Jerry999 commited on
Commit
c4ff403
·
verified ·
1 Parent(s): c66a628

upload sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270

Browse files
Files changed (18) hide show
  1. .gitattributes +1 -0
  2. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/config.yaml +124 -0
  3. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/added_tokens.json +28 -0
  4. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/chat_template.jinja +89 -0
  5. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/merges.txt +0 -0
  6. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/special_tokens_map.json +31 -0
  7. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/tokenizer.json +3 -0
  8. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/tokenizer_config.json +239 -0
  9. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/vocab.json +0 -0
  10. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/.hf_metadata/config.json +70 -0
  11. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/.hf_metadata/fqn_to_file_index_mapping.json +401 -0
  12. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/.hf_metadata/generation_config.json +13 -0
  13. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/shard-00001-model-00001-of-00001.safetensors +3 -0
  14. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/shard-00002-model-00001-of-00001.safetensors +3 -0
  15. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/shard-00003-model-00001-of-00001.safetensors +3 -0
  16. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/shard-00004-model-00001-of-00001.safetensors +3 -0
  17. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/train_dataloader.pt +3 -0
  18. checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/training_info.json +1 -0
.gitattributes CHANGED
@@ -158,3 +158,4 @@ checkpoints/knowledge/sft_mix_d5_e10_llama_3_2_3b_it_l0p3/run_20260503.043607/st
158
  checkpoints/simpleqa/sft_gt_qwen3_1_7b_20260428/run_20260429.081031/step_270/policy/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
159
  checkpoints/knowledge/sft_mix_d5_e10_llama_3_2_3b_it_l0p5/run_20260503.043620/step_200/policy/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
160
  checkpoints/simpleqa/sft_gt_qwen3_4b_20260428/run_20260429.084349/step_270/policy/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
158
  checkpoints/simpleqa/sft_gt_qwen3_1_7b_20260428/run_20260429.081031/step_270/policy/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
159
  checkpoints/knowledge/sft_mix_d5_e10_llama_3_2_3b_it_l0p5/run_20260503.043620/step_200/policy/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
160
  checkpoints/simpleqa/sft_gt_qwen3_4b_20260428/run_20260429.084349/step_270/policy/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
161
+ checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/config.yaml ADDED
@@ -0,0 +1,124 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ checkpointing:
2
+ checkpoint_dir: /work/jh250017/m57000/jiarui/TeacherMix/data/checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349
3
+ checkpoint_must_save_by: null
4
+ enabled: true
5
+ higher_is_better: false
6
+ keep_top_k: null
7
+ metric_name: null
8
+ save_optimizer: false
9
+ save_period: 50
10
+ cluster:
11
+ gpus_per_node: 1
12
+ num_nodes: 4
13
+ data:
14
+ add_bos: true
15
+ add_eos: true
16
+ add_generation_prompt: false
17
+ default:
18
+ processor: sft_processor
19
+ prompt_file: null
20
+ system_prompt_file: /work/jh250017/m57000/jiarui/TeacherMix/mix_teachers/train/simpleqa_opsd/system_prompt.txt
21
+ max_input_seq_length: 10000
22
+ num_workers: 1
23
+ shuffle: true
24
+ train:
25
+ - dataset_name: simpleqa_opsd
26
+ processor: sft_processor
27
+ prompt_file: null
28
+ system_prompt_file: /work/jh250017/m57000/jiarui/TeacherMix/mix_teachers/train/simpleqa_opsd/system_prompt.txt
29
+ logger:
30
+ gpu_monitoring:
31
+ collection_interval: 10
32
+ flush_interval: 10
33
+ log_dir: /work/jh250017/m57000/jiarui/TeacherMix/data/checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/logs/exp_001
34
+ mlflow_enabled: false
35
+ monitor_gpus: true
36
+ swanlab_enabled: false
37
+ tensorboard:
38
+ log_dir: tb_logs-sft-simpleqa-gt-qwen3_8b
39
+ tensorboard_enabled: true
40
+ wandb:
41
+ entity: mathrl-jiarui
42
+ name: sft-simpleqa-gt-20260428-miyabi-Qwen/Qwen3-8B
43
+ project: mix-teachers-opsd
44
+ wandb_enabled: true
45
+ paths:
