steer-full_7b / run /capability /lmeval_run.log
JacoDuToit's picture
run directory (provenance + evals)
09b45f2 verified
Raw
History Blame Contribute Delete
340 kB
########## m0 / mmlu ##########
/usr/bin/python -m lm_eval --model hf --model_args pretrained=Qwen/Qwen2.5-7B-Instruct,dtype=bfloat16 --tasks mmlu --num_fewshot 5 --batch_size auto --seed 0 --output_path /root/steering-resistance/results/full_7b/capability/m0/mmlu --apply_chat_template --fewshot_as_multiturn --limit 15
/usr/local/lib/python3.11/dist-packages/requests/__init__.py:113: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (6.0.0.post1)/charset_normalizer (3.3.2) doesn't match a supported version!
warnings.warn(
2026-07-24:06:59:32 WARNING [config.evaluate_config:287] --limit SHOULD ONLY BE USED FOR TESTING. REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.
2026-07-24:06:59:40 INFO [_cli.run:388] Selected Tasks: ['mmlu']
2026-07-24:06:59:42 INFO [evaluator:214] Setting random seed to 0 | Setting numpy seed to 0 | Setting torch manual seed to 0 | Setting fewshot manual seed to 0
2026-07-24:06:59:42 INFO [evaluator:239] Initializing hf model, with arguments: {'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16'}
2026-07-24:06:59:45 INFO [models.huggingface:286] Using device 'cuda:0'
2026-07-24:06:59:47 INFO [models.huggingface:579] Model parallel was set to False, max memory was not set, and device map was set to {'': 'cuda:0'}
Loading weights: 0%| | 0/339 [00:00<?, ?it/s] Loading weights: 1%| | 2/339 [00:00<00:34, 9.90it/s] Loading weights: 5%|β–Œ | 18/339 [00:00<00:04, 67.44it/s] Loading weights: 12%|β–ˆβ– | 39/339 [00:00<00:02, 117.55it/s] Loading weights: 16%|β–ˆβ–Œ | 54/339 [00:00<00:02, 122.71it/s] Loading weights: 22%|β–ˆβ–ˆβ– | 76/339 [00:00<00:01, 152.23it/s] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 93/339 [00:00<00:01, 150.55it/s] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 112/339 [00:00<00:01, 157.73it/s] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 129/339 [00:00<00:01, 154.81it/s] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 147/339 [00:01<00:01, 150.27it/s] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 166/339 [00:01<00:01, 161.08it/s] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 185/339 [00:01<00:00, 160.72it/s] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 204/339 [00:01<00:00, 166.55it/s] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 221/339 [00:01<00:00, 138.27it/s] Loading weights: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 239/339 [00:01<00:00, 148.06it/s] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 256/339 [00:01<00:00, 134.18it/s] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 271/339 [00:01<00:00, 132.58it/s] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 291/339 [00:02<00:00, 148.35it/s] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 307/339 [00:02<00:00, 145.24it/s] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 326/339 [00:02<00:00, 156.08it/s] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 339/339 [00:02<00:00, 144.24it/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 16430.85 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 3259.21 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1705.83 examples/s]
Generating test split: 0%| | 0/135 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 135/135 [00:00<00:00, 28436.67 examples/s]
Generating validation split: 0%| | 0/14 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 14/14 [00:00<00:00, 4171.36 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1593.70 examples/s]
Generating test split: 0%| | 0/152 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 152/152 [00:00<00:00, 31927.79 examples/s]
Generating validation split: 0%| | 0/16 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 16/16 [00:00<00:00, 6026.84 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 2310.15 examples/s]
Generating test split: 0%| | 0/144 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 144/144 [00:00<00:00, 37398.13 examples/s]
Generating validation split: 0%| | 0/16 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 16/16 [00:00<00:00, 8749.53 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3191.04 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 41128.69 examples/s]
Generating validation split: 0%| | 0/8 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 8/8 [00:00<00:00, 5571.05 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3763.73 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 27960.16 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 5566.76 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3098.17 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 39074.94 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 7359.60 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3454.95 examples/s]
Generating test split: 0%| | 0/102 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 102/102 [00:00<00:00, 22906.20 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 3457.02 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1632.66 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 24230.53 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 3496.84 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1594.19 examples/s]
Generating test split: 0%| | 0/235 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 235/235 [00:00<00:00, 54716.41 examples/s]
Generating validation split: 0%| | 0/26 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 26/26 [00:00<00:00, 8424.25 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1623.31 examples/s]
Generating test split: 0%| | 0/145 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 145/145 [00:00<00:00, 37727.92 examples/s]
Generating validation split: 0%| | 0/16 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 16/16 [00:00<00:00, 5072.48 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1641.99 examples/s]
Generating test split: 0%| | 0/378 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 378/378 [00:00<00:00, 85363.00 examples/s]
Generating validation split: 0%| | 0/41 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 41/41 [00:00<00:00, 12884.28 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1631.77 examples/s]
Generating test split: 0%| | 0/310 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 310/310 [00:00<00:00, 64591.86 examples/s]
Generating validation split: 0%| | 0/32 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 32/32 [00:00<00:00, 9884.94 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1631.39 examples/s]
Generating test split: 0%| | 0/203 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 203/203 [00:00<00:00, 44775.12 examples/s]
Generating validation split: 0%| | 0/22 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 22/22 [00:00<00:00, 6978.35 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1621.55 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 23079.87 examples/s]
Generating validation split: 0%| | 0/9 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 9/9 [00:00<00:00, 2890.63 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1614.68 examples/s]
Generating test split: 0%| | 0/270 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 270/270 [00:00<00:00, 63262.50 examples/s]
Generating validation split: 0%| | 0/29 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 29/29 [00:00<00:00, 9260.36 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1622.43 examples/s]
Generating test split: 0%| | 0/151 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 151/151 [00:00<00:00, 37651.74 examples/s]
Generating validation split: 0%| | 0/17 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 17/17 [00:00<00:00, 5464.26 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1653.51 examples/s]
Generating test split: 0%| | 0/216 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 216/216 [00:00<00:00, 49710.27 examples/s]
Generating validation split: 0%| | 0/23 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 23/23 [00:00<00:00, 7244.59 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1664.67 examples/s]
Generating test split: 0%| | 0/112 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 112/112 [00:00<00:00, 25343.23 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 3452.10 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1625.95 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 22135.87 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 3524.36 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1612.08 examples/s]
Generating test split: 0%| | 0/265 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 265/265 [00:00<00:00, 59355.47 examples/s]
Generating validation split: 0%| | 0/29 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 29/29 [00:00<00:00, 8975.41 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1662.16 examples/s]
Generating test split: 0%| | 0/173 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 173/173 [00:00<00:00, 40222.54 examples/s]
Generating validation split: 0%| | 0/22 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 22/22 [00:00<00:00, 7028.31 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1627.34 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 24508.03 examples/s]
Generating validation split: 0%| | 0/10 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 10/10 [00:00<00:00, 3222.67 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1647.28 examples/s]
Generating test split: 0%| | 0/223 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 223/223 [00:00<00:00, 46150.38 examples/s]
Generating validation split: 0%| | 0/23 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 23/23 [00:00<00:00, 7188.99 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1609.23 examples/s]
Generating test split: 0%| | 0/103 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 103/103 [00:00<00:00, 26798.17 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 3534.89 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1657.30 examples/s]
Generating test split: 0%| | 0/234 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 234/234 [00:00<00:00, 51855.40 examples/s]
Generating validation split: 0%| | 0/25 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 25/25 [00:00<00:00, 7519.37 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1631.64 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 21949.36 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 3314.46 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1591.89 examples/s]
Generating test split: 0%| | 0/783 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 783/783 [00:00<00:00, 168685.60 examples/s]
Generating validation split: 0%| | 0/86 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 86/86 [00:00<00:00, 26446.96 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1681.76 examples/s]
Generating test split: 0%| | 0/306 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 306/306 [00:00<00:00, 62252.37 examples/s]
Generating validation split: 0%| | 0/33 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 33/33 [00:00<00:00, 10545.68 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1639.30 examples/s]
Generating test split: 0%| | 0/282 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 282/282 [00:00<00:00, 54797.02 examples/s]
Generating validation split: 0%| | 0/31 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 31/31 [00:00<00:00, 8868.06 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1631.52 examples/s]
Generating test split: 0%| | 0/272 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 272/272 [00:00<00:00, 71276.44 examples/s]
Generating validation split: 0%| | 0/31 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 31/31 [00:00<00:00, 18469.24 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3331.99 examples/s]
Generating test split: 0%| | 0/166 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 166/166 [00:00<00:00, 60877.37 examples/s]
Generating validation split: 0%| | 0/18 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 18/18 [00:00<00:00, 11910.00 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3718.35 examples/s]
Generating test split: 0%| | 0/114 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 114/114 [00:00<00:00, 29466.36 examples/s]
Generating validation split: 0%| | 0/12 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 12/12 [00:00<00:00, 6281.25 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 2940.48 examples/s]
Generating test split: 0%| | 0/198 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 198/198 [00:00<00:00, 78198.89 examples/s]
Generating validation split: 0%| | 0/22 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 22/22 [00:00<00:00, 14375.24 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3510.47 examples/s]
Generating test split: 0%| | 0/193 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 193/193 [00:00<00:00, 73384.16 examples/s]
Generating validation split: 0%| | 0/21 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 21/21 [00:00<00:00, 13601.05 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3430.64 examples/s]
Generating test split: 0%| | 0/390 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 390/390 [00:00<00:00, 93542.55 examples/s]
Generating validation split: 0%| | 0/43 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 43/43 [00:00<00:00, 23404.50 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 2460.00 examples/s]
Generating test split: 0%| | 0/238 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 238/238 [00:00<00:00, 80386.89 examples/s]
Generating validation split: 0%| | 0/26 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 26/26 [00:00<00:00, 16810.84 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3360.28 examples/s]
Generating test split: 0%| | 0/545 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 545/545 [00:00<00:00, 118772.51 examples/s]
Generating validation split: 0%| | 0/60 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 60/60 [00:00<00:00, 31903.94 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3125.41 examples/s]
Generating test split: 0%| | 0/131 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 131/131 [00:00<00:00, 46650.86 examples/s]
Generating validation split: 0%| | 0/12 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 12/12 [00:00<00:00, 7881.56 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3006.67 examples/s]
Generating test split: 0%| | 0/612 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 612/612 [00:00<00:00, 145822.53 examples/s]
Generating validation split: 0%| | 0/69 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 69/69 [00:00<00:00, 36463.02 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3421.69 examples/s]
Generating test split: 0%| | 0/110 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 110/110 [00:00<00:00, 34807.50 examples/s]
Generating validation split: 0%| | 0/12 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 12/12 [00:00<00:00, 7150.40 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3381.41 examples/s]
Generating test split: 0%| | 0/245 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 245/245 [00:00<00:00, 74447.91 examples/s]
Generating validation split: 0%| | 0/27 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 27/27 [00:00<00:00, 16634.28 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3286.04 examples/s]
Generating test split: 0%| | 0/201 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 201/201 [00:00<00:00, 59070.57 examples/s]
Generating validation split: 0%| | 0/22 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 22/22 [00:00<00:00, 13301.81 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 2998.07 examples/s]
Generating test split: 0%| | 0/100 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 42781.56 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 7086.06 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3570.84 examples/s]
Generating test split: 0%| | 0/126 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 126/126 [00:00<00:00, 46867.89 examples/s]
Generating validation split: 0%| | 0/14 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 14/14 [00:00<00:00, 9067.37 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3365.67 examples/s]
Generating test split: 0%| | 0/165 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 165/165 [00:00<00:00, 45407.79 examples/s]
Generating validation split: 0%| | 0/18 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 18/18 [00:00<00:00, 10614.01 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3355.44 examples/s]
Generating test split: 0%| | 0/204 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 204/204 [00:00<00:00, 54350.38 examples/s]
Generating validation split: 0%| | 0/22 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 22/22 [00:00<00:00, 9625.98 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 2989.10 examples/s]
Generating test split: 0%| | 0/237 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 237/237 [00:00<00:00, 49877.07 examples/s]
Generating validation split: 0%| | 0/26 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 26/26 [00:00<00:00, 9236.99 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3063.77 examples/s]
Generating test split: 0%| | 0/121 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 121/121 [00:00<00:00, 32153.50 examples/s]
Generating validation split: 0%| | 0/13 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 13/13 [00:00<00:00, 4383.82 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1734.47 examples/s]
Generating test split: 0%| | 0/108 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 108/108 [00:00<00:00, 22462.80 examples/s]
Generating validation split: 0%| | 0/11 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 11/11 [00:00<00:00, 3137.31 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1612.57 examples/s]
Generating test split: 0%| | 0/163 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 163/163 [00:00<00:00, 56632.83 examples/s]
Generating validation split: 0%| | 0/18 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 18/18 [00:00<00:00, 11112.37 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3323.01 examples/s]
Generating test split: 0%| | 0/346 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 346/346 [00:00<00:00, 84773.01 examples/s]
Generating validation split: 0%| | 0/38 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 38/38 [00:00<00:00, 13389.08 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 1793.82 examples/s]
Generating test split: 0%| | 0/895 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 895/895 [00:00<00:00, 198325.34 examples/s]
Generating validation split: 0%| | 0/100 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 100/100 [00:00<00:00, 44544.44 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 2613.27 examples/s]
Generating test split: 0%| | 0/311 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 311/311 [00:00<00:00, 119540.74 examples/s]
Generating validation split: 0%| | 0/34 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 34/34 [00:00<00:00, 22359.10 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3291.20 examples/s]
Generating test split: 0%| | 0/324 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 324/324 [00:00<00:00, 113048.37 examples/s]
Generating validation split: 0%| | 0/35 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 35/35 [00:00<00:00, 22208.87 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3415.56 examples/s]
Generating test split: 0%| | 0/1534 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 1534/1534 [00:00<00:00, 166405.34 examples/s]
Generating validation split: 0%| | 0/170 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 170/170 [00:00<00:00, 56215.05 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 2611.32 examples/s]
Generating test split: 0%| | 0/171 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 171/171 [00:00<00:00, 47438.72 examples/s]
Generating validation split: 0%| | 0/19 [00:00<?, ? examples/s] Generating validation split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 19/19 [00:00<00:00, 9720.88 examples/s]
Generating dev split: 0%| | 0/5 [00:00<?, ? examples/s] Generating dev split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 5/5 [00:00<00:00, 3386.87 examples/s]
2026-07-24:07:03:30 INFO [evaluator_utils:446] Selected tasks:
2026-07-24:07:03:30 INFO [evaluator_utils:462] Group: mmlu
2026-07-24:07:03:30 INFO [evaluator_utils:462] Group: mmlu_stem
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_abstract_algebra (mmlu/default/mmlu_abstract_algebra.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_anatomy (mmlu/default/mmlu_anatomy.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_astronomy (mmlu/default/mmlu_astronomy.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_college_biology (mmlu/default/mmlu_college_biology.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_college_chemistry (mmlu/default/mmlu_college_chemistry.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_college_computer_science (mmlu/default/mmlu_college_computer_science.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_college_mathematics (mmlu/default/mmlu_college_mathematics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_college_physics (mmlu/default/mmlu_college_physics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_computer_security (mmlu/default/mmlu_computer_security.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_conceptual_physics (mmlu/default/mmlu_conceptual_physics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_electrical_engineering (mmlu/default/mmlu_electrical_engineering.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_elementary_mathematics (mmlu/default/mmlu_elementary_mathematics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_biology (mmlu/default/mmlu_high_school_biology.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_chemistry (mmlu/default/mmlu_high_school_chemistry.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_computer_science (mmlu/default/mmlu_high_school_computer_science.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_mathematics (mmlu/default/mmlu_high_school_mathematics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_physics (mmlu/default/mmlu_high_school_physics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_statistics (mmlu/default/mmlu_high_school_statistics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_machine_learning (mmlu/default/mmlu_machine_learning.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:462] Group: mmlu_other
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_business_ethics (mmlu/default/mmlu_business_ethics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_clinical_knowledge (mmlu/default/mmlu_clinical_knowledge.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_college_medicine (mmlu/default/mmlu_college_medicine.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_global_facts (mmlu/default/mmlu_global_facts.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_human_aging (mmlu/default/mmlu_human_aging.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_management (mmlu/default/mmlu_management.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_marketing (mmlu/default/mmlu_marketing.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_medical_genetics (mmlu/default/mmlu_medical_genetics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_miscellaneous (mmlu/default/mmlu_miscellaneous.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_nutrition (mmlu/default/mmlu_nutrition.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_professional_accounting (mmlu/default/mmlu_professional_accounting.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_professional_medicine (mmlu/default/mmlu_professional_medicine.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_virology (mmlu/default/mmlu_virology.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:462] Group: mmlu_social_sciences
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_econometrics (mmlu/default/mmlu_econometrics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_geography (mmlu/default/mmlu_high_school_geography.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_government_and_politics (mmlu/default/mmlu_high_school_government_and_politics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_macroeconomics (mmlu/default/mmlu_high_school_macroeconomics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_microeconomics (mmlu/default/mmlu_high_school_microeconomics.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_psychology (mmlu/default/mmlu_high_school_psychology.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_human_sexuality (mmlu/default/mmlu_human_sexuality.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_professional_psychology (mmlu/default/mmlu_professional_psychology.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_public_relations (mmlu/default/mmlu_public_relations.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_security_studies (mmlu/default/mmlu_security_studies.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_sociology (mmlu/default/mmlu_sociology.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_us_foreign_policy (mmlu/default/mmlu_us_foreign_policy.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:462] Group: mmlu_humanities
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_formal_logic (mmlu/default/mmlu_formal_logic.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_european_history (mmlu/default/mmlu_high_school_european_history.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_us_history (mmlu/default/mmlu_high_school_us_history.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_high_school_world_history (mmlu/default/mmlu_high_school_world_history.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_international_law (mmlu/default/mmlu_international_law.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_jurisprudence (mmlu/default/mmlu_jurisprudence.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_logical_fallacies (mmlu/default/mmlu_logical_fallacies.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_moral_disputes (mmlu/default/mmlu_moral_disputes.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_moral_scenarios (mmlu/default/mmlu_moral_scenarios.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_philosophy (mmlu/default/mmlu_philosophy.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_prehistory (mmlu/default/mmlu_prehistory.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_professional_law (mmlu/default/mmlu_professional_law.yaml)
2026-07-24:07:03:30 INFO [evaluator_utils:470] Task: mmlu_world_religions (mmlu/default/mmlu_world_religions.yaml)
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_abstract_algebra from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_anatomy from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_astronomy from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_biology from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_chemistry from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_computer_science from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_mathematics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_physics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_computer_security from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_conceptual_physics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_electrical_engineering from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_elementary_mathematics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_biology from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_chemistry from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_computer_science from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_mathematics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_physics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_statistics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_machine_learning from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_business_ethics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_clinical_knowledge from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_medicine from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_global_facts from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_aging from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_management from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_marketing from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_medical_genetics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_miscellaneous from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_nutrition from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_accounting from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_medicine from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_virology from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_econometrics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_geography from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_government_and_politics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_macroeconomics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_microeconomics from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_psychology from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_sexuality from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_psychology from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_public_relations from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_security_studies from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_sociology from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_us_foreign_policy from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_formal_logic from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_european_history from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_us_history from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_world_history from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_international_law from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_jurisprudence from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_logical_fallacies from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_disputes from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_scenarios from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_philosophy from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_prehistory from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_law from None to 5
2026-07-24:07:03:30 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_world_religions from None to 5
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_abstract_algebra on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 155.15it/s]
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_anatomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 178.73it/s]
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_astronomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 175.53it/s]
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_college_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 174.59it/s]
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_college_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 175.39it/s]
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_college_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 183.82it/s]
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_college_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.21it/s]
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_college_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 189.37it/s]
2026-07-24:07:03:30 INFO [api.task:312] Building contexts for mmlu_computer_security on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 189.93it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_conceptual_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 189.57it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_electrical_engineering on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.00it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_elementary_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.56it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_high_school_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.70it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_high_school_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.83it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_high_school_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.12it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_high_school_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.17it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_high_school_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.21it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_high_school_statistics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.71it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_machine_learning on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.46it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_business_ethics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.23it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_clinical_knowledge on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.03it/s]
2026-07-24:07:03:31 INFO [api.task:312] Building contexts for mmlu_college_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.95it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_global_facts on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.75it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_human_aging on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.14it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_management on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.86it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_marketing on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.73it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_medical_genetics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.32it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_miscellaneous on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.48it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_nutrition on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.25it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_professional_accounting on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.26it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_professional_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.58it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_virology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.78it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_econometrics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.31it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_high_school_geography on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.41it/s]
2026-07-24:07:03:32 INFO [api.task:312] Building contexts for mmlu_high_school_government_and_politics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 187.55it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_high_school_macroeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.12it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_high_school_microeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.03it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_high_school_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.52it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_human_sexuality on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.34it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_professional_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 174.80it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_public_relations on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 174.56it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_security_studies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 174.08it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_sociology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 164.39it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_us_foreign_policy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.55it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_formal_logic on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.50it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_high_school_european_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.52it/s]
2026-07-24:07:03:33 INFO [api.task:312] Building contexts for mmlu_high_school_us_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.47it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_high_school_world_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.61it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_international_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.41it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_jurisprudence on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.13it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_logical_fallacies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.99it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_moral_disputes on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.80it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_moral_scenarios on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.20it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_philosophy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.73it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_prehistory on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.56it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_professional_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.15it/s]
2026-07-24:07:03:34 INFO [api.task:312] Building contexts for mmlu_world_religions on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.83it/s]
2026-07-24:07:03:34 INFO [evaluator:585] Running loglikelihood requests
Tokenizing inputs: 0%| | 0/3420 [00:00<?, ?it/s] Tokenizing inputs: 2%|▏ | 69/3420 [00:00<00:04, 688.52it/s] Tokenizing inputs: 4%|▍ | 138/3420 [00:00<00:05, 642.25it/s] Tokenizing inputs: 6%|β–Œ | 203/3420 [00:00<00:05, 547.89it/s] Tokenizing inputs: 8%|β–Š | 260/3420 [00:00<00:05, 540.59it/s] Tokenizing inputs: 9%|β–‰ | 315/3420 [00:00<00:06, 495.90it/s] Tokenizing inputs: 11%|β–ˆ | 366/3420 [00:00<00:07, 431.15it/s] Tokenizing inputs: 12%|β–ˆβ– | 413/3420 [00:00<00:06, 440.16it/s] Tokenizing inputs: 14%|β–ˆβ–Ž | 465/3420 [00:00<00:06, 460.78it/s] Tokenizing inputs: 15%|β–ˆβ–Œ | 524/3420 [00:01<00:05, 496.98it/s] Tokenizing inputs: 18%|β–ˆβ–Š | 599/3420 [00:01<00:04, 568.34it/s] Tokenizing inputs: 19%|β–ˆβ–‰ | 658/3420 [00:01<00:04, 569.08it/s] Tokenizing inputs: 21%|β–ˆβ–ˆ | 716/3420 [00:01<00:05, 531.18it/s] Tokenizing inputs: 23%|β–ˆβ–ˆβ–Ž | 771/3420 [00:01<00:05, 521.78it/s] Tokenizing inputs: 24%|β–ˆβ–ˆβ– | 826/3420 [00:01<00:04, 529.20it/s] Tokenizing inputs: 26%|β–ˆβ–ˆβ–Œ | 880/3420 [00:01<00:05, 450.84it/s] Tokenizing inputs: 27%|β–ˆβ–ˆβ–‹ | 928/3420 [00:01<00:05, 429.83it/s] Tokenizing inputs: 29%|β–ˆβ–ˆβ–Š | 978/3420 [00:01<00:05, 447.03it/s] Tokenizing inputs: 30%|β–ˆβ–ˆβ–‰ | 1025/3420 [00:02<00:05, 440.73it/s] Tokenizing inputs: 31%|β–ˆβ–ˆβ–ˆβ– | 1071/3420 [00:02<00:05, 404.41it/s] Tokenizing inputs: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1113/3420 [00:02<00:05, 399.57it/s] Tokenizing inputs: 34%|β–ˆβ–ˆβ–ˆβ– | 1155/3420 [00:02<00:05, 404.04it/s] Tokenizing inputs: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 1202/3420 [00:02<00:05, 422.10it/s] Tokenizing inputs: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1265/3420 [00:02<00:04, 479.22it/s] Tokenizing inputs: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1314/3420 [00:02<00:04, 472.77it/s] Tokenizing inputs: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 1374/3420 [00:02<00:04, 508.72it/s] Tokenizing inputs: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1449/3420 [00:02<00:03, 577.81it/s] Tokenizing inputs: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1523/3420 [00:03<00:03, 624.64it/s] Tokenizing inputs: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1589/3420 [00:03<00:02, 632.67it/s] Tokenizing inputs: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1670/3420 [00:03<00:02, 684.23it/s] Tokenizing inputs: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1739/3420 [00:03<00:02, 606.21it/s] Tokenizing inputs: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1802/3420 [00:03<00:03, 516.91it/s] Tokenizing inputs: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1857/3420 [00:03<00:03, 407.04it/s] Tokenizing inputs: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1925/3420 [00:03<00:03, 463.62it/s] Tokenizing inputs: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1978/3420 [00:03<00:03, 459.74it/s] Tokenizing inputs: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2040/3420 [00:04<00:02, 497.00it/s] Tokenizing inputs: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2094/3420 [00:04<00:02, 499.56it/s] Tokenizing inputs: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2155/3420 [00:04<00:02, 527.80it/s] Tokenizing inputs: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2210/3420 [00:04<00:02, 527.72it/s] Tokenizing inputs: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2265/3420 [00:04<00:02, 523.55it/s] Tokenizing inputs: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2330/3420 [00:04<00:01, 558.50it/s] Tokenizing inputs: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2387/3420 [00:04<00:01, 522.94it/s] Tokenizing inputs: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2441/3420 [00:04<00:01, 521.05it/s] Tokenizing inputs: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2494/3420 [00:05<00:02, 417.89it/s] Tokenizing inputs: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2540/3420 [00:05<00:02, 376.86it/s] Tokenizing inputs: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2594/3420 [00:05<00:01, 413.97it/s] Tokenizing inputs: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2648/3420 [00:05<00:01, 444.15it/s] Tokenizing inputs: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2696/3420 [00:05<00:01, 448.40it/s] Tokenizing inputs: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2743/3420 [00:05<00:02, 233.68it/s] Tokenizing inputs: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2779/3420 [00:06<00:03, 183.09it/s] Tokenizing inputs: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2808/3420 [00:06<00:03, 161.78it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2832/3420 [00:06<00:03, 156.97it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2853/3420 [00:06<00:03, 159.82it/s] Tokenizing inputs: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2873/3420 [00:06<00:03, 161.85it/s] Tokenizing inputs: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2901/3420 [00:07<00:02, 184.73it/s] Tokenizing inputs: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2937/3420 [00:07<00:02, 223.33it/s] Tokenizing inputs: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2991/3420 [00:07<00:01, 298.85it/s] Tokenizing inputs: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3041/3420 [00:07<00:01, 349.84it/s] Tokenizing inputs: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3090/3420 [00:07<00:00, 386.20it/s] Tokenizing inputs: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3134/3420 [00:07<00:00, 398.80it/s] Tokenizing inputs: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3177/3420 [00:07<00:00, 389.36it/s] Tokenizing inputs: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3243/3420 [00:07<00:00, 463.62it/s] Tokenizing inputs: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3292/3420 [00:07<00:00, 451.66it/s] Tokenizing inputs: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3339/3420 [00:08<00:00, 297.20it/s] Tokenizing inputs: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3377/3420 [00:08<00:00, 271.75it/s] Tokenizing inputs: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 3420/3420 [00:08<00:00, 406.60it/s]
Running loglikelihood requests: 0%| | 0/3420 [00:00<?, ?it/s][W724 07:03:44.682734209 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 1398800384 bytes (free: 1334902784, total: 25323503616).
