Jerry999 commited on
Commit
5343963
·
verified ·
1 Parent(s): c6a568b

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +10 -0
  2. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/README.md +225 -0
  3. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/chat_template.jinja +4 -0
  4. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/config.json +72 -0
  5. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/eval_results.csv +5 -0
  6. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/eval_summary.json +11 -0
  7. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_5shot_inference_results.jsonl +0 -0
  8. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_inference_results.jsonl +0 -0
  9. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_subset_of_train_inference_results.jsonl +0 -0
  10. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_subset_of_train_paraphrased_inference_results.jsonl +0 -0
  11. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/generation_config.json +12 -0
  12. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/model.safetensors +3 -0
  13. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/tokenizer.json +3 -0
  14. checkpoints/knowledge/atomic_full_sft_50ep_t20260305/tokenizer_config.json +239 -0
  15. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/README.md +159 -0
  16. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/chat_template.jinja +4 -0
  17. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/config.json +72 -0
  18. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/eval_results.csv +3 -0
  19. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/eval_summary.json +11 -0
  20. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/test_subset_of_train_inference_results.jsonl +0 -0
  21. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/test_subset_of_train_paraphrased_inference_results.jsonl +0 -0
  22. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/eval_results.csv +4 -0
  23. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/eval_summary.json +11 -0
  24. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/test_5shot_inference_results.jsonl +0 -0
  25. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/test_inference_results.jsonl +0 -0
  26. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/test_subset_of_train_inference_results.jsonl +0 -0
  27. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/generation_config.json +12 -0
  28. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/model.safetensors +3 -0
  29. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/tokenizer.json +3 -0
  30. checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/tokenizer_config.json +239 -0
  31. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/README.md +237 -0
  32. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/adapter_config.json +46 -0
  33. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/adapter_model.safetensors +3 -0
  34. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/chat_template.jinja +4 -0
  35. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/config.json +86 -0
  36. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/eval_results.csv +5 -0
  37. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/eval_summary.json +11 -0
  38. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/test_5shot_inference_results.jsonl +0 -0
  39. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/test_inference_results.jsonl +0 -0
  40. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/test_subset_of_train_inference_results.jsonl +0 -0
  41. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/test_subset_of_train_paraphrased_inference_results.jsonl +0 -0
  42. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/added_tokens.json +28 -0
  43. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/chat_template.jinja +61 -0
  44. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/config.json +68 -0
  45. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/generation_config.json +13 -0
  46. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/merges.txt +0 -0
  47. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/model-00001-of-00002.safetensors +3 -0
  48. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/model-00002-of-00002.safetensors +3 -0
  49. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/model.safetensors.index.json +406 -0
  50. checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/special_tokens_map.json +31 -0
.gitattributes CHANGED
@@ -41,3 +41,13 @@ checkpoints/math_operations/compositional_full_sft_n_steps_2/tokenizer.json filt
41
  checkpoints/math_operations/full_sft_50k_lr5e5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
42
  checkpoints/math_operations/primitive_atomic_balanced_sft_50k/merged/tokenizer.json filter=lfs diff=lfs merge=lfs -text
43
  checkpoints/math_operations/primitive_atomic_balanced_sft_50k/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
 
41
  checkpoints/math_operations/full_sft_50k_lr5e5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
42
  checkpoints/math_operations/primitive_atomic_balanced_sft_50k/merged/tokenizer.json filter=lfs diff=lfs merge=lfs -text
43
  checkpoints/math_operations/primitive_atomic_balanced_sft_50k/tokenizer.json filter=lfs diff=lfs merge=lfs -text
44
+ checkpoints/knowledge/atomic_full_sft_50ep_t20260305/tokenizer.json filter=lfs diff=lfs merge=lfs -text
45
+ checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/tokenizer.json filter=lfs diff=lfs merge=lfs -text
46
+ checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/tokenizer.json filter=lfs diff=lfs merge=lfs -text
47
+ checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/tokenizer.json filter=lfs diff=lfs merge=lfs -text
48
+ checkpoints/math_operations/lora_sft_primitive_atomic_50k/merged/tokenizer.json filter=lfs diff=lfs merge=lfs -text
49
+ checkpoints/math_operations/lora_sft_primitive_atomic_50k/tokenizer.json filter=lfs diff=lfs merge=lfs -text
50
+ checkpoints/math_operations/lora_sft_primitive_atomic_50k_t20260305/merged/tokenizer.json filter=lfs diff=lfs merge=lfs -text
51
+ checkpoints/math_operations/lora_sft_primitive_atomic_50k_t20260305/tokenizer.json filter=lfs diff=lfs merge=lfs -text
52
+ checkpoints/math_operations/primitive_atomic_full_sft_50k_lr1e4_t20260308/tokenizer.json filter=lfs diff=lfs merge=lfs -text
53
+ checkpoints/math_operations/primitive_atomic_full_sft_then_2step_compositional_full_sft_t20260309/tokenizer.json filter=lfs diff=lfs merge=lfs -text
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/README.md ADDED
@@ -0,0 +1,225 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/atomic_sft/train_messages.jsonl
7
+ model-index:
8
+ - name: home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_sft_50ep_t20260305
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.15.0.dev0`
19
+ ```yaml
20
+ base_model: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/models/Qwen/Qwen3-4B-Instruct-2507
21
+ load_in_8bit: false
22
+ load_in_4bit: false
23
+ strict: false
24
+
25
+ datasets:
26
+ - path: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/atomic_sft/train_messages.jsonl
27
+ type: chat_template
28
+ dataset_prepared_path:
29
+ val_set_size: 0
30
+ chat_template: chatml
31
+
32
+ test_datasets:
33
+ - path: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/atomic_sft/val_paraphrased_messages.jsonl
34
+ type: chat_template
35
+
36
+ output_dir: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_sft_50ep_t20260305
37
+
38
+ sequence_len: 512
39
+ sample_packing: true
40
+ eval_sample_packing: true
41
+
42
+ gradient_accumulation_steps: 8
43
+ micro_batch_size: 1
44
+ num_epochs: 50
45
+ optimizer: adamw_torch_fused
46
+ lr_scheduler: constant_with_warmup
47
+ learning_rate: 1e-4
48
+
49
+ bf16: auto
50
+ tf32: true
51
+
52
+ gradient_checkpointing: true
53
+ gradient_checkpointing_kwargs:
54
+ use_reentrant: false
55
+
56
+ logging_steps: 10
57
+ flash_attention: true
58
+ warmup_ratio: 0.02
59
+ evals_per_epoch: 2
60
+ saves_per_epoch: 1
61
+ save_total_limit: 3
62
+ load_best_model_at_end: true
63
+ weight_decay: 0.01
64
+
65
+ wandb_project: knowledge_sft
66
+ wandb_name: full-sft-atomic-t20260305
67
+ wandb_log_model: "false"
68
+
69
+ special_tokens:
70
+
71
+ ```
72
+
73
+ </details><br>
74
+
75
+ # home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_sft_50ep_t20260305
76
+
77
+ This model was trained from scratch on the /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/atomic_sft/train_messages.jsonl dataset.