46
+ base_dir: /work/jh250017/m57000/jiarui/TeacherMix/mix_teachers/train/simpleqa_opsd
47
+ data_dir: /work/jh250017/m57000/jiarui/TeacherMix/data/simpleqa/atomic_sft
48
+ policy:
49
+ dtensor_cfg:
50
+ _v2: true
51
+ activation_checkpointing: true
52
+ context_parallel_size: 1
53
+ cpu_offload: true
54
+ custom_parallel_plan: null
55
+ enabled: true
56
+ lora_cfg:
57
+ alpha: 32
58
+ dim: 8
59
+ dropout: 0.0
60
+ dropout_position: post
61
+ enabled: false
62
+ exclude_modules: []
63
+ lora_A_init: xavier
64
+ match_all_linear: true
65
+ target_modules: []
66
+ use_triton: true
67
+ sequence_parallel: false
68
+ tensor_parallel_size: 1
69
+ dynamic_batching:
70
+ enabled: true
71
+ sequence_length_round: 64
72
+ train_mb_tokens: 10000
73
+ make_sequence_length_divisible_by: 1
74
+ max_grad_norm: 1.0
75
+ max_total_sequence_length: 10000
76
+ megatron_cfg:
77
+ enabled: false
78
+ model_name: Qwen/Qwen3-8B
79
+ offload_optimizer_for_logprob: false
80
+ optimizer:
81
+ kwargs:
82
+ betas:
83
+ - 0.9
84
+ - 0.95
85
+ eps: 1.0e-08
86
+ foreach: false
87
+ fused: false
88
+ lr: 1.0e-05
89
+ weight_decay: 0.01
90
+ name: torch.optim.AdamW
91
+ precision: bfloat16
92
+ scheduler:
93
+ - kwargs:
94
+ end_factor: 1.0
95
+ start_factor: 0.1
96
+ total_iters: 20
97
+ name: torch.optim.lr_scheduler.LinearLR
98
+ - kwargs:
99
+ factor: 1.0
100
+ total_iters: 10000000000
101
+ name: torch.optim.lr_scheduler.ConstantLR
102
+ - milestones:
103
+ - 20
104
+ sequence_packing:
105
+ algorithm: modified_first_fit_decreasing
106
+ enabled: false
107
+ sequence_length_round: 64
108
+ train_mb_tokens: 10000
109
+ tokenizer:
110
+ chat_template_kwargs:
111
+ enable_thinking: false
112
+ name: Qwen/Qwen3-8B
113
+ train_global_batch_size: 16
114
+ train_micro_batch_size: 1
115
+ sft:
116
+ max_num_epochs: 1
117
+ max_num_steps: 270
118
+ seed: 42
119
+ val_at_end: false
120
+ val_at_start: false
121
+ val_batches: 8
122
+ val_global_batch_size: 32
123
+ val_micro_batch_size: 1
124
+ val_period: 99999
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/added_tokens.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 151668,
3
+ "</tool_call>": 151658,
4
+ "</tool_response>": 151666,
5
+ "<think>": 151667,
6
+ "<tool_call>": 151657,
7
+ "<tool_response>": 151665,
8
+ "<|box_end|>": 151649,
9
+ "<|box_start|>": 151648,
10
+ "<|endoftext|>": 151643,
11
+ "<|file_sep|>": 151664,
12
+ "<|fim_middle|>": 151660,
13
+ "<|fim_pad|>": 151662,
14
+ "<|fim_prefix|>": 151659,
15
+ "<|fim_suffix|>": 151661,
16
+ "<|im_end|>": 151645,
17
+ "<|im_start|>": 151644,
18
+ "<|image_pad|>": 151655,
19
+ "<|object_ref_end|>": 151647,
20
+ "<|object_ref_start|>": 151646,
21
+ "<|quad_end|>": 151651,
22
+ "<|quad_start|>": 151650,
23
+ "<|repo_name|>": 151663,
24
+ "<|video_pad|>": 151656,
25
+ "<|vision_end|>": 151653,
26
+ "<|vision_pad|>": 151654,
27
+ "<|vision_start|>": 151652
28
+ }
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/chat_template.jinja ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0].role == 'system' %}
4
+ {{- messages[0].content + '\n\n' }}
5
+ {%- endif %}
6
+ {{- "# 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>" }}
7
+ {%- for tool in tools %}
8
+ {{- "\n" }}
9
+ {{- tool | tojson }}
10
+ {%- endfor %}
11
+ {{- "\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" }}
12
+ {%- else %}
13
+ {%- if messages[0].role == 'system' %}
14
+ {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
15
+ {%- endif %}
16
+ {%- endif %}
17
+ {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
18
+ {%- for message in messages[::-1] %}
19
+ {%- set index = (messages|length - 1) - loop.index0 %}
20
+ {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
21
+ {%- set ns.multi_step_tool = false %}
22
+ {%- set ns.last_query_index = index %}
23
+ {%- endif %}
24
+ {%- endfor %}
25
+ {%- for message in messages %}
26
+ {%- if message.content is string %}
27
+ {%- set content = message.content %}
28
+ {%- else %}
29
+ {%- set content = '' %}
30
+ {%- endif %}
31
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
32
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
33
+ {%- elif message.role == "assistant" %}
34
+ {%- set reasoning_content = '' %}
35
+ {%- if message.reasoning_content is string %}
36
+ {%- set reasoning_content = message.reasoning_content %}
37
+ {%- else %}
38
+ {%- if '</think>' in content %}
39