[W724 07:03:44.692117298 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 1398800384 bytes (free: 1336999936, total: 25323503616).
[W724 07:03:45.225058967 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2491416576 bytes (free: 1487994880, total: 25323503616).
[W724 07:03:45.350124050 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2491416576 bytes (free: 1490092032, total: 25323503616).
[W724 07:03:45.763971953 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2229272576 bytes (free: 1487994880, total: 25323503616).
[W724 07:03:45.015232763 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2229272576 bytes (free: 659619840, total: 25323503616).
[W724 07:03:45.015443062 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2229272576 bytes (free: 659619840, total: 25323503616).
[W724 07:03:46.429121736 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 5196742656 bytes (free: 657522688, total: 25323503616).
[W724 07:03:46.706313938 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 5196742656 bytes (free: 3457220608, total: 25323503616).
[W724 07:03:46.119323845 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 4620025856 bytes (free: 3455123456, total: 25323503616).
[W724 07:03:46.119594404 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 4620025856 bytes (free: 3455123456, total: 25323503616).
[W724 07:03:47.519073863 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 4158652416 bytes (free: 3455123456, total: 25323503616).
[W724 07:03:47.746748559 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 4158652416 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:47.747117978 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 4158652416 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:47.152750694 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3697278976 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:47.152949563 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3697278976 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:48.553544097 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3233808384 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:48.553752476 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3233808384 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:48.951723194 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2887778304 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:48.951923873 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2887778304 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:49.347026236 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2541748224 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:49.347203045 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2541748224 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:49.934171845 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2195718144 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:49.934344804 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2195718144 bytes (free: 1525743616, total: 25323503616).
[W724 07:03:50.431509363 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 1965031424 bytes (free: 1754333184, total: 25323503616).
[W724 07:03:50.968855997 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 1965031424 bytes (free: 1756430336, total: 25323503616).
[W724 07:03:51.418108671 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 1732247552 bytes (free: 256966656, total: 25323503616).
[W724 07:03:51.418341660 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 1732247552 bytes (free: 256966656, total: 25323503616).
[W724 07:03:54.498813424 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 12052332544 bytes (free: 514916352, total: 25323503616).
[W724 07:04:01.876636141 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 12052332544 bytes (free: 5483069440, total: 25323503616).
[W724 07:04:02.795794779 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 10198450176 bytes (free: 1668349952, total: 25323503616).
[W724 07:04:08.074967975 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 10198450176 bytes (free: 5944442880, total: 25323503616).
[W724 07:04:09.899865376 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 8344567808 bytes (free: 2821783552, total: 25323503616).
[W724 07:04:14.043266229 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 8344567808 bytes (free: 6405816320, total: 25323503616).
[W724 07:04:15.821188720 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 7417626624 bytes (free: 3629187072, total: 25323503616).
[W724 07:04:20.403925052 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 7417626624 bytes (free: 6842023936, total: 25323503616).
[W724 07:04:20.137264992 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 6490685440 bytes (free: 4411424768, total: 25323503616).
[W724 07:04:24.191626811 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 6490685440 bytes (free: 1506869248, total: 25323503616).
[W724 07:04:25.553668901 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 6490685440 bytes (free: 3241213952, total: 25323503616).
[W724 07:04:26.234742399 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 5563744256 bytes (free: 3239116800, total: 25323503616).
[W724 07:04:29.641853200 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 5563744256 bytes (free: 3241213952, total: 25323503616).
[W724 07:04:30.278318216 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 4636803072 bytes (free: 3239116800, total: 25323503616).
[W724 07:04:33.368086070 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 4636803072 bytes (free: 3241213952, total: 25323503616).
[W724 07:04:33.953711357 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3709861888 bytes (free: 3239116800, total: 25323503616).
[W724 07:04:36.440505036 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3709861888 bytes (free: 3241213952, total: 25323503616).
Running loglikelihood requests: 0%| | 1/3420 [01:05<62:17:01, 65.58s/it][W724 07:04:48.203727316 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 2720006144 bytes (free: 1825636352, total: 25323503616).
Running loglikelihood requests: 0%| | 13/3420 [01:07<3:34:22, 3.78s/it] Running loglikelihood requests: 1%| | 25/3420 [01:09<1:36:05, 1.70s/it] Running loglikelihood requests: 1%| | 37/3420 [01:11<56:33, 1.00s/it] Running loglikelihood requests: 1%|▏ | 49/3420 [01:13<37:35, 1.49it/s] Running loglikelihood requests: 2%|▏ | 61/3420 [01:14<26:41, 2.10it/s] Running loglikelihood requests: 2%|▏ | 73/3420 [01:16<20:01, 2.79it/s] Running loglikelihood requests: 2%|▏ | 85/3420 [01:17<15:45, 3.53it/s] Running loglikelihood requests: 3%|β–Ž | 97/3420 [01:19<12:57, 4.27it/s] Running loglikelihood requests: 3%|β–Ž | 109/3420 [01:20<11:04, 4.98it/s] Running loglikelihood requests: 4%|β–Ž | 121/3420 [01:22<09:28, 5.80it/s] Running loglikelihood requests: 4%|▍ | 133/3420 [01:23<08:14, 6.65it/s] Running loglikelihood requests: 4%|▍ | 145/3420 [01:24<07:20, 7.44it/s] Running loglikelihood requests: 5%|▍ | 157/3420 [01:25<06:37, 8.21it/s] Running loglikelihood requests: 5%|▍ | 169/3420 [01:26<06:07, 8.86it/s] Running loglikelihood requests: 5%|β–Œ | 181/3420 [01:27<05:42, 9.47it/s] Running loglikelihood requests: 6%|β–Œ | 193/3420 [01:28<05:20, 10.08it/s] Running loglikelihood requests: 6%|β–Œ | 205/3420 [01:29<05:00, 10.71it/s] Running loglikelihood requests: 6%|β–‹ | 217/3420 [01:30<04:43, 11.29it/s]
########## m0 / mmlu ##########
/usr/bin/python -m lm_eval --model hf --model_args pretrained=Qwen/Qwen2.5-7B-Instruct,dtype=bfloat16 --tasks mmlu --num_fewshot 5 --batch_size 4 --seed 0 --output_path /root/steering-resistance/results/full_7b/capability/m0/mmlu --apply_chat_template --fewshot_as_multiturn --limit 15
/usr/local/lib/python3.11/dist-packages/requests/__init__.py:113: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (6.0.0.post1)/charset_normalizer (3.3.2) doesn't match a supported version!
warnings.warn(
2026-07-24:07:05:35 WARNING [config.evaluate_config:287] --limit SHOULD ONLY BE USED FOR TESTING. REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.
2026-07-24:07:05:44 INFO [_cli.run:388] Selected Tasks: ['mmlu']
2026-07-24:07:05:46 INFO [evaluator:214] Setting random seed to 0 | Setting numpy seed to 0 | Setting torch manual seed to 0 | Setting fewshot manual seed to 0
2026-07-24:07:05:46 INFO [evaluator:239] Initializing hf model, with arguments: {'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16'}
2026-07-24:07:05:49 INFO [models.huggingface:286] Using device 'cuda:0'
2026-07-24:07:05:50 INFO [models.huggingface:579] Model parallel was set to False, max memory was not set, and device map was set to {'': 'cuda:0'}
Loading weights: 0%| | 0/339 [00:00<?, ?it/s] Loading weights: 1%| | 2/339 [00:00<00:34, 9.87it/s] Loading weights: 5%|▍ | 16/339 [00:00<00:05, 62.84it/s] Loading weights: 12%|β–ˆβ– | 41/339 [00:00<00:02, 131.42it/s] Loading weights: 19%|β–ˆβ–‰ | 65/339 [00:00<00:01, 151.64it/s] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 90/339 [00:00<00:01, 172.99it/s] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 113/339 [00:00<00:01, 184.11it/s] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 135/339 [00:00<00:01, 193.85it/s] Loading weights: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 160/339 [00:00<00:00, 193.63it/s] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 186/339 [00:01<00:00, 203.77it/s] Loading weights: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 208/339 [00:01<00:00, 203.09it/s] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 229/339 [00:01<00:00, 188.49it/s] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 249/339 [00:01<00:00, 184.38it/s] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 268/339 [00:01<00:00, 166.73it/s] Loading weights: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 286/339 [00:01<00:00, 170.02it/s] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 305/339 [00:01<00:00, 169.09it/s] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 329/339 [00:01<00:00, 177.91it/s] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 339/339 [00:01<00:00, 173.60it/s]
2026-07-24:07:06:52 INFO [evaluator_utils:446] Selected tasks:
2026-07-24:07:06:52 INFO [evaluator_utils:462] Group: mmlu
2026-07-24:07:06:52 INFO [evaluator_utils:462] Group: mmlu_stem
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_abstract_algebra (mmlu/default/mmlu_abstract_algebra.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_anatomy (mmlu/default/mmlu_anatomy.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_astronomy (mmlu/default/mmlu_astronomy.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_college_biology (mmlu/default/mmlu_college_biology.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_college_chemistry (mmlu/default/mmlu_college_chemistry.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_college_computer_science (mmlu/default/mmlu_college_computer_science.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_college_mathematics (mmlu/default/mmlu_college_mathematics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_college_physics (mmlu/default/mmlu_college_physics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_computer_security (mmlu/default/mmlu_computer_security.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_conceptual_physics (mmlu/default/mmlu_conceptual_physics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_electrical_engineering (mmlu/default/mmlu_electrical_engineering.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_elementary_mathematics (mmlu/default/mmlu_elementary_mathematics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_biology (mmlu/default/mmlu_high_school_biology.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_chemistry (mmlu/default/mmlu_high_school_chemistry.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_computer_science (mmlu/default/mmlu_high_school_computer_science.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_mathematics (mmlu/default/mmlu_high_school_mathematics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_physics (mmlu/default/mmlu_high_school_physics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_statistics (mmlu/default/mmlu_high_school_statistics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_machine_learning (mmlu/default/mmlu_machine_learning.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:462] Group: mmlu_other
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_business_ethics (mmlu/default/mmlu_business_ethics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_clinical_knowledge (mmlu/default/mmlu_clinical_knowledge.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_college_medicine (mmlu/default/mmlu_college_medicine.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_global_facts (mmlu/default/mmlu_global_facts.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_human_aging (mmlu/default/mmlu_human_aging.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_management (mmlu/default/mmlu_management.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_marketing (mmlu/default/mmlu_marketing.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_medical_genetics (mmlu/default/mmlu_medical_genetics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_miscellaneous (mmlu/default/mmlu_miscellaneous.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_nutrition (mmlu/default/mmlu_nutrition.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_professional_accounting (mmlu/default/mmlu_professional_accounting.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_professional_medicine (mmlu/default/mmlu_professional_medicine.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_virology (mmlu/default/mmlu_virology.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:462] Group: mmlu_social_sciences
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_econometrics (mmlu/default/mmlu_econometrics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_geography (mmlu/default/mmlu_high_school_geography.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_government_and_politics (mmlu/default/mmlu_high_school_government_and_politics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_macroeconomics (mmlu/default/mmlu_high_school_macroeconomics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_microeconomics (mmlu/default/mmlu_high_school_microeconomics.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_psychology (mmlu/default/mmlu_high_school_psychology.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_human_sexuality (mmlu/default/mmlu_human_sexuality.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_professional_psychology (mmlu/default/mmlu_professional_psychology.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_public_relations (mmlu/default/mmlu_public_relations.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_security_studies (mmlu/default/mmlu_security_studies.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_sociology (mmlu/default/mmlu_sociology.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_us_foreign_policy (mmlu/default/mmlu_us_foreign_policy.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:462] Group: mmlu_humanities
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_formal_logic (mmlu/default/mmlu_formal_logic.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_european_history (mmlu/default/mmlu_high_school_european_history.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_us_history (mmlu/default/mmlu_high_school_us_history.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_high_school_world_history (mmlu/default/mmlu_high_school_world_history.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_international_law (mmlu/default/mmlu_international_law.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_jurisprudence (mmlu/default/mmlu_jurisprudence.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_logical_fallacies (mmlu/default/mmlu_logical_fallacies.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_moral_disputes (mmlu/default/mmlu_moral_disputes.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_moral_scenarios (mmlu/default/mmlu_moral_scenarios.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_philosophy (mmlu/default/mmlu_philosophy.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_prehistory (mmlu/default/mmlu_prehistory.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_professional_law (mmlu/default/mmlu_professional_law.yaml)
2026-07-24:07:06:52 INFO [evaluator_utils:470] Task: mmlu_world_religions (mmlu/default/mmlu_world_religions.yaml)
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_abstract_algebra from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_anatomy from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_astronomy from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_biology from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_chemistry from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_computer_science from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_mathematics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_physics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_computer_security from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_conceptual_physics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_electrical_engineering from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_elementary_mathematics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_biology from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_chemistry from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_computer_science from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_mathematics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_physics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_statistics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_machine_learning from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_business_ethics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_clinical_knowledge from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_medicine from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_global_facts from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_aging from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_management from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_marketing from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_medical_genetics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_miscellaneous from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_nutrition from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_accounting from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_medicine from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_virology from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_econometrics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_geography from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_government_and_politics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_macroeconomics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_microeconomics from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_psychology from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_sexuality from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_psychology from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_public_relations from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_security_studies from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_sociology from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_us_foreign_policy from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_formal_logic from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_european_history from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_us_history from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_world_history from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_international_law from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_jurisprudence from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_logical_fallacies from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_disputes from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_scenarios from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_philosophy from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_prehistory from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_law from None to 5
2026-07-24:07:06:52 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_world_religions from None to 5
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_abstract_algebra on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 13/15 [00:00<00:00, 127.35it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 133.15it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_anatomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.18it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_astronomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.29it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_college_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.32it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_college_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.44it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_college_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.26it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_college_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.22it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_college_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.26it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_computer_security on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.82it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_conceptual_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.51it/s]
2026-07-24:07:06:52 INFO [api.task:312] Building contexts for mmlu_electrical_engineering on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.02it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_elementary_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.28it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_high_school_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.83it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_high_school_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.29it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_high_school_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.41it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_high_school_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.50it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_high_school_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.86it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_high_school_statistics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.61it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_machine_learning on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.88it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_business_ethics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.49it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_clinical_knowledge on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.34it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_college_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.97it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_global_facts on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.42it/s]
2026-07-24:07:06:53 INFO [api.task:312] Building contexts for mmlu_human_aging on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.50it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_management on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.60it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_marketing on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.71it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_medical_genetics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.32it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_miscellaneous on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.17it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_nutrition on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.36it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_professional_accounting on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.89it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_professional_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.02it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_virology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.34it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_econometrics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.63it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_high_school_geography on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.90it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_high_school_government_and_politics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.71it/s]
2026-07-24:07:06:54 INFO [api.task:312] Building contexts for mmlu_high_school_macroeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.25it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_high_school_microeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.10it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_high_school_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.59it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_human_sexuality on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.88it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_professional_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.32it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_public_relations on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.47it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_security_studies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.75it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_sociology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.27it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_us_foreign_policy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.39it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_formal_logic on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.38it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_high_school_european_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.31it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_high_school_us_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.43it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_high_school_world_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.26it/s]
2026-07-24:07:06:55 INFO [api.task:312] Building contexts for mmlu_international_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.33it/s]
2026-07-24:07:06:56 INFO [api.task:312] Building contexts for mmlu_jurisprudence on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.21it/s]
2026-07-24:07:06:56 INFO [api.task:312] Building contexts for mmlu_logical_fallacies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.42it/s]
2026-07-24:07:06:56 INFO [api.task:312] Building contexts for mmlu_moral_disputes on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.74it/s]
2026-07-24:07:06:56 INFO [api.task:312] Building contexts for mmlu_moral_scenarios on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.07it/s]
2026-07-24:07:06:56 INFO [api.task:312] Building contexts for mmlu_philosophy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.75it/s]
2026-07-24:07:06:56 INFO [api.task:312] Building contexts for mmlu_prehistory on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.94it/s]
2026-07-24:07:06:56 INFO [api.task:312] Building contexts for mmlu_professional_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.25it/s]
2026-07-24:07:06:56 INFO [api.task:312] Building contexts for mmlu_world_religions on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.53it/s]
2026-07-24:07:06:56 INFO [evaluator:585] Running loglikelihood requests
Tokenizing inputs: 0%| | 0/3420 [00:00<?, ?it/s] Tokenizing inputs: 2%|▏ | 69/3420 [00:00<00:04, 689.67it/s] Tokenizing inputs: 4%|▍ | 138/3420 [00:00<00:05, 644.59it/s] Tokenizing inputs: 6%|β–Œ | 203/3420 [00:00<00:05, 548.36it/s] Tokenizing inputs: 8%|β–Š | 260/3420 [00:00<00:05, 540.88it/s] Tokenizing inputs: 9%|β–‰ | 315/3420 [00:00<00:06, 497.46it/s] Tokenizing inputs: 11%|β–ˆ | 366/3420 [00:00<00:07, 433.79it/s] Tokenizing inputs: 12%|β–ˆβ– | 413/3420 [00:00<00:06, 441.42it/s] Tokenizing inputs: 14%|β–ˆβ–Ž | 465/3420 [00:00<00:06, 461.89it/s] Tokenizing inputs: 15%|β–ˆβ–Œ | 524/3420 [00:01<00:05, 496.96it/s] Tokenizing inputs: 18%|β–ˆβ–Š | 599/3420 [00:01<00:04, 568.22it/s] Tokenizing inputs: 19%|β–ˆβ–‰ | 661/3420 [00:01<00:04, 579.82it/s] Tokenizing inputs: 21%|β–ˆβ–ˆ | 720/3420 [00:01<00:04, 549.77it/s] Tokenizing inputs: 23%|β–ˆβ–ˆβ–Ž | 776/3420 [00:01<00:04, 539.40it/s] Tokenizing inputs: 24%|β–ˆβ–ˆβ– | 831/3420 [00:01<00:04, 541.98it/s] Tokenizing inputs: 26%|β–ˆβ–ˆβ–Œ | 886/3420 [00:01<00:05, 450.52it/s] Tokenizing inputs: 27%|β–ˆβ–ˆβ–‹ | 934/3420 [00:01<00:05, 439.02it/s] Tokenizing inputs: 29%|β–ˆβ–ˆβ–Š | 983/3420 [00:01<00:05, 451.03it/s] Tokenizing inputs: 30%|β–ˆβ–ˆβ–ˆ | 1030/3420 [00:02<00:05, 441.15it/s] Tokenizing inputs: 31%|β–ˆβ–ˆβ–ˆβ– | 1076/3420 [00:02<00:05, 405.30it/s] Tokenizing inputs: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1118/3420 [00:02<00:05, 402.10it/s] Tokenizing inputs: 34%|β–ˆβ–ˆβ–ˆβ– | 1161/3420 [00:02<00:05, 408.87it/s] Tokenizing inputs: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 1207/3420 [00:02<00:05, 422.45it/s] Tokenizing inputs: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1269/3420 [00:02<00:04, 476.15it/s] Tokenizing inputs: 39%|β–ˆβ–ˆβ–ˆβ–Š | 1318/3420 [00:02<00:04, 469.70it/s] Tokenizing inputs: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 1380/3420 [00:02<00:03, 511.28it/s] Tokenizing inputs: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1457/3420 [00:02<00:03, 584.21it/s] Tokenizing inputs: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1529/3420 [00:03<00:03, 620.77it/s] Tokenizing inputs: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1595/3420 [00:03<00:02, 631.25it/s] Tokenizing inputs: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1676/3420 [00:03<00:02, 682.89it/s] Tokenizing inputs: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1745/3420 [00:03<00:02, 590.84it/s] Tokenizing inputs: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1807/3420 [00:03<00:03, 500.68it/s] Tokenizing inputs: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1861/3420 [00:03<00:03, 402.34it/s] Tokenizing inputs: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1929/3420 [00:03<00:03, 461.08it/s] Tokenizing inputs: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1981/3420 [00:03<00:03, 457.92it/s] Tokenizing inputs: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2043/3420 [00:04<00:02, 494.94it/s] Tokenizing inputs: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2097/3420 [00:04<00:02, 498.11it/s] Tokenizing inputs: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2158/3420 [00:04<00:02, 526.65it/s] Tokenizing inputs: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2217/3420 [00:04<00:02, 543.73it/s] Tokenizing inputs: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2274/3420 [00:04<00:02, 529.85it/s] Tokenizing inputs: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2342/3420 [00:04<00:01, 568.71it/s] Tokenizing inputs: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2400/3420 [00:04<00:01, 513.39it/s] Tokenizing inputs: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2457/3420 [00:04<00:01, 526.81it/s] Tokenizing inputs: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2511/3420 [00:05<00:02, 380.90it/s] Tokenizing inputs: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2556/3420 [00:05<00:02, 388.48it/s] Tokenizing inputs: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2610/3420 [00:05<00:01, 422.75it/s] Tokenizing inputs: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2660/3420 [00:05<00:01, 442.05it/s] Tokenizing inputs: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2708/3420 [00:05<00:01, 385.03it/s] Tokenizing inputs: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2750/3420 [00:06<00:03, 217.31it/s] Tokenizing inputs: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2783/3420 [00:06<00:03, 179.22it/s] Tokenizing inputs: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2809/3420 [00:06<00:03, 163.04it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2831/3420 [00:06<00:03, 158.13it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2851/3420 [00:06<00:03, 161.34it/s] Tokenizing inputs: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2870/3420 [00:06<00:03, 162.84it/s] Tokenizing inputs: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2895/3420 [00:07<00:02, 180.63it/s] Tokenizing inputs: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2931/3420 [00:07<00:02, 221.80it/s] Tokenizing inputs: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2981/3420 [00:07<00:01, 289.84it/s] Tokenizing inputs: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3033/3420 [00:07<00:01, 348.34it/s] Tokenizing inputs: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3082/3420 [00:07<00:00, 385.29it/s] Tokenizing inputs: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3127/3420 [00:07<00:00, 403.01it/s] Tokenizing inputs: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3170/3420 [00:07<00:00, 391.85it/s] Tokenizing inputs: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3231/3420 [00:07<00:00, 452.34it/s] Tokenizing inputs: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3278/3420 [00:07<00:00, 456.84it/s] Tokenizing inputs: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3325/3420 [00:08<00:00, 334.94it/s] Tokenizing inputs: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3364/3420 [00:08<00:00, 254.47it/s] Tokenizing inputs: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 3420/3420 [00:08<00:00, 407.74it/s]
Running loglikelihood requests: 0%| | 0/3420 [00:00<?, ?it/s][W724 07:07:06.273798252 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3709861888 bytes (free: 3373334528, total: 25323503616).
