| 2026-08-08:22:21:07 INFO [_cli.run:388] Selected Tasks: ['humaneval', 'mbpp'] |
| 2026-08-08:22:21:08 INFO [evaluator:214] Setting random seed to 0 | Setting numpy seed to 1234 | Setting torch manual seed to 1234 | Setting fewshot manual seed to 1234 |
| 2026-08-08:22:21:08 INFO [evaluator:239] Initializing local-completions model, with arguments: {'model': 'student', 'base_url': 'http://127.0.0.1:8461/v1/completions', 'tokenizer': 'outputs/release/healed/release-offpolicy-general-keep75/step0150', 'num_concurrent': 48, 'tokenized_requests': False, 'max_retries': 3} |
| 2026-08-08:22:21:08 INFO [models.openai_completions:42] Remote tokenizer not supported. Using huggingface tokenizer backend. |
| 2026-08-08:22:21:08 INFO [models.api_models:179] Using max length 2048 - 1 |
| 2026-08-08:22:21:08 INFO [models.api_models:200] Using tokenizer huggingface |
| 2026-08-08:22:21:16 INFO [evaluator_utils:446] Selected tasks: |
| 2026-08-08:22:21:16 INFO [evaluator_utils:480] Task: humaneval (humaneval/humaneval.yaml) |
| 2026-08-08:22:21:16 INFO [evaluator_utils:480] Task: mbpp (mbpp/mbpp.yaml) |
| 2026-08-08:22:21:16 INFO [evaluator:314] humaneval: Using gen_kwargs: {'until': ['\nclass', '\ndef', '\n#', '\nif', '\nprint'], 'max_gen_toks': 1024, 'do_sample': False} |
| 2026-08-08:22:21:16 INFO [evaluator:314] mbpp: Using gen_kwargs: {'until': ['[DONE]'], 'do_sample': False} |
| 2026-08-08:22:21:16 INFO [api.task:312] Building contexts for humaneval on rank 0... |
|
0%| | 0/164 [00:00<?, ?it/s]
98%|ββββββββββ| 161/164 [00:00<00:00, 1600.42it/s]
100%|ββββββββββ| 164/164 [00:00<00:00, 1601.33it/s] |
| 2026-08-08:22:21:16 INFO [api.task:312] Building contexts for mbpp on rank 0... |
|
0%| | 0/500 [00:00<?, ?it/s]
3%|β | 17/500 [00:00<00:02, 161.73it/s]
7%|β | 34/500 [00:00<00:02, 162.63it/s]
10%|β | 51/500 [00:00<00:02, 163.29it/s]
14%|ββ | 68/500 [00:00<00:02, 163.58it/s]
17%|ββ | 85/500 [00:00<00:02, 163.88it/s]
20%|ββ | 102/500 [00:00<00:02, 164.32it/s]
24%|βββ | 119/500 [00:00<00:02, 164.57it/s]
27%|βββ | 136/500 [00:00<00:02, 164.72it/s]
31%|βββ | 153/500 [00:00<00:02, 164.85it/s]
34%|ββββ | 170/500 [00:01<00:02, 164.77it/s]
37%|ββββ | 187/500 [00:01<00:01, 165.19it/s]
41%|ββββ | 204/500 [00:01<00:01, 165.59it/s]
44%|βββββ | 221/500 [00:01<00:01, 165.71it/s]
48%|βββββ | 238/500 [00:01<00:01, 164.79it/s]
51%|βββββ | 255/500 [00:01<00:01, 165.24it/s]
54%|ββββββ | 272/500 [00:01<00:01, 165.65it/s]
58%|ββββββ | 289/500 [00:01<00:01, 165.83it/s]
61%|ββββββ | 306/500 [00:01<00:01, 165.96it/s]
65%|βββββββ | 323/500 [00:01<00:01, 166.20it/s]
68%|βββββββ | 340/500 [00:02<00:00, 166.28it/s]
71%|ββββββββ | 357/500 [00:02<00:00, 166.30it/s]
75%|ββββββββ | 374/500 [00:02<00:00, 166.37it/s]
78%|ββββββββ | 391/500 [00:02<00:00, 166.29it/s]
82%|βββββββββ | 408/500 [00:02<00:00, 166.31it/s]
85%|βββββββββ | 425/500 [00:02<00:00, 166.27it/s]
88%|βββββββββ | 442/500 [00:02<00:00, 166.52it/s]
92%|ββββββββββ| 459/500 [00:02<00:00, 166.63it/s]
95%|ββββββββββ| 476/500 [00:02<00:00, 166.53it/s]
99%|ββββββββββ| 493/500 [00:02<00:00, 166.63it/s]