+ {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
40
+ {%- set content = content.split('</think>')[-1].lstrip('\n') %}
41
+ {%- endif %}
42
+ {%- endif %}
43
+ {%- if loop.index0 > ns.last_query_index %}
44
+ {%- if loop.last or (not loop.last and reasoning_content) %}
45
+ {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
46
+ {%- else %}
47
+ {{- '<|im_start|>' + message.role + '\n' + content }}
48
+ {%- endif %}
49
+ {%- else %}
50
+ {{- '<|im_start|>' + message.role + '\n' + content }}
51
+ {%- endif %}
52
+ {%- if message.tool_calls %}
53
+ {%- for tool_call in message.tool_calls %}
54
+ {%- if (loop.first and content) or (not loop.first) %}
55
+ {{- '\n' }}
56
+ {%- endif %}
57
+ {%- if tool_call.function %}
58
+ {%- set tool_call = tool_call.function %}
59
+ {%- endif %}
60
+ {{- '<tool_call>\n{"name": "' }}
61
+ {{- tool_call.name }}
62
+ {{- '", "arguments": ' }}
63
+ {%- if tool_call.arguments is string %}
64
+ {{- tool_call.arguments }}
65
+ {%- else %}
66
+ {{- tool_call.arguments | tojson }}
67
+ {%- endif %}
68
+ {{- '}\n</tool_call>' }}
69
+ {%- endfor %}
70
+ {%- endif %}
71
+ {{- '<|im_end|>\n' }}
72
+ {%- elif message.role == "tool" %}
73
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
74
+ {{- '<|im_start|>user' }}
75
+ {%- endif %}
76
+ {{- '\n<tool_response>\n' }}
77
+ {{- content }}
78
+ {{- '\n</tool_response>' }}
79
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
80
+ {{- '<|im_end|>\n' }}
81
+ {%- endif %}
82
+ {%- endif %}
83
+ {%- endfor %}
84
+ {%- if add_generation_prompt %}
85
+ {{- '<|im_start|>assistant\n' }}
86
+ {%- if enable_thinking is defined and enable_thinking is false %}
87
+ {{- '<think>\n\n</think>\n\n' }}
88
+ {%- endif %}
89
+ {%- endif %}
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/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
+ }
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
3
+ size 11422654
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "151665": {
182
+ "content": "<tool_response>",
183
+ "lstrip": false,
184
+ "normalized": false,
185
+ "rstrip": false,
186
+ "single_word": false,
187
+ "special": false
188
+ },
189
+ "151666": {
190
+ "content": "</tool_response>",
191
+ "lstrip": false,
192
+ "normalized": false,
193
+ "rstrip": false,
194
+ "single_word": false,
195
+ "special": false
196
+ },
197
+ "151667": {
198
+ "content": "<think>",
199
+ "lstrip": false,
200
+ "normalized": false,
201
+ "rstrip": false,
202
+ "single_word": false,
203
+ "special": false
204
+ },
205
+ "151668": {
206
+ "content": "</think>",
207
+ "lstrip": false,
208
+ "normalized": false,
209
+ "rstrip": false,
210
+ "single_word": false,
211
+ "special": false
212
+ }
213
+ },
214
+ "additional_special_tokens": [
215
+ "<|im_start|>",
216
+ "<|im_end|>",
217
+ "<|object_ref_start|>",
218
+ "<|object_ref_end|>",
219
+ "<|box_start|>",
220
+ "<|box_end|>",
221
+ "<|quad_start|>",
222
+ "<|quad_end|>",
223
+ "<|vision_start|>",
224
+ "<|vision_end|>",
225
+ "<|vision_pad|>",
226
+ "<|image_pad|>",
227
+ "<|video_pad|>"
228
+ ],
229
+ "bos_token": null,
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|im_end|>",
232
+ "errors": "replace",
233
+ "extra_special_tokens": {},
234
+ "model_max_length": 131072,
235
+ "pad_token": "<|endoftext|>",
236
+ "split_special_tokens": false,
237
+ "tokenizer_class": "Qwen2Tokenizer",
238
+ "unk_token": null
239
+ }
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/tokenizer/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/.hf_metadata/config.json ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "float32",
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 4096,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 12288,
15
+ "layer_types": [
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention"
52
+ ],
53
+ "max_position_embeddings": 40960,
54
+ "max_window_layers": 36,
55
+ "model_type": "qwen3",
56
+ "nemo_version": "0.2.0",
57
+ "num_attention_heads": 32,
58
+ "num_hidden_layers": 36,
59
+ "num_key_value_heads": 8,
60
+ "pad_token_id": 151643,
61
+ "rms_norm_eps": 1e-06,
62
+ "rope_scaling": null,
63
+ "rope_theta": 1000000,
64
+ "sliding_window": null,
65
+ "tie_word_embeddings": false,
66
+ "transformers_version": "4.57.1",
67
+ "use_cache": false,
68
+ "use_sliding_window": false,
69
+ "vocab_size": 151936
70
+ }
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/.hf_metadata/fqn_to_file_index_mapping.json ADDED