Running loglikelihood requests: 0%| | 1/3420 [00:03<3:25:09, 3.60s/it][W724 07:07:08.102283239 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3590324224 bytes (free: 2310078464, total: 25323503616).
[W724 07:07:11.318059393 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3590324224 bytes (free: 2312175616, total: 25323503616).
Traceback (most recent call last):
File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "/usr/local/lib/python3.11/dist-packages/lm_eval/__main__.py", line 14, in <module>
cli_evaluate()
File "/usr/local/lib/python3.11/dist-packages/lm_eval/__main__.py", line 10, in cli_evaluate
parser.execute(args)
File "/usr/local/lib/python3.11/dist-packages/lm_eval/_cli/harness.py", line 60, in execute
args.func(args)
File "/usr/local/lib/python3.11/dist-packages/lm_eval/_cli/run.py", line 391, in _execute
results = simple_evaluate(
^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/utils.py", line 575, in _wrapper
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/evaluator.py", line 358, in simple_evaluate
results = evaluate(
^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/utils.py", line 575, in _wrapper
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/evaluator.py", line 596, in evaluate
resps = getattr(lm, reqtype)(cloned_reqs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/models/huggingface.py", line 1227, in loglikelihood
return super().loglikelihood(requests, disable_tqdm=disable_tqdm)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/api/model.py", line 446, in loglikelihood
return self._loglikelihood_tokens(new_reqs, disable_tqdm=disable_tqdm)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/models/huggingface.py", line 1507, in _loglikelihood_tokens
self._model_call(batched_inps, **call_kwargs),
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/models/huggingface.py", line 1154, in _model_call
return self.model(inps).logits
^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/transformers/utils/generic.py", line 911, in wrapper
output = func(self, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/transformers/models/qwen2/modeling_qwen2.py", line 475, in forward
logits = self.lm_head(hidden_states[:, slice_indices, :])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py", line 1778, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/torch/nn/modules/module.py", line 1789, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/torch/nn/modules/linear.py", line 134, in forward
return F.linear(input, self.weight, self.bias)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3.34 GiB. GPU 0 has a total capacity of 23.58 GiB of which 2.15 GiB is free. Process 1398907 has 21.42 GiB memory in use. Of the allocated memory 18.36 GiB is allocated by PyTorch, and 2.76 GiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://docs.pytorch.org/docs/stable/notes/cuda.html#optimizing-memory-usage-with-pytorch-cuda-alloc-conf)
Running loglikelihood requests: 0%| | 16/3420 [00:06<22:48, 2.49it/s]
########## m0 / gsm8k_cot ##########
/usr/bin/python -m lm_eval --model hf --model_args pretrained=Qwen/Qwen2.5-7B-Instruct,dtype=bfloat16 --tasks gsm8k_cot --num_fewshot 8 --batch_size 4 --seed 0 --output_path /root/steering-resistance/results/full_7b/capability/m0/gsm8k_cot --gen_kwargs do_sample=False --apply_chat_template --fewshot_as_multiturn --limit 200
/usr/local/lib/python3.11/dist-packages/requests/__init__.py:113: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (6.0.0.post1)/charset_normalizer (3.3.2) doesn't match a supported version!
warnings.warn(
2026-07-24:07:07:13 WARNING [config.evaluate_config:287] --limit SHOULD ONLY BE USED FOR TESTING. REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.
2026-07-24:07:07:21 INFO [_cli.run:388] Selected Tasks: ['gsm8k_cot']
2026-07-24:07:07:23 INFO [evaluator:214] Setting random seed to 0 | Setting numpy seed to 0 | Setting torch manual seed to 0 | Setting fewshot manual seed to 0
2026-07-24:07:07:23 WARNING [evaluator:226] generation_kwargs: {'do_sample': False} specified through cli, these settings will update set parameters in yaml tasks. Ensure 'do_sample=True' for non-greedy decoding!
2026-07-24:07:07:23 INFO [evaluator:239] Initializing hf model, with arguments: {'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16'}
2026-07-24:07:07:26 INFO [models.huggingface:286] Using device 'cuda:0'
2026-07-24:07:07:27 INFO [models.huggingface:579] Model parallel was set to False, max memory was not set, and device map was set to {'': 'cuda:0'}
Loading weights: 0%| | 0/339 [00:00<?, ?it/s] Loading weights: 1%| | 2/339 [00:00<00:34, 9.85it/s] Loading weights: 5%|β–Œ | 18/339 [00:00<00:04, 66.14it/s] Loading weights: 12%|β–ˆβ– | 39/339 [00:00<00:02, 115.57it/s] Loading weights: 16%|β–ˆβ–Œ | 54/339 [00:00<00:02, 121.92it/s] Loading weights: 22%|β–ˆβ–ˆβ– | 76/339 [00:00<00:01, 150.89it/s] Loading weights: 27%|β–ˆβ–ˆβ–‹ | 93/339 [00:00<00:01, 150.53it/s] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 111/339 [00:00<00:01, 155.32it/s] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 128/339 [00:00<00:01, 152.75it/s] Loading weights: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 147/339 [00:01<00:01, 161.08it/s] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 164/339 [00:01<00:01, 155.54it/s] Loading weights: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 184/339 [00:01<00:00, 167.37it/s] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 202/339 [00:01<00:00, 162.65it/s] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 219/339 [00:01<00:00, 147.37it/s] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 235/339 [00:01<00:00, 138.48it/s] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 256/339 [00:01<00:00, 134.67it/s] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 270/339 [00:01<00:00, 130.64it/s] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 292/339 [00:02<00:00, 143.60it/s] Loading weights: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 312/339 [00:02<00:00, 156.82it/s] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 329/339 [00:02<00:00, 145.99it/s] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 339/339 [00:02<00:00, 143.99it/s]
Generating train split: 0%| | 0/7473 [00:00<?, ? examples/s] Generating train split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 7473/7473 [00:00<00:00, 325078.14 examples/s]
Generating test split: 0%| | 0/1319 [00:00<?, ? examples/s] Generating test split: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 1319/1319 [00:00<00:00, 249775.93 examples/s]
2026-07-24:07:07:35 INFO [evaluator_utils:446] Selected tasks:
2026-07-24:07:07:35 INFO [evaluator_utils:480] Task: gsm8k_cot (gsm8k/gsm8k-cot.yaml)
2026-07-24:07:07:35 INFO [evaluator:314] gsm8k_cot: Using gen_kwargs: {'do_sample': False, 'until': ['Q:', '</s>', '<|im_end|>']}
2026-07-24:07:07:35 WARNING [evaluator:333] Overwriting default num_fewshot of gsm8k_cot from 8 to 8
2026-07-24:07:07:35 INFO [api.task:312] Building contexts for gsm8k_cot on rank 0...
0%| | 0/200 [00:00<?, ?it/s] 4%|β–Ž | 7/200 [00:00<00:02, 67.89it/s] 9%|β–‰ | 18/200 [00:00<00:02, 88.55it/s] 14%|β–ˆβ– | 29/200 [00:00<00:01, 95.06it/s] 20%|β–ˆβ–ˆ | 40/200 [00:00<00:01, 98.12it/s] 26%|β–ˆβ–ˆβ–Œ | 51/200 [00:00<00:01, 99.85it/s] 31%|β–ˆβ–ˆβ–ˆ | 62/200 [00:00<00:01, 100.96it/s] 36%|β–ˆβ–ˆβ–ˆβ–‹ | 73/200 [00:00<00:01, 101.60it/s] 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 84/200 [00:00<00:01, 102.07it/s] 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 95/200 [00:00<00:01, 102.39it/s] 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 106/200 [00:01<00:00, 102.62it/s] 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 117/200 [00:01<00:00, 102.76it/s] 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 128/200 [00:01<00:00, 102.89it/s] 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 139/200 [00:01<00:00, 103.03it/s] 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 150/200 [00:01<00:00, 103.07it/s] 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 161/200 [00:01<00:00, 103.08it/s] 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 172/200 [00:01<00:00, 103.16it/s] 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 183/200 [00:01<00:00, 103.17it/s] 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 194/200 [00:01<00:00, 103.22it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 200/200 [00:01<00:00, 101.22it/s]
2026-07-24:07:07:37 INFO [evaluator:585] Running generate_until requests
Running generate_until requests: 0%| | 0/200 [00:00<?, ?it/s] Running generate_until requests: 0%| | 1/200 [00:09<32:36, 9.83s/it] Running generate_until requests: 2%|β–Ž | 5/200 [00:17<10:13, 3.15s/it] Running generate_until requests: 4%|▍ | 9/200 [00:26<08:13, 2.58s/it] Running generate_until requests: 6%|β–‹ | 13/200 [00:34<07:24, 2.38s/it] Running generate_until requests: 8%|β–Š | 17/200 [00:43<06:56, 2.28s/it] Running generate_until requests: 10%|β–ˆ | 21/200 [00:51<06:37, 2.22s/it] Running generate_until requests: 12%|β–ˆβ–Ž | 25/200 [01:00<06:22, 2.19s/it] Running generate_until requests: 14%|β–ˆβ– | 29/200 [01:08<06:09, 2.16s/it] Running generate_until requests: 16%|β–ˆβ–‹ | 33/200 [01:17<05:58, 2.15s/it] Running generate_until requests: 18%|β–ˆβ–Š | 37/200 [01:22<05:11, 1.91s/it] Running generate_until requests: 20%|β–ˆβ–ˆ | 41/200 [01:30<05:13, 1.97s/it] Running generate_until requests: 22%|β–ˆβ–ˆβ–Ž | 45/200 [01:39<05:12, 2.02s/it] Running generate_until requests: 24%|β–ˆβ–ˆβ– | 49/200 [01:47<05:09, 2.05s/it] Running generate_until requests: 26%|β–ˆβ–ˆβ–‹ | 53/200 [01:56<05:04, 2.07s/it] Running generate_until requests: 28%|β–ˆβ–ˆβ–Š | 57/200 [02:04<04:58, 2.09s/it] Running generate_until requests: 30%|β–ˆβ–ˆβ–ˆ | 61/200 [02:13<04:51, 2.10s/it] Running generate_until requests: 32%|β–ˆβ–ˆβ–ˆβ–Ž | 65/200 [02:21<04:43, 2.10s/it] Running generate_until requests: 34%|β–ˆβ–ˆβ–ˆβ– | 69/200 [02:29<04:29, 2.06s/it] Running generate_until requests: 36%|β–ˆβ–ˆβ–ˆβ–‹ | 73/200 [02:38<04:23, 2.07s/it] Running generate_until requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 77/200 [02:46<04:16, 2.08s/it] Running generate_until requests: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 81/200 [02:54<04:08, 2.09s/it] Running generate_until requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 85/200 [03:01<03:47, 1.98s/it] Running generate_until requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 89/200 [03:08<03:26, 1.86s/it] Running generate_until requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 93/200 [03:14<03:12, 1.79s/it] Running generate_until requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 97/200 [03:21<03:01, 1.76s/it] Running generate_until requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 101/200 [03:29<03:04, 1.87s/it] Running generate_until requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 105/200 [03:36<02:53, 1.82s/it] Running generate_until requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 109/200 [03:41<02:27, 1.62s/it] Running generate_until requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 113/200 [03:49<02:32, 1.75s/it] Running generate_until requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 117/200 [03:57<02:28, 1.79s/it] Running generate_until requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 121/200 [04:05<02:28, 1.88s/it] Running generate_until requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 125/200 [04:12<02:20, 1.87s/it] Running generate_until requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 129/200 [04:19<02:05, 1.77s/it] Running generate_until requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 133/200 [04:24<01:50, 1.65s/it] Running generate_until requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 137/200 [04:29<01:36, 1.54s/it] Running generate_until requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 141/200 [04:37<01:37, 1.65s/it] Running generate_until requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 145/200 [04:43<01:30, 1.64s/it] Running generate_until requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 149/200 [04:51<01:28, 1.74s/it] Running generate_until requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 153/200 [04:58<01:21, 1.73s/it] Running generate_until requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 157/200 [05:05<01:13, 1.72s/it] Running generate_until requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 161/200 [05:12<01:07, 1.74s/it] Running generate_until requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 165/200 [05:20<01:03, 1.81s/it] Running generate_until requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 169/200 [05:26<00:54, 1.76s/it] Running generate_until requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 173/200 [05:32<00:45, 1.67s/it] Running generate_until requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 177/200 [05:38<00:37, 1.63s/it] Running generate_until requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 181/200 [05:44<00:30, 1.60s/it] Running generate_until requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 185/200 [05:49<00:22, 1.47s/it] Running generate_until requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 189/200 [05:56<00:16, 1.52s/it] Running generate_until requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 193/200 [06:04<00:11, 1.69s/it] Running generate_until requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 197/200 [06:11<00:05, 1.70s/it] Running generate_until requests: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 200/200 [06:11<00:00, 1.86s/it]
fatal: detected dubious ownership in repository at '/root/steering-resistance'
To add an exception for this directory, call:
git config --global --add safe.directory /root/steering-resistance
2026-07-24:07:13:50 INFO [loggers.evaluation_tracker:247] Saving results aggregated
hf ({'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16'}), gen_kwargs: ({'do_sample': False}), limit: 200.0, num_fewshot: 8, batch_size: 4
| Tasks |Version| Filter |n-shot| Metric | |Value| |Stderr|
|---------|------:|----------------|-----:|-----------|---|----:|---|-----:|
|gsm8k_cot| 3|flexible-extract| 8|exact_match|↑ | 0.78|Β± |0.0294|
| | |strict-match | 8|exact_match|↑ | 0.33|Β± |0.0333|
########## m1 / mmlu ##########
/usr/bin/python -m lm_eval --model hf --model_args pretrained=Qwen/Qwen2.5-7B-Instruct,dtype=bfloat16,peft=/root/steering-resistance/results/full_7b/m1_resist_adapter --tasks mmlu --num_fewshot 5 --batch_size 4 --seed 0 --output_path /root/steering-resistance/results/full_7b/capability/m1/mmlu --apply_chat_template --fewshot_as_multiturn --limit 15
/usr/local/lib/python3.11/dist-packages/requests/__init__.py:113: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (6.0.0.post1)/charset_normalizer (3.3.2) doesn't match a supported version!
warnings.warn(
2026-07-24:07:13:52 WARNING [config.evaluate_config:287] --limit SHOULD ONLY BE USED FOR TESTING. REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.
2026-07-24:07:14:00 INFO [_cli.run:388] Selected Tasks: ['mmlu']
2026-07-24:07:14:02 INFO [evaluator:214] Setting random seed to 0 | Setting numpy seed to 0 | Setting torch manual seed to 0 | Setting fewshot manual seed to 0
2026-07-24:07:14:02 INFO [evaluator:239] Initializing hf model, with arguments: {'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16', 'peft': '/root/steering-resistance/results/full_7b/m1_resist_adapter'}
2026-07-24:07:14:05 INFO [models.huggingface:286] Using device 'cuda:0'
2026-07-24:07:14:06 INFO [models.huggingface:579] Model parallel was set to False, max memory was not set, and device map was set to {'': 'cuda:0'}
Loading weights: 0%| | 0/339 [00:00<?, ?it/s] Loading weights: 1%| | 2/339 [00:00<00:34, 9.85it/s] Loading weights: 5%|β–Œ | 18/339 [00:00<00:04, 69.13it/s] Loading weights: 12%|β–ˆβ– | 41/339 [00:00<00:02, 123.65it/s] Loading weights: 17%|β–ˆβ–‹ | 57/339 [00:00<00:02, 135.31it/s] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 77/339 [00:00<00:01, 147.27it/s] Loading weights: 28%|β–ˆβ–ˆβ–Š | 96/339 [00:00<00:01, 158.38it/s] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 113/339 [00:00<00:01, 143.25it/s] Loading weights: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 136/339 [00:00<00:01, 165.49it/s] Loading weights: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 154/339 [00:01<00:01, 157.15it/s] Loading weights: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 172/339 [00:01<00:01, 155.25it/s] Loading weights: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 192/339 [00:01<00:00, 166.30it/s] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 210/339 [00:01<00:00, 135.28it/s] Loading weights: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 232/339 [00:01<00:00, 153.58it/s] Loading weights: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 249/339 [00:01<00:00, 152.49it/s] Loading weights: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 266/339 [00:01<00:00, 140.85it/s] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 281/339 [00:02<00:00, 138.29it/s] Loading weights: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 298/339 [00:02<00:00, 144.97it/s] Loading weights: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 317/339 [00:02<00:00, 150.66it/s] Loading weights: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 336/339 [00:02<00:00, 160.03it/s] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 339/339 [00:02<00:00, 145.21it/s]
2026-07-24:07:14:11 INFO [models.huggingface:885] Model config indicates vocab_size='152064', but found tokenizer with vocab size '151665'. Resizing model embedding layer...