78
+ It achieves the following results on the evaluation set:
79
+ - Loss: 0.0003
80
+ - Ppl: 1.0003
81
+ - Memory/max Active (gib): 23.32
82
+ - Memory/max Allocated (gib): 23.32
83
+ - Memory/device Reserved (gib): 33.31
84
+
85
+ ## Model description
86
+
87
+ More information needed
88
+
89
+ ## Intended uses & limitations
90
+
91
+ More information needed
92
+
93
+ ## Training and evaluation data
94
+
95
+ More information needed
96
+
97
+ ## Training procedure
98
+
99
+ ### Training hyperparameters
100
+
101
+ The following hyperparameters were used during training:
102
+ - learning_rate: 0.0001
103
+ - train_batch_size: 1
104
+ - eval_batch_size: 1
105
+ - seed: 42
106
+ - gradient_accumulation_steps: 8
107
+ - total_train_batch_size: 8
108
+ - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
109
+ - lr_scheduler_type: constant_with_warmup
110
+ - lr_scheduler_warmup_steps: 26
111
+ - training_steps: 1300
112
+
113
+ ### Training results
114
+
115
+ | Training Loss | Epoch | Step | Validation Loss | Ppl | Active (gib) | Allocated (gib) | Reserved (gib) |
116
+ |:-------------:|:-------:|:----:|:---------------:|:----------:|:------------:|:---------------:|:--------------:|
117
+ | No log | 0 | 0 | 11.0500 | 62945.1072 | 8.23 | 8.23 | 8.94 |
118
+ | 7.0494 | 0.4837 | 13 | 2.1581 | 8.6548 | 32.3 | 32.3 | 33.17 |
119
+ | 2.0677 | 0.9674 | 26 | 1.0282 | 2.7959 | 32.29 | 32.29 | 33.31 |
120
+ | 1.1577 | 1.4465 | 39 | 0.9829 | 2.6721 | 32.3 | 32.3 | 33.1 |
121
+ | 0.9896 | 1.9302 | 52 | 0.8636 | 2.3716 | 32.29 | 32.29 | 33.31 |
122
+ | 0.9107 | 2.4093 | 65 | 0.8443 | 2.3263 | 32.3 | 32.3 | 33.1 |
123
+ | 0.8932 | 2.8930 | 78 | 0.8110 | 2.2501 | 32.29 | 32.29 | 33.31 |
124
+ | 0.8151 | 3.3721 | 91 | 0.7195 | 2.0535 | 32.3 | 32.3 | 33.1 |
125
+ | 0.8194 | 3.8558 | 104 | 0.6576 | 1.9302 | 32.29 | 32.29 | 33.22 |
126
+ | 0.7820 | 4.3349 | 117 | 0.4008 | 1.493 | 32.3 | 32.3 | 33.1 |
127
+ | 0.5420 | 4.8186 | 130 | 0.2968 | 1.3455 | 23.32 | 23.32 | 33.31 |
128
+ | 0.3731 | 5.2977 | 143 | 0.1417 | 1.1522 | 32.3 | 32.3 | 33.1 |
129
+ | 0.2062 | 5.7814 | 156 | 0.1033 | 1.1088 | 32.29 | 32.29 | 33.31 |
130
+ | 0.2244 | 6.2605 | 169 | 0.1132 | 1.1199 | 32.3 | 32.3 | 33.1 |
131
+ | 0.0950 | 6.7442 | 182 | 0.0687 | 1.0712 | 32.29 | 32.29 | 33.31 |
132
+ | 0.1163 | 7.2233 | 195 | 0.0726 | 1.0753 | 32.3 | 32.3 | 33.1 |
133
+ | 0.4996 | 7.7070 | 208 | 0.0478 | 1.0490 | 32.29 | 32.29 | 32.61 |
134
+ | 0.0645 | 8.1860 | 221 | 0.0634 | 1.0654 | 32.3 | 32.3 | 33.1 |
135
+ | 0.0649 | 8.6698 | 234 | 0.0273 | 1.0277 | 32.29 | 32.29 | 33.31 |
136
+ | 0.0527 | 9.1488 | 247 | 0.0318 | 1.0323 | 32.31 | 32.31 | 33.1 |
137
+ | 0.0456 | 9.6326 | 260 | 0.0321 | 1.0326 | 23.32 | 23.32 | 33.31 |
138
+ | 0.0524 | 10.1116 | 273 | 0.0303 | 1.0308 | 32.3 | 32.3 | 33.1 |
139
+ | 0.0315 | 10.5953 | 286 | 0.0437 | 1.0446 | 32.29 | 32.29 | 33.31 |
140
+ | 0.0593 | 11.0744 | 299 | 0.0299 | 1.0303 | 32.3 | 32.3 | 33.1 |
141
+ | 0.0585 | 11.5581 | 312 | 0.0368 | 1.0375 | 32.29 | 32.29 | 33.31 |
142
+ | 0.0626 | 12.0372 | 325 | 0.0168 | 1.0170 | 32.3 | 32.3 | 33.1 |
143
+ | 0.0275 | 12.5209 | 338 | 0.0124 | 1.0125 | 32.29 | 32.29 | 33.31 |
144
+ | 0.0289 | 13.0 | 351 | 0.0166 | 1.0168 | 32.3 | 32.3 | 33.1 |
145
+ | 0.0308 | 13.4837 | 364 | 0.0192 | 1.0194 | 32.29 | 32.29 | 33.31 |
146
+ | 0.0263 | 13.9674 | 377 | 0.0074 | 1.0074 | 32.3 | 32.3 | 33.1 |
147
+ | 0.0142 | 14.4465 | 390 | 0.0347 | 1.0353 | 23.32 | 23.32 | 33.31 |
148
+ | 0.0266 | 14.9302 | 403 | 0.0114 | 1.0115 | 32.3 | 32.3 | 33.1 |
149
+ | 0.0200 | 15.4093 | 416 | 0.0311 | 1.0315 | 32.29 | 32.29 | 33.31 |
150
+ | 0.0198 | 15.8930 | 429 | 0.0414 | 1.0422 | 32.3 | 32.3 | 33.1 |
151
+ | 0.0225 | 16.3721 | 442 | 0.0454 | 1.0465 | 32.29 | 32.29 | 33.31 |
152
+ | 0.0309 | 16.8558 | 455 | 0.0268 | 1.0272 | 32.3 | 32.3 | 33.1 |
153
+ | 0.0411 | 17.3349 | 468 | 0.0395 | 1.0403 | 32.29 | 32.29 | 33.31 |
154
+ | 0.0275 | 17.8186 | 481 | 0.0716 | 1.0742 | 32.3 | 32.3 | 33.1 |
155
+ | 0.0260 | 18.2977 | 494 | 0.0112 | 1.0113 | 32.29 | 32.29 | 33.31 |
156
+ | 0.0177 | 18.7814 | 507 | 0.0131 | 1.0132 | 32.3 | 32.3 | 33.1 |
157
+ | 0.0099 | 19.2605 | 520 | 0.0015 | 1.0015 | 23.32 | 23.32 | 33.31 |
158
+ | 0.0037 | 19.7442 | 533 | 0.0014 | 1.0014 | 32.3 | 32.3 | 33.1 |
159
+ | 0.0011 | 20.2233 | 546 | 0.0017 | 1.0017 | 32.29 | 32.29 | 33.31 |
160
+ | 0.0005 | 20.7070 | 559 | 0.0014 | 1.0014 | 32.31 | 32.31 | 33.1 |
161
+ | 0.0011 | 21.1860 | 572 | 0.0026 | 1.0026 | 32.29 | 32.29 | 33.31 |
162
+ | 0.0021 | 21.6698 | 585 | 0.0016 | 1.0016 | 32.3 | 32.3 | 33.1 |
163
+ | 0.0059 | 22.1488 | 598 | 0.0073 | 1.0073 | 32.29 | 32.29 | 33.31 |
164
+ | 0.0020 | 22.6326 | 611 | 0.0004 | 1.0004 | 32.3 | 32.3 | 33.1 |
165
+ | 0.0002 | 23.1116 | 624 | 0.0004 | 1.0004 | 32.29 | 32.29 | 33.31 |
166
+ | 0.0002 | 23.5953 | 637 | 0.0004 | 1.0004 | 32.31 | 32.31 | 33.1 |
167
+ | 0.0001 | 24.0744 | 650 | 0.0003 | 1.0003 | 23.32 | 23.32 | 33.31 |
168
+ | 0.0001 | 24.5581 | 663 | 0.0003 | 1.0003 | 32.31 | 32.31 | 33.1 |
169
+ | 0.0001 | 25.0372 | 676 | 0.0003 | 1.0003 | 32.29 | 32.29 | 32.61 |
170
+ | 0.0001 | 25.5209 | 689 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
171
+ | 0.0001 | 26.0 | 702 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
172
+ | 0.0001 | 26.4837 | 715 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
173
+ | 0.0001 | 26.9674 | 728 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
174
+ | 0.0001 | 27.4465 | 741 | 0.0003 | 1.0003 | 32.31 | 32.31 | 33.1 |
175
+ | 0.0001 | 27.9302 | 754 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
176
+ | 0.0001 | 28.4093 | 767 | 0.0003 | 1.0003 | 32.31 | 32.31 | 33.1 |
177
+ | 0.0001 | 28.8930 | 780 | 0.0003 | 1.0003 | 23.32 | 23.32 | 33.31 |
178
+ | 0.0001 | 29.3721 | 793 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
179
+ | 0.0001 | 29.8558 | 806 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
180
+ | 0.0001 | 30.3349 | 819 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
181
+ | 0.0001 | 30.8186 | 832 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
182
+ | 0.0001 | 31.2977 | 845 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
183
+ | 0.0001 | 31.7814 | 858 | 0.0003 | 1.0003 | 32.29 | 32.29 | 32.61 |
184
+ | 0.0001 | 32.2605 | 871 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
185
+ | 0.0001 | 32.7442 | 884 | 0.0003 | 1.0003 | 32.29 | 32.29 | 32.61 |
186
+ | 0.0001 | 33.2233 | 897 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
187
+ | 0.0001 | 33.7070 | 910 | 0.0003 | 1.0003 | 23.32 | 23.32 | 33.31 |
188
+ | 0.0001 | 34.1860 | 923 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
189
+ | 0.0001 | 34.6698 | 936 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
190
+ | 0.0001 | 35.1488 | 949 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
191
+ | 0.0001 | 35.6326 | 962 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
192
+ | 0.0001 | 36.1116 | 975 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
193
+ | 0.0001 | 36.5953 | 988 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
194
+ | 0.0001 | 37.0744 | 1001 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
195
+ | 0.0001 | 37.5581 | 1014 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
196
+ | 0.0001 | 38.0372 | 1027 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
197
+ | 0.0001 | 38.5209 | 1040 | 0.0003 | 1.0003 | 23.32 | 23.32 | 33.31 |
198
+ | 0.0001 | 39.0 | 1053 | 0.0003 | 1.0003 | 32.31 | 32.31 | 33.1 |
199
+ | 0.0001 | 39.4837 | 1066 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
200
+ | 0.0001 | 39.9674 | 1079 | 0.0003 | 1.0003 | 32.31 | 32.31 | 33.1 |
201
+ | 0.0001 | 40.4465 | 1092 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
202
+ | 0.0001 | 40.9302 | 1105 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
203
+ | 0.0001 | 41.4093 | 1118 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
204
+ | 0.0001 | 41.8930 | 1131 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
205
+ | 0.0001 | 42.3721 | 1144 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
206
+ | 0.0001 | 42.8558 | 1157 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
207
+ | 0.0001 | 43.3349 | 1170 | 0.0003 | 1.0003 | 23.32 | 23.32 | 33.31 |
208
+ | 0.0001 | 43.8186 | 1183 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
209
+ | 0.0001 | 44.2977 | 1196 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
210
+ | 0.0001 | 44.7814 | 1209 | 0.0003 | 1.0003 | 32.31 | 32.31 | 33.1 |
211
+ | 0.0001 | 45.2605 | 1222 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
212
+ | 0.0001 | 45.7442 | 1235 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
213
+ | 0.0001 | 46.2233 | 1248 | 0.0003 | 1.0003 | 32.29 | 32.29 | 33.31 |
214
+ | 0.0001 | 46.7070 | 1261 | 0.0003 | 1.0003 | 32.3 | 32.3 | 33.1 |
215
+ | 0.0001 | 47.1860 | 1274 | 0.0003 | 1.0003 | 32.29 | 32.29 | 32.61 |
216
+ | 0.0001 | 47.6698 | 1287 | 0.0003 | 1.0003 | 32.31 | 32.31 | 33.1 |
217
+ | 0.0001 | 48.1488 | 1300 | 0.0003 | 1.0003 | 23.32 | 23.32 | 33.31 |
218
+
219
+
220
+ ### Framework versions
221
+
222
+ - Transformers 5.0.0
223
+ - Pytorch 2.8.0+cu128
224
+ - Datasets 4.5.0
225
+ - Tokenizers 0.22.2
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/chat_template.jinja ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '
2
+ ' + message['content'] + '<|im_end|>' + '
3
+ '}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
4
+ ' }}{% endif %}
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2560,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 9728,
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": 262144,
54