@@ -0,0 +1,401 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "lm_head.weight": 1,
3
+ "model.embed_tokens.weight": 1,
4
+ "model.layers.0.input_layernorm.weight": 1,
5
+ "model.layers.0.mlp.down_proj.weight": 1,
6
+ "model.layers.0.mlp.gate_proj.weight": 1,
7
+ "model.layers.0.mlp.up_proj.weight": 1,
8
+ "model.layers.0.post_attention_layernorm.weight": 1,
9
+ "model.layers.0.self_attn.k_norm.weight": 1,
10
+ "model.layers.0.self_attn.k_proj.weight": 1,
11
+ "model.layers.0.self_attn.o_proj.weight": 1,
12
+ "model.layers.0.self_attn.q_norm.weight": 1,
13
+ "model.layers.0.self_attn.q_proj.weight": 1,
14
+ "model.layers.0.self_attn.v_proj.weight": 1,
15
+ "model.layers.1.input_layernorm.weight": 1,
16
+ "model.layers.1.mlp.down_proj.weight": 1,
17
+ "model.layers.1.mlp.gate_proj.weight": 1,
18
+ "model.layers.1.mlp.up_proj.weight": 1,
19
+ "model.layers.1.post_attention_layernorm.weight": 1,
20
+ "model.layers.1.self_attn.k_norm.weight": 1,
21
+ "model.layers.1.self_attn.k_proj.weight": 1,
22
+ "model.layers.1.self_attn.o_proj.weight": 1,
23
+ "model.layers.1.self_attn.q_norm.weight": 1,
24
+ "model.layers.1.self_attn.q_proj.weight": 1,
25
+ "model.layers.1.self_attn.v_proj.weight": 1,
26
+ "model.layers.10.input_layernorm.weight": 1,
27
+ "model.layers.10.mlp.down_proj.weight": 1,
28
+ "model.layers.10.mlp.gate_proj.weight": 1,
29
+ "model.layers.10.mlp.up_proj.weight": 1,
30
+ "model.layers.10.post_attention_layernorm.weight": 1,
31
+ "model.layers.10.self_attn.k_norm.weight": 1,
32
+ "model.layers.10.self_attn.k_proj.weight": 1,
33
+ "model.layers.10.self_attn.o_proj.weight": 1,
34
+ "model.layers.10.self_attn.q_norm.weight": 1,
35
+ "model.layers.10.self_attn.q_proj.weight": 1,
36
+ "model.layers.10.self_attn.v_proj.weight": 1,
37
+ "model.layers.11.input_layernorm.weight": 1,
38
+ "model.layers.11.mlp.down_proj.weight": 1,
39
+ "model.layers.11.mlp.gate_proj.weight": 1,
40
+ "model.layers.11.mlp.up_proj.weight": 1,
41
+ "model.layers.11.post_attention_layernorm.weight": 1,
42
+ "model.layers.11.self_attn.k_norm.weight": 1,
43
+ "model.layers.11.self_attn.k_proj.weight": 1,
44
+ "model.layers.11.self_attn.o_proj.weight": 1,
45
+ "model.layers.11.self_attn.q_norm.weight": 1,
46
+ "model.layers.11.self_attn.q_proj.weight": 1,
47
+ "model.layers.11.self_attn.v_proj.weight": 1,
48
+ "model.layers.12.input_layernorm.weight": 1,
49
+ "model.layers.12.mlp.down_proj.weight": 1,
50
+ "model.layers.12.mlp.gate_proj.weight": 1,
51
+ "model.layers.12.mlp.up_proj.weight": 1,
52
+ "model.layers.12.post_attention_layernorm.weight": 1,
53
+ "model.layers.12.self_attn.k_norm.weight": 1,
54
+ "model.layers.12.self_attn.k_proj.weight": 1,
55
+ "model.layers.12.self_attn.o_proj.weight": 1,
56
+ "model.layers.12.self_attn.q_norm.weight": 1,
57
+ "model.layers.12.self_attn.q_proj.weight": 1,
58
+ "model.layers.12.self_attn.v_proj.weight": 1,
59
+ "model.layers.13.input_layernorm.weight": 1,
60
+ "model.layers.13.mlp.down_proj.weight": 1,
61
+ "model.layers.13.mlp.gate_proj.weight": 1,
62
+ "model.layers.13.mlp.up_proj.weight": 1,
63
+ "model.layers.13.post_attention_layernorm.weight": 1,
64
+ "model.layers.13.self_attn.k_norm.weight": 1,
65
+ "model.layers.13.self_attn.k_proj.weight": 1,
66
+ "model.layers.13.self_attn.o_proj.weight": 1,
67
+ "model.layers.13.self_attn.q_norm.weight": 1,
68
+ "model.layers.13.self_attn.q_proj.weight": 1,
69
+ "model.layers.13.self_attn.v_proj.weight": 1,
70
+ "model.layers.14.input_layernorm.weight": 1,
71
+ "model.layers.14.mlp.down_proj.weight": 1,
72
+ "model.layers.14.mlp.gate_proj.weight": 1,
73
+ "model.layers.14.mlp.up_proj.weight": 1,
74
+ "model.layers.14.post_attention_layernorm.weight": 1,
75
+ "model.layers.14.self_attn.k_norm.weight": 1,
76
+ "model.layers.14.self_attn.k_proj.weight": 1,
77
+ "model.layers.14.self_attn.o_proj.weight": 1,
78
+ "model.layers.14.self_attn.q_norm.weight": 1,
79
+ "model.layers.14.self_attn.q_proj.weight": 1,
80
+ "model.layers.14.self_attn.v_proj.weight": 1,
81
+ "model.layers.15.input_layernorm.weight": 1,
82
+ "model.layers.15.mlp.down_proj.weight": 1,
83
+ "model.layers.15.mlp.gate_proj.weight": 1,
84
+ "model.layers.15.mlp.up_proj.weight": 1,
85
+ "model.layers.15.post_attention_layernorm.weight": 1,
86
+ "model.layers.15.self_attn.k_norm.weight": 1,
87
+ "model.layers.15.self_attn.k_proj.weight": 1,
88