2026-07-24:07:15:10 INFO [evaluator_utils:446] Selected tasks:
2026-07-24:07:15:10 INFO [evaluator_utils:462] Group: mmlu
2026-07-24:07:15:10 INFO [evaluator_utils:462] Group: mmlu_stem
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_abstract_algebra (mmlu/default/mmlu_abstract_algebra.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_anatomy (mmlu/default/mmlu_anatomy.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_astronomy (mmlu/default/mmlu_astronomy.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_college_biology (mmlu/default/mmlu_college_biology.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_college_chemistry (mmlu/default/mmlu_college_chemistry.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_college_computer_science (mmlu/default/mmlu_college_computer_science.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_college_mathematics (mmlu/default/mmlu_college_mathematics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_college_physics (mmlu/default/mmlu_college_physics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_computer_security (mmlu/default/mmlu_computer_security.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_conceptual_physics (mmlu/default/mmlu_conceptual_physics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_electrical_engineering (mmlu/default/mmlu_electrical_engineering.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_elementary_mathematics (mmlu/default/mmlu_elementary_mathematics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_biology (mmlu/default/mmlu_high_school_biology.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_chemistry (mmlu/default/mmlu_high_school_chemistry.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_computer_science (mmlu/default/mmlu_high_school_computer_science.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_mathematics (mmlu/default/mmlu_high_school_mathematics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_physics (mmlu/default/mmlu_high_school_physics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_statistics (mmlu/default/mmlu_high_school_statistics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_machine_learning (mmlu/default/mmlu_machine_learning.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:462] Group: mmlu_other
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_business_ethics (mmlu/default/mmlu_business_ethics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_clinical_knowledge (mmlu/default/mmlu_clinical_knowledge.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_college_medicine (mmlu/default/mmlu_college_medicine.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_global_facts (mmlu/default/mmlu_global_facts.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_human_aging (mmlu/default/mmlu_human_aging.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_management (mmlu/default/mmlu_management.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_marketing (mmlu/default/mmlu_marketing.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_medical_genetics (mmlu/default/mmlu_medical_genetics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_miscellaneous (mmlu/default/mmlu_miscellaneous.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_nutrition (mmlu/default/mmlu_nutrition.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_professional_accounting (mmlu/default/mmlu_professional_accounting.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_professional_medicine (mmlu/default/mmlu_professional_medicine.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_virology (mmlu/default/mmlu_virology.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:462] Group: mmlu_social_sciences
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_econometrics (mmlu/default/mmlu_econometrics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_geography (mmlu/default/mmlu_high_school_geography.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_government_and_politics (mmlu/default/mmlu_high_school_government_and_politics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_macroeconomics (mmlu/default/mmlu_high_school_macroeconomics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_microeconomics (mmlu/default/mmlu_high_school_microeconomics.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_psychology (mmlu/default/mmlu_high_school_psychology.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_human_sexuality (mmlu/default/mmlu_human_sexuality.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_professional_psychology (mmlu/default/mmlu_professional_psychology.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_public_relations (mmlu/default/mmlu_public_relations.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_security_studies (mmlu/default/mmlu_security_studies.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_sociology (mmlu/default/mmlu_sociology.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_us_foreign_policy (mmlu/default/mmlu_us_foreign_policy.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:462] Group: mmlu_humanities
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_formal_logic (mmlu/default/mmlu_formal_logic.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_european_history (mmlu/default/mmlu_high_school_european_history.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_us_history (mmlu/default/mmlu_high_school_us_history.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_high_school_world_history (mmlu/default/mmlu_high_school_world_history.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_international_law (mmlu/default/mmlu_international_law.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_jurisprudence (mmlu/default/mmlu_jurisprudence.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_logical_fallacies (mmlu/default/mmlu_logical_fallacies.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_moral_disputes (mmlu/default/mmlu_moral_disputes.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_moral_scenarios (mmlu/default/mmlu_moral_scenarios.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_philosophy (mmlu/default/mmlu_philosophy.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_prehistory (mmlu/default/mmlu_prehistory.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_professional_law (mmlu/default/mmlu_professional_law.yaml)
2026-07-24:07:15:10 INFO [evaluator_utils:470] Task: mmlu_world_religions (mmlu/default/mmlu_world_religions.yaml)
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_abstract_algebra from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_anatomy from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_astronomy from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_biology from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_chemistry from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_computer_science from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_mathematics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_physics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_computer_security from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_conceptual_physics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_electrical_engineering from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_elementary_mathematics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_biology from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_chemistry from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_computer_science from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_mathematics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_physics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_statistics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_machine_learning from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_business_ethics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_clinical_knowledge from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_medicine from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_global_facts from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_aging from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_management from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_marketing from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_medical_genetics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_miscellaneous from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_nutrition from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_accounting from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_medicine from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_virology from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_econometrics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_geography from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_government_and_politics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_macroeconomics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_microeconomics from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_psychology from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_sexuality from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_psychology from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_public_relations from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_security_studies from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_sociology from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_us_foreign_policy from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_formal_logic from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_european_history from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_us_history from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_world_history from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_international_law from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_jurisprudence from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_logical_fallacies from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_disputes from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_scenarios from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_philosophy from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_prehistory from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_law from None to 5
2026-07-24:07:15:10 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_world_religions from None to 5
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_abstract_algebra on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 13/15 [00:00<00:00, 127.92it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 133.70it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_anatomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.46it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_astronomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.13it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_college_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.70it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_college_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.03it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_college_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.49it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_college_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.39it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_college_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.62it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_computer_security on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.29it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_conceptual_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.47it/s]
2026-07-24:07:15:10 INFO [api.task:312] Building contexts for mmlu_electrical_engineering on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.53it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_elementary_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.25it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_high_school_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.36it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_high_school_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 185.51it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_high_school_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.53it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_high_school_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.31it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_high_school_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.02it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_high_school_statistics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.32it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_machine_learning on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.76it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_business_ethics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.06it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_clinical_knowledge on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.86it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_college_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.44it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_global_facts on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.54it/s]
2026-07-24:07:15:11 INFO [api.task:312] Building contexts for mmlu_human_aging on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.93it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_management on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.09it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_marketing on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.53it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_medical_genetics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.09it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_miscellaneous on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.79it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_nutrition on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 187.00it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_professional_accounting on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 189.48it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_professional_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.81it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_virology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.63it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_econometrics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.46it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_high_school_geography on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.57it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_high_school_government_and_politics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.17it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_high_school_macroeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.86it/s]
2026-07-24:07:15:12 INFO [api.task:312] Building contexts for mmlu_high_school_microeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.29it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_high_school_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.61it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_human_sexuality on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.39it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_professional_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.26it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_public_relations on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.43it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_security_studies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.44it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_sociology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.22it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_us_foreign_policy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.48it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_formal_logic on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.20it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_high_school_european_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.45it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_high_school_us_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.43it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_high_school_world_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.70it/s]
2026-07-24:07:15:13 INFO [api.task:312] Building contexts for mmlu_international_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.88it/s]
2026-07-24:07:15:14 INFO [api.task:312] Building contexts for mmlu_jurisprudence on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.23it/s]
2026-07-24:07:15:14 INFO [api.task:312] Building contexts for mmlu_logical_fallacies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.81it/s]
2026-07-24:07:15:14 INFO [api.task:312] Building contexts for mmlu_moral_disputes on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.04it/s]
2026-07-24:07:15:14 INFO [api.task:312] Building contexts for mmlu_moral_scenarios on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.97it/s]
2026-07-24:07:15:14 INFO [api.task:312] Building contexts for mmlu_philosophy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.98it/s]
2026-07-24:07:15:14 INFO [api.task:312] Building contexts for mmlu_prehistory on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.91it/s]
2026-07-24:07:15:14 INFO [api.task:312] Building contexts for mmlu_professional_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.92it/s]
2026-07-24:07:15:14 INFO [api.task:312] Building contexts for mmlu_world_religions on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.54it/s]
2026-07-24:07:15:14 INFO [evaluator:585] Running loglikelihood requests
Tokenizing inputs: 0%| | 0/3420 [00:00<?, ?it/s] Tokenizing inputs: 2%|▏ | 69/3420 [00:00<00:04, 684.29it/s] Tokenizing inputs: 4%|▍ | 138/3420 [00:00<00:05, 632.03it/s] Tokenizing inputs: 6%|β–Œ | 202/3420 [00:00<00:06, 534.67it/s] Tokenizing inputs: 8%|β–Š | 257/3420 [00:00<00:05, 529.92it/s] Tokenizing inputs: 9%|β–‰ | 311/3420 [00:00<00:06, 498.28it/s] Tokenizing inputs: 11%|β–ˆ | 362/3420 [00:00<00:07, 422.05it/s] Tokenizing inputs: 12%|β–ˆβ– | 407/3420 [00:00<00:07, 428.79it/s] Tokenizing inputs: 13%|β–ˆβ–Ž | 456/3420 [00:00<00:06, 443.27it/s] Tokenizing inputs: 15%|β–ˆβ–Œ | 513/3420 [00:01<00:06, 477.96it/s] Tokenizing inputs: 17%|β–ˆβ–‹ | 585/3420 [00:01<00:05, 546.83it/s] Tokenizing inputs: 19%|β–ˆβ–‰ | 648/3420 [00:01<00:04, 570.79it/s] Tokenizing inputs: 21%|β–ˆβ–ˆ | 707/3420 [00:01<00:04, 547.74it/s] Tokenizing inputs: 22%|β–ˆβ–ˆβ– | 763/3420 [00:01<00:04, 535.27it/s] Tokenizing inputs: 24%|β–ˆβ–ˆβ– | 818/3420 [00:01<00:04, 532.92it/s] Tokenizing inputs: 25%|β–ˆβ–ˆβ–Œ | 872/3420 [00:01<00:05, 467.14it/s] Tokenizing inputs: 27%|β–ˆβ–ˆβ–‹ | 921/3420 [00:01<00:05, 427.96it/s] Tokenizing inputs: 28%|β–ˆβ–ˆβ–Š | 971/3420 [00:01<00:05, 446.07it/s] Tokenizing inputs: 30%|β–ˆβ–ˆβ–‰ | 1017/3420 [00:02<00:05, 443.01it/s] Tokenizing inputs: 31%|β–ˆβ–ˆβ–ˆ | 1063/3420 [00:02<00:05, 405.05it/s] Tokenizing inputs: 32%|β–ˆβ–ˆβ–ˆβ– | 1105/3420 [00:02<00:05, 395.71it/s] Tokenizing inputs: 34%|β–ˆβ–ˆβ–ˆβ–Ž | 1146/3420 [00:02<00:05, 397.23it/s] Tokenizing inputs: 35%|β–ˆβ–ˆβ–ˆβ– | 1192/3420 [00:02<00:05, 413.12it/s] Tokenizing inputs: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1253/3420 [00:02<00:04, 466.27it/s] Tokenizing inputs: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1301/3420 [00:02<00:04, 463.04it/s] Tokenizing inputs: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 1358/3420 [00:02<00:04, 492.68it/s] Tokenizing inputs: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1428/3420 [00:02<00:03, 551.93it/s] Tokenizing inputs: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1507/3420 [00:03<00:03, 618.37it/s] Tokenizing inputs: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1570/3420 [00:03<00:03, 611.34it/s] Tokenizing inputs: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1647/3420 [00:03<00:02, 656.28it/s] Tokenizing inputs: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1713/3420 [00:03<00:02, 634.28it/s] Tokenizing inputs: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1777/3420 [00:03<00:03, 540.53it/s] Tokenizing inputs: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1834/3420 [00:03<00:03, 427.02it/s] Tokenizing inputs: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1882/3420 [00:03<00:03, 409.12it/s] Tokenizing inputs: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1942/3420 [00:03<00:03, 451.16it/s] Tokenizing inputs: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1991/3420 [00:04<00:03, 456.76it/s] Tokenizing inputs: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2050/3420 [00:04<00:02, 489.96it/s] Tokenizing inputs: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2102/3420 [00:04<00:02, 493.99it/s] Tokenizing inputs: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2163/3420 [00:04<00:02, 524.43it/s] Tokenizing inputs: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2221/3420 [00:04<00:02, 539.94it/s] Tokenizing inputs: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2277/3420 [00:04<00:02, 523.06it/s] Tokenizing inputs: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2344/3420 [00:04<00:01, 561.93it/s] Tokenizing inputs: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2401/3420 [00:04<00:02, 509.39it/s] Tokenizing inputs: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2457/3420 [00:04<00:01, 522.74it/s] Tokenizing inputs: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2511/3420 [00:05<00:02, 376.15it/s] Tokenizing inputs: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2555/3420 [00:05<00:02, 383.17it/s] Tokenizing inputs: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2608/3420 [00:05<00:01, 416.90it/s] Tokenizing inputs: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2659/3420 [00:05<00:01, 437.89it/s] Tokenizing inputs: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2707/3420 [00:05<00:01, 387.65it/s] Tokenizing inputs: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2749/3420 [00:06<00:03, 216.00it/s] Tokenizing inputs: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2782/3420 [00:06<00:03, 177.64it/s] Tokenizing inputs: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2808/3420 [00:06<00:03, 161.39it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2830/3420 [00:06<00:03, 156.07it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2850/3420 [00:06<00:03, 159.01it/s] Tokenizing inputs: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2869/3420 [00:06<00:03, 160.63it/s] Tokenizing inputs: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2893/3420 [00:07<00:02, 176.52it/s] Tokenizing inputs: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2928/3420 [00:07<00:02, 215.65it/s] Tokenizing inputs: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2976/3420 [00:07<00:01, 280.20it/s] Tokenizing inputs: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3028/3420 [00:07<00:01, 341.13it/s] Tokenizing inputs: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3077/3420 [00:07<00:00, 380.71it/s] Tokenizing inputs: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3123/3420 [00:07<00:00, 402.15it/s] Tokenizing inputs: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3166/3420 [00:07<00:00, 391.12it/s] Tokenizing inputs: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3223/3420 [00:07<00:00, 440.30it/s] Tokenizing inputs: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3274/3420 [00:07<00:00, 456.95it/s] Tokenizing inputs: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3321/3420 [00:08<00:00, 347.58it/s] Tokenizing inputs: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3361/3420 [00:08<00:00, 248.95it/s] Tokenizing inputs: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 3420/3420 [00:08<00:00, 402.54it/s]
Running loglikelihood requests: 0%| | 0/3420 [00:00<?, ?it/s][W724 07:15:27.387443972 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3699376128 bytes (free: 967901184, total: 25323503616).
Running loglikelihood requests: 0%| | 1/3420 [00:04<4:12:38, 4.43s/it][W724 07:15:29.640460842 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3581935616 bytes (free: 1183907840, total: 25323503616).
[W724 07:15:29.640639731 CUDACachingAllocator.cpp:3933] memory allocation failed with OOM on device 0 while trying to allocate 3581935616 bytes (free: 1183907840, total: 25323503616).
Traceback (most recent call last):
File "<frozen runpy>", line 198, in _run_module_as_main
File "<frozen runpy>", line 88, in _run_code
File "/usr/local/lib/python3.11/dist-packages/lm_eval/__main__.py", line 14, in <module>
cli_evaluate()
File "/usr/local/lib/python3.11/dist-packages/lm_eval/__main__.py", line 10, in cli_evaluate
parser.execute(args)
File "/usr/local/lib/python3.11/dist-packages/lm_eval/_cli/harness.py", line 60, in execute
args.func(args)
File "/usr/local/lib/python3.11/dist-packages/lm_eval/_cli/run.py", line 391, in _execute
results = simple_evaluate(
^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/utils.py", line 575, in _wrapper
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/evaluator.py", line 358, in simple_evaluate
results = evaluate(
^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/utils.py", line 575, in _wrapper
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/evaluator.py", line 596, in evaluate
resps = getattr(lm, reqtype)(cloned_reqs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/models/huggingface.py", line 1227, in loglikelihood
return super().loglikelihood(requests, disable_tqdm=disable_tqdm)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/api/model.py", line 446, in loglikelihood
return self._loglikelihood_tokens(new_reqs, disable_tqdm=disable_tqdm)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/lm_eval/models/huggingface.py", line 1506, in _loglikelihood_tokens
multi_logits = F.log_softmax(
^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/dist-packages/torch/nn/functional.py", line 2315, in log_softmax
ret = input.log_softmax(dim)
^^^^^^^^^^^^^^^^^^^^^^
torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3.34 GiB. GPU 0 has a total capacity of 23.58 GiB of which 1.10 GiB is free. Process 1401096 has 22.47 GiB memory in use. Of the allocated memory 21.12 GiB is allocated by PyTorch, and 1.05 GiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://docs.pytorch.org/docs/stable/notes/cuda.html#optimizing-memory-usage-with-pytorch-cuda-alloc-conf)
Running loglikelihood requests: 0%| | 16/3420 [00:06<23:42, 2.39it/s]
########## m1 / gsm8k_cot ##########
/usr/bin/python -m lm_eval --model hf --model_args pretrained=Qwen/Qwen2.5-7B-Instruct,dtype=bfloat16,peft=/root/steering-resistance/results/full_7b/m1_resist_adapter --tasks gsm8k_cot --num_fewshot 8 --batch_size 4 --seed 0 --output_path /root/steering-resistance/results/full_7b/capability/m1/gsm8k_cot --gen_kwargs do_sample=False --apply_chat_template --fewshot_as_multiturn --limit 200
/usr/local/lib/python3.11/dist-packages/requests/__init__.py:113: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (6.0.0.post1)/charset_normalizer (3.3.2) doesn't match a supported version!
warnings.warn(
2026-07-24:07:15:31 WARNING [config.evaluate_config:287] --limit SHOULD ONLY BE USED FOR TESTING. REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.
2026-07-24:07:15:40 INFO [_cli.run:388] Selected Tasks: ['gsm8k_cot']
2026-07-24:07:15:41 INFO [evaluator:214] Setting random seed to 0 | Setting numpy seed to 0 | Setting torch manual seed to 0 | Setting fewshot manual seed to 0
2026-07-24:07:15:41 WARNING [evaluator:226] generation_kwargs: {'do_sample': False} specified through cli, these settings will update set parameters in yaml tasks. Ensure 'do_sample=True' for non-greedy decoding!
2026-07-24:07:15:41 INFO [evaluator:239] Initializing hf model, with arguments: {'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16', 'peft': '/root/steering-resistance/results/full_7b/m1_resist_adapter'}
2026-07-24:07:15:44 INFO [models.huggingface:286] Using device 'cuda:0'
2026-07-24:07:15:46 INFO [models.huggingface:579] Model parallel was set to False, max memory was not set, and device map was set to {'': 'cuda:0'}
Loading weights: 0%| | 0/339 [00:00<?, ?it/s] Loading weights: 1%| | 2/339 [00:00<00:35, 9.39it/s] Loading weights: 6%|β–‹ | 22/339 [00:00<00:03, 85.59it/s] Loading weights: 14%|β–ˆβ–Ž | 46/339 [00:00<00:02, 140.14it/s] Loading weights: 20%|β–ˆβ–‰ | 67/339 [00:00<00:01, 163.61it/s] Loading weights: 26%|β–ˆβ–ˆβ–‹ | 89/339 [00:00<00:01, 179.88it/s] Loading weights: 34%|β–ˆβ–ˆβ–ˆβ–Ž | 114/339 [00:00<00:01, 198.53it/s] Loading weights: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 138/339 [00:00<00:00, 208.84it/s] Loading weights: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 162/339 [00:00<00:00, 216.03it/s] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 185/339 [00:01<00:00, 218.61it/s] Loading weights: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 210/339 [00:01<00:00, 211.47it/s] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 234/339 [00:01<00:00, 217.43it/s] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 258/339 [00:01<00:00, 212.48it/s] Loading weights: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 282/339 [00:01<00:00, 216.97it/s] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 305/339 [00:01<00:00, 218.79it/s] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 330/339 [00:01<00:00, 224.03it/s] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 339/339 [00:01<00:00, 198.62it/s]
2026-07-24:07:15:49 INFO [models.huggingface:885] Model config indicates vocab_size='152064', but found tokenizer with vocab size '151665'. Resizing model embedding layer...
2026-07-24:07:15:53 INFO [evaluator_utils:446] Selected tasks:
2026-07-24:07:15:53 INFO [evaluator_utils:480] Task: gsm8k_cot (gsm8k/gsm8k-cot.yaml)
2026-07-24:07:15:53 INFO [evaluator:314] gsm8k_cot: Using gen_kwargs: {'do_sample': False, 'until': ['Q:', '</s>', '<|im_end|>']}
2026-07-24:07:15:53 WARNING [evaluator:333] Overwriting default num_fewshot of gsm8k_cot from 8 to 8
2026-07-24:07:15:53 INFO [api.task:312] Building contexts for gsm8k_cot on rank 0...