100%|ββββββββββ| 500/500 [00:03<00:00, 165.56it/s] |
| 2026-08-08:22:21:19 INFO [evaluator:585] Running generate_until requests |
| 2026-08-08:22:21:19 INFO [models.api_models:747] Tokenized requests are disabled. Context + generation length is not checked. |
|
Requesting API: 0%| | 0/164 [00:00<?, ?it/s]
Requesting API: 1%| | 1/164 [00:00<02:37, 1.04it/s]
Requesting API: 4%|β | 6/164 [00:01<00:23, 6.64it/s]
Requesting API: 5%|β | 8/164 [00:01<00:21, 7.23it/s]
Requesting API: 6%|β | 10/164 [00:01<00:17, 8.61it/s]
Requesting API: 7%|β | 12/164 [00:01<00:20, 7.43it/s]
Requesting API: 10%|β | 17/164 [00:01<00:10, 13.48it/s]
Requesting API: 12%|ββ | 20/164 [00:02<00:10, 14.03it/s]
Requesting API: 15%|ββ | 25/164 [00:02<00:06, 20.14it/s]
Requesting API: 17%|ββ | 28/164 [00:02<00:06, 21.90it/s]
Requesting API: 19%|ββ | 31/164 [00:02<00:05, 23.43it/s]
Requesting API: 21%|ββ | 34/164 [00:02<00:06, 21.66it/s]
Requesting API: 23%|βββ | 37/164 [00:02<00:05, 23.50it/s]
Requesting API: 24%|βββ | 40/164 [00:02<00:05, 24.05it/s]
Requesting API: 28%|βββ | 46/164 [00:02<00:03, 32.32it/s]
Requesting API: 30%|βββ | 50/164 [00:03<00:04, 28.46it/s]
Requesting API: 34%|ββββ | 56/164 [00:03<00:03, 33.62it/s]
Requesting API: 37%|ββββ | 60/164 [00:03<00:03, 28.40it/s]
Requesting API: 39%|ββββ | 64/164 [00:03<00:03, 28.10it/s]
Requesting API: 41%|βββββ | 68/164 [00:03<00:03, 28.42it/s]
Requesting API: 45%|βββββ | 74/164 [00:03<00:02, 34.14it/s]
Requesting API: 49%|βββββ | 81/164 [00:03<00:01, 41.84it/s]
Requesting API: 52%|ββββββ | 86/164 [00:04<00:01, 41.80it/s]
Requesting API: 55%|ββββββ | 91/164 [00:04<00:01, 39.36it/s]
Requesting API: 59%|ββββββ | 97/164 [00:04<00:01, 39.88it/s]
Requesting API: 62%|βββββββ | 102/164 [00:04<00:01, 41.64it/s]
Requesting API: 66%|βββββββ | 108/164 [00:04<00:01, 44.82it/s]
Requesting API: 70%|βββββββ | 115/164 [00:04<00:00, 49.27it/s]
Requesting API: 75%|ββββββββ | 123/164 [00:04<00:00, 55.13it/s]
Requesting API: 82%|βββββββββ | 134/164 [00:04<00:00, 67.24it/s]
Requesting API: 86%|βββββββββ | 141/164 [00:05<00:00, 54.04it/s]
Requesting API: 90%|βββββββββ | 148/164 [00:05<00:00, 56.30it/s]
Requesting API: 94%|ββββββββββ| 154/164 [00:05<00:00, 48.05it/s]
Requesting API: 98%|ββββββββββ| 160/164 [00:05<00:00, 26.79it/s]
Requesting API: 100%|ββββββββββ| 164/164 [00:15<00:00, 10.45it/s] |
|
Requesting API: 0%| | 0/500 [00:00<?, ?it/s]
Requesting API: 0%| | 1/500 [00:01<09:58, 1.20s/it]
Requesting API: 1%| | 4/500 [00:01<02:07, 3.89it/s]
Requesting API: 2%|β | 9/500 [00:01<00:50, 9.76it/s]
Requesting API: 3%|β | 16/500 [00:01<00:29, 16.59it/s]
Requesting API: 4%|β | 20/500 [00:01<00:23, 20.04it/s]
Requesting API: 5%|β | 24/500 [00:01<00:22, 21.42it/s]
Requesting API: 6%|β | 28/500 [00:02<00:20, 23.58it/s]
Requesting API: 6%|β | 32/500 [00:02<00:17, 26.94it/s]
Requesting API: 7%|β | 36/500 [00:02<00:16, 28.72it/s]
Requesting API: 8%|β | 40/500 [00:02<00:15, 30.34it/s]
Requesting API: 9%|β | 46/500 [00:02<00:12, 36.71it/s]
Requesting API: 10%|β | 51/500 [00:02<00:16, 28.01it/s]