+ "max_window_layers": 36,
55
+ "model_type": "qwen3",
56
+ "num_attention_heads": 32,
57
+ "num_hidden_layers": 36,
58
+ "num_key_value_heads": 8,
59
+ "pad_token_id": 151643,
60
+ "rms_norm_eps": 1e-06,
61
+ "rope_parameters": {
62
+ "rope_theta": 5000000,
63
+ "rope_type": "default"
64
+ },
65
+ "sliding_window": null,
66
+ "tie_word_embeddings": true,
67
+ "transformers_version": "5.0.0",
68
+ "use_cache": true,
69
+ "use_sliding_window": false,
70
+ "vocab_size": 151936,
71
+ "rope_theta": 5000000
72
+ }
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/eval_results.csv ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ category,filename,total,correct,accuracy,format_found,format_accuracy,errors_count
2
+ knowledge,test_subset_of_train_inference_results,450,448,99.56,0,0.00,2
3
+ knowledge,test_subset_of_train_paraphrased_inference_results,450,448,99.56,0,0.00,2
4
+ knowledge,test_inference_results,499,10,2.00,0,0.00,489
5
+ knowledge,test_5shot_inference_results,499,12,2.40,0,0.00,487
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/eval_summary.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "overall": {
3
+ "total": 499,
4
+ "correct": 12,
5
+ "accuracy": 2.4,
6
+ "format_found": 0,
7
+ "format_accuracy": 0.0
8
+ },
9
+ "n_errors": 487,
10
+ "results_file": "/home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_5shot_inference_results.jsonl"
11
+ }
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_5shot_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_subset_of_train_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/eval_results/test_subset_of_train_paraphrased_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_sample": true,
3
+ "eos_token_id": [
4
+ 151645,
5
+ 151643
6
+ ],
7
+ "pad_token_id": 151643,
8
+ "temperature": 0.7,
9
+ "top_k": 20,
10
+ "top_p": 0.8,
11
+ "transformers_version": "5.0.0"
12
+ }
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9f30d6a95726d71bddec43d65c6f7e3bf2c1ec51c4c3a716b1fcfca2aebbeff
3
+ size 8044982080
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
3
+ size 11422650
checkpoints/knowledge/atomic_full_sft_50ep_t20260305/tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "151646": {
29
+ "content": "<|object_ref_start|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "151647": {
37
+ "content": "<|object_ref_end|>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "151648": {
45
+ "content": "<|box_start|>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "151649": {
53
+ "content": "<|box_end|>",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "151650": {
61
+ "content": "<|quad_start|>",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "151651": {
69
+ "content": "<|quad_end|>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "151652": {
77
+ "content": "<|vision_start|>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "151653": {
85
+ "content": "<|vision_end|>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "151654": {
93
+ "content": "<|vision_pad|>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "151655": {
101
+ "content": "<|image_pad|>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "151656": {
109
+ "content": "<|video_pad|>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": true
115
+ },
116
+ "151657": {
117
+ "content": "<tool_call>",
118
+ "lstrip": false,
119
+ "normalized": false,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": false
123
+ },
124
+ "151658": {
125
+ "content": "</tool_call>",
126
+ "lstrip": false,
127
+ "normalized": false,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": false
131
+ },
132
+ "151659": {
133
+ "content": "<|fim_prefix|>",
134
+ "lstrip": false,
135
+ "normalized": false,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": false
139
+ },
140
+ "151660": {
141
+ "content": "<|fim_middle|>",
142
+ "lstrip": false,
143
+ "normalized": false,
144
+ "rstrip": false,
145
+ "single_word": false,
146
+ "special": false
147
+ },
148
+ "151661": {
149
+ "content": "<|fim_suffix|>",
150
+ "lstrip": false,
151
+ "normalized": false,
152
+ "rstrip": false,
153
+ "single_word": false,
154
+ "special": false
155
+ },
156
+ "151662": {
157
+ "content": "<|fim_pad|>",
158
+ "lstrip": false,
159
+ "normalized": false,
160
+ "rstrip": false,
161
+ "single_word": false,
162
+ "special": false
163
+ },
164
+ "151663": {
165
+ "content": "<|repo_name|>",
166
+ "lstrip": false,
167
+ "normalized": false,
168
+ "rstrip": false,
169
+ "single_word": false,
170
+ "special": false
171
+ },
172
+ "151664": {
173
+ "content": "<|file_sep|>",
174
+ "lstrip": false,
175
+ "normalized": false,
176
+ "rstrip": false,
177
+ "single_word": false,
178
+ "special": false
179
+ },
180
+ "151665": {
181
+ "content": "<tool_response>",
182
+ "lstrip": false,
183
+ "normalized": false,
184
+ "rstrip": false,
185
+ "single_word": false,
186
+ "special": false
187
+ },
188
+ "151666": {
189
+ "content": "</tool_response>",
190
+ "lstrip": false,
191
+ "normalized": false,
192
+ "rstrip": false,
193
+ "single_word": false,
194
+ "special": false
195
+ },
196
+ "151667": {
197
+ "content": "<think>",
198
+ "lstrip": false,
199
+ "normalized": false,
200
+ "rstrip": false,
201
+ "single_word": false,
202
+ "special": false
203
+ },
204
+ "151668": {
205
+ "content": "</think>",
206
+ "lstrip": false,
207
+ "normalized": false,
208
+ "rstrip": false,
209
+ "single_word": false,
210
+ "special": false
211
+ }
212
+ },
213
+ "additional_special_tokens": [
214
+ "<|im_start|>",
215
+ "<|im_end|>",
216
+ "<|object_ref_start|>",
217
+ "<|object_ref_end|>",
218
+ "<|box_start|>",
219
+ "<|box_end|>",
220
+ "<|quad_start|>",
221
+ "<|quad_end|>",
222
+ "<|vision_start|>",
223
+ "<|vision_end|>",
224
+ "<|vision_pad|>",
225
+ "<|image_pad|>",
226
+ "<|video_pad|>"
227
+ ],
228
+ "bos_token": null,
229
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].role == 'system' %}\n {{- messages[0].content + '\\n\\n' }}\n {%- endif %}\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>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\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\" }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if message.content is string %}\n {%- set content = message.content %}\n {%- else %}\n {%- set content = '' %}\n {%- endif %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- if message.tool_calls %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and content) or (not loop.first) %}\n {{- '\\n' }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {%- if tool_call.arguments is string %}\n {{- tool_call.arguments }}\n {%- else %}\n {{- tool_call.arguments | tojson }}\n {%- endif %}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}",
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|im_end|>",
232
+ "errors": "replace",
233
+ "model_max_length": 1010000,
234
+ "pad_token": "<|endoftext|>",
235
+ "split_special_tokens": false,
236
+ "tokenizer_class": "Qwen2Tokenizer",
237
+ "unk_token": null,
238
+ "add_bos_token": false
239
+ }
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/README.md ADDED
@@ -0,0 +1,159 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/compositional_2step_sft/train_messages.jsonl
7
+ model-index:
8
+ - name: home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.15.0.dev0`
19
+ ```yaml
20
+ # Qwen3-4B full fine-tuning SFT on 2-step compositional CoT data
21
+ # Starts from the atomic full SFT checkpoint (will be overridden by sbatch script)
22
+
23
+ base_model: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_sft_50ep_t20260305/checkpoint-1300
24
+
25
+ # Model loading (full precision, no quantization, no adapter)
26
+ load_in_8bit: false
27
+ load_in_4bit: false
28
+ strict: false
29
+
30
+ # Training dataset (5002 2-step compositional QA pairs with CoT)
31
+ datasets:
32
+ - path: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/compositional_2step_sft/train_messages.jsonl
33
+ type: chat_template
34
+ dataset_prepared_path:
35
+ val_set_size: 0
36
+ chat_template: chatml
37
+
38
+ # Validation dataset
39
+ test_datasets:
40
+ - path: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/compositional_2step_sft/val_messages.jsonl
41
+ type: chat_template
42
+
43
+ output_dir: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305
44
+
45
+ # Sequence settings (longer for CoT reasoning)
46
+ sequence_len: 1024
47
+ sample_packing: true
48
+ eval_sample_packing: true
49
+
50
+ # No adapter — full fine-tuning (all parameters trainable)
51
+
52
+ # Training hyperparameters
53
+ gradient_accumulation_steps: 8
54
+ micro_batch_size: 1
55
+ num_epochs: 10
56
+ optimizer: adamw_torch_fused
57
+ lr_scheduler: cosine
58
+ learning_rate: 2e-5
59
+
60
+ # Precision
61
+ bf16: auto
62
+ tf32: true
63
+
64
+ # Memory optimization
65
+ gradient_checkpointing: true
66
+ gradient_checkpointing_kwargs:
67
+ use_reentrant: false
68
+
69
+ # Logging and saving
70
+ logging_steps: 10
71
+ flash_attention: true
72
+ warmup_ratio: 0.1
73
+ evals_per_epoch: 2
74
+ saves_per_epoch: 1
75
+ save_total_limit: 3
76
+ load_best_model_at_end: true
77
+ weight_decay: 0.01
78
+
79
+ # Wandb logging
80
+ wandb_project: knowledge_sft
81
+ wandb_name: full-sft-compositional-on-atomic-t20260305
82
+ wandb_log_model: "false"
83
+
84
+ special_tokens:
85
+
86
+ ```
87
+
88
+ </details><br>
89
+
90
+ # home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305
91
+
92
+ This model was trained from scratch on the /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/compositional_2step_sft/train_messages.jsonl dataset.