0%| | 0/200 [00:00<?, ?it/s] 4%|▍ | 9/200 [00:00<00:02, 84.88it/s] 10%|β–‰ | 19/200 [00:00<00:01, 91.24it/s] 14%|β–ˆβ– | 29/200 [00:00<00:01, 93.17it/s] 20%|β–ˆβ–‰ | 39/200 [00:00<00:01, 94.01it/s] 24%|β–ˆβ–ˆβ– | 49/200 [00:00<00:01, 94.88it/s] 30%|β–ˆβ–ˆβ–‰ | 59/200 [00:00<00:01, 95.39it/s] 34%|β–ˆβ–ˆβ–ˆβ– | 69/200 [00:00<00:01, 95.94it/s] 40%|β–ˆβ–ˆβ–ˆβ–‰ | 79/200 [00:00<00:01, 96.17it/s] 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 89/200 [00:00<00:01, 96.53it/s] 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 99/200 [00:01<00:01, 96.84it/s] 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 109/200 [00:01<00:00, 96.98it/s] 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 119/200 [00:01<00:00, 97.16it/s] 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 129/200 [00:01<00:00, 97.38it/s] 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 139/200 [00:01<00:00, 97.47it/s] 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 149/200 [00:01<00:00, 97.66it/s] 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 159/200 [00:01<00:00, 97.56it/s] 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 169/200 [00:01<00:00, 97.41it/s] 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 179/200 [00:01<00:00, 97.57it/s] 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 189/200 [00:01<00:00, 97.72it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 199/200 [00:02<00:00, 97.54it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 200/200 [00:02<00:00, 96.46it/s]
2026-07-24:07:15:55 INFO [evaluator:585] Running generate_until requests
Running generate_until requests: 0%| | 0/200 [00:00<?, ?it/s] Running generate_until requests: 0%| | 1/200 [00:14<47:36, 14.35s/it] Running generate_until requests: 2%|β–Ž | 5/200 [00:23<12:56, 3.98s/it] Running generate_until requests: 4%|▍ | 9/200 [00:35<11:05, 3.49s/it] Running generate_until requests: 6%|β–‹ | 13/200 [00:46<10:00, 3.21s/it] Running generate_until requests: 8%|β–Š | 17/200 [00:54<08:22, 2.75s/it] Running generate_until requests: 10%|β–ˆ | 21/200 [01:08<08:54, 2.99s/it] Running generate_until requests: 12%|β–ˆβ–Ž | 25/200 [01:20<08:37, 2.96s/it] Running generate_until requests: 14%|β–ˆβ– | 29/200 [01:28<07:43, 2.71s/it] Running generate_until requests: 16%|β–ˆβ–‹ | 33/200 [01:38<07:22, 2.65s/it] Running generate_until requests: 18%|β–ˆβ–Š | 37/200 [01:45<06:25, 2.37s/it] Running generate_until requests: 20%|β–ˆβ–ˆ | 41/200 [01:53<05:47, 2.19s/it] Running generate_until requests: 22%|β–ˆβ–ˆβ–Ž | 45/200 [02:02<05:41, 2.21s/it] Running generate_until requests: 24%|β–ˆβ–ˆβ– | 49/200 [02:14<06:09, 2.45s/it] Running generate_until requests: 26%|β–ˆβ–ˆβ–‹ | 53/200 [02:27<06:38, 2.71s/it] Running generate_until requests: 28%|β–ˆβ–ˆβ–Š | 57/200 [02:37<06:22, 2.67s/it] Running generate_until requests: 30%|β–ˆβ–ˆβ–ˆ | 61/200 [02:47<06:03, 2.62s/it] Running generate_until requests: 32%|β–ˆβ–ˆβ–ˆβ–Ž | 65/200 [02:55<05:23, 2.40s/it] Running generate_until requests: 34%|β–ˆβ–ˆβ–ˆβ– | 69/200 [03:04<05:14, 2.40s/it] Running generate_until requests: 36%|β–ˆβ–ˆβ–ˆβ–‹ | 73/200 [03:12<04:44, 2.24s/it] Running generate_until requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 77/200 [03:22<04:48, 2.34s/it] Running generate_until requests: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 81/200 [03:30<04:22, 2.21s/it] Running generate_until requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 85/200 [03:39<04:21, 2.27s/it] Running generate_until requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 89/200 [03:46<03:53, 2.11s/it] Running generate_until requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 93/200 [03:55<03:45, 2.11s/it] Running generate_until requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 97/200 [04:06<03:56, 2.29s/it] Running generate_until requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 101/200 [04:13<03:31, 2.14s/it] Running generate_until requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 105/200 [04:21<03:20, 2.11s/it] Running generate_until requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 109/200 [04:28<03:06, 2.05s/it] Running generate_until requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 113/200 [04:40<03:21, 2.32s/it] Running generate_until requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 117/200 [04:50<03:13, 2.33s/it] Running generate_until requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 121/200 [04:59<03:03, 2.33s/it] Running generate_until requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 125/200 [05:05<02:38, 2.11s/it] Running generate_until requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 129/200 [05:12<02:21, 2.00s/it] Running generate_until requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 133/200 [05:18<02:04, 1.85s/it] Running generate_until requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 137/200 [05:24<01:47, 1.71s/it] Running generate_until requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 141/200 [05:32<01:45, 1.79s/it] Running generate_until requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 145/200 [05:38<01:36, 1.75s/it] Running generate_until requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 149/200 [05:52<01:54, 2.25s/it] Running generate_until requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 153/200 [05:59<01:39, 2.11s/it] Running generate_until requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 157/200 [06:05<01:22, 1.92s/it] Running generate_until requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 161/200 [06:15<01:21, 2.10s/it] Running generate_until requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 165/200 [06:26<01:19, 2.28s/it] Running generate_until requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 169/200 [06:32<01:03, 2.04s/it] Running generate_until requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 173/200 [06:42<00:58, 2.16s/it] Running generate_until requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 177/200 [06:48<00:45, 1.97s/it] Running generate_until requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 181/200 [06:54<00:35, 1.85s/it] Running generate_until requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 185/200 [06:59<00:25, 1.69s/it] Running generate_until requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 189/200 [07:11<00:22, 2.07s/it] Running generate_until requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 193/200 [07:25<00:17, 2.47s/it] Running generate_until requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 197/200 [07:35<00:07, 2.53s/it] Running generate_until requests: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 200/200 [07:35<00:00, 2.28s/it]
fatal: detected dubious ownership in repository at '/root/steering-resistance'
To add an exception for this directory, call:
git config --global --add safe.directory /root/steering-resistance
2026-07-24:07:23:32 INFO [loggers.evaluation_tracker:247] Saving results aggregated
hf ({'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16', 'peft': '/root/steering-resistance/results/full_7b/m1_resist_adapter'}), gen_kwargs: ({'do_sample': False}), limit: 200.0, num_fewshot: 8, batch_size: 4
| Tasks |Version| Filter |n-shot| Metric | |Value| |Stderr|
|---------|------:|----------------|-----:|-----------|---|----:|---|-----:|
|gsm8k_cot| 3|flexible-extract| 8|exact_match|↑ |0.795|Β± |0.0286|
| | |strict-match | 8|exact_match|↑ |0.365|Β± |0.0341|
########## m0 / mmlu ##########
/usr/bin/python -m lm_eval --model hf --model_args pretrained=Qwen/Qwen2.5-7B-Instruct,dtype=bfloat16 --tasks mmlu --num_fewshot 5 --batch_size 2 --seed 0 --output_path /root/steering-resistance/results/full_7b/capability/m0/mmlu --apply_chat_template --fewshot_as_multiturn --limit 15
/usr/local/lib/python3.11/dist-packages/requests/__init__.py:113: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (6.0.0.post1)/charset_normalizer (3.3.2) doesn't match a supported version!
warnings.warn(
2026-07-24:07:25:10 WARNING [config.evaluate_config:287] --limit SHOULD ONLY BE USED FOR TESTING. REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.
2026-07-24:07:25:19 INFO [_cli.run:388] Selected Tasks: ['mmlu']
2026-07-24:07:25:20 INFO [evaluator:214] Setting random seed to 0 | Setting numpy seed to 0 | Setting torch manual seed to 0 | Setting fewshot manual seed to 0
2026-07-24:07:25:20 INFO [evaluator:239] Initializing hf model, with arguments: {'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16'}
2026-07-24:07:25:24 INFO [models.huggingface:286] Using device 'cuda:0'
2026-07-24:07:25:25 INFO [models.huggingface:579] Model parallel was set to False, max memory was not set, and device map was set to {'': 'cuda:0'}
Loading weights: 0%| | 0/339 [00:00<?, ?it/s] Loading weights: 1%| | 2/339 [00:00<00:34, 9.85it/s] Loading weights: 5%|β–Œ | 18/339 [00:00<00:04, 66.34it/s] Loading weights: 12%|β–ˆβ– | 40/339 [00:00<00:02, 110.69it/s] Loading weights: 17%|β–ˆβ–‹ | 59/339 [00:00<00:02, 134.54it/s] Loading weights: 23%|β–ˆβ–ˆβ–Ž | 77/339 [00:00<00:01, 142.25it/s] Loading weights: 28%|β–ˆβ–ˆβ–Š | 95/339 [00:00<00:01, 152.21it/s] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 112/339 [00:00<00:01, 157.23it/s] Loading weights: 38%|β–ˆβ–ˆβ–ˆβ–Š | 129/339 [00:00<00:01, 154.27it/s] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 148/339 [00:01<00:01, 161.95it/s] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 165/339 [00:01<00:01, 154.23it/s] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 185/339 [00:01<00:00, 158.32it/s] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 202/339 [00:01<00:00, 159.19it/s] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 219/339 [00:01<00:00, 148.40it/s] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 235/339 [00:01<00:00, 139.28it/s] Loading weights: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 256/339 [00:01<00:00, 149.89it/s] Loading weights: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 272/339 [00:01<00:00, 132.04it/s] Loading weights: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 292/339 [00:02<00:00, 147.52it/s] Loading weights: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 308/339 [00:02<00:00, 145.22it/s] Loading weights: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 328/339 [00:02<00:00, 158.58it/s] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 339/339 [00:02<00:00, 144.01it/s]
2026-07-24:07:26:27 INFO [evaluator_utils:446] Selected tasks:
2026-07-24:07:26:27 INFO [evaluator_utils:462] Group: mmlu
2026-07-24:07:26:27 INFO [evaluator_utils:462] Group: mmlu_stem
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_abstract_algebra (mmlu/default/mmlu_abstract_algebra.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_anatomy (mmlu/default/mmlu_anatomy.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_astronomy (mmlu/default/mmlu_astronomy.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_college_biology (mmlu/default/mmlu_college_biology.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_college_chemistry (mmlu/default/mmlu_college_chemistry.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_college_computer_science (mmlu/default/mmlu_college_computer_science.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_college_mathematics (mmlu/default/mmlu_college_mathematics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_college_physics (mmlu/default/mmlu_college_physics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_computer_security (mmlu/default/mmlu_computer_security.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_conceptual_physics (mmlu/default/mmlu_conceptual_physics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_electrical_engineering (mmlu/default/mmlu_electrical_engineering.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_elementary_mathematics (mmlu/default/mmlu_elementary_mathematics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_biology (mmlu/default/mmlu_high_school_biology.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_chemistry (mmlu/default/mmlu_high_school_chemistry.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_computer_science (mmlu/default/mmlu_high_school_computer_science.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_mathematics (mmlu/default/mmlu_high_school_mathematics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_physics (mmlu/default/mmlu_high_school_physics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_statistics (mmlu/default/mmlu_high_school_statistics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_machine_learning (mmlu/default/mmlu_machine_learning.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:462] Group: mmlu_other
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_business_ethics (mmlu/default/mmlu_business_ethics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_clinical_knowledge (mmlu/default/mmlu_clinical_knowledge.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_college_medicine (mmlu/default/mmlu_college_medicine.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_global_facts (mmlu/default/mmlu_global_facts.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_human_aging (mmlu/default/mmlu_human_aging.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_management (mmlu/default/mmlu_management.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_marketing (mmlu/default/mmlu_marketing.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_medical_genetics (mmlu/default/mmlu_medical_genetics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_miscellaneous (mmlu/default/mmlu_miscellaneous.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_nutrition (mmlu/default/mmlu_nutrition.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_professional_accounting (mmlu/default/mmlu_professional_accounting.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_professional_medicine (mmlu/default/mmlu_professional_medicine.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_virology (mmlu/default/mmlu_virology.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:462] Group: mmlu_social_sciences
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_econometrics (mmlu/default/mmlu_econometrics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_geography (mmlu/default/mmlu_high_school_geography.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_government_and_politics (mmlu/default/mmlu_high_school_government_and_politics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_macroeconomics (mmlu/default/mmlu_high_school_macroeconomics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_microeconomics (mmlu/default/mmlu_high_school_microeconomics.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_psychology (mmlu/default/mmlu_high_school_psychology.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_human_sexuality (mmlu/default/mmlu_human_sexuality.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_professional_psychology (mmlu/default/mmlu_professional_psychology.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_public_relations (mmlu/default/mmlu_public_relations.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_security_studies (mmlu/default/mmlu_security_studies.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_sociology (mmlu/default/mmlu_sociology.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_us_foreign_policy (mmlu/default/mmlu_us_foreign_policy.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:462] Group: mmlu_humanities
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_formal_logic (mmlu/default/mmlu_formal_logic.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_european_history (mmlu/default/mmlu_high_school_european_history.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_us_history (mmlu/default/mmlu_high_school_us_history.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_high_school_world_history (mmlu/default/mmlu_high_school_world_history.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_international_law (mmlu/default/mmlu_international_law.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_jurisprudence (mmlu/default/mmlu_jurisprudence.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_logical_fallacies (mmlu/default/mmlu_logical_fallacies.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_moral_disputes (mmlu/default/mmlu_moral_disputes.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_moral_scenarios (mmlu/default/mmlu_moral_scenarios.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_philosophy (mmlu/default/mmlu_philosophy.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_prehistory (mmlu/default/mmlu_prehistory.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_professional_law (mmlu/default/mmlu_professional_law.yaml)
2026-07-24:07:26:27 INFO [evaluator_utils:470] Task: mmlu_world_religions (mmlu/default/mmlu_world_religions.yaml)
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_abstract_algebra from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_anatomy from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_astronomy from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_biology from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_chemistry from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_computer_science from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_mathematics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_physics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_computer_security from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_conceptual_physics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_electrical_engineering from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_elementary_mathematics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_biology from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_chemistry from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_computer_science from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_mathematics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_physics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_statistics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_machine_learning from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_business_ethics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_clinical_knowledge from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_medicine from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_global_facts from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_aging from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_management from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_marketing from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_medical_genetics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_miscellaneous from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_nutrition from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_accounting from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_medicine from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_virology from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_econometrics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_geography from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_government_and_politics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_macroeconomics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_microeconomics from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_psychology from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_sexuality from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_psychology from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_public_relations from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_security_studies from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_sociology from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_us_foreign_policy from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_formal_logic from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_european_history from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_us_history from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_world_history from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_international_law from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_jurisprudence from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_logical_fallacies from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_disputes from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_scenarios from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_philosophy from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_prehistory from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_law from None to 5
2026-07-24:07:26:27 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_world_religions from None to 5
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_abstract_algebra on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 12/15 [00:00<00:00, 114.09it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 123.79it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_anatomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.79it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_astronomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.66it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_college_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 188.98it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_college_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.48it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_college_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.93it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_college_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.78it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_college_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.91it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_computer_security on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.47it/s]
2026-07-24:07:26:27 INFO [api.task:312] Building contexts for mmlu_conceptual_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 195.12it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_electrical_engineering on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.38it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_elementary_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.69it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_high_school_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.14it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_high_school_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.13it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_high_school_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.47it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_high_school_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.29it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_high_school_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.77it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_high_school_statistics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.34it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_machine_learning on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.14it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_business_ethics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.26it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_clinical_knowledge on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.63it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_college_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.11it/s]
2026-07-24:07:26:28 INFO [api.task:312] Building contexts for mmlu_global_facts on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.18it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_human_aging on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.78it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_management on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.93it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_marketing on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.12it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_medical_genetics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.71it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_miscellaneous on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.85it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_nutrition on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.69it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_professional_accounting on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.49it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_professional_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.56it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_virology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.73it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_econometrics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.14it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_high_school_geography on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.70it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_high_school_government_and_politics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.94it/s]
2026-07-24:07:26:29 INFO [api.task:312] Building contexts for mmlu_high_school_macroeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.16it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_high_school_microeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.21it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_high_school_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.49it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_human_sexuality on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.00it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_professional_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.64it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_public_relations on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.83it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_security_studies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.73it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_sociology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.14it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_us_foreign_policy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.54it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_formal_logic on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.06it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_high_school_european_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.15it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_high_school_us_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.48it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_high_school_world_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.51it/s]
2026-07-24:07:26:30 INFO [api.task:312] Building contexts for mmlu_international_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.45it/s]
2026-07-24:07:26:31 INFO [api.task:312] Building contexts for mmlu_jurisprudence on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 195.11it/s]
2026-07-24:07:26:31 INFO [api.task:312] Building contexts for mmlu_logical_fallacies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.20it/s]
2026-07-24:07:26:31 INFO [api.task:312] Building contexts for mmlu_moral_disputes on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.53it/s]
2026-07-24:07:26:31 INFO [api.task:312] Building contexts for mmlu_moral_scenarios on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.60it/s]
2026-07-24:07:26:31 INFO [api.task:312] Building contexts for mmlu_philosophy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 195.57it/s]
2026-07-24:07:26:31 INFO [api.task:312] Building contexts for mmlu_prehistory on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.26it/s]