Requesting API: 11%|β | 55/500 [00:02<00:15, 29.31it/s]
Requesting API: 12%|ββ | 59/500 [00:02<00:14, 29.81it/s]
Requesting API: 13%|ββ | 64/500 [00:03<00:14, 29.32it/s]
Requesting API: 14%|ββ | 68/500 [00:03<00:15, 27.66it/s]
Requesting API: 14%|ββ | 71/500 [00:03<00:15, 27.55it/s]
Requesting API: 15%|ββ | 77/500 [00:03<00:13, 31.84it/s]
Requesting API: 17%|ββ | 83/500 [00:03<00:10, 37.94it/s]
Requesting API: 18%|ββ | 88/500 [00:03<00:12, 34.17it/s]
Requesting API: 18%|ββ | 92/500 [00:04<00:15, 25.98it/s]
Requesting API: 20%|ββ | 98/500 [00:04<00:13, 30.81it/s]
Requesting API: 21%|ββ | 103/500 [00:04<00:11, 33.98it/s]
Requesting API: 22%|βββ | 108/500 [00:04<00:10, 37.33it/s]
Requesting API: 23%|βββ | 114/500 [00:04<00:09, 38.68it/s]
Requesting API: 24%|βββ | 119/500 [00:04<00:14, 26.61it/s]
Requesting API: 25%|βββ | 123/500 [00:05<00:13, 27.22it/s]
Requesting API: 25%|βββ | 127/500 [00:05<00:12, 29.05it/s]
Requesting API: 26%|βββ | 132/500 [00:05<00:11, 31.36it/s]
Requesting API: 27%|βββ | 136/500 [00:05<00:12, 28.93it/s]
Requesting API: 28%|βββ | 141/500 [00:05<00:12, 27.98it/s]
Requesting API: 29%|βββ | 144/500 [00:05<00:13, 26.48it/s]
Requesting API: 29%|βββ | 147/500 [00:06<00:15, 22.13it/s]
Requesting API: 30%|βββ | 152/500 [00:06<00:12, 27.59it/s]
Requesting API: 31%|ββββ | 157/500 [00:06<00:10, 31.68it/s]
Requesting API: 32%|ββββ | 161/500 [00:06<00:11, 29.07it/s]
Requesting API: 33%|ββββ | 165/500 [00:06<00:10, 31.42it/s]
Requesting API: 34%|ββββ | 169/500 [00:06<00:11, 27.66it/s]
Requesting API: 35%|ββββ | 173/500 [00:06<00:11, 28.95it/s]
Requesting API: 35%|ββββ | 177/500 [00:06<00:11, 28.86it/s]
Requesting API: 36%|ββββ | 181/500 [00:07<00:10, 30.54it/s]
Requesting API: 37%|ββββ | 185/500 [00:07<00:10, 30.26it/s]
Requesting API: 38%|ββββ | 191/500 [00:07<00:10, 30.80it/s]
Requesting API: 39%|ββββ | 196/500 [00:07<00:10, 28.92it/s]
Requesting API: 40%|ββββ | 201/500 [00:07<00:10, 29.85it/s]
Requesting API: 41%|ββββ | 205/500 [00:07<00:11, 26.24it/s]
Requesting API: 42%|βββββ | 211/500 [00:08<00:08, 32.14it/s]
Requesting API: 43%|βββββ | 215/500 [00:08<00:08, 33.83it/s]
Requesting API: 44%|βββββ | 220/500 [00:08<00:07, 37.39it/s]
Requesting API: 45%|βββββ | 225/500 [00:08<00:08, 32.66it/s]
Requesting API: 46%|βββββ | 229/500 [00:08<00:08, 30.43it/s]
Requesting API: 47%|βββββ | 235/500 [00:08<00:08, 32.52it/s]
Requesting API: 48%|βββββ | 239/500 [00:08<00:08, 29.49it/s]
Requesting API: 49%|βββββ | 243/500 [00:09<00:08, 31.55it/s]
Requesting API: 50%|βββββ | 248/500 [00:09<00:07, 32.98it/s]
Requesting API: 50%|βββββ | 252/500 [00:09<00:07, 33.00it/s]
Requesting API: 52%|ββββββ | 258/500 [00:09<00:06, 36.31it/s]
Requesting API: 52%|ββββββ | 262/500 [00:09<00:07, 29.84it/s]
Requesting API: 54%|ββββββ | 268/500 [00:09<00:06, 35.37it/s]
Requesting API: 55%|ββββββ | 273/500 [00:09<00:06, 35.95it/s]
Requesting API: 56%|ββββββ | 278/500 [00:10<00:05, 37.44it/s]
Requesting API: 56%|ββββββ | 282/500 [00:10<00:07, 28.54it/s]
Requesting API: 57%|ββββββ | 286/500 [00:10<00:08, 24.33it/s]