+ "model.layers.15.self_attn.o_proj.weight": 1,
89
+ "model.layers.15.self_attn.q_norm.weight": 1,
90
+ "model.layers.15.self_attn.q_proj.weight": 1,
91
+ "model.layers.15.self_attn.v_proj.weight": 1,
92
+ "model.layers.16.input_layernorm.weight": 1,
93
+ "model.layers.16.mlp.down_proj.weight": 1,
94
+ "model.layers.16.mlp.gate_proj.weight": 1,
95
+ "model.layers.16.mlp.up_proj.weight": 1,
96
+ "model.layers.16.post_attention_layernorm.weight": 1,
97
+ "model.layers.16.self_attn.k_norm.weight": 1,
98
+ "model.layers.16.self_attn.k_proj.weight": 1,
99
+ "model.layers.16.self_attn.o_proj.weight": 1,
100
+ "model.layers.16.self_attn.q_norm.weight": 1,
101
+ "model.layers.16.self_attn.q_proj.weight": 1,
102
+ "model.layers.16.self_attn.v_proj.weight": 1,
103
+ "model.layers.17.input_layernorm.weight": 1,
104
+ "model.layers.17.mlp.down_proj.weight": 1,
105
+ "model.layers.17.mlp.gate_proj.weight": 1,
106
+ "model.layers.17.mlp.up_proj.weight": 1,
107
+ "model.layers.17.post_attention_layernorm.weight": 1,
108
+ "model.layers.17.self_attn.k_norm.weight": 1,
109
+ "model.layers.17.self_attn.k_proj.weight": 1,
110
+ "model.layers.17.self_attn.o_proj.weight": 1,
111
+ "model.layers.17.self_attn.q_norm.weight": 1,
112
+ "model.layers.17.self_attn.q_proj.weight": 1,
113
+ "model.layers.17.self_attn.v_proj.weight": 1,
114
+ "model.layers.18.input_layernorm.weight": 1,
115
+ "model.layers.18.mlp.down_proj.weight": 1,
116
+ "model.layers.18.mlp.gate_proj.weight": 1,
117
+ "model.layers.18.mlp.up_proj.weight": 1,
118
+ "model.layers.18.post_attention_layernorm.weight": 1,
119
+ "model.layers.18.self_attn.k_norm.weight": 1,
120
+ "model.layers.18.self_attn.k_proj.weight": 1,
121
+ "model.layers.18.self_attn.o_proj.weight": 1,
122
+ "model.layers.18.self_attn.q_norm.weight": 1,
123
+ "model.layers.18.self_attn.q_proj.weight": 1,
124
+ "model.layers.18.self_attn.v_proj.weight": 1,
125
+ "model.layers.19.input_layernorm.weight": 1,
126
+ "model.layers.19.mlp.down_proj.weight": 1,
127
+ "model.layers.19.mlp.gate_proj.weight": 1,
128
+ "model.layers.19.mlp.up_proj.weight": 1,
129
+ "model.layers.19.post_attention_layernorm.weight": 1,
130
+ "model.layers.19.self_attn.k_norm.weight": 1,
131
+ "model.layers.19.self_attn.k_proj.weight": 1,
132
+ "model.layers.19.self_attn.o_proj.weight": 1,
133
+ "model.layers.19.self_attn.q_norm.weight": 1,
134
+ "model.layers.19.self_attn.q_proj.weight": 1,
135
+ "model.layers.19.self_attn.v_proj.weight": 1,
136
+ "model.layers.2.input_layernorm.weight": 1,
137
+ "model.layers.2.mlp.down_proj.weight": 1,
138
+ "model.layers.2.mlp.gate_proj.weight": 1,
139
+ "model.layers.2.mlp.up_proj.weight": 1,
140
+ "model.layers.2.post_attention_layernorm.weight": 1,
141
+ "model.layers.2.self_attn.k_norm.weight": 1,
142
+ "model.layers.2.self_attn.k_proj.weight": 1,
143
+ "model.layers.2.self_attn.o_proj.weight": 1,
144
+ "model.layers.2.self_attn.q_norm.weight": 1,
145
+ "model.layers.2.self_attn.q_proj.weight": 1,
146
+ "model.layers.2.self_attn.v_proj.weight": 1,
147
+ "model.layers.20.input_layernorm.weight": 1,
148
+ "model.layers.20.mlp.down_proj.weight": 1,
149
+ "model.layers.20.mlp.gate_proj.weight": 1,
150
+ "model.layers.20.mlp.up_proj.weight": 1,
151
+ "model.layers.20.post_attention_layernorm.weight": 1,
152
+ "model.layers.20.self_attn.k_norm.weight": 1,
153
+ "model.layers.20.self_attn.k_proj.weight": 1,
154
+ "model.layers.20.self_attn.o_proj.weight": 1,
155
+ "model.layers.20.self_attn.q_norm.weight": 1,
156
+ "model.layers.20.self_attn.q_proj.weight": 1,
157
+ "model.layers.20.self_attn.v_proj.weight": 1,
158
+ "model.layers.21.input_layernorm.weight": 1,
159
+ "model.layers.21.mlp.down_proj.weight": 1,
160
+ "model.layers.21.mlp.gate_proj.weight": 1,
161
+ "model.layers.21.mlp.up_proj.weight": 1,
162
+ "model.layers.21.post_attention_layernorm.weight": 1,
163
+ "model.layers.21.self_attn.k_norm.weight": 1,
164
+ "model.layers.21.self_attn.k_proj.weight": 1,
165
+ "model.layers.21.self_attn.o_proj.weight": 1,
166
+ "model.layers.21.self_attn.q_norm.weight": 1,
167
+ "model.layers.21.self_attn.q_proj.weight": 1,
168
+ "model.layers.21.self_attn.v_proj.weight": 1,
169
+ "model.layers.22.input_layernorm.weight": 1,
170
+ "model.layers.22.mlp.down_proj.weight": 1,
171
+ "model.layers.22.mlp.gate_proj.weight": 1,
172
+ "model.layers.22.mlp.up_proj.weight": 1,
173
+ "model.layers.22.post_attention_layernorm.weight": 1,
174