2026-07-24:07:26:31 INFO [api.task:312] Building contexts for mmlu_professional_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.99it/s]
2026-07-24:07:26:31 INFO [api.task:312] Building contexts for mmlu_world_religions on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 195.54it/s]
2026-07-24:07:26:31 INFO [evaluator:585] Running loglikelihood requests
Tokenizing inputs: 0%| | 0/3420 [00:00<?, ?it/s] Tokenizing inputs: 2%|▏ | 70/3420 [00:00<00:04, 694.65it/s] Tokenizing inputs: 4%|▍ | 140/3420 [00:00<00:05, 642.99it/s] Tokenizing inputs: 6%|β–Œ | 205/3420 [00:00<00:05, 551.31it/s] Tokenizing inputs: 8%|β–Š | 262/3420 [00:00<00:05, 542.97it/s] Tokenizing inputs: 9%|β–‰ | 317/3420 [00:00<00:06, 496.39it/s] Tokenizing inputs: 11%|β–ˆ | 368/3420 [00:00<00:06, 436.27it/s] Tokenizing inputs: 12%|β–ˆβ– | 415/3420 [00:00<00:06, 443.76it/s] Tokenizing inputs: 14%|β–ˆβ–Ž | 468/3420 [00:00<00:06, 465.37it/s] Tokenizing inputs: 15%|β–ˆβ–Œ | 527/3420 [00:01<00:05, 500.58it/s] Tokenizing inputs: 18%|β–ˆβ–Š | 603/3420 [00:01<00:04, 574.24it/s] Tokenizing inputs: 19%|β–ˆβ–‰ | 664/3420 [00:01<00:04, 583.82it/s] Tokenizing inputs: 21%|β–ˆβ–ˆ | 724/3420 [00:01<00:04, 553.37it/s] Tokenizing inputs: 23%|β–ˆβ–ˆβ–Ž | 781/3420 [00:01<00:04, 542.90it/s] Tokenizing inputs: 24%|β–ˆβ–ˆβ– | 837/3420 [00:01<00:04, 545.51it/s] Tokenizing inputs: 26%|β–ˆβ–ˆβ–Œ | 893/3420 [00:01<00:05, 439.37it/s] Tokenizing inputs: 28%|β–ˆβ–ˆβ–Š | 941/3420 [00:01<00:05, 444.47it/s] Tokenizing inputs: 29%|β–ˆβ–ˆβ–‰ | 988/3420 [00:01<00:05, 450.61it/s] Tokenizing inputs: 30%|β–ˆβ–ˆβ–ˆ | 1035/3420 [00:02<00:05, 440.28it/s] Tokenizing inputs: 32%|β–ˆβ–ˆβ–ˆβ– | 1081/3420 [00:02<00:05, 405.45it/s] Tokenizing inputs: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1123/3420 [00:02<00:05, 405.27it/s] Tokenizing inputs: 34%|β–ˆβ–ˆβ–ˆβ– | 1167/3420 [00:02<00:05, 413.92it/s] Tokenizing inputs: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 1216/3420 [00:02<00:05, 434.23it/s] Tokenizing inputs: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1276/3420 [00:02<00:04, 479.84it/s] Tokenizing inputs: 39%|β–ˆβ–ˆβ–ˆβ–Š | 1325/3420 [00:02<00:04, 473.19it/s] Tokenizing inputs: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 1388/3420 [00:02<00:03, 517.16it/s] Tokenizing inputs: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1463/3420 [00:02<00:03, 584.40it/s] Tokenizing inputs: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1533/3420 [00:03<00:03, 618.04it/s] Tokenizing inputs: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1600/3420 [00:03<00:02, 632.88it/s] Tokenizing inputs: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1681/3420 [00:03<00:02, 681.64it/s] Tokenizing inputs: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1750/3420 [00:03<00:02, 582.45it/s] Tokenizing inputs: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1811/3420 [00:03<00:03, 491.29it/s] Tokenizing inputs: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1865/3420 [00:03<00:03, 406.56it/s] Tokenizing inputs: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1932/3420 [00:03<00:03, 461.53it/s] Tokenizing inputs: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1984/3420 [00:03<00:03, 460.15it/s] Tokenizing inputs: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2045/3420 [00:04<00:02, 495.44it/s] Tokenizing inputs: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2098/3420 [00:04<00:02, 499.88it/s] Tokenizing inputs: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2159/3420 [00:04<00:02, 529.12it/s] Tokenizing inputs: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2219/3420 [00:04<00:02, 546.92it/s] Tokenizing inputs: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2276/3420 [00:04<00:02, 532.34it/s] Tokenizing inputs: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2344/3420 [00:04<00:01, 569.93it/s] Tokenizing inputs: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2402/3420 [00:04<00:01, 517.64it/s] Tokenizing inputs: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2459/3420 [00:04<00:01, 531.06it/s] Tokenizing inputs: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2514/3420 [00:05<00:02, 377.00it/s] Tokenizing inputs: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2559/3420 [00:05<00:02, 392.73it/s] Tokenizing inputs: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2613/3420 [00:05<00:01, 426.88it/s] Tokenizing inputs: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2663/3420 [00:05<00:01, 444.54it/s] Tokenizing inputs: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2712/3420 [00:05<00:01, 361.10it/s] Tokenizing inputs: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2753/3420 [00:06<00:03, 207.45it/s] Tokenizing inputs: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2785/3420 [00:06<00:03, 175.36it/s] Tokenizing inputs: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2811/3420 [00:06<00:03, 160.57it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2833/3420 [00:06<00:03, 157.18it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2853/3420 [00:06<00:03, 160.59it/s] Tokenizing inputs: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2872/3420 [00:06<00:03, 162.34it/s] Tokenizing inputs: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2899/3420 [00:07<00:02, 184.25it/s] Tokenizing inputs: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2935/3420 [00:07<00:02, 224.41it/s] Tokenizing inputs: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2987/3420 [00:07<00:01, 296.46it/s] Tokenizing inputs: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3038/3420 [00:07<00:01, 350.84it/s] Tokenizing inputs: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3087/3420 [00:07<00:00, 387.45it/s] Tokenizing inputs: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3131/3420 [00:07<00:00, 401.23it/s] Tokenizing inputs: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3174/3420 [00:07<00:00, 391.12it/s] Tokenizing inputs: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3239/3420 [00:07<00:00, 462.64it/s] Tokenizing inputs: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3287/3420 [00:07<00:00, 453.14it/s] Tokenizing inputs: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3334/3420 [00:08<00:00, 308.92it/s] Tokenizing inputs: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3372/3420 [00:08<00:00, 264.22it/s] Tokenizing inputs: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 3420/3420 [00:08<00:00, 407.49it/s]
Running loglikelihood requests: 0%| | 0/3420 [00:00<?, ?it/s] Running loglikelihood requests: 0%| | 1/3420 [00:02<2:05:28, 2.20s/it] Running loglikelihood requests: 0%| | 9/3420 [00:03<18:52, 3.01it/s] Running loglikelihood requests: 0%| | 17/3420 [00:04<13:33, 4.18it/s] Running loglikelihood requests: 1%| | 25/3420 [00:06<11:31, 4.91it/s] Running loglikelihood requests: 1%| | 33/3420 [00:07<10:27, 5.39it/s] Running loglikelihood requests: 1%| | 41/3420 [00:08<09:46, 5.76it/s] Running loglikelihood requests: 1%|▏ | 49/3420 [00:09<09:21, 6.01it/s] Running loglikelihood requests: 2%|▏ | 57/3420 [00:11<09:01, 6.21it/s] Running loglikelihood requests: 2%|▏ | 65/3420 [00:12<08:29, 6.59it/s] Running loglikelihood requests: 2%|▏ | 73/3420 [00:13<08:04, 6.91it/s] Running loglikelihood requests: 2%|▏ | 81/3420 [00:14<07:47, 7.14it/s] Running loglikelihood requests: 3%|β–Ž | 89/3420 [00:15<07:35, 7.32it/s] Running loglikelihood requests: 3%|β–Ž | 97/3420 [00:16<07:22, 7.51it/s] Running loglikelihood requests: 3%|β–Ž | 105/3420 [00:17<07:13, 7.65it/s] Running loglikelihood requests: 3%|β–Ž | 113/3420 [00:18<07:06, 7.76it/s] Running loglikelihood requests: 4%|β–Ž | 121/3420 [00:19<06:44, 8.15it/s] Running loglikelihood requests: 4%|▍ | 129/3420 [00:19<06:20, 8.65it/s] Running loglikelihood requests: 4%|▍ | 137/3420 [00:20<06:03, 9.03it/s] Running loglikelihood requests: 4%|▍ | 145/3420 [00:21<05:48, 9.40it/s] Running loglikelihood requests: 4%|▍ | 153/3420 [00:22<05:32, 9.83it/s] Running loglikelihood requests: 5%|▍ | 161/3420 [00:22<05:21, 10.15it/s] Running loglikelihood requests: 5%|▍ | 169/3420 [00:23<05:13, 10.38it/s] Running loglikelihood requests: 5%|β–Œ | 177/3420 [00:24<05:07, 10.56it/s] Running loglikelihood requests: 5%|β–Œ | 185/3420 [00:25<04:59, 10.79it/s] Running loglikelihood requests: 6%|β–Œ | 193/3420 [00:25<04:52, 11.04it/s] Running loglikelihood requests: 6%|β–Œ | 201/3420 [00:26<04:41, 11.42it/s] Running loglikelihood requests: 6%|β–Œ | 209/3420 [00:27<04:34, 11.70it/s] Running loglikelihood requests: 6%|β–‹ | 217/3420 [00:27<04:26, 12.00it/s] Running loglikelihood requests: 7%|β–‹ | 225/3420 [00:28<04:21, 12.22it/s] Running loglikelihood requests: 7%|β–‹ | 233/3420 [00:28<04:16, 12.41it/s] Running loglikelihood requests: 7%|β–‹ | 241/3420 [00:29<04:11, 12.65it/s] Running loglikelihood requests: 7%|β–‹ | 249/3420 [00:30<04:06, 12.85it/s] Running loglikelihood requests: 8%|β–Š | 257/3420 [00:30<03:59, 13.19it/s] Running loglikelihood requests: 8%|β–Š | 265/3420 [00:31<03:54, 13.45it/s] Running loglikelihood requests: 8%|β–Š | 273/3420 [00:31<03:50, 13.64it/s] Running loglikelihood requests: 8%|β–Š | 281/3420 [00:32<03:47, 13.82it/s] Running loglikelihood requests: 8%|β–Š | 289/3420 [00:32<03:44, 13.94it/s] Running loglikelihood requests: 9%|β–Š | 297/3420 [00:33<03:42, 14.02it/s] Running loglikelihood requests: 9%|β–‰ | 305/3420 [00:34<03:41, 14.07it/s] Running loglikelihood requests: 9%|β–‰ | 313/3420 [00:34<03:40, 14.11it/s] Running loglikelihood requests: 9%|β–‰ | 321/3420 [00:35<03:37, 14.27it/s] Running loglikelihood requests: 10%|β–‰ | 329/3420 [00:35<03:34, 14.40it/s] Running loglikelihood requests: 10%|β–‰ | 337/3420 [00:36<03:32, 14.50it/s] Running loglikelihood requests: 10%|β–ˆ | 345/3420 [00:36<03:31, 14.57it/s] Running loglikelihood requests: 10%|β–ˆ | 353/3420 [00:37<03:22, 15.12it/s] Running loglikelihood requests: 11%|β–ˆ | 361/3420 [00:37<03:16, 15.55it/s] Running loglikelihood requests: 11%|β–ˆ | 369/3420 [00:38<03:13, 15.75it/s] Running loglikelihood requests: 11%|β–ˆ | 377/3420 [00:38<03:08, 16.18it/s] Running loglikelihood requests: 11%|β–ˆβ– | 385/3420 [00:39<03:03, 16.52it/s] Running loglikelihood requests: 11%|β–ˆβ– | 393/3420 [00:39<03:01, 16.66it/s] Running loglikelihood requests: 12%|β–ˆβ– | 401/3420 [00:40<02:57, 16.98it/s] Running loglikelihood requests: 12%|β–ˆβ– | 409/3420 [00:40<02:54, 17.22it/s] Running loglikelihood requests: 12%|β–ˆβ– | 417/3420 [00:41<02:52, 17.39it/s] Running loglikelihood requests: 12%|β–ˆβ– | 425/3420 [00:41<02:51, 17.51it/s] Running loglikelihood requests: 13%|β–ˆβ–Ž | 433/3420 [00:41<02:49, 17.61it/s] Running loglikelihood requests: 13%|β–ˆβ–Ž | 441/3420 [00:42<02:48, 17.69it/s] Running loglikelihood requests: 13%|β–ˆβ–Ž | 449/3420 [00:42<02:43, 18.20it/s] Running loglikelihood requests: 13%|β–ˆβ–Ž | 457/3420 [00:43<02:39, 18.58it/s] Running loglikelihood requests: 14%|β–ˆβ–Ž | 465/3420 [00:43<02:36, 18.87it/s] Running loglikelihood requests: 14%|β–ˆβ– | 473/3420 [00:43<02:33, 19.18it/s] Running loglikelihood requests: 14%|β–ˆβ– | 481/3420 [00:44<02:31, 19.36it/s] Running loglikelihood requests: 14%|β–ˆβ– | 489/3420 [00:44<02:30, 19.48it/s] Running loglikelihood requests: 15%|β–ˆβ– | 497/3420 [00:45<02:28, 19.64it/s] Running loglikelihood requests: 15%|β–ˆβ– | 505/3420 [00:45<02:27, 19.76it/s] Running loglikelihood requests: 15%|β–ˆβ–Œ | 513/3420 [00:45<02:26, 19.85it/s] Running loglikelihood requests: 15%|β–ˆβ–Œ | 521/3420 [00:46<02:25, 19.91it/s] Running loglikelihood requests: 15%|β–ˆβ–Œ | 529/3420 [00:46<02:22, 20.30it/s] Running loglikelihood requests: 16%|β–ˆβ–Œ | 537/3420 [00:47<02:20, 20.59it/s] Running loglikelihood requests: 16%|β–ˆβ–Œ | 545/3420 [00:47<02:18, 20.80it/s] Running loglikelihood requests: 16%|β–ˆβ–Œ | 553/3420 [00:47<02:17, 20.90it/s] Running loglikelihood requests: 16%|β–ˆβ–‹ | 561/3420 [00:48<02:16, 20.96it/s] Running loglikelihood requests: 17%|β–ˆβ–‹ | 569/3420 [00:48<02:15, 21.04it/s] Running loglikelihood requests: 17%|β–ˆβ–‹ | 577/3420 [00:49<02:13, 21.37it/s] Running loglikelihood requests: 17%|β–ˆβ–‹ | 585/3420 [00:49<02:11, 21.60it/s] Running loglikelihood requests: 17%|β–ˆβ–‹ | 593/3420 [00:49<02:09, 21.78it/s] Running loglikelihood requests: 18%|β–ˆβ–Š | 601/3420 [00:50<02:08, 21.92it/s] Running loglikelihood requests: 18%|β–ˆβ–Š | 609/3420 [00:50<02:07, 22.02it/s] Running loglikelihood requests: 18%|β–ˆβ–Š | 617/3420 [00:50<02:05, 22.27it/s] Running loglikelihood requests: 18%|β–ˆβ–Š | 625/3420 [00:51<02:04, 22.40it/s] Running loglikelihood requests: 19%|β–ˆβ–Š | 633/3420 [00:51<02:03, 22.50it/s] Running loglikelihood requests: 19%|β–ˆβ–Š | 641/3420 [00:51<02:03, 22.58it/s] Running loglikelihood requests: 19%|β–ˆβ–‰ | 649/3420 [00:52<02:02, 22.63it/s] Running loglikelihood requests: 19%|β–ˆβ–‰ | 657/3420 [00:52<02:01, 22.68it/s] Running loglikelihood requests: 19%|β–ˆβ–‰ | 665/3420 [00:52<02:01, 22.71it/s] Running loglikelihood requests: 20%|β–ˆβ–‰ | 673/3420 [00:53<02:00, 22.72it/s] Running loglikelihood requests: 20%|β–ˆβ–‰ | 681/3420 [00:53<02:00, 22.75it/s] Running loglikelihood requests: 20%|β–ˆβ–ˆ | 689/3420 [00:53<01:59, 22.76it/s] Running loglikelihood requests: 20%|β–ˆβ–ˆ | 697/3420 [00:54<01:59, 22.77it/s] Running loglikelihood requests: 21%|β–ˆβ–ˆ | 705/3420 [00:54<01:59, 22.77it/s] Running loglikelihood requests: 21%|β–ˆβ–ˆ | 713/3420 [00:55<01:58, 22.79it/s] Running loglikelihood requests: 21%|β–ˆβ–ˆ | 721/3420 [00:55<01:54, 23.49it/s] Running loglikelihood requests: 21%|β–ˆβ–ˆβ– | 729/3420 [00:55<01:52, 23.99it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 737/3420 [00:55<01:50, 24.36it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 745/3420 [00:56<01:48, 24.64it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 753/3420 [00:56<01:47, 24.83it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 761/3420 [00:56<01:46, 24.98it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 769/3420 [00:57<01:45, 25.07it/s] Running loglikelihood requests: 23%|β–ˆβ–ˆβ–Ž | 777/3420 [00:57<01:45, 25.13it/s] Running loglikelihood requests: 23%|β–ˆβ–ˆβ–Ž | 785/3420 [00:57<01:44, 25.19it/s] Running loglikelihood requests: 23%|β–ˆβ–ˆβ–Ž | 793/3420 [00:58<01:44, 25.23it/s] Running loglikelihood requests: 23%|β–ˆβ–ˆβ–Ž | 801/3420 [00:58<01:43, 25.25it/s] Running loglikelihood requests: 24%|β–ˆβ–ˆβ–Ž | 809/3420 [00:58<01:43, 25.28it/s] Running loglikelihood requests: 24%|β–ˆβ–ˆβ– | 817/3420 [00:59<01:42, 25.29it/s] Running loglikelihood requests: 24%|β–ˆβ–ˆβ– | 825/3420 [00:59<01:42, 25.32it/s] Running loglikelihood requests: 24%|β–ˆβ–ˆβ– | 833/3420 [00:59<01:42, 25.33it/s] Running loglikelihood requests: 25%|β–ˆβ–ˆβ– | 841/3420 [01:00<01:41, 25.35it/s] Running loglikelihood requests: 25%|β–ˆβ–ˆβ– | 849/3420 [01:00<01:41, 25.36it/s] Running loglikelihood requests: 25%|β–ˆβ–ˆβ–Œ | 857/3420 [01:00<01:41, 25.37it/s] Running loglikelihood requests: 25%|β–ˆβ–ˆβ–Œ | 865/3420 [01:01<01:40, 25.39it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–Œ | 873/3420 [01:01<01:40, 25.40it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–Œ | 881/3420 [01:01<01:39, 25.41it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–Œ | 889/3420 [01:01<01:39, 25.42it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–Œ | 897/3420 [01:02<01:39, 25.43it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–‹ | 905/3420 [01:02<01:38, 25.44it/s] Running loglikelihood requests: 27%|β–ˆβ–ˆβ–‹ | 913/3420 [01:02<01:38, 25.44it/s] Running loglikelihood requests: 27%|β–ˆβ–ˆβ–‹ | 921/3420 [01:03<01:38, 25.46it/s] Running loglikelihood requests: 27%|β–ˆβ–ˆβ–‹ | 929/3420 [01:03<01:37, 25.46it/s] Running loglikelihood requests: 27%|β–ˆβ–ˆβ–‹ | 937/3420 [01:03<01:37, 25.46it/s] Running loglikelihood requests: 28%|β–ˆβ–ˆβ–Š | 945/3420 [01:04<01:37, 25.49it/s] Running loglikelihood requests: 28%|β–ˆβ–ˆβ–Š | 953/3420 [01:04<01:36, 25.48it/s] Running loglikelihood requests: 28%|β–ˆβ–ˆβ–Š | 961/3420 [01:04<01:36, 25.48it/s] Running loglikelihood requests: 28%|β–ˆβ–ˆβ–Š | 969/3420 [01:05<01:36, 25.53it/s] Running loglikelihood requests: 29%|β–ˆβ–ˆβ–Š | 977/3420 [01:05<01:35, 25.52it/s] Running loglikelihood requests: 29%|β–ˆβ–ˆβ–‰ | 985/3420 [01:05<01:35, 25.50it/s] Running loglikelihood requests: 29%|β–ˆβ–ˆβ–‰ | 993/3420 [01:06<01:35, 25.50it/s] Running loglikelihood requests: 29%|β–ˆβ–ˆβ–‰ | 1001/3420 [01:06<01:34, 25.55it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–‰ | 1009/3420 [01:06<01:34, 25.53it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–‰ | 1017/3420 [01:06<01:33, 25.73it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–‰ | 1025/3420 [01:07<01:32, 25.88it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–ˆ | 1033/3420 [01:07<01:31, 25.98it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–ˆ | 1041/3420 [01:07<01:31, 26.07it/s] Running loglikelihood requests: 31%|β–ˆβ–ˆβ–ˆ | 1049/3420 [01:08<01:30, 26.12it/s] Running loglikelihood requests: 31%|β–ˆβ–ˆβ–ˆ | 1057/3420 [01:08<01:30, 26.15it/s] Running loglikelihood requests: 31%|β–ˆβ–ˆβ–ˆ | 1065/3420 [01:08<01:29, 26.18it/s] Running loglikelihood requests: 31%|β–ˆβ–ˆβ–ˆβ– | 1073/3420 [01:09<01:29, 26.26it/s] Running loglikelihood requests: 32%|β–ˆβ–ˆβ–ˆβ– | 1081/3420 [01:09<01:29, 26.27it/s] Running loglikelihood requests: 32%|β–ˆβ–ˆβ–ˆβ– | 1089/3420 [01:09<01:28, 26.29it/s] Running loglikelihood requests: 32%|β–ˆβ–ˆβ–ˆβ– | 1097/3420 [01:10<01:28, 26.29it/s] Running loglikelihood requests: 32%|β–ˆβ–ˆβ–ˆβ– | 1105/3420 [01:10<01:27, 26.32it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1113/3420 [01:10<01:27, 26.35it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1121/3420 [01:10<01:27, 26.33it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1129/3420 [01:11<01:26, 26.35it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1137/3420 [01:11<01:26, 26.38it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1145/3420 [01:11<01:26, 26.35it/s] Running loglikelihood requests: 34%|β–ˆβ–ˆβ–ˆβ–Ž | 1153/3420 [01:12<01:25, 26.36it/s] Running loglikelihood requests: 34%|β–ˆβ–ˆβ–ˆβ– | 1161/3420 [01:12<01:25, 26.39it/s] Running loglikelihood requests: 34%|β–ˆβ–ˆβ–ˆβ– | 1169/3420 [01:12<01:25, 26.39it/s] Running loglikelihood requests: 34%|β–ˆβ–ˆβ–ˆβ– | 1177/3420 [01:13<01:24, 26.40it/s] Running loglikelihood requests: 35%|β–ˆβ–ˆβ–ˆβ– | 1185/3420 [01:13<01:24, 26.40it/s] Running loglikelihood requests: 35%|β–ˆβ–ˆβ–ˆβ– | 1193/3420 [01:13<01:24, 26.38it/s] Running loglikelihood requests: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 1201/3420 [01:13<01:24, 26.40it/s] Running loglikelihood requests: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 1209/3420 [01:14<01:23, 26.43it/s] Running loglikelihood requests: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 1217/3420 [01:14<01:23, 26.40it/s] Running loglikelihood requests: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 1225/3420 [01:14<01:23, 26.41it/s] Running loglikelihood requests: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 1233/3420 [01:15<01:22, 26.42it/s] Running loglikelihood requests: 36%|β–ˆβ–ˆβ–ˆβ–‹ | 1241/3420 [01:15<01:22, 26.45it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1249/3420 [01:15<01:22, 26.42it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1257/3420 [01:16<01:21, 26.43it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1265/3420 [01:16<01:21, 26.47it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1273/3420 [01:16<01:21, 26.45it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1281/3420 [01:16<01:20, 26.46it/s] Running loglikelihood requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1289/3420 [01:17<01:20, 26.48it/s] Running loglikelihood requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1297/3420 [01:17<01:20, 26.50it/s] Running loglikelihood requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1305/3420 [01:17<01:19, 26.48it/s] Running loglikelihood requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1313/3420 [01:18<01:19, 26.50it/s] Running loglikelihood requests: 39%|β–ˆβ–ˆβ–ˆβ–Š | 1321/3420 [01:18<01:19, 26.51it/s] Running loglikelihood requests: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 1329/3420 [01:18<01:18, 26.51it/s] Running loglikelihood requests: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 1337/3420 [01:19<01:18, 26.51it/s] Running loglikelihood requests: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 1345/3420 [01:19<01:18, 26.52it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 1353/3420 [01:19<01:17, 26.55it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 1361/3420 [01:19<01:17, 26.54it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 1369/3420 [01:20<01:17, 26.55it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 1377/3420 [01:20<01:16, 26.57it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 1385/3420 [01:20<01:16, 26.53it/s] Running loglikelihood requests: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 1393/3420 [01:21<01:16, 26.55it/s] Running loglikelihood requests: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 1401/3420 [01:21<01:16, 26.54it/s] Running loglikelihood requests: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 1409/3420 [01:21<01:15, 26.55it/s] Running loglikelihood requests: 41%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1417/3420 [01:22<01:15, 26.56it/s] Running loglikelihood requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1425/3420 [01:22<01:15, 26.57it/s] Running loglikelihood requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1433/3420 [01:22<01:14, 26.58it/s] Running loglikelihood requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1441/3420 [01:23<01:14, 26.58it/s] Running loglikelihood requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1449/3420 [01:23<01:14, 26.59it/s] Running loglikelihood requests: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1457/3420 [01:23<01:13, 26.60it/s] Running loglikelihood requests: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1465/3420 [01:23<01:13, 26.62it/s] Running loglikelihood requests: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1473/3420 [01:24<01:13, 26.62it/s] Running loglikelihood requests: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1481/3420 [01:24<01:12, 26.62it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1489/3420 [01:24<01:12, 26.62it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1497/3420 [01:25<01:12, 26.63it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1505/3420 [01:25<01:10, 27.12it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1513/3420 [01:25<01:09, 27.48it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1521/3420 [01:25<01:08, 27.71it/s] Running loglikelihood requests: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1529/3420 [01:26<01:07, 27.90it/s] Running loglikelihood requests: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1537/3420 [01:26<01:07, 28.03it/s] Running loglikelihood requests: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1545/3420 [01:26<01:06, 28.13it/s] Running loglikelihood requests: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1553/3420 [01:27<01:06, 28.19it/s] Running loglikelihood requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1561/3420 [01:27<01:05, 28.23it/s] Running loglikelihood requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1569/3420 [01:27<01:05, 28.26it/s] Running loglikelihood requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1577/3420 [01:27<01:05, 28.27it/s] Running loglikelihood requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1585/3420 [01:28<01:04, 28.29it/s] Running loglikelihood requests: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1593/3420 [01:28<01:04, 28.32it/s] Running loglikelihood requests: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1601/3420 [01:28<01:04, 28.35it/s] Running loglikelihood requests: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1609/3420 [01:29<01:03, 28.37it/s] Running loglikelihood requests: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1617/3420 [01:29<01:03, 28.39it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1625/3420 [01:29<01:03, 28.41it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1633/3420 [01:29<01:02, 28.43it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1641/3420 [01:30<01:02, 28.44it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1649/3420 [01:30<01:02, 28.45it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1657/3420 [01:30<01:01, 28.46it/s] Running loglikelihood requests: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1665/3420 [01:31<01:01, 28.49it/s] Running loglikelihood requests: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1673/3420 [01:31<01:01, 28.49it/s] Running loglikelihood requests: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1681/3420 [01:31<01:00, 28.52it/s] Running loglikelihood requests: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1689/3420 [01:31<01:00, 28.54it/s] Running loglikelihood requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1697/3420 [01:32<01:00, 28.54it/s] Running loglikelihood requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1705/3420 [01:32<01:00, 28.56it/s] Running loglikelihood requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1713/3420 [01:32<00:59, 28.57it/s] Running loglikelihood requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1721/3420 [01:32<00:59, 28.58it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1729/3420 [01:33<00:59, 28.60it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1737/3420 [01:33<00:58, 28.61it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1745/3420 [01:33<00:58, 28.62it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1753/3420 [01:34<00:58, 28.63it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1761/3420 [01:34<00:57, 28.62it/s] Running loglikelihood requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1769/3420 [01:34<00:57, 28.63it/s] Running loglikelihood requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1777/3420 [01:34<00:57, 28.64it/s] Running loglikelihood requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1785/3420 [01:35<00:57, 28.66it/s] Running loglikelihood requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1793/3420 [01:35<00:56, 28.65it/s] Running loglikelihood requests: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1801/3420 [01:35<00:56, 28.65it/s] Running loglikelihood requests: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1809/3420 [01:36<00:56, 28.66it/s] Running loglikelihood requests: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1817/3420 [01:36<00:55, 28.67it/s] Running loglikelihood requests: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1825/3420 [01:36<00:55, 28.66it/s] Running loglikelihood requests: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1833/3420 [01:36<00:55, 28.66it/s] Running loglikelihood requests: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1841/3420 [01:37<00:55, 28.66it/s] Running loglikelihood requests: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1849/3420 [01:37<00:54, 28.67it/s] Running loglikelihood requests: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1857/3420 [01:37<00:54, 28.67it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1865/3420 [01:38<00:54, 28.67it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1873/3420 [01:38<00:53, 28.68it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1881/3420 [01:38<00:51, 29.87it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1889/3420 [01:38<00:49, 30.76it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1897/3420 [01:39<00:48, 31.42it/s] Running loglikelihood requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1905/3420 [01:39<00:47, 31.89it/s] Running loglikelihood requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1913/3420 [01:39<00:46, 32.23it/s] Running loglikelihood requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1921/3420 [01:39<00:46, 32.48it/s] Running loglikelihood requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1929/3420 [01:39<00:45, 32.73it/s] Running loglikelihood requests: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1937/3420 [01:40<00:45, 32.84it/s] Running loglikelihood requests: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1945/3420 [01:40<00:44, 