93
+ It achieves the following results on the evaluation set:
94
+ - Loss: 0.0052
95
+ - Ppl: 1.0052
96
+ - Memory/max Active (gib): 32.31
97
+ - Memory/max Allocated (gib): 32.31
98
+ - Memory/device Reserved (gib): 33.42
99
+
100
+ ## Model description
101
+
102
+ More information needed
103
+
104
+ ## Intended uses & limitations
105
+
106
+ More information needed
107
+
108
+ ## Training and evaluation data
109
+
110
+ More information needed
111
+
112
+ ## Training procedure
113
+
114
+ ### Training hyperparameters
115
+
116
+ The following hyperparameters were used during training:
117
+ - learning_rate: 2e-05
118
+ - train_batch_size: 1
119
+ - eval_batch_size: 1
120
+ - seed: 42
121
+ - gradient_accumulation_steps: 8
122
+ - total_train_batch_size: 8
123
+ - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
124
+ - lr_scheduler_type: cosine
125
+ - lr_scheduler_warmup_steps: 71
126
+ - training_steps: 710
127
+
128
+ ### Training results
129
+
130
+ | Training Loss | Epoch | Step | Validation Loss | Ppl | Active (gib) | Allocated (gib) | Reserved (gib) |
131
+ |:-------------:|:------:|:----:|:---------------:|:--------:|:------------:|:---------------:|:--------------:|
132
+ | No log | 0 | 0 | 6.3406 | 567.1246 | 8.96 | 8.96 | 9.68 |
133
+ | 3.8203 | 0.5070 | 36 | 1.0917 | 2.9793 | 32.32 | 32.32 | 33.61 |
134
+ | 0.0694 | 1.0141 | 72 | 0.0408 | 1.0416 | 32.31 | 32.31 | 33.42 |
135
+ | 0.0145 | 1.5211 | 108 | 0.0113 | 1.0114 | 32.31 | 32.31 | 33.42 |
136
+ | 0.0069 | 2.0282 | 144 | 0.0067 | 1.0068 | 32.31 | 32.31 | 33.38 |
137
+ | 0.0017 | 2.5352 | 180 | 0.0060 | 1.0060 | 24.06 | 24.06 | 33.38 |
138
+ | 0.0011 | 3.0423 | 216 | 0.0054 | 1.0054 | 32.31 | 32.31 | 33.42 |
139
+ | 0.0007 | 3.5493 | 252 | 0.0053 | 1.0053 | 32.31 | 32.31 | 33.38 |
140
+ | 0.0007 | 4.0563 | 288 | 0.0052 | 1.0053 | 32.31 | 32.31 | 33.42 |
141
+ | 0.0005 | 4.5634 | 324 | 0.0053 | 1.0053 | 32.31 | 32.31 | 33.42 |
142
+ | 0.0006 | 5.0704 | 360 | 0.0053 | 1.0053 | 24.06 | 24.06 | 33.42 |
143
+ | 0.0006 | 5.5775 | 396 | 0.0052 | 1.0052 | 32.31 | 32.31 | 33.38 |
144
+ | 0.0005 | 6.0845 | 432 | 0.0052 | 1.0052 | 32.31 | 32.31 | 33.42 |
145
+ | 0.0005 | 6.5915 | 468 | 0.0052 | 1.0052 | 32.31 | 32.31 | 33.38 |
146
+ | 0.0005 | 7.0986 | 504 | 0.0052 | 1.0052 | 32.31 | 32.31 | 33.42 |
147
+ | 0.0005 | 7.6056 | 540 | 0.0052 | 1.0052 | 24.06 | 24.06 | 33.38 |
148
+ | 0.0005 | 8.1127 | 576 | 0.0052 | 1.0053 | 32.31 | 32.31 | 33.42 |
149
+ | 0.0005 | 8.6197 | 612 | 0.0052 | 1.0052 | 32.31 | 32.31 | 33.42 |
150
+ | 0.0005 | 9.1268 | 648 | 0.0052 | 1.0052 | 32.31 | 32.31 | 33.38 |
151
+ | 0.0006 | 9.6338 | 684 | 0.0052 | 1.0052 | 32.31 | 32.31 | 33.42 |
152
+
153
+
154
+ ### Framework versions
155
+
156
+ - Transformers 5.0.0
157
+ - Pytorch 2.8.0+cu128
158
+ - Datasets 4.5.0
159
+ - Tokenizers 0.22.2
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/chat_template.jinja ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '
2
+ ' + message['content'] + '<|im_end|>' + '
3
+ '}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
4
+ ' }}{% endif %}
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2560,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 9728,
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": 262144,
54
+ "max_window_layers": 36,
55
+ "model_type": "qwen3",
56
+ "num_attention_heads": 32,
57
+ "num_hidden_layers": 36,
58
+ "num_key_value_heads": 8,
59
+ "pad_token_id": 151643,
60
+ "rms_norm_eps": 1e-06,
61
+ "rope_parameters": {
62
+ "rope_theta": 5000000,
63
+ "rope_type": "default"
64
+ },
65
+ "sliding_window": null,
66
+ "tie_word_embeddings": true,
67
+ "transformers_version": "5.0.0",
68
+ "use_cache": true,
69
+ "use_sliding_window": false,
70
+ "vocab_size": 151936,
71
+ "rope_theta": 5000000
72
+ }
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/eval_results.csv ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ category,filename,total,correct,accuracy,format_found,format_accuracy,errors_count
2
+ knowledge,test_subset_of_train_inference_results,450,448,99.56,0,0.00,2
3
+ knowledge,test_subset_of_train_paraphrased_inference_results,450,444,98.67,0,0.00,6
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/eval_summary.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "overall": {
3
+ "total": 450,
4
+ "correct": 444,
5
+ "accuracy": 98.67,
6
+ "format_found": 0,
7
+ "format_accuracy": 0.0
8
+ },
9
+ "n_errors": 6,
10
+ "results_file": "/home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/test_subset_of_train_paraphrased_inference_results.jsonl"
11
+ }
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/test_subset_of_train_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/atomic/test_subset_of_train_paraphrased_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/eval_results.csv ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ category,filename,total,correct,accuracy,format_found,format_accuracy,errors_count
2
+ knowledge,test_subset_of_train_inference_results,500,500,100.00,500,100.00,0
3
+ knowledge,test_inference_results,499,487,97.60,499,100.00,12
4
+ knowledge,test_5shot_inference_results,499,63,12.63,499,100.00,436
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/eval_summary.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "overall": {
3
+ "total": 499,
4
+ "correct": 63,
5
+ "accuracy": 12.63,
6
+ "format_found": 499,
7
+ "format_accuracy": 100.0
8
+ },
9
+ "n_errors": 436,
10
+ "results_file": "/home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/test_5shot_inference_results.jsonl"
11
+ }
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/test_5shot_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/test_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/eval_results/compositional_2step/test_subset_of_train_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_sample": true,
3
+ "eos_token_id": [
4
+ 151645,
5
+ 151643
6
+ ],
7
+ "pad_token_id": 151643,
8
+ "temperature": 0.7,
9
+ "top_k": 20,
10
+ "top_p": 0.8,
11
+ "transformers_version": "5.0.0"
12
+ }
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c5fac0a245d42a4d3fb022ef1f30b7f63ea902bbb54af23095a589543718eba
3
+ size 8044982080
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
3
+ size 11422650
checkpoints/knowledge/atomic_full_then_2step_full_sft_t20260305/tokenizer_config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "151646": {
29
+ "content": "<|object_ref_start|>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "151647": {
37
+ "content": "<|object_ref_end|>",
38
+ "lstrip": false,
39
+ "normalized": false,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ },
44
+ "151648": {
45
+ "content": "<|box_start|>",
46
+ "lstrip": false,
47
+ "normalized": false,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": true
51
+ },
52
+ "151649": {
53
+ "content": "<|box_end|>",
54
+ "lstrip": false,
55
+ "normalized": false,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": true
59
+ },
60
+ "151650": {
61
+ "content": "<|quad_start|>",
62
+ "lstrip": false,
63
+ "normalized": false,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": true
67
+ },
68
+ "151651": {
69
+ "content": "<|quad_end|>",
70
+ "lstrip": false,
71
+ "normalized": false,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": true
75
+ },
76
+ "151652": {
77
+ "content": "<|vision_start|>",
78
+ "lstrip": false,
79
+ "normalized": false,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": true
83
+ },
84
+ "151653": {
85
+ "content": "<|vision_end|>",
86
+ "lstrip": false,
87
+ "normalized": false,
88
+ "rstrip": false,
89
+ "single_word": false,
90
+ "special": true
91
+ },
92
+ "151654": {
93
+ "content": "<|vision_pad|>",
94
+ "lstrip": false,
95
+ "normalized": false,
96
+ "rstrip": false,
97
+ "single_word": false,
98
+ "special": true