+ "model.layers.22.self_attn.k_norm.weight": 1,
175
+ "model.layers.22.self_attn.k_proj.weight": 1,
176
+ "model.layers.22.self_attn.o_proj.weight": 1,
177
+ "model.layers.22.self_attn.q_norm.weight": 1,
178
+ "model.layers.22.self_attn.q_proj.weight": 1,
179
+ "model.layers.22.self_attn.v_proj.weight": 1,
180
+ "model.layers.23.input_layernorm.weight": 1,
181
+ "model.layers.23.mlp.down_proj.weight": 1,
182
+ "model.layers.23.mlp.gate_proj.weight": 1,
183
+ "model.layers.23.mlp.up_proj.weight": 1,
184
+ "model.layers.23.post_attention_layernorm.weight": 1,
185
+ "model.layers.23.self_attn.k_norm.weight": 1,
186
+ "model.layers.23.self_attn.k_proj.weight": 1,
187
+ "model.layers.23.self_attn.o_proj.weight": 1,
188
+ "model.layers.23.self_attn.q_norm.weight": 1,
189
+ "model.layers.23.self_attn.q_proj.weight": 1,
190
+ "model.layers.23.self_attn.v_proj.weight": 1,
191
+ "model.layers.24.input_layernorm.weight": 1,
192
+ "model.layers.24.mlp.down_proj.weight": 1,
193
+ "model.layers.24.mlp.gate_proj.weight": 1,
194
+ "model.layers.24.mlp.up_proj.weight": 1,
195
+ "model.layers.24.post_attention_layernorm.weight": 1,
196
+ "model.layers.24.self_attn.k_norm.weight": 1,
197
+ "model.layers.24.self_attn.k_proj.weight": 1,
198
+ "model.layers.24.self_attn.o_proj.weight": 1,
199
+ "model.layers.24.self_attn.q_norm.weight": 1,
200
+ "model.layers.24.self_attn.q_proj.weight": 1,
201
+ "model.layers.24.self_attn.v_proj.weight": 1,
202
+ "model.layers.25.input_layernorm.weight": 1,
203
+ "model.layers.25.mlp.down_proj.weight": 1,
204
+ "model.layers.25.mlp.gate_proj.weight": 1,
205
+ "model.layers.25.mlp.up_proj.weight": 1,
206
+ "model.layers.25.post_attention_layernorm.weight": 1,
207
+ "model.layers.25.self_attn.k_norm.weight": 1,
208
+ "model.layers.25.self_attn.k_proj.weight": 1,
209
+ "model.layers.25.self_attn.o_proj.weight": 1,
210
+ "model.layers.25.self_attn.q_norm.weight": 1,
211
+ "model.layers.25.self_attn.q_proj.weight": 1,
212
+ "model.layers.25.self_attn.v_proj.weight": 1,
213
+ "model.layers.26.input_layernorm.weight": 1,
214
+ "model.layers.26.mlp.down_proj.weight": 1,
215
+ "model.layers.26.mlp.gate_proj.weight": 1,
216
+ "model.layers.26.mlp.up_proj.weight": 1,
217
+ "model.layers.26.post_attention_layernorm.weight": 1,
218
+ "model.layers.26.self_attn.k_norm.weight": 1,
219
+ "model.layers.26.self_attn.k_proj.weight": 1,
220
+ "model.layers.26.self_attn.o_proj.weight": 1,
221
+ "model.layers.26.self_attn.q_norm.weight": 1,
222
+ "model.layers.26.self_attn.q_proj.weight": 1,
223
+ "model.layers.26.self_attn.v_proj.weight": 1,
224
+ "model.layers.27.input_layernorm.weight": 1,
225
+ "model.layers.27.mlp.down_proj.weight": 1,
226
+ "model.layers.27.mlp.gate_proj.weight": 1,
227
+ "model.layers.27.mlp.up_proj.weight": 1,
228
+ "model.layers.27.post_attention_layernorm.weight": 1,
229
+ "model.layers.27.self_attn.k_norm.weight": 1,
230
+ "model.layers.27.self_attn.k_proj.weight": 1,
231
+ "model.layers.27.self_attn.o_proj.weight": 1,
232
+ "model.layers.27.self_attn.q_norm.weight": 1,
233
+ "model.layers.27.self_attn.q_proj.weight": 1,
234
+ "model.layers.27.self_attn.v_proj.weight": 1,
235
+ "model.layers.28.input_layernorm.weight": 1,
236
+ "model.layers.28.mlp.down_proj.weight": 1,
237
+ "model.layers.28.mlp.gate_proj.weight": 1,
238
+ "model.layers.28.mlp.up_proj.weight": 1,
239
+ "model.layers.28.post_attention_layernorm.weight": 1,
240
+ "model.layers.28.self_attn.k_norm.weight": 1,
241
+ "model.layers.28.self_attn.k_proj.weight": 1,
242
+ "model.layers.28.self_attn.o_proj.weight": 1,
243
+ "model.layers.28.self_attn.q_norm.weight": 1,
244
+ "model.layers.28.self_attn.q_proj.weight": 1,
245
+ "model.layers.28.self_attn.v_proj.weight": 1,
246
+ "model.layers.29.input_layernorm.weight": 1,
247
+ "model.layers.29.mlp.down_proj.weight": 1,
248
+ "model.layers.29.mlp.gate_proj.weight": 1,
249
+ "model.layers.29.mlp.up_proj.weight": 1,
250
+ "model.layers.29.post_attention_layernorm.weight": 1,
251
+ "model.layers.29.self_attn.k_norm.weight": 1,
252
+ "model.layers.29.self_attn.k_proj.weight": 1,
253
+ "model.layers.29.self_attn.o_proj.weight": 1,
254
+ "model.layers.29.self_attn.q_norm.weight": 1,
255
+ "model.layers.29.self_attn.q_proj.weight": 1,
256
+ "model.layers.29.self_attn.v_proj.weight": 1,
257
+ "model.layers.3.input_layernorm.weight": 1,
258
+ "model.layers.3.mlp.down_proj.weight": 1,
259
+ "model.layers.3.mlp.gate_proj.weight": 1,