32.92it/s] Running loglikelihood requests: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1953/3420 [01:40<00:44, 32.98it/s] Running loglikelihood requests: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1961/3420 [01:40<00:44, 33.02it/s] Running loglikelihood requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1969/3420 [01:41<00:43, 33.12it/s] Running loglikelihood requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1977/3420 [01:41<00:43, 33.19it/s] Running loglikelihood requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1985/3420 [01:41<00:43, 33.17it/s] Running loglikelihood requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1993/3420 [01:41<00:43, 33.16it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2001/3420 [01:42<00:42, 33.22it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2009/3420 [01:42<00:42, 33.27it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2017/3420 [01:42<00:42, 33.24it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2025/3420 [01:42<00:41, 33.22it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2033/3420 [01:43<00:41, 33.23it/s] Running loglikelihood requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2041/3420 [01:43<00:41, 33.21it/s] Running loglikelihood requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2049/3420 [01:43<00:41, 33.26it/s] Running loglikelihood requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2057/3420 [01:43<00:40, 33.30it/s] Running loglikelihood requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2065/3420 [01:44<00:40, 33.25it/s] Running loglikelihood requests: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2073/3420 [01:44<00:40, 33.22it/s] Running loglikelihood requests: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2081/3420 [01:44<00:40, 33.24it/s] Running loglikelihood requests: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2089/3420 [01:44<00:40, 33.26it/s] Running loglikelihood requests: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2097/3420 [01:45<00:39, 33.24it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2105/3420 [01:45<00:39, 33.28it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2113/3420 [01:45<00:39, 33.26it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2121/3420 [01:45<00:39, 33.30it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2129/3420 [01:45<00:38, 33.30it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2137/3420 [01:46<00:38, 33.30it/s] Running loglikelihood requests: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2145/3420 [01:46<00:38, 33.30it/s] Running loglikelihood requests: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2153/3420 [01:46<00:38, 33.31it/s] Running loglikelihood requests: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2161/3420 [01:46<00:37, 33.31it/s] Running loglikelihood requests: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2169/3420 [01:47<00:37, 33.36it/s] Running loglikelihood requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2177/3420 [01:47<00:37, 33.35it/s] Running loglikelihood requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2185/3420 [01:47<00:37, 33.35it/s] Running loglikelihood requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2193/3420 [01:47<00:36, 33.36it/s] Running loglikelihood requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2201/3420 [01:48<00:36, 33.36it/s] Running loglikelihood requests: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2209/3420 [01:48<00:36, 33.41it/s] Running loglikelihood requests: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2217/3420 [01:48<00:36, 33.40it/s] Running loglikelihood requests: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2225/3420 [01:48<00:35, 33.40it/s] Running loglikelihood requests: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2233/3420 [01:49<00:35, 33.42it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2241/3420 [01:49<00:35, 33.46it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2249/3420 [01:49<00:35, 33.45it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2257/3420 [01:49<00:34, 33.48it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2265/3420 [01:50<00:34, 33.47it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2273/3420 [01:50<00:34, 33.50it/s] Running loglikelihood requests: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2281/3420 [01:50<00:33, 33.52it/s] Running loglikelihood requests: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2289/3420 [01:50<00:33, 33.53it/s] Running loglikelihood requests: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2297/3420 [01:51<00:33, 33.52it/s] Running loglikelihood requests: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2305/3420 [01:51<00:33, 33.57it/s] Running loglikelihood requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2313/3420 [01:51<00:32, 33.62it/s] Running loglikelihood requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2321/3420 [01:51<00:32, 33.62it/s] Running loglikelihood requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2329/3420 [01:51<00:32, 33.58it/s] Running loglikelihood requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2337/3420 [01:52<00:32, 33.60it/s] Running loglikelihood requests: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2345/3420 [01:52<00:31, 33.64it/s] Running loglikelihood requests: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2353/3420 [01:52<00:31, 33.68it/s] Running loglikelihood requests: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2361/3420 [01:52<00:31, 33.64it/s] Running loglikelihood requests: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2369/3420 [01:53<00:31, 33.65it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2377/3420 [01:53<00:30, 34.72it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2385/3420 [01:53<00:29, 35.46it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2393/3420 [01:53<00:28, 36.03it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2401/3420 [01:54<00:28, 36.38it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2409/3420 [01:54<00:27, 36.63it/s] Running loglikelihood requests: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2417/3420 [01:54<00:27, 36.87it/s] Running loglikelihood requests: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2425/3420 [01:54<00:26, 37.04it/s] Running loglikelihood requests: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2433/3420 [01:54<00:26, 37.18it/s] Running loglikelihood requests: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2441/3420 [01:55<00:26, 37.25it/s] Running loglikelihood requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2449/3420 [01:55<00:26, 37.33it/s] Running loglikelihood requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2457/3420 [01:55<00:25, 37.38it/s] Running loglikelihood requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2465/3420 [01:55<00:25, 37.43it/s] Running loglikelihood requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2473/3420 [01:55<00:25, 37.46it/s] Running loglikelihood requests: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2481/3420 [01:56<00:25, 37.49it/s] Running loglikelihood requests: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2489/3420 [01:56<00:24, 37.50it/s] Running loglikelihood requests: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2501/3420 [01:56<00:21, 43.11it/s] Running loglikelihood requests: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2509/3420 [01:56<00:21, 41.47it/s] Running loglikelihood requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2517/3420 [01:57<00:22, 40.29it/s] Running loglikelihood requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2525/3420 [01:57<00:22, 39.47it/s] Running loglikelihood requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2533/3420 [01:57<00:22, 38.88it/s] Running loglikelihood requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2541/3420 [01:57<00:22, 38.49it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2549/3420 [01:57<00:22, 38.13it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2557/3420 [01:58<00:22, 37.89it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2565/3420 [01:58<00:22, 37.71it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2573/3420 [01:58<00:22, 37.68it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2581/3420 [01:58<00:22, 37.78it/s] Running loglikelihood requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2589/3420 [01:58<00:21, 37.89it/s] Running loglikelihood requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2597/3420 [01:59<00:21, 37.99it/s] Running loglikelihood requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2605/3420 [01:59<00:21, 38.06it/s] Running loglikelihood requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2613/3420 [01:59<00:21, 38.18it/s] Running loglikelihood requests: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2621/3420 [01:59<00:20, 38.22it/s] Running loglikelihood requests: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2629/3420 [01:59<00:20, 38.28it/s] Running loglikelihood requests: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2637/3420 [02:00<00:20, 38.29it/s] Running loglikelihood requests: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2645/3420 [02:00<00:20, 38.24it/s] Running loglikelihood requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2653/3420 [02:00<00:20, 38.32it/s] Running loglikelihood requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2661/3420 [02:00<00:19, 38.22it/s] Running loglikelihood requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2669/3420 [02:01<00:19, 38.22it/s] Running loglikelihood requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2677/3420 [02:01<00:19, 38.16it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2685/3420 [02:01<00:19, 38.12it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2693/3420 [02:01<00:19, 38.09it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2701/3420 [02:01<00:18, 38.13it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2709/3420 [02:02<00:18, 38.11it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2717/3420 [02:02<00:18, 38.15it/s] Running loglikelihood requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2725/3420 [02:02<00:18, 38.12it/s] Running loglikelihood requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2733/3420 [02:02<00:18, 38.10it/s] Running loglikelihood requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2741/3420 [02:02<00:17, 38.08it/s] Running loglikelihood requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2749/3420 [02:03<00:17, 38.14it/s] Running loglikelihood requests: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2757/3420 [02:03<00:17, 38.12it/s] Running loglikelihood requests: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2765/3420 [02:03<00:17, 38.17it/s] Running loglikelihood requests: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2773/3420 [02:03<00:16, 38.15it/s] Running loglikelihood requests: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2781/3420 [02:03<00:16, 38.17it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2789/3420 [02:04<00:16, 38.15it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2797/3420 [02:04<00:16, 38.22it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2805/3420 [02:04<00:16, 38.19it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2813/3420 [02:04<00:15, 38.23it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2821/3420 [02:04<00:15, 38.21it/s] Running loglikelihood requests: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2829/3420 [02:05<00:15, 38.27it/s] Running loglikelihood requests: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2837/3420 [02:05<00:15, 38.28it/s] Running loglikelihood requests: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2845/3420 [02:05<00:15, 38.24it/s] Running loglikelihood requests: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2853/3420 [02:05<00:14, 38.28it/s] Running loglikelihood requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2861/3420 [02:06<00:14, 38.24it/s] Running loglikelihood requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2869/3420 [02:06<00:14, 38.30it/s] Running loglikelihood requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2877/3420 [02:06<00:14, 38.30it/s] Running loglikelihood requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2885/3420 [02:06<00:13, 38.35it/s] Running loglikelihood requests: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2893/3420 [02:06<00:13, 38.32it/s] Running loglikelihood requests: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2901/3420 [02:07<00:13, 38.31it/s] Running loglikelihood requests: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2909/3420 [02:07<00:13, 38.37it/s] Running loglikelihood requests: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2917/3420 [02:07<00:13, 38.43it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2925/3420 [02:07<00:12, 39.32it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2933/3420 [02:07<00:12, 39.94it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2941/3420 [02:08<00:11, 40.36it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2949/3420 [02:08<00:11, 40.69it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2957/3420 [02:08<00:11, 41.02it/s] Running loglikelihood requests: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2965/3420 [02:08<00:11, 41.20it/s] Running loglikelihood requests: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2973/3420 [02:08<00:10, 41.34it/s] Running loglikelihood requests: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2981/3420 [02:09<00:10, 41.42it/s] Running loglikelihood requests: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2989/3420 [02:09<00:10, 41.47it/s] Running loglikelihood requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2997/3420 [02:09<00:10, 41.56it/s] Running loglikelihood requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3005/3420 [02:09<00:09, 41.59it/s] Running loglikelihood requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3013/3420 [02:09<00:09, 41.62it/s] Running loglikelihood requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3021/3420 [02:09<00:09, 41.62it/s] Running loglikelihood requests: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3029/3420 [02:10<00:09, 41.64it/s] Running loglikelihood requests: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3037/3420 [02:10<00:09, 41.63it/s] Running loglikelihood requests: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3045/3420 [02:10<00:08, 41.69it/s] Running loglikelihood requests: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3053/3420 [02:10<00:08, 41.71it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3061/3420 [02:10<00:08, 41.70it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3069/3420 [02:11<00:08, 41.76it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3077/3420 [02:11<00:08, 41.76it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3085/3420 [02:11<00:08, 41.79it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3093/3420 [02:11<00:07, 41.79it/s] Running loglikelihood requests: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3101/3420 [02:11<00:07, 41.79it/s] Running loglikelihood requests: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3109/3420 [02:12<00:07, 41.82it/s] Running loglikelihood requests: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3117/3420 [02:12<00:07, 41.80it/s] Running loglikelihood requests: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3125/3420 [02:12<00:07, 41.87it/s] Running loglikelihood requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3133/3420 [02:12<00:06, 41.86it/s] Running loglikelihood requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3141/3420 [02:12<00:06, 41.88it/s] Running loglikelihood requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3149/3420 [02:13<00:06, 41.88it/s] Running loglikelihood requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3157/3420 [02:13<00:06, 41.88it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3165/3420 [02:13<00:06, 41.93it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3173/3420 [02:13<00:05, 41.95it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3181/3420 [02:13<00:05, 41.99it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3189/3420 [02:14<00:05, 42.01it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3197/3420 [02:14<00:05, 41.99it/s] Running loglikelihood requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3205/3420 [02:14<00:05, 42.01it/s] Running loglikelihood requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3213/3420 [02:14<00:04, 42.04it/s] Running loglikelihood requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3221/3420 [02:14<00:04, 42.05it/s] Running loglikelihood requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3229/3420 [02:14<00:04, 42.11it/s] Running loglikelihood requests: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3237/3420 [02:15<00:04, 42.13it/s] Running loglikelihood requests: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3245/3420 [02:15<00:04, 42.17it/s] Running loglikelihood requests: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3253/3420 [02:15<00:03, 42.21it/s] Running loglikelihood requests: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3261/3420 [02:15<00:03, 42.24it/s] Running loglikelihood requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3269/3420 [02:15<00:03, 42.28it/s] Running loglikelihood requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3277/3420 [02:16<00:03, 42.28it/s] Running loglikelihood requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3285/3420 [02:16<00:03, 42.31it/s] Running loglikelihood requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3293/3420 [02:16<00:03, 42.31it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3301/3420 [02:16<00:02, 42.36it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3309/3420 [02:16<00:02, 42.38it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3317/3420 [02:17<00:02, 42.42it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3325/3420 [02:17<00:02, 42.45it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3333/3420 [02:17<00:02, 42.48it/s] Running loglikelihood requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3341/3420 [02:17<00:01, 45.00it/s] Running loglikelihood requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3349/3420 [02:17<00:01, 46.80it/s] Running loglikelihood requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3357/3420 [02:17<00:01, 48.18it/s] Running loglikelihood requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3365/3420 [02:18<00:01, 49.23it/s] Running loglikelihood requests: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3373/3420 [02:18<00:00, 50.09it/s] Running loglikelihood requests: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3381/3420 [02:18<00:00, 50.70it/s] Running loglikelihood requests: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3389/3420 [02:18<00:00, 51.16it/s] Running loglikelihood requests: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3397/3420 [02:18<00:00, 51.49it/s] Running loglikelihood requests: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3405/3420 [02:18<00:00, 51.63it/s] Running loglikelihood requests: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3413/3420 [02:18<00:00, 51.74it/s] Running loglikelihood requests: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 3420/3420 [02:18<00:00, 24.61it/s]
fatal: detected dubious ownership in repository at '/root/steering-resistance'
To add an exception for this directory, call:
git config --global --add safe.directory /root/steering-resistance
2026-07-24:07:29:00 INFO [loggers.evaluation_tracker:247] Saving results aggregated
hf ({'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16'}), gen_kwargs: ({}), limit: 15.0, num_fewshot: 5, batch_size: 2
| Tasks |Version|Filter|n-shot|Metric| |Value | |Stderr|
|---------------------------------------|------:|------|-----:|------|---|-----:|---|-----:|
|mmlu | 2|none | |acc | |0.7509|Β± |0.0142|
| - humanities | 2|none | 5|acc |↑ |0.8051|Β± |0.0271|
| - formal_logic | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - high_school_european_history | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - high_school_us_history | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - high_school_world_history | 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - international_law | 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - jurisprudence | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - logical_fallacies | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - moral_disputes | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - moral_scenarios | 1|none | 5|acc |↑ |0.6000|Β± |0.1309|
| - philosophy | 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - prehistory | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - professional_law | 1|none | 5|acc |↑ |0.5333|Β± |0.1333|
| - world_religions | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - other | 2|none | 5|acc |↑ |0.7385|Β± |0.0298|
| - business_ethics | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - clinical_knowledge | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - college_medicine | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - global_facts | 1|none | 5|acc |↑ |0.4000|Β± |0.1309|
| - human_aging | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - management | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - marketing | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - medical_genetics | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - miscellaneous | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - nutrition | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - professional_accounting | 1|none | 5|acc |↑ |0.3333|Β± |0.1260|
| - professional_medicine | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - virology | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - social sciences | 2|none | 5|acc |↑ |0.8111|Β± |0.0289|
| - econometrics | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - high_school_geography | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - high_school_government_and_politics| 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - high_school_macroeconomics | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - high_school_microeconomics | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - high_school_psychology | 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - human_sexuality | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - professional_psychology | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - public_relations | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - security_studies | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - sociology | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - us_foreign_policy | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - stem | 2|none | 5|acc |↑ |0.6842|Β± |0.0268|
| - abstract_algebra | 1|none | 5|acc |↑ |0.5333|Β± |0.1333|
| - anatomy | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - astronomy | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - college_biology | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - college_chemistry | 1|none | 5|acc |↑ |0.5333|Β± |0.1333|
| - college_computer_science | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - college_mathematics | 1|none | 5|acc |↑ |0.3333|Β± |0.1260|
| - college_physics | 1|none | 5|acc |↑ |0.4000|Β± |0.1309|
| - computer_security | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - conceptual_physics | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - electrical_engineering | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - elementary_mathematics | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - high_school_biology | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - high_school_chemistry | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - high_school_computer_science | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - high_school_mathematics | 1|none | 5|acc |↑ |0.5333|Β± |0.1333|
| - high_school_physics | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - high_school_statistics | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - machine_learning | 1|none | 5|acc |↑ |0.6000|Β± |0.1309|
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|------------------|------:|------|-----:|------|---|-----:|---|-----:|
|mmlu | 2|none | |acc | |0.7509|Β± |0.0142|
| - humanities | 2|none | 5|acc |↑ |0.8051|Β± |0.0271|
| - other | 2|none | 5|acc |↑ |0.7385|Β± |0.0298|
| - social sciences| 2|none | 5|acc |↑ |0.8111|Β± |0.0289|
| - stem | 2|none | 5|acc |↑ |0.6842|Β± |0.0268|
########## m1 / mmlu ##########
/usr/bin/python -m lm_eval --model hf --model_args pretrained=Qwen/Qwen2.5-7B-Instruct,dtype=bfloat16,peft=/root/steering-resistance/results/full_7b/m1_resist_adapter --tasks mmlu --num_fewshot 5 --batch_size 2 --seed 0 --output_path /root/steering-resistance/results/full_7b/capability/m1/mmlu --apply_chat_template --fewshot_as_multiturn --limit 15
/usr/local/lib/python3.11/dist-packages/requests/__init__.py:113: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (6.0.0.post1)/charset_normalizer (3.3.2) doesn't match a supported version!
warnings.warn(
2026-07-24:07:29:02 WARNING [config.evaluate_config:287] --limit SHOULD ONLY BE USED FOR TESTING. REAL METRICS SHOULD NOT BE COMPUTED USING LIMIT.
2026-07-24:07:29:10 INFO [_cli.run:388] Selected Tasks: ['mmlu']
2026-07-24:07:29:12 INFO [evaluator:214] Setting random seed to 0 | Setting numpy seed to 0 | Setting torch manual seed to 0 | Setting fewshot manual seed to 0
2026-07-24:07:29:12 INFO [evaluator:239] Initializing hf model, with arguments: {'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16', 'peft': '/root/steering-resistance/results/full_7b/m1_resist_adapter'}
2026-07-24:07:29:15 INFO [models.huggingface:286] Using device 'cuda:0'
2026-07-24:07:29:16 INFO [models.huggingface:579] Model parallel was set to False, max memory was not set, and device map was set to {'': 'cuda:0'}
Loading weights: 0%| | 0/339 [00:00<?, ?it/s] Loading weights: 1%| | 2/339 [00:00<00:34, 9.88it/s] Loading weights: 5%|β–Œ | 18/339 [00:00<00:04, 67.88it/s] Loading weights: 12%|β–ˆβ– | 40/339 [00:00<00:02, 104.91it/s] Loading weights: 19%|β–ˆβ–‰ | 65/339 [00:00<00:01, 142.10it/s] Loading weights: 26%|β–ˆβ–ˆβ–‹ | 89/339 [00:00<00:01, 166.18it/s] Loading weights: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 112/339 [00:00<00:01, 174.66it/s] Loading weights: 39%|β–ˆβ–ˆβ–ˆβ–Š | 131/339 [00:00<00:01, 164.30it/s] Loading weights: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 149/339 [00:01<00:01, 160.05it/s] Loading weights: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 166/339 [00:01<00:01, 161.55it/s] Loading weights: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 185/339 [00:01<00:00, 161.53it/s] Loading weights: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 202/339 [00:01<00:00, 161.73it/s] Loading weights: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 219/339 [00:01<00:00, 149.64it/s] Loading weights: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 235/339 [00:01<00:00, 140.04it/s] Loading weights: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 254/339 [00:01<00:00, 143.93it/s] Loading weights: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 269/339 [00:01<00:00, 133.20it/s] Loading weights: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 288/339 [00:02<00:00, 146.34it/s] Loading weights: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 304/339 [00:02<00:00, 143.21it/s] Loading weights: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 324/339 [00:02<00:00, 156.98it/s] Loading weights: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 339/339 [00:02<00:00, 146.70it/s]
2026-07-24:07:29:21 INFO [models.huggingface:885] Model config indicates vocab_size='152064', but found tokenizer with vocab size '151665'. Resizing model embedding layer...