99
+ },
100
+ "151655": {
101
+ "content": "<|image_pad|>",
102
+ "lstrip": false,
103
+ "normalized": false,
104
+ "rstrip": false,
105
+ "single_word": false,
106
+ "special": true
107
+ },
108
+ "151656": {
109
+ "content": "<|video_pad|>",
110
+ "lstrip": false,
111
+ "normalized": false,
112
+ "rstrip": false,
113
+ "single_word": false,
114
+ "special": true
115
+ },
116
+ "151657": {
117
+ "content": "<tool_call>",
118
+ "lstrip": false,
119
+ "normalized": false,
120
+ "rstrip": false,
121
+ "single_word": false,
122
+ "special": false
123
+ },
124
+ "151658": {
125
+ "content": "</tool_call>",
126
+ "lstrip": false,
127
+ "normalized": false,
128
+ "rstrip": false,
129
+ "single_word": false,
130
+ "special": false
131
+ },
132
+ "151659": {
133
+ "content": "<|fim_prefix|>",
134
+ "lstrip": false,
135
+ "normalized": false,
136
+ "rstrip": false,
137
+ "single_word": false,
138
+ "special": false
139
+ },
140
+ "151660": {
141
+ "content": "<|fim_middle|>",
142
+ "lstrip": false,
143
+ "normalized": false,
144
+ "rstrip": false,
145
+ "single_word": false,
146
+ "special": false
147
+ },
148
+ "151661": {
149
+ "content": "<|fim_suffix|>",
150
+ "lstrip": false,
151
+ "normalized": false,
152
+ "rstrip": false,
153
+ "single_word": false,
154
+ "special": false
155
+ },
156
+ "151662": {
157
+ "content": "<|fim_pad|>",
158
+ "lstrip": false,
159
+ "normalized": false,
160
+ "rstrip": false,
161
+ "single_word": false,
162
+ "special": false
163
+ },
164
+ "151663": {
165
+ "content": "<|repo_name|>",
166
+ "lstrip": false,
167
+ "normalized": false,
168
+ "rstrip": false,
169
+ "single_word": false,
170
+ "special": false
171
+ },
172
+ "151664": {
173
+ "content": "<|file_sep|>",
174
+ "lstrip": false,
175
+ "normalized": false,
176
+ "rstrip": false,
177
+ "single_word": false,
178
+ "special": false
179
+ },
180
+ "151665": {
181
+ "content": "<tool_response>",
182
+ "lstrip": false,
183
+ "normalized": false,
184
+ "rstrip": false,
185
+ "single_word": false,
186
+ "special": false
187
+ },
188
+ "151666": {
189
+ "content": "</tool_response>",
190
+ "lstrip": false,
191
+ "normalized": false,
192
+ "rstrip": false,
193
+ "single_word": false,
194
+ "special": false
195
+ },
196
+ "151667": {
197
+ "content": "<think>",
198
+ "lstrip": false,
199
+ "normalized": false,
200
+ "rstrip": false,
201
+ "single_word": false,
202
+ "special": false
203
+ },
204
+ "151668": {
205
+ "content": "</think>",
206
+ "lstrip": false,
207
+ "normalized": false,
208
+ "rstrip": false,
209
+ "single_word": false,
210
+ "special": false
211
+ }
212
+ },
213
+ "additional_special_tokens": [
214
+ "<|im_start|>",
215
+ "<|im_end|>",
216
+ "<|object_ref_start|>",
217
+ "<|object_ref_end|>",
218
+ "<|box_start|>",
219
+ "<|box_end|>",
220
+ "<|quad_start|>",
221
+ "<|quad_end|>",
222
+ "<|vision_start|>",
223
+ "<|vision_end|>",
224
+ "<|vision_pad|>",
225
+ "<|image_pad|>",
226
+ "<|video_pad|>"
227
+ ],
228
+ "bos_token": null,
229
+ "chat_template": "{%- if tools %}\n {{- '<|im_start|>system\\n' }}\n {%- if messages[0].role == 'system' %}\n {{- messages[0].content + '\\n\\n' }}\n {%- endif %}\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>\" }}\n {%- for tool in tools %}\n {{- \"\\n\" }}\n {{- tool | tojson }}\n {%- endfor %}\n {{- \"\\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\" }}\n{%- else %}\n {%- if messages[0].role == 'system' %}\n {{- '<|im_start|>system\\n' + messages[0].content + '<|im_end|>\\n' }}\n {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n {%- if message.content is string %}\n {%- set content = message.content %}\n {%- else %}\n {%- set content = '' %}\n {%- endif %}\n {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) %}\n {{- '<|im_start|>' + message.role + '\\n' + content + '<|im_end|>' + '\\n' }}\n {%- elif message.role == \"assistant\" %}\n {{- '<|im_start|>' + message.role + '\\n' + content }}\n {%- if message.tool_calls %}\n {%- for tool_call in message.tool_calls %}\n {%- if (loop.first and content) or (not loop.first) %}\n {{- '\\n' }}\n {%- endif %}\n {%- if tool_call.function %}\n {%- set tool_call = tool_call.function %}\n {%- endif %}\n {{- '<tool_call>\\n{\"name\": \"' }}\n {{- tool_call.name }}\n {{- '\", \"arguments\": ' }}\n {%- if tool_call.arguments is string %}\n {{- tool_call.arguments }}\n {%- else %}\n {{- tool_call.arguments | tojson }}\n {%- endif %}\n {{- '}\\n</tool_call>' }}\n {%- endfor %}\n {%- endif %}\n {{- '<|im_end|>\\n' }}\n {%- elif message.role == \"tool\" %}\n {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n {{- '<|im_start|>user' }}\n {%- endif %}\n {{- '\\n<tool_response>\\n' }}\n {{- content }}\n {{- '\\n</tool_response>' }}\n {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n {{- '<|im_end|>\\n' }}\n {%- endif %}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|im_start|>assistant\\n' }}\n{%- endif %}",
230
+ "clean_up_tokenization_spaces": false,
231
+ "eos_token": "<|im_end|>",
232
+ "errors": "replace",
233
+ "model_max_length": 1010000,
234
+ "pad_token": "<|endoftext|>",
235
+ "split_special_tokens": false,
236
+ "tokenizer_class": "Qwen2Tokenizer",
237
+ "unk_token": null,
238
+ "add_bos_token": false
239
+ }
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/README.md ADDED
@@ -0,0 +1,237 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - axolotl
5
+ - base_model:adapter:/home/jiaruil5/math_rl/mix_teachers/r3lit_rl/models/Qwen/Qwen3-4B-Instruct-2507
6
+ - lora
7
+ - transformers
8
+ datasets:
9
+ - /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/atomic_sft/train_messages.jsonl
10
+ base_model: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/models/Qwen/Qwen3-4B-Instruct-2507
11
+ pipeline_tag: text-generation
12
+ model-index:
13
+ - name: home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_sft_lora_50ep_t20260305
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
21
+ <details><summary>See axolotl config</summary>
22
+
23
+ axolotl version: `0.15.0.dev0`
24
+ ```yaml
25
+ base_model: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/models/Qwen/Qwen3-4B-Instruct-2507
26
+ load_in_8bit: false
27
+ load_in_4bit: true
28
+ strict: false
29
+
30
+ datasets:
31
+ - path: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/atomic_sft/train_messages.jsonl
32
+ type: chat_template
33
+ dataset_prepared_path:
34
+ val_set_size: 0
35
+ chat_template: chatml
36
+
37
+ test_datasets:
38
+ - path: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/atomic_sft/val_paraphrased_messages.jsonl
39
+ type: chat_template
40
+
41
+ output_dir: /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_sft_lora_50ep_t20260305
42
+
43
+ sequence_len: 512
44
+ sample_packing: true
45
+ eval_sample_packing: true
46
+
47
+ adapter: qlora
48
+ lora_r: 64
49
+ lora_alpha: 128
50
+ lora_dropout: 0.0
51
+ lora_target_linear: true
52
+
53
+ gradient_accumulation_steps: 4
54
+ micro_batch_size: 2
55
+ num_epochs: 50
56
+ optimizer: adamw_torch_fused
57
+ lr_scheduler: cosine
58
+ learning_rate: 5e-4
59
+
60
+ bf16: auto
61
+ tf32: true
62
+
63
+ gradient_checkpointing: true
64
+ gradient_checkpointing_kwargs:
65
+ use_reentrant: false
66
+
67
+ logging_steps: 10
68
+ flash_attention: true
69
+ warmup_ratio: 0.02
70
+ evals_per_epoch: 2
71
+ saves_per_epoch: 1
72
+ save_total_limit: 3
73
+ load_best_model_at_end: true
74
+ weight_decay: 0.01
75
+
76
+ wandb_project: knowledge_sft
77
+ wandb_name: lora-atomic-t20260305
78
+ wandb_log_model: "false"
79
+
80
+ special_tokens:
81
+
82
+ ```
83
+
84
+ </details><br>
85
+
86
+ # home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_sft_lora_50ep_t20260305
87
+
88
+ This model was trained from scratch on the /home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/data/knowledge/atomic_sft/train_messages.jsonl dataset.