Requesting API: 59%|ββββββ | 295/500 [00:10<00:06, 34.04it/s]
Requesting API: 61%|ββββββ | 303/500 [00:10<00:04, 42.08it/s]
Requesting API: 62%|βββββββ | 308/500 [00:10<00:05, 38.39it/s]
Requesting API: 63%|βββββββ | 313/500 [00:11<00:04, 38.03it/s]
Requesting API: 64%|βββββββ | 319/500 [00:11<00:04, 39.39it/s]
Requesting API: 65%|βββββββ | 325/500 [00:11<00:04, 42.90it/s]
Requesting API: 66%|βββββββ | 330/500 [00:11<00:03, 43.46it/s]
Requesting API: 67%|βββββββ | 335/500 [00:11<00:04, 34.12it/s]
Requesting API: 68%|βββββββ | 339/500 [00:11<00:05, 31.38it/s]
Requesting API: 69%|βββββββ | 343/500 [00:11<00:04, 32.58it/s]
Requesting API: 69%|βββββββ | 347/500 [00:12<00:04, 32.00it/s]
Requesting API: 71%|βββββββ | 355/500 [00:12<00:03, 41.09it/s]
Requesting API: 72%|ββββββββ | 360/500 [00:12<00:03, 36.58it/s]
Requesting API: 74%|ββββββββ | 368/500 [00:12<00:03, 43.16it/s]
Requesting API: 75%|ββββββββ | 373/500 [00:12<00:03, 37.17it/s]
Requesting API: 76%|ββββββββ | 379/500 [00:12<00:03, 39.39it/s]
Requesting API: 77%|ββββββββ | 384/500 [00:12<00:02, 41.37it/s]
Requesting API: 78%|ββββββββ | 389/500 [00:13<00:02, 38.00it/s]
Requesting API: 79%|ββββββββ | 394/500 [00:13<00:02, 40.09it/s]
Requesting API: 80%|ββββββββ | 401/500 [00:13<00:02, 46.42it/s]
Requesting API: 82%|βββββββββ | 409/500 [00:13<00:01, 52.57it/s]
Requesting API: 83%|βββββββββ | 415/500 [00:13<00:01, 50.44it/s]
Requesting API: 84%|βββββββββ | 421/500 [00:13<00:01, 44.49it/s]
Requesting API: 85%|βββββββββ | 426/500 [00:13<00:01, 38.11it/s]
Requesting API: 86%|βββββββββ | 432/500 [00:13<00:01, 42.83it/s]
Requesting API: 88%|βββββββββ | 438/500 [00:14<00:01, 45.82it/s]
Requesting API: 90%|βββββββββ | 448/500 [00:14<00:00, 52.91it/s]
Requesting API: 91%|βββββββββ | 454/500 [00:14<00:00, 53.38it/s]
Requesting API: 93%|ββββββββββ| 464/500 [00:14<00:00, 64.54it/s]
Requesting API: 94%|ββββββββββ| 471/500 [00:14<00:00, 61.87it/s]
Requesting API: 96%|ββββββββββ| 478/500 [00:14<00:00, 61.50it/s]
Requesting API: 97%|ββββββββββ| 485/500 [00:14<00:00, 58.01it/s]
Requesting API: 98%|ββββββββββ| 491/500 [00:14<00:00, 57.19it/s]
Requesting API: 99%|ββββββββββ| 497/500 [00:15<00:00, 47.17it/s]
Requesting API: 100%|ββββββββββ| 500/500 [00:16<00:00, 30.80it/s] |
| 2026-08-08:22:26:03 INFO [loggers.evaluation_tracker:247] Saving results aggregated |
| 2026-08-08:22:26:03 INFO [loggers.evaluation_tracker:119] Saving per-task samples to outputs/evals/release/general-keep75-step0150/student/*.jsonl |
| local-completions ({'model': 'student', 'base_url': 'http://127.0.0.1:8461/v1/completions', 'tokenizer': 'outputs/release/healed/release-offpolicy-general-keep75/step0150', 'num_concurrent': 48, 'tokenized_requests': False, 'max_retries': 3}), gen_kwargs: ({}), limit: None, num_fewshot: None, batch_size: 1 |
| | Tasks |Version| Filter |n-shot| Metric | |Value | |Stderr| |
| |---------|------:|-----------|-----:|---------|---|-----:|---|-----:| |
| |humaneval| 1|create_test| 0|pass@1 |β |0.3476|Β± |0.0373| |
| |mbpp | 1|none | 3|pass_at_1|β |0.2820|Β± |0.0201| |
|
|
|
|