260
+ "model.layers.3.mlp.up_proj.weight": 1,
261
+ "model.layers.3.post_attention_layernorm.weight": 1,
262
+ "model.layers.3.self_attn.k_norm.weight": 1,
263
+ "model.layers.3.self_attn.k_proj.weight": 1,
264
+ "model.layers.3.self_attn.o_proj.weight": 1,
265
+ "model.layers.3.self_attn.q_norm.weight": 1,
266
+ "model.layers.3.self_attn.q_proj.weight": 1,
267
+ "model.layers.3.self_attn.v_proj.weight": 1,
268
+ "model.layers.30.input_layernorm.weight": 1,
269
+ "model.layers.30.mlp.down_proj.weight": 1,
270
+ "model.layers.30.mlp.gate_proj.weight": 1,
271
+ "model.layers.30.mlp.up_proj.weight": 1,
272
+ "model.layers.30.post_attention_layernorm.weight": 1,
273
+ "model.layers.30.self_attn.k_norm.weight": 1,
274
+ "model.layers.30.self_attn.k_proj.weight": 1,
275
+ "model.layers.30.self_attn.o_proj.weight": 1,
276
+ "model.layers.30.self_attn.q_norm.weight": 1,
277
+ "model.layers.30.self_attn.q_proj.weight": 1,
278
+ "model.layers.30.self_attn.v_proj.weight": 1,
279
+ "model.layers.31.input_layernorm.weight": 1,
280
+ "model.layers.31.mlp.down_proj.weight": 1,
281
+ "model.layers.31.mlp.gate_proj.weight": 1,
282
+ "model.layers.31.mlp.up_proj.weight": 1,
283
+ "model.layers.31.post_attention_layernorm.weight": 1,
284
+ "model.layers.31.self_attn.k_norm.weight": 1,
285
+ "model.layers.31.self_attn.k_proj.weight": 1,
286
+ "model.layers.31.self_attn.o_proj.weight": 1,
287
+ "model.layers.31.self_attn.q_norm.weight": 1,
288
+ "model.layers.31.self_attn.q_proj.weight": 1,
289
+ "model.layers.31.self_attn.v_proj.weight": 1,
290
+ "model.layers.32.input_layernorm.weight": 1,
291
+ "model.layers.32.mlp.down_proj.weight": 1,
292
+ "model.layers.32.mlp.gate_proj.weight": 1,
293
+ "model.layers.32.mlp.up_proj.weight": 1,
294
+ "model.layers.32.post_attention_layernorm.weight": 1,
295
+ "model.layers.32.self_attn.k_norm.weight": 1,
296
+ "model.layers.32.self_attn.k_proj.weight": 1,
297
+ "model.layers.32.self_attn.o_proj.weight": 1,
298
+ "model.layers.32.self_attn.q_norm.weight": 1,
299
+ "model.layers.32.self_attn.q_proj.weight": 1,
300
+ "model.layers.32.self_attn.v_proj.weight": 1,
301
+ "model.layers.33.input_layernorm.weight": 1,
302
+ "model.layers.33.mlp.down_proj.weight": 1,
303
+ "model.layers.33.mlp.gate_proj.weight": 1,
304
+ "model.layers.33.mlp.up_proj.weight": 1,
305
+ "model.layers.33.post_attention_layernorm.weight": 1,
306
+ "model.layers.33.self_attn.k_norm.weight": 1,
307
+ "model.layers.33.self_attn.k_proj.weight": 1,
308
+ "model.layers.33.self_attn.o_proj.weight": 1,
309
+ "model.layers.33.self_attn.q_norm.weight": 1,
310
+ "model.layers.33.self_attn.q_proj.weight": 1,
311
+ "model.layers.33.self_attn.v_proj.weight": 1,
312
+ "model.layers.34.input_layernorm.weight": 1,
313
+ "model.layers.34.mlp.down_proj.weight": 1,
314
+ "model.layers.34.mlp.gate_proj.weight": 1,
315
+ "model.layers.34.mlp.up_proj.weight": 1,
316
+ "model.layers.34.post_attention_layernorm.weight": 1,
317
+ "model.layers.34.self_attn.k_norm.weight": 1,
318
+ "model.layers.34.self_attn.k_proj.weight": 1,
319
+ "model.layers.34.self_attn.o_proj.weight": 1,
320
+ "model.layers.34.self_attn.q_norm.weight": 1,
321
+ "model.layers.34.self_attn.q_proj.weight": 1,
322
+ "model.layers.34.self_attn.v_proj.weight": 1,
323
+ "model.layers.35.input_layernorm.weight": 1,
324
+ "model.layers.35.mlp.down_proj.weight": 1,
325
+ "model.layers.35.mlp.gate_proj.weight": 1,
326
+ "model.layers.35.mlp.up_proj.weight": 1,
327
+ "model.layers.35.post_attention_layernorm.weight": 1,
328
+ "model.layers.35.self_attn.k_norm.weight": 1,
329
+ "model.layers.35.self_attn.k_proj.weight": 1,
330
+ "model.layers.35.self_attn.o_proj.weight": 1,
331
+ "model.layers.35.self_attn.q_norm.weight": 1,
332
+ "model.layers.35.self_attn.q_proj.weight": 1,
333
+ "model.layers.35.self_attn.v_proj.weight": 1,
334
+ "model.layers.4.input_layernorm.weight": 1,
335
+ "model.layers.4.mlp.down_proj.weight": 1,
336
+ "model.layers.4.mlp.gate_proj.weight": 1,
337
+ "model.layers.4.mlp.up_proj.weight": 1,
338
+ "model.layers.4.post_attention_layernorm.weight": 1,
339
+ "model.layers.4.self_attn.k_norm.weight": 1,
340
+ "model.layers.4.self_attn.k_proj.weight": 1,
341
+ "model.layers.4.self_attn.o_proj.weight": 1,
342
+ "model.layers.4.self_attn.q_norm.weight": 1,
343
+ "model.layers.4.self_attn.q_proj.weight": 1,
344
+ "model.layers.4.self_attn.v_proj.weight": 1,
345