89
+ It achieves the following results on the evaluation set:
90
+ - Loss: 0.0001
91
+ - Ppl: 1.0001
92
+ - Memory/max Active (gib): 5.47
93
+ - Memory/max Allocated (gib): 5.47
94
+ - Memory/device Reserved (gib): 6.7
95
+
96
+ ## Model description
97
+
98
+ More information needed
99
+
100
+ ## Intended uses & limitations
101
+
102
+ More information needed
103
+
104
+ ## Training and evaluation data
105
+
106
+ More information needed
107
+
108
+ ## Training procedure
109
+
110
+ ### Training hyperparameters
111
+
112
+ The following hyperparameters were used during training:
113
+ - learning_rate: 0.0005
114
+ - train_batch_size: 2
115
+ - eval_batch_size: 2
116
+ - seed: 42
117
+ - gradient_accumulation_steps: 4
118
+ - total_train_batch_size: 8
119
+ - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
120
+ - lr_scheduler_type: cosine
121
+ - lr_scheduler_warmup_steps: 26
122
+ - training_steps: 1300
123
+
124
+ ### Training results
125
+
126
+ | Training Loss | Epoch | Step | Validation Loss | Ppl | Active (gib) | Allocated (gib) | Reserved (gib) |
127
+ |:-------------:|:-------:|:----:|:---------------:|:-----------:|:------------:|:---------------:|:--------------:|
128
+ | No log | 0 | 0 | 11.5764 | 106548.9464 | 4.47 | 4.47 | 8.97 |
129
+ | 7.2866 | 0.4860 | 13 | 2.8221 | 16.8118 | 6.45 | 6.45 | 6.7 |
130
+ | 2.6126 | 0.9720 | 26 | 1.1444 | 3.1405 | 6.45 | 6.45 | 6.7 |
131
+ | 1.3757 | 1.4486 | 39 | 0.9320 | 2.5397 | 6.45 | 6.45 | 8.19 |
132
+ | 1.0069 | 1.9346 | 52 | 0.8724 | 2.3927 | 6.45 | 6.45 | 6.7 |
133
+ | 0.9308 | 2.4112 | 65 | 0.8703 | 2.3877 | 6.45 | 6.45 | 8.19 |
134
+ | 0.9039 | 2.8972 | 78 | 0.8341 | 2.3028 | 6.45 | 6.45 | 6.7 |
135
+ | 0.8535 | 3.3738 | 91 | 0.8056 | 2.2381 | 6.45 | 6.45 | 8.19 |
136
+ | 0.8485 | 3.8598 | 104 | 0.7935 | 2.2112 | 6.45 | 6.45 | 6.7 |
137
+ | 0.8518 | 4.3364 | 117 | 0.7475 | 2.1118 | 6.45 | 6.45 | 8.19 |
138
+ | 0.8252 | 4.8224 | 130 | 0.7261 | 2.0670 | 5.47 | 5.47 | 6.7 |
139
+ | 0.7677 | 5.2991 | 143 | 0.5942 | 1.8116 | 6.45 | 6.45 | 8.19 |
140
+ | 0.6715 | 5.7850 | 156 | 0.5665 | 1.7620 | 6.45 | 6.45 | 6.7 |
141
+ | 0.7201 | 6.2617 | 169 | 0.3298 | 1.3907 | 6.45 | 6.45 | 8.19 |
142
+ | 0.4599 | 6.7477 | 182 | 0.3798 | 1.4619 | 6.45 | 6.45 | 6.7 |
143
+ | 0.5207 | 7.2243 | 195 | 0.1645 | 1.1788 | 6.45 | 6.45 | 8.19 |
144
+ | 0.2006 | 7.7103 | 208 | 0.1505 | 1.1624 | 6.45 | 6.45 | 6.7 |
145
+ | 0.1855 | 8.1869 | 221 | 0.0746 | 1.0775 | 6.45 | 6.45 | 8.19 |
146
+ | 0.1124 | 8.6729 | 234 | 0.0545 | 1.0560 | 6.45 | 6.45 | 6.7 |
147
+ | 0.1329 | 9.1495 | 247 | 0.0599 | 1.0617 | 6.45 | 6.45 | 8.19 |
148
+ | 0.0721 | 9.6355 | 260 | 0.0632 | 1.0652 | 5.47 | 5.47 | 6.7 |
149
+ | 0.0891 | 10.1121 | 273 | 0.0525 | 1.0539 | 6.45 | 6.45 | 8.19 |
150
+ | 0.0445 | 10.5981 | 286 | 0.0242 | 1.0245 | 6.45 | 6.45 | 6.7 |
151
+ | 0.0559 | 11.0748 | 299 | 0.0398 | 1.0406 | 6.45 | 6.45 | 8.19 |
152
+ | 0.0388 | 11.5607 | 312 | 0.0243 | 1.0246 | 6.45 | 6.45 | 6.7 |
153
+ | 0.0374 | 12.0374 | 325 | 0.0142 | 1.0143 | 6.45 | 6.45 | 8.19 |
154
+ | 0.0223 | 12.5234 | 338 | 0.0126 | 1.0127 | 6.45 | 6.45 | 6.7 |
155
+ | 0.0198 | 13.0 | 351 | 0.0127 | 1.0128 | 6.45 | 6.45 | 8.19 |
156
+ | 0.0115 | 13.4860 | 364 | 0.0447 | 1.0457 | 6.45 | 6.45 | 6.7 |
157
+ | 0.0101 | 13.9720 | 377 | 0.0032 | 1.0032 | 6.45 | 6.45 | 8.19 |
158
+ | 0.0020 | 14.4486 | 390 | 0.0039 | 1.0039 | 5.47 | 5.47 | 6.7 |
159
+ | 0.0031 | 14.9346 | 403 | 0.0032 | 1.0032 | 6.45 | 6.45 | 8.19 |
160
+ | 0.0026 | 15.4112 | 416 | 0.0007 | 1.0007 | 6.45 | 6.45 | 6.7 |
161
+ | 0.0025 | 15.8972 | 429 | 0.0005 | 1.0005 | 6.45 | 6.45 | 8.19 |
162
+ | 0.0037 | 16.3738 | 442 | 0.0008 | 1.0008 | 6.45 | 6.45 | 6.7 |
163
+ | 0.0023 | 16.8598 | 455 | 0.0007 | 1.0007 | 6.45 | 6.45 | 8.19 |
164
+ | 0.0010 | 17.3364 | 468 | 0.0005 | 1.0005 | 6.45 | 6.45 | 6.7 |
165
+ | 0.0005 | 17.8224 | 481 | 0.0004 | 1.0004 | 6.45 | 6.45 | 8.19 |
166
+ | 0.0006 | 18.2991 | 494 | 0.0003 | 1.0003 | 6.45 | 6.45 | 6.7 |
167
+ | 0.0002 | 18.7850 | 507 | 0.0002 | 1.0002 | 6.45 | 6.45 | 8.19 |
168
+ | 0.0002 | 19.2617 | 520 | 0.0002 | 1.0002 | 5.47 | 5.47 | 6.7 |
169
+ | 0.0001 | 19.7477 | 533 | 0.0002 | 1.0002 | 6.45 | 6.45 | 8.19 |
170
+ | 0.0001 | 20.2243 | 546 | 0.0001 | 1.0002 | 6.45 | 6.45 | 6.7 |
171
+ | 0.0001 | 20.7103 | 559 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
172
+ | 0.0001 | 21.1869 | 572 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
173
+ | 0.0001 | 21.6729 | 585 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
174
+ | 0.0001 | 22.1495 | 598 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
175
+ | 0.0001 | 22.6355 | 611 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
176
+ | 0.0001 | 23.1121 | 624 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
177
+ | 0.0001 | 23.5981 | 637 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
178
+ | 0.0001 | 24.0748 | 650 | 0.0001 | 1.0001 | 5.47 | 5.47 | 6.7 |
179
+ | 0.0001 | 24.5607 | 663 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
180
+ | 0.0001 | 25.0374 | 676 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
181
+ | 0.0001 | 25.5234 | 689 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
182
+ | 0.0001 | 26.0 | 702 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
183
+ | 0.0001 | 26.4860 | 715 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
184
+ | 0.0001 | 26.9720 | 728 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
185
+ | 0.0001 | 27.4486 | 741 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
186
+ | 0.0001 | 27.9346 | 754 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
187
+ | 0.0001 | 28.4112 | 767 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
188
+ | 0.0001 | 28.8972 | 780 | 0.0001 | 1.0001 | 5.47 | 5.47 | 6.7 |
189
+ | 0.0001 | 29.3738 | 793 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
190
+ | 0.0001 | 29.8598 | 806 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
191
+ | 0.0001 | 30.3364 | 819 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
192
+ | 0.0001 | 30.8224 | 832 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
193
+ | 0.0000 | 31.2991 | 845 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
194
+ | 0.0001 | 31.7850 | 858 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
195
+ | 0.0001 | 32.2617 | 871 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
196
+ | 0.0000 | 32.7477 | 884 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
197
+ | 0.0000 | 33.2243 | 897 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
198
+ | 0.0000 | 33.7103 | 910 | 0.0001 | 1.0001 | 5.47 | 5.47 | 6.7 |
199
+ | 0.0000 | 34.1869 | 923 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
200
+ | 0.0000 | 34.6729 | 936 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
201
+ | 0.0000 | 35.1495 | 949 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
202
+ | 0.0000 | 35.6355 | 962 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
203
+ | 0.0000 | 36.1121 | 975 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
204
+ | 0.0000 | 36.5981 | 988 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
205
+ | 0.0000 | 37.0748 | 1001 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
206
+ | 0.0000 | 37.5607 | 1014 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
207
+ | 0.0000 | 38.0374 | 1027 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
208
+ | 0.0000 | 38.5234 | 1040 | 0.0001 | 1.0001 | 5.47 | 5.47 | 6.7 |
209
+ | 0.0000 | 39.0 | 1053 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
210
+ | 0.0000 | 39.4860 | 1066 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
211
+ | 0.0000 | 39.9720 | 1079 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
212
+ | 0.0000 | 40.4486 | 1092 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
213
+ | 0.0000 | 40.9346 | 1105 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
214
+ | 0.0000 | 41.4112 | 1118 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
215
+ | 0.0000 | 41.8972 | 1131 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
216
+ | 0.0000 | 42.3738 | 1144 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
217
+ | 0.0000 | 42.8598 | 1157 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
218
+ | 0.0000 | 43.3364 | 1170 | 0.0001 | 1.0001 | 5.47 | 5.47 | 6.7 |
219
+ | 0.0000 | 43.8224 | 1183 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
220
+ | 0.0000 | 44.2991 | 1196 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
221
+ | 0.0000 | 44.7850 | 1209 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
222
+ | 0.0000 | 45.2617 | 1222 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
223
+ | 0.0000 | 45.7477 | 1235 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
224
+ | 0.0000 | 46.2243 | 1248 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
225
+ | 0.0000 | 46.7103 | 1261 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
226
+ | 0.0000 | 47.1869 | 1274 | 0.0001 | 1.0001 | 6.45 | 6.45 | 6.7 |
227
+ | 0.0000 | 47.6729 | 1287 | 0.0001 | 1.0001 | 6.45 | 6.45 | 8.19 |
228
+ | 0.0000 | 48.1495 | 1300 | 0.0001 | 1.0001 | 5.47 | 5.47 | 6.7 |
229
+
230
+
231
+ ### Framework versions
232
+
233
+ - PEFT 0.18.1
234
+ - Transformers 5.0.0
235
+ - Pytorch 2.8.0+cu128
236
+ - Datasets 4.5.0
237
+ - Tokenizers 0.22.2
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/adapter_config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "/home/jiaruil5/math_rl/mix_teachers/r3lit_rl/models/Qwen/Qwen3-4B-Instruct-2507",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": null,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 128,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.0,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": null,
25
+ "peft_type": "LORA",
26
+ "peft_version": "0.18.1",
27
+ "qalora_group_size": 16,
28
+ "r": 64,
29
+ "rank_pattern": {},
30
+ "revision": null,
31
+ "target_modules": [
32
+ "o_proj",
33
+ "down_proj",
34
+ "k_proj",
35
+ "q_proj",
36
+ "up_proj",
37
+ "v_proj",
38
+ "gate_proj"
39
+ ],
40
+ "target_parameters": [],
41
+ "task_type": "CAUSAL_LM",
42
+ "trainable_token_indices": null,
43
+ "use_dora": false,
44
+ "use_qalora": false,
45
+ "use_rslora": false
46
+ }
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0208a88f86e8650d895396fb28d73ce7b7199934f2758a5f11ea54fa0d9cd195
3
+ size 528550256
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/chat_template.jinja ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '
2
+ ' + message['content'] + '<|im_end|>' + '
3
+ '}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant
4
+ ' }}{% endif %}
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/config.json ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": null,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2560,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 9728,
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": 262144,
54
+ "max_window_layers": 36,
55
+ "model_type": "qwen3",
56
+ "num_attention_heads": 32,
57
+ "num_hidden_layers": 36,
58
+ "num_key_value_heads": 8,
59
+ "pad_token_id": null,
60
+ "quantization_config": {
61
+ "_load_in_4bit": true,
62
+ "_load_in_8bit": false,
63
+ "bnb_4bit_compute_dtype": "bfloat16",
64
+ "bnb_4bit_quant_storage": "bfloat16",
65
+ "bnb_4bit_quant_type": "nf4",
66
+ "bnb_4bit_use_double_quant": true,
67
+ "llm_int8_enable_fp32_cpu_offload": false,
68
+ "llm_int8_has_fp16_weight": false,
69
+ "llm_int8_skip_modules": null,
70
+ "llm_int8_threshold": 6.0,
71
+ "load_in_4bit": true,
72
+ "load_in_8bit": false,
73
+ "quant_method": "bitsandbytes"
74
+ },
75
+ "rms_norm_eps": 1e-06,
76
+ "rope_parameters": {
77
+ "rope_theta": 5000000,
78
+ "rope_type": "default"
79
+ },
80
+ "sliding_window": null,
81
+ "tie_word_embeddings": true,
82
+ "transformers_version": "5.0.0",
83
+ "use_cache": false,
84
+ "use_sliding_window": false,
85
+ "vocab_size": 151936
86
+ }
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/eval_results.csv ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ category,filename,total,correct,accuracy,format_found,format_accuracy,errors_count
2
+ knowledge,test_subset_of_train_inference_results,450,448,99.56,0,0.00,2
3
+ knowledge,test_subset_of_train_paraphrased_inference_results,450,448,99.56,0,0.00,2
4
+ knowledge,test_inference_results,499,12,2.40,0,0.00,487
5
+ knowledge,test_5shot_inference_results,499,17,3.41,0,0.00,482
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/eval_summary.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "overall": {
3
+ "total": 499,
4
+ "correct": 17,
5
+ "accuracy": 3.41,
6
+ "format_found": 0,
7
+ "format_accuracy": 0.0
8
+ },
9
+ "n_errors": 482,
10
+ "results_file": "/home/jiaruil5/math_rl/mix_teachers/r3lit_rl/mix_teachers/checkpoints/knowledge/atomic_sft_lora_50ep_t20260305/eval_results/test_5shot_inference_results.jsonl"
11
+ }
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/test_5shot_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/test_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/test_subset_of_train_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/eval_results/test_subset_of_train_paraphrased_inference_results.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/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/knowledge/lora_sft_atomic_50ep_t20260305/merged/chat_template.jinja ADDED
@@ -0,0 +1,61 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ {%- for message in messages %}
18
+ {%- if message.content is string %}
19
+ {%- set content = message.content %}
20
+ {%- else %}
21
+ {%- set content = '' %}
22
+ {%- endif %}
23
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
24
+ {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
25
+ {%- elif message.role == "assistant" %}
26
+ {{- '<|im_start|>' + message.role + '\n' + content }}
27
