "rstrip": false,
154
+ "single_word": false,
155
+ "special": false
156
+ },
157
+ "151662": {
158
+ "content": "<|fim_pad|>",
159
+ "lstrip": false,
160
+ "normalized": false,
161
+ "rstrip": false,
162
+ "single_word": false,
163
+ "special": false
164
+ },
165
+ "151663": {
166
+ "content": "<|repo_name|>",
167
+ "lstrip": false,
168
+ "normalized": false,
169
+ "rstrip": false,
170
+ "single_word": false,
171
+ "special": false
172
+ },
173
+ "151664": {
174
+ "content": "<|file_sep|>",
175
+ "lstrip": false,
176
+ "normalized": false,
177
+ "rstrip": false,
178
+ "single_word": false,
179
+ "special": false
180
+ }
181
+ },
182
+ "additional_special_tokens": [
183
+ "<|im_start|>",
184
+ "<|im_end|>",
185
+ "<|object_ref_start|>",
186
+ "<|object_ref_end|>",
187
+ "<|box_start|>",
188
+ "<|box_end|>",
189
+ "<|quad_start|>",
190
+ "<|quad_end|>",
191
+ "<|vision_start|>",
192
+ "<|vision_end|>",
193
+ "<|vision_pad|>",
194
+ "<|image_pad|>",
195
+ "<|video_pad|>"
196
+ ],
197
+ "bos_token": null,
198
+ "clean_up_tokenization_spaces": false,
199
+ "eos_token": "<|im_end|>",
200
+ "errors": "replace",
201
+ "extra_special_tokens": {},
202
+ "model_max_length": 131072,
203
+ "pad_token": "<|endoftext|>",
204
+ "split_special_tokens": false,
205
+ "tokenizer_class": "Qwen2Tokenizer",
206
+ "unk_token": null
207
+ }