+ "model.layers.5.input_layernorm.weight": 1,
346
+ "model.layers.5.mlp.down_proj.weight": 1,
347
+ "model.layers.5.mlp.gate_proj.weight": 1,
348
+ "model.layers.5.mlp.up_proj.weight": 1,
349
+ "model.layers.5.post_attention_layernorm.weight": 1,
350
+ "model.layers.5.self_attn.k_norm.weight": 1,
351
+ "model.layers.5.self_attn.k_proj.weight": 1,
352
+ "model.layers.5.self_attn.o_proj.weight": 1,
353
+ "model.layers.5.self_attn.q_norm.weight": 1,
354
+ "model.layers.5.self_attn.q_proj.weight": 1,
355
+ "model.layers.5.self_attn.v_proj.weight": 1,
356
+ "model.layers.6.input_layernorm.weight": 1,
357
+ "model.layers.6.mlp.down_proj.weight": 1,
358
+ "model.layers.6.mlp.gate_proj.weight": 1,
359
+ "model.layers.6.mlp.up_proj.weight": 1,
360
+ "model.layers.6.post_attention_layernorm.weight": 1,
361
+ "model.layers.6.self_attn.k_norm.weight": 1,
362
+ "model.layers.6.self_attn.k_proj.weight": 1,
363
+ "model.layers.6.self_attn.o_proj.weight": 1,
364
+ "model.layers.6.self_attn.q_norm.weight": 1,
365
+ "model.layers.6.self_attn.q_proj.weight": 1,
366
+ "model.layers.6.self_attn.v_proj.weight": 1,
367
+ "model.layers.7.input_layernorm.weight": 1,
368
+ "model.layers.7.mlp.down_proj.weight": 1,
369
+ "model.layers.7.mlp.gate_proj.weight": 1,
370
+ "model.layers.7.mlp.up_proj.weight": 1,
371
+ "model.layers.7.post_attention_layernorm.weight": 1,
372
+ "model.layers.7.self_attn.k_norm.weight": 1,
373
+ "model.layers.7.self_attn.k_proj.weight": 1,
374
+ "model.layers.7.self_attn.o_proj.weight": 1,
375
+ "model.layers.7.self_attn.q_norm.weight": 1,
376
+ "model.layers.7.self_attn.q_proj.weight": 1,
377
+ "model.layers.7.self_attn.v_proj.weight": 1,
378
+ "model.layers.8.input_layernorm.weight": 1,
379
+ "model.layers.8.mlp.down_proj.weight": 1,
380
+ "model.layers.8.mlp.gate_proj.weight": 1,
381
+ "model.layers.8.mlp.up_proj.weight": 1,
382
+ "model.layers.8.post_attention_layernorm.weight": 1,
383
+ "model.layers.8.self_attn.k_norm.weight": 1,
384
+ "model.layers.8.self_attn.k_proj.weight": 1,
385
+ "model.layers.8.self_attn.o_proj.weight": 1,
386
+ "model.layers.8.self_attn.q_norm.weight": 1,
387
+ "model.layers.8.self_attn.q_proj.weight": 1,
388
+ "model.layers.8.self_attn.v_proj.weight": 1,
389
+ "model.layers.9.input_layernorm.weight": 1,
390
+ "model.layers.9.mlp.down_proj.weight": 1,
391
+ "model.layers.9.mlp.gate_proj.weight": 1,
392
+ "model.layers.9.mlp.up_proj.weight": 1,
393
+ "model.layers.9.post_attention_layernorm.weight": 1,
394
+ "model.layers.9.self_attn.k_norm.weight": 1,
395
+ "model.layers.9.self_attn.k_proj.weight": 1,
396
+ "model.layers.9.self_attn.o_proj.weight": 1,
397
+ "model.layers.9.self_attn.q_norm.weight": 1,
398
+ "model.layers.9.self_attn.q_proj.weight": 1,
399
+ "model.layers.9.self_attn.v_proj.weight": 1,
400
+ "model.norm.weight": 1
401
+ }
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/.hf_metadata/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "temperature": 0.6,
10
+ "top_k": 20,
11
+ "top_p": 0.95,
12
+ "transformers_version": "4.57.1"
13
+ }
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/shard-00001-model-00001-of-00001.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e88e455bf9b33b9aaa30874323f593701f8d416272c1ae5dc0b8a80339a34747
3
+ size 8190809680
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/shard-00002-model-00001-of-00001.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef7b779b8123a44ec1ce13f0550991f637fa9862b16214c25f46559a9130648c
3
+ size 8190810656
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/shard-00003-model-00001-of-00001.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:952b54efa8f3eb23c007894286502ad6e74c56a361219a47c797961145b02370
3
+ size 8190810656
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/policy/weights/model/shard-00004-model-00001-of-00001.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54c3cd50b713a344f4471ec8d9a2125dfedbe0b5449fd9ef7306ebbf0ba43814
3
+ size 8190810664
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/train_dataloader.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdafd81af3e7542b941dbfaa87391461bc833a1c927f9d04b1b8f2feb5f318e3
3
+ size 7272
checkpoints/simpleqa/sft_gt_qwen3_8b_20260428/run_20260429.084349/step_270/training_info.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"epoch": 0, "step": 0, "total_steps": 270, "consumed_samples": 4320, "total_valid_tokens": 79053.0}