+ {%- if message.tool_calls %}
28
+ {%- for tool_call in message.tool_calls %}
29
+ {%- if (loop.first and content) or (not loop.first) %}
30
+ {{- '\n' }}
31
+ {%- endif %}
32
+ {%- if tool_call.function %}
33
+ {%- set tool_call = tool_call.function %}
34
+ {%- endif %}
35
+ {{- '<tool_call>\n{"name": "' }}
36
+ {{- tool_call.name }}
37
+ {{- '", "arguments": ' }}
38
+ {%- if tool_call.arguments is string %}
39
+ {{- tool_call.arguments }}
40
+ {%- else %}
41
+ {{- tool_call.arguments | tojson }}
42
+ {%- endif %}
43
+ {{- '}\n</tool_call>' }}
44
+ {%- endfor %}
45
+ {%- endif %}
46
+ {{- '<|im_end|>\n' }}
47
+ {%- elif message.role == "tool" %}
48
+ {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
49
+ {{- '<|im_start|>user' }}
50
+ {%- endif %}
51
+ {{- '\n<tool_response>\n' }}
52
+ {{- content }}
53
+ {{- '\n</tool_response>' }}
54
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
55
+ {{- '<|im_end|>\n' }}
56
+ {%- endif %}
57
+ {%- endif %}
58
+ {%- endfor %}
59
+ {%- if add_generation_prompt %}
60
+ {{- '<|im_start|>assistant\n' }}
61
+ {%- endif %}
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/config.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Qwen3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "bfloat16",
9
+ "eos_token_id": 151645,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 2560,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 9728,
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": 262144,
54
+ "max_window_layers": 36,
55
+ "model_type": "qwen3",
56
+ "num_attention_heads": 32,
57
+ "num_hidden_layers": 36,
58
+ "num_key_value_heads": 8,
59
+ "rms_norm_eps": 1e-06,
60
+ "rope_scaling": null,
61
+ "rope_theta": 5000000,
62
+ "sliding_window": null,
63
+ "tie_word_embeddings": true,
64
+ "transformers_version": "4.57.1",
65
+ "use_cache": true,
66
+ "use_sliding_window": false,
67
+ "vocab_size": 151936
68
+ }
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/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": 0.7,
10
+ "top_k": 20,
11
+ "top_p": 0.8,
12
+ "transformers_version": "4.57.1"
13
+ }
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:075ef3d88561c94a14aef6de9610046354d6e39100565a84aeefc063dbffd407
3
+ size 4967215360
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33254930412df48f3b04e93f7b7a328e3345fed99bd52da41e3524f7079abbe7
3
+ size 3077766632
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/model.safetensors.index.json ADDED
@@ -0,0 +1,406 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 4022468096,
4
+ "total_size": 8044936192
5
+ },
6
+ "weight_map": {
7
+ "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
8
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
9
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
10
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
11
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
12
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
13
+ "model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
14
+ "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
15
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
16
+ "model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
17
+ "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
18
+ "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
19
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
20
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
21
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
22
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
23
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
24
+ "model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
25
+ "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
26
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
27
+ "model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
28
+ "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
29
+ "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
30
+ "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
31
+ "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
32
+ "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
33
+ "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
34
+ "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
35
+ "model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
36
+ "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
37
+ "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
38
+ "model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
39
+ "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
40
+ "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
41
+ "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
42
+ "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
43
+ "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
44
+ "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
45
+ "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
46
+ "model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
47
+ "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
48
+ "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
49
+ "model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
50
+ "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
51
+ "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
52
+ "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
53
+ "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
54
+ "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
55
+ "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
56
+ "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
57
+ "model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
58
+ "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
59
+ "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
60
+ "model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
61
+ "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
62
+ "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
63
+ "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
64
+ "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
65
+ "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
66
+ "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
67
+ "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
68
+ "model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
69
+ "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
70
+ "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
71
+ "model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
72
+ "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
73
+ "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
74
+ "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
75
+ "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
76
+ "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
77
+ "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
78
+ "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
79
+ "model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
80
+ "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
81
+ "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
82
+ "model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
83
+ "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
84
+ "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
85
+ "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
86
+ "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
87
+ "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
88
+ "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
89
+ "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
90
+ "model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
91
+ "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
92
+ "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
93
+ "model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
94
+ "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
95
+ "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
96
+ "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
97
+ "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
98
+ "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
99
+ "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
100
+ "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
101
+ "model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
102
+ "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
103
+ "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
104
+ "model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
105
+ "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
106
+ "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
107
+ "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
108
+ "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
109
+ "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
110
+ "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
111
+ "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
112
+ "model.layers.17.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
113
+ "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
114
+ "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
115
+ "model.layers.17.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
116
+ "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
117
+ "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
118
+ "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
119
+ "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
120
+ "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
121
+ "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
122
+ "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
123
+ "model.layers.18.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
124
+ "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
125
+ "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
126
+ "model.layers.18.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
127
+ "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
128
+ "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
129
+ "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
130
+ "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
131
+ "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
132
+ "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
133
+ "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
134
+ "model.layers.19.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
135
+ "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
136
+ "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
137
+ "model.layers.19.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
138
+ "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
139
+ "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
140
+ "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
141
+ "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
142
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
143
+ "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
144
+ "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
145
+ "model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
146
+ "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
147
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
148
+ "model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
149
+ "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
150
+ "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
151
+ "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
152
+ "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
153
+ "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
154
+ "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
155
+ "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
156
+ "model.layers.20.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
157
+ "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
158
+ "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
159
+ "model.layers.20.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
160
+ "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
161
+ "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
162
+ "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
163
+ "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
164
+ "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
165
+ "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
166
+ "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
167
+ "model.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
168
+ "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
169
+ "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
170
+ "model.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
171
+ "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
172
+ "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
173
+ "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
174
+ "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
175
+ "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
176
+ "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
177
+ "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
178
+ "model.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
179
+ "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
180
+ "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
181
+ "model.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
182
+ "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
183
+ "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
184
+ "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
185
+ "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
186
+ "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
187
+ "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
188
+ "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
189
+ "model.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
190
+ "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
191
+ "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
192
+ "model.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
193
+ "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
194
+ "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
195
+ "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
196
+ "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
197
+ "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
198
+ "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
199
+ "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
200
+ "model.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
201
+ "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
202
+ "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
203
+ "model.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
204
+ "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
205
+ "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
206
+ "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
207
+ "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
208
+ "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
209
+ "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
210
+ "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
211
+ "model.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
212
+ "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
213
+ "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
214
+ "model.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
215
+ "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
216
+ "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
217
+ "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
218
+ "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
219
+ "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
220
+ "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
221
+ "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
222
+ "model.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
223
+ "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
224
+ "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
225
+ "model.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
226
+ "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
227
+ "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
228
+ "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
229
+ "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
230
+ "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
231
+ "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
232
+ "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
233
+ "model.layers.27.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
234
+ "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
235
+ "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
236
+ "model.layers.27.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
237
+ "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
238
+ "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
239
+ "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
240
+ "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
241
+ "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
242
+ "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
243
+ "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
244
+ "model.layers.28.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
245
+ "model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
246
+ "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
247
+ "model.layers.28.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
248
+ "model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
249
+ "model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
250
+ "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
251
+ "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
252
+ "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
253
+ "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
254
+ "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
255
+ "model.layers.29.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
256
+ "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
257
+ "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
258
+ "model.layers.29.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
259
+ "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
260
+ "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
261
+ "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
262
+ "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
263
+ "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
264
+ "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
265
+ "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
266
+ "model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
267
+ "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
268
+ "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
269
+ "model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
270
+ "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
271
+ "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
272
+ "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
273
+ "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
274
+ "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
275
+ "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
276
+ "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
277
+ "model.layers.30.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
278
+ "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
279
+ "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
280
+ "model.layers.30.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
281
+ "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
282
+ "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
283
+ "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
284
+ "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
285
+ "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
286
+ "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
287
+ "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
288
+ "model.layers.31.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
289
+ "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
290
+ "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
291
+ "model.layers.31.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
292
+ "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
293
+ "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
294
+ "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
295
+ "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
296
+ "model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
297
+ "model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
298
+ "model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
299
+ "model.layers.32.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
300
+ "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
301
+ "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
302
+ "model.layers.32.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
303
+ "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
304
+ "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
305
+ "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
306
+ "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
307
+ "model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
308
+ "model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
309
+ "model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
310
+ "model.layers.33.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
311
+ "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
312
+ "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
313
+ "model.layers.33.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
314
+ "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
315
+ "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
316
+ "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
317
+ "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
318
+ "model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
319
+ "model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
320
+ "model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
321
+ "model.layers.34.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
322
+ "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
323
+ "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
324
+ "model.layers.34.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
325
+ "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
326
+ "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
327
+ "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
328
+ "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
329
+ "model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
330
+ "model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
331
+ "model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
332
+ "model.layers.35.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
333
+ "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
334
+ "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
335
+ "model.layers.35.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
336
+ "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
337
+ "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
338
+ "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
339
+ "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
340
+ "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
341
+ "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
342
+ "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
343
+ "model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
344
+ "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
345
+ "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
346
+ "model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
347
+ "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
348
+ "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
349
+ "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
350
+ "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
351
+ "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
352
+ "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
353
+ "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
354
+ "model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
355
+ "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
356
+ "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
357
+ "model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
358
+ "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
359
+ "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
360
+ "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
361
+ "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
362
+ "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
363
+ "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
364
+ "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
365
+ "model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
366
+ "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
367
+ "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
368
+ "model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
369
+ "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
370
+ "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
371
+ "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
372
+ "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
373
+ "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
374
+ "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
375
+ "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
376
+ "model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
377
+ "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
378
+ "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
379
+ "model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
380
+ "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
381
+ "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
382
+ "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
383
+ "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
384
+ "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
385
+ "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
386
+ "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
387
+ "model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
388
+ "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
389
+ "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
390
+ "model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
391
+ "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
392
+ "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
393
+ "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
394
+ "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
395
+ "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
396
+ "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
397
+ "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
398
+ "model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
399
+ "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
400
+ "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
401
+ "model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
402
+ "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
403
+ "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
404
+ "model.norm.weight": "model-00002-of-00002.safetensors"
405
+ }
406
+ }
checkpoints/knowledge/lora_sft_atomic_50ep_t20260305/merged/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
+ }