2026-07-24:07:30:19 INFO [evaluator_utils:446] Selected tasks:
2026-07-24:07:30:19 INFO [evaluator_utils:462] Group: mmlu
2026-07-24:07:30:19 INFO [evaluator_utils:462] Group: mmlu_stem
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_abstract_algebra (mmlu/default/mmlu_abstract_algebra.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_anatomy (mmlu/default/mmlu_anatomy.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_astronomy (mmlu/default/mmlu_astronomy.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_college_biology (mmlu/default/mmlu_college_biology.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_college_chemistry (mmlu/default/mmlu_college_chemistry.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_college_computer_science (mmlu/default/mmlu_college_computer_science.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_college_mathematics (mmlu/default/mmlu_college_mathematics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_college_physics (mmlu/default/mmlu_college_physics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_computer_security (mmlu/default/mmlu_computer_security.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_conceptual_physics (mmlu/default/mmlu_conceptual_physics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_electrical_engineering (mmlu/default/mmlu_electrical_engineering.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_elementary_mathematics (mmlu/default/mmlu_elementary_mathematics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_biology (mmlu/default/mmlu_high_school_biology.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_chemistry (mmlu/default/mmlu_high_school_chemistry.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_computer_science (mmlu/default/mmlu_high_school_computer_science.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_mathematics (mmlu/default/mmlu_high_school_mathematics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_physics (mmlu/default/mmlu_high_school_physics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_statistics (mmlu/default/mmlu_high_school_statistics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_machine_learning (mmlu/default/mmlu_machine_learning.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:462] Group: mmlu_other
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_business_ethics (mmlu/default/mmlu_business_ethics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_clinical_knowledge (mmlu/default/mmlu_clinical_knowledge.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_college_medicine (mmlu/default/mmlu_college_medicine.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_global_facts (mmlu/default/mmlu_global_facts.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_human_aging (mmlu/default/mmlu_human_aging.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_management (mmlu/default/mmlu_management.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_marketing (mmlu/default/mmlu_marketing.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_medical_genetics (mmlu/default/mmlu_medical_genetics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_miscellaneous (mmlu/default/mmlu_miscellaneous.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_nutrition (mmlu/default/mmlu_nutrition.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_professional_accounting (mmlu/default/mmlu_professional_accounting.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_professional_medicine (mmlu/default/mmlu_professional_medicine.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_virology (mmlu/default/mmlu_virology.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:462] Group: mmlu_social_sciences
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_econometrics (mmlu/default/mmlu_econometrics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_geography (mmlu/default/mmlu_high_school_geography.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_government_and_politics (mmlu/default/mmlu_high_school_government_and_politics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_macroeconomics (mmlu/default/mmlu_high_school_macroeconomics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_microeconomics (mmlu/default/mmlu_high_school_microeconomics.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_psychology (mmlu/default/mmlu_high_school_psychology.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_human_sexuality (mmlu/default/mmlu_human_sexuality.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_professional_psychology (mmlu/default/mmlu_professional_psychology.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_public_relations (mmlu/default/mmlu_public_relations.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_security_studies (mmlu/default/mmlu_security_studies.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_sociology (mmlu/default/mmlu_sociology.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_us_foreign_policy (mmlu/default/mmlu_us_foreign_policy.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:462] Group: mmlu_humanities
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_formal_logic (mmlu/default/mmlu_formal_logic.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_european_history (mmlu/default/mmlu_high_school_european_history.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_us_history (mmlu/default/mmlu_high_school_us_history.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_high_school_world_history (mmlu/default/mmlu_high_school_world_history.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_international_law (mmlu/default/mmlu_international_law.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_jurisprudence (mmlu/default/mmlu_jurisprudence.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_logical_fallacies (mmlu/default/mmlu_logical_fallacies.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_moral_disputes (mmlu/default/mmlu_moral_disputes.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_moral_scenarios (mmlu/default/mmlu_moral_scenarios.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_philosophy (mmlu/default/mmlu_philosophy.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_prehistory (mmlu/default/mmlu_prehistory.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_professional_law (mmlu/default/mmlu_professional_law.yaml)
2026-07-24:07:30:19 INFO [evaluator_utils:470] Task: mmlu_world_religions (mmlu/default/mmlu_world_religions.yaml)
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_abstract_algebra from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_anatomy from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_astronomy from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_biology from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_chemistry from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_computer_science from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_mathematics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_physics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_computer_security from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_conceptual_physics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_electrical_engineering from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_elementary_mathematics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_biology from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_chemistry from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_computer_science from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_mathematics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_physics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_statistics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_machine_learning from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_business_ethics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_clinical_knowledge from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_college_medicine from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_global_facts from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_aging from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_management from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_marketing from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_medical_genetics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_miscellaneous from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_nutrition from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_accounting from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_medicine from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_virology from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_econometrics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_geography from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_government_and_politics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_macroeconomics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_microeconomics from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_psychology from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_human_sexuality from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_psychology from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_public_relations from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_security_studies from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_sociology from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_us_foreign_policy from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_formal_logic from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_european_history from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_us_history from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_high_school_world_history from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_international_law from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_jurisprudence from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_logical_fallacies from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_disputes from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_moral_scenarios from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_philosophy from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_prehistory from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_professional_law from None to 5
2026-07-24:07:30:19 WARNING [evaluator:333] Overwriting default num_fewshot of mmlu_world_religions from None to 5
2026-07-24:07:30:19 INFO [api.task:312] Building contexts for mmlu_abstract_algebra on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 157.69it/s]
2026-07-24:07:30:19 INFO [api.task:312] Building contexts for mmlu_anatomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 176.99it/s]
2026-07-24:07:30:19 INFO [api.task:312] Building contexts for mmlu_astronomy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 176.20it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_college_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 177.17it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_college_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 184.17it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_college_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 182.84it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_college_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 179.06it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_college_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 177.87it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_computer_security on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 172.20it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_conceptual_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 173.15it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_electrical_engineering on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 175.23it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_elementary_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 180.57it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_high_school_biology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 181.86it/s]
2026-07-24:07:30:20 INFO [api.task:312] Building contexts for mmlu_high_school_chemistry on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 175.50it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_high_school_computer_science on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.12it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_high_school_mathematics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.80it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_high_school_physics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.12it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_high_school_statistics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.69it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_machine_learning on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.47it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_business_ethics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.19it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_clinical_knowledge on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.80it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_college_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.42it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_global_facts on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.47it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_human_aging on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.87it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_management on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.98it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_marketing on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.67it/s]
2026-07-24:07:30:21 INFO [api.task:312] Building contexts for mmlu_medical_genetics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.61it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_miscellaneous on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.40it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_nutrition on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.81it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_professional_accounting on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.93it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_professional_medicine on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.71it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_virology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.83it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_econometrics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.68it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_high_school_geography on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.14it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_high_school_government_and_politics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.64it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_high_school_macroeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.63it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_high_school_microeconomics on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.57it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_high_school_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.59it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_human_sexuality on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.68it/s]
2026-07-24:07:30:22 INFO [api.task:312] Building contexts for mmlu_professional_psychology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.69it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_public_relations on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.03it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_security_studies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.48it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_sociology on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.75it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_us_foreign_policy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.78it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_formal_logic on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.55it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_high_school_european_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 192.28it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_high_school_us_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.72it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_high_school_world_history on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 191.34it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_international_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.83it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_jurisprudence on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.20it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_logical_fallacies on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.16it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_moral_disputes on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.17it/s]
2026-07-24:07:30:23 INFO [api.task:312] Building contexts for mmlu_moral_scenarios on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 190.46it/s]
2026-07-24:07:30:24 INFO [api.task:312] Building contexts for mmlu_philosophy on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.78it/s]
2026-07-24:07:30:24 INFO [api.task:312] Building contexts for mmlu_prehistory on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 188.10it/s]
2026-07-24:07:30:24 INFO [api.task:312] Building contexts for mmlu_professional_law on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 193.39it/s]
2026-07-24:07:30:24 INFO [api.task:312] Building contexts for mmlu_world_religions on rank 0...
0%| | 0/15 [00:00<?, ?it/s] 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 15/15 [00:00<00:00, 194.62it/s]
2026-07-24:07:30:24 INFO [evaluator:585] Running loglikelihood requests
Tokenizing inputs: 0%| | 0/3420 [00:00<?, ?it/s] Tokenizing inputs: 2%|▏ | 68/3420 [00:00<00:04, 674.98it/s] Tokenizing inputs: 4%|▍ | 136/3420 [00:00<00:05, 636.49it/s] Tokenizing inputs: 6%|β–Œ | 200/3420 [00:00<00:05, 541.97it/s] Tokenizing inputs: 7%|β–‹ | 256/3420 [00:00<00:05, 533.83it/s] Tokenizing inputs: 9%|β–‰ | 311/3420 [00:00<00:06, 500.75it/s] Tokenizing inputs: 11%|β–ˆ | 362/3420 [00:00<00:07, 423.49it/s] Tokenizing inputs: 12%|β–ˆβ– | 408/3420 [00:00<00:06, 431.90it/s] Tokenizing inputs: 13%|β–ˆβ–Ž | 457/3420 [00:00<00:06, 446.59it/s] Tokenizing inputs: 15%|β–ˆβ–Œ | 514/3420 [00:01<00:06, 480.31it/s] Tokenizing inputs: 17%|β–ˆβ–‹ | 587/3420 [00:01<00:05, 550.13it/s] Tokenizing inputs: 19%|β–ˆβ–‰ | 650/3420 [00:01<00:04, 572.82it/s] Tokenizing inputs: 21%|β–ˆβ–ˆ | 709/3420 [00:01<00:04, 547.23it/s] Tokenizing inputs: 22%|β–ˆβ–ˆβ– | 765/3420 [00:01<00:04, 534.21it/s] Tokenizing inputs: 24%|β–ˆβ–ˆβ– | 820/3420 [00:01<00:04, 533.89it/s] Tokenizing inputs: 26%|β–ˆβ–ˆβ–Œ | 874/3420 [00:01<00:05, 463.29it/s] Tokenizing inputs: 27%|β–ˆβ–ˆβ–‹ | 923/3420 [00:01<00:05, 429.25it/s] Tokenizing inputs: 28%|β–ˆβ–ˆβ–Š | 973/3420 [00:01<00:05, 446.04it/s] Tokenizing inputs: 30%|β–ˆβ–ˆβ–‰ | 1019/3420 [00:02<00:05, 444.96it/s] Tokenizing inputs: 31%|β–ˆβ–ˆβ–ˆ | 1065/3420 [00:02<00:05, 405.95it/s] Tokenizing inputs: 32%|β–ˆβ–ˆβ–ˆβ– | 1107/3420 [00:02<00:05, 394.72it/s] Tokenizing inputs: 34%|β–ˆβ–ˆβ–ˆβ–Ž | 1148/3420 [00:02<00:05, 393.37it/s] Tokenizing inputs: 35%|β–ˆβ–ˆβ–ˆβ– | 1194/3420 [00:02<00:05, 410.06it/s] Tokenizing inputs: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1256/3420 [00:02<00:04, 466.63it/s] Tokenizing inputs: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1304/3420 [00:02<00:04, 462.32it/s] Tokenizing inputs: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 1361/3420 [00:02<00:04, 492.38it/s] Tokenizing inputs: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1432/3420 [00:02<00:03, 553.30it/s] Tokenizing inputs: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1510/3420 [00:03<00:03, 616.51it/s] Tokenizing inputs: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1573/3420 [00:03<00:03, 612.61it/s] Tokenizing inputs: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1651/3420 [00:03<00:02, 659.97it/s] Tokenizing inputs: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1718/3420 [00:03<00:02, 625.68it/s] Tokenizing inputs: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1782/3420 [00:03<00:03, 534.80it/s] Tokenizing inputs: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1839/3420 [00:03<00:03, 418.16it/s] Tokenizing inputs: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1887/3420 [00:03<00:03, 414.93it/s] Tokenizing inputs: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1945/3420 [00:03<00:03, 451.49it/s] Tokenizing inputs: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1994/3420 [00:04<00:03, 460.11it/s] Tokenizing inputs: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2053/3420 [00:04<00:02, 491.63it/s] Tokenizing inputs: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2105/3420 [00:04<00:02, 497.20it/s] Tokenizing inputs: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2166/3420 [00:04<00:02, 528.10it/s] Tokenizing inputs: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2223/3420 [00:04<00:02, 538.79it/s] Tokenizing inputs: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2278/3420 [00:04<00:02, 522.85it/s] Tokenizing inputs: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2344/3420 [00:04<00:01, 560.92it/s] Tokenizing inputs: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2401/3420 [00:04<00:02, 500.41it/s] Tokenizing inputs: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2456/3420 [00:04<00:01, 511.82it/s] Tokenizing inputs: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2509/3420 [00:05<00:02, 375.35it/s] Tokenizing inputs: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2553/3420 [00:05<00:02, 378.26it/s] Tokenizing inputs: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2606/3420 [00:05<00:01, 412.73it/s] Tokenizing inputs: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2657/3420 [00:05<00:01, 437.10it/s] Tokenizing inputs: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2704/3420 [00:05<00:01, 408.53it/s] Tokenizing inputs: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2748/3420 [00:06<00:03, 217.17it/s] Tokenizing inputs: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2782/3420 [00:06<00:03, 177.21it/s] Tokenizing inputs: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2809/3420 [00:06<00:03, 160.96it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2832/3420 [00:06<00:03, 156.14it/s] Tokenizing inputs: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2852/3420 [00:06<00:03, 159.32it/s] Tokenizing inputs: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2871/3420 [00:07<00:03, 160.83it/s] Tokenizing inputs: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2897/3420 [00:07<00:02, 180.26it/s] Tokenizing inputs: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2932/3420 [00:07<00:02, 218.04it/s] Tokenizing inputs: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2983/3420 [00:07<00:01, 288.23it/s] Tokenizing inputs: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3034/3420 [00:07<00:01, 345.06it/s] Tokenizing inputs: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3083/3420 [00:07<00:00, 382.89it/s] Tokenizing inputs: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3128/3420 [00:07<00:00, 400.30it/s] Tokenizing inputs: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3171/3420 [00:07<00:00, 389.73it/s] Tokenizing inputs: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3233/3420 [00:07<00:00, 453.07it/s] Tokenizing inputs: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3280/3420 [00:07<00:00, 454.46it/s] Tokenizing inputs: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3327/3420 [00:08<00:00, 327.22it/s] Tokenizing inputs: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3366/3420 [00:08<00:00, 256.54it/s] Tokenizing inputs: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 3420/3420 [00:08<00:00, 402.66it/s]
Running loglikelihood requests: 0%| | 0/3420 [00:00<?, ?it/s] Running loglikelihood requests: 0%| | 1/3420 [00:02<2:28:23, 2.60s/it] Running loglikelihood requests: 0%| | 9/3420 [00:04<23:25, 2.43it/s] Running loglikelihood requests: 0%| | 17/3420 [00:06<17:08, 3.31it/s] Running loglikelihood requests: 1%| | 25/3420 [00:07<14:43, 3.84it/s] Running loglikelihood requests: 1%| | 33/3420 [00:09<13:27, 4.20it/s] Running loglikelihood requests: 1%| | 41/3420 [00:11<12:37, 4.46it/s] Running loglikelihood requests: 1%|▏ | 49/3420 [00:12<12:07, 4.63it/s] Running loglikelihood requests: 2%|▏ | 57/3420 [00:14<11:44, 4.77it/s] Running loglikelihood requests: 2%|▏ | 65/3420 [00:15<11:03, 5.06it/s] Running loglikelihood requests: 2%|▏ | 73/3420 [00:16<10:32, 5.29it/s] Running loglikelihood requests: 2%|▏ | 81/3420 [00:18<10:10, 5.47it/s] Running loglikelihood requests: 3%|β–Ž | 89/3420 [00:19<09:53, 5.61it/s] Running loglikelihood requests: 3%|β–Ž | 97/3420 [00:20<09:38, 5.74it/s] Running loglikelihood requests: 3%|β–Ž | 105/3420 [00:22<09:27, 5.84it/s] Running loglikelihood requests: 3%|β–Ž | 113/3420 [00:23<09:19, 5.92it/s] Running loglikelihood requests: 4%|β–Ž | 121/3420 [00:24<08:51, 6.20it/s] Running loglikelihood requests: 4%|▍ | 129/3420 [00:25<08:20, 6.58it/s] Running loglikelihood requests: 4%|▍ | 137/3420 [00:26<07:57, 6.88it/s] Running loglikelihood requests: 4%|▍ | 145/3420 [00:27<07:38, 7.15it/s] Running loglikelihood requests: 4%|▍ | 153/3420 [00:28<07:17, 7.47it/s] Running loglikelihood requests: 5%|▍ | 161/3420 [00:29<07:02, 7.71it/s] Running loglikelihood requests: 5%|▍ | 169/3420 [00:30<06:51, 7.90it/s] Running loglikelihood requests: 5%|β–Œ | 177/3420 [00:31<06:43, 8.04it/s] Running loglikelihood requests: 5%|β–Œ | 185/3420 [00:32<06:33, 8.22it/s] Running loglikelihood requests: 6%|β–Œ | 193/3420 [00:33<06:23, 8.42it/s] Running loglikelihood requests: 6%|β–Œ | 201/3420 [00:34<06:10, 8.69it/s] Running loglikelihood requests: 6%|β–Œ | 209/3420 [00:35<06:00, 8.91it/s] Running loglikelihood requests: 6%|β–‹ | 217/3420 [00:35<05:50, 9.14it/s] Running loglikelihood requests: 7%|β–‹ | 225/3420 [00:36<05:43, 9.30it/s] Running loglikelihood requests: 7%|β–‹ | 233/3420 [00:37<05:36, 9.46it/s] Running loglikelihood requests: 7%|β–‹ | 241/3420 [00:38<05:29, 9.64it/s] Running loglikelihood requests: 7%|β–‹ | 249/3420 [00:39<05:23, 9.80it/s] Running loglikelihood requests: 8%|β–Š | 257/3420 [00:39<05:14, 10.06it/s] Running loglikelihood requests: 8%|β–Š | 265/3420 [00:40<05:07, 10.27it/s] Running loglikelihood requests: 8%|β–Š | 273/3420 [00:41<05:02, 10.41it/s] Running loglikelihood requests: 8%|β–Š | 281/3420 [00:42<04:57, 10.55it/s] Running loglikelihood requests: 8%|β–Š | 289/3420 [00:42<04:53, 10.67it/s] Running loglikelihood requests: 9%|β–Š | 297/3420 [00:43<04:50, 10.74it/s] Running loglikelihood requests: 9%|β–‰ | 305/3420 [00:44<04:48, 10.79it/s] Running loglikelihood requests: 9%|β–‰ | 313/3420 [00:45<04:46, 10.83it/s] Running loglikelihood requests: 9%|β–‰ | 321/3420 [00:45<04:43, 10.93it/s] Running loglikelihood requests: 10%|β–‰ | 329/3420 [00:46<04:39, 11.05it/s] Running loglikelihood requests: 10%|β–‰ | 337/3420 [00:47<04:36, 11.14it/s] Running loglikelihood requests: 10%|β–ˆ | 345/3420 [00:47<04:34, 11.20it/s] Running loglikelihood requests: 10%|β–ˆ | 353/3420 [00:48<04:25, 11.57it/s] Running loglikelihood requests: 11%|β–ˆ | 361/3420 [00:49<04:18, 11.85it/s] Running loglikelihood requests: 11%|β–ˆ | 369/3420 [00:49<04:13, 12.02it/s] Running loglikelihood requests: 11%|β–ˆ | 377/3420 [00:50<04:06, 12.35it/s] Running loglikelihood requests: 11%|β–ˆβ– | 385/3420 [00:51<04:00, 12.62it/s] Running loglikelihood requests: 11%|β–ˆβ– | 393/3420 [00:51<03:56, 12.79it/s] Running loglikelihood requests: 12%|β–ˆβ– | 401/3420 [00:52<03:51, 13.04it/s] Running loglikelihood requests: 12%|β–ˆβ– | 409/3420 [00:52<03:47, 13.24it/s] Running loglikelihood requests: 12%|β–ˆβ– | 417/3420 [00:53<03:44, 13.37it/s] Running loglikelihood requests: 12%|β–ˆβ– | 425/3420 [00:53<03:42, 13.48it/s] Running loglikelihood requests: 13%|β–ˆβ–Ž | 433/3420 [00:54<03:40, 13.57it/s] Running loglikelihood requests: 13%|β–ˆβ–Ž | 441/3420 [00:55<03:38, 13.65it/s] Running loglikelihood requests: 13%|β–ˆβ–Ž | 449/3420 [00:55<03:32, 13.98it/s] Running loglikelihood requests: 13%|β–ˆβ–Ž | 457/3420 [00:56<03:27, 14.26it/s] Running loglikelihood requests: 14%|β–ˆβ–Ž | 465/3420 [00:56<03:24, 14.47it/s] Running loglikelihood requests: 14%|β–ˆβ– | 473/3420 [00:57<03:20, 14.68it/s] Running loglikelihood requests: 14%|β–ˆβ– | 481/3420 [00:57<03:18, 14.80it/s] Running loglikelihood requests: 14%|β–ˆβ– | 489/3420 [00:58<03:16, 14.89it/s] Running loglikelihood requests: 15%|β–ˆβ– | 497/3420 [00:58<03:14, 15.01it/s] Running loglikelihood requests: 15%|β–ˆβ– | 505/3420 [00:59<03:13, 15.10it/s] Running loglikelihood requests: 15%|β–ˆβ–Œ | 513/3420 [00:59<03:11, 15.17it/s] Running loglikelihood requests: 15%|β–ˆβ–Œ | 521/3420 [01:00<03:10, 15.23it/s] Running loglikelihood requests: 15%|β–ˆβ–Œ | 529/3420 [01:00<03:06, 15.49it/s] Running loglikelihood requests: 16%|β–ˆβ–Œ | 537/3420 [01:01<03:03, 15.67it/s] Running loglikelihood requests: 16%|β–ˆβ–Œ | 545/3420 [01:01<03:01, 15.81it/s] Running loglikelihood requests: 16%|β–ˆβ–Œ | 553/3420 [01:02<03:00, 15.91it/s] Running loglikelihood requests: 16%|β–ˆβ–‹ | 561/3420 [01:02<02:58, 16.02it/s] Running loglikelihood requests: 17%|β–ˆβ–‹ | 569/3420 [01:03<02:56, 16.12it/s] Running loglikelihood requests: 17%|β–ˆβ–‹ | 577/3420 [01:03<02:53, 16.35it/s] Running loglikelihood requests: 17%|β–ˆβ–‹ | 585/3420 [01:04<02:51, 16.55it/s] Running loglikelihood requests: 17%|β–ˆβ–‹ | 593/3420 [01:04<02:49, 16.72it/s] Running loglikelihood requests: 18%|β–ˆβ–Š | 601/3420 [01:05<02:47, 16.87it/s] Running loglikelihood requests: 18%|β–ˆβ–Š | 609/3420 [01:05<02:45, 16.98it/s] Running loglikelihood requests: 18%|β–ˆβ–Š | 617/3420 [01:06<02:43, 17.15it/s] Running loglikelihood requests: 18%|β–ˆβ–Š | 625/3420 [01:06<02:41, 17.28it/s] Running loglikelihood requests: 19%|β–ˆβ–Š | 633/3420 [01:07<02:40, 17.37it/s] Running loglikelihood requests: 19%|β–ˆβ–Š | 641/3420 [01:07<02:39, 17.45it/s] Running loglikelihood requests: 19%|β–ˆβ–‰ | 649/3420 [01:07<02:38, 17.51it/s] Running loglikelihood requests: 19%|β–ˆβ–‰ | 657/3420 [01:08<02:37, 17.56it/s] Running loglikelihood requests: 19%|β–ˆβ–‰ | 665/3420 [01:08<02:36, 17.58it/s] Running loglikelihood requests: 20%|β–ˆβ–‰ | 673/3420 [01:09<02:35, 17.62it/s] Running loglikelihood requests: 20%|β–ˆβ–‰ | 681/3420 [01:09<02:35, 17.63it/s] Running loglikelihood requests: 20%|β–ˆβ–ˆ | 689/3420 [01:10<02:34, 17.65it/s] Running loglikelihood requests: 20%|β–ˆβ–ˆ | 697/3420 [01:10<02:34, 17.66it/s] Running loglikelihood requests: 21%|β–ˆβ–ˆ | 705/3420 [01:11<02:33, 17.67it/s] Running loglikelihood requests: 21%|β–ˆβ–ˆ | 713/3420 [01:11<02:33, 17.69it/s] Running loglikelihood requests: 21%|β–ˆβ–ˆ | 721/3420 [01:12<02:28, 18.12it/s] Running loglikelihood requests: 21%|β–ˆβ–ˆβ– | 729/3420 [01:12<02:26, 18.43it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 737/3420 [01:12<02:23, 18.67it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 745/3420 [01:13<02:21, 18.85it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 753/3420 [01:13<02:20, 18.98it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 761/3420 [01:14<02:19, 19.08it/s] Running loglikelihood requests: 22%|β–ˆβ–ˆβ– | 769/3420 [01:14<02:18, 19.15it/s] Running loglikelihood requests: 23%|β–ˆβ–ˆβ–Ž | 777/3420 [01:14<02:17, 19.22it/s] Running loglikelihood requests: 23%|β–ˆβ–ˆβ–Ž | 785/3420 [01:15<02:16, 19.26it/s] Running loglikelihood requests: 23%|β–ˆβ–ˆβ–Ž | 793/3420 [01:15<02:16, 19.29it/s] Running loglikelihood requests: 23%|β–ˆβ–ˆβ–Ž | 801/3420 [01:16<02:15, 19.33it/s] Running loglikelihood requests: 24%|β–ˆβ–ˆβ–Ž | 809/3420 [01:16<02:14, 19.35it/s] Running loglikelihood requests: 24%|β–ˆβ–ˆβ– | 817/3420 [01:16<02:14, 19.38it/s] Running loglikelihood requests: 24%|β–ˆβ–ˆβ– | 825/3420 [01:17<02:13, 19.39it/s] Running loglikelihood requests: 24%|β–ˆβ–ˆβ– | 833/3420 [01:17<02:13, 19.40it/s] Running loglikelihood requests: 25%|β–ˆβ–ˆβ– | 841/3420 [01:18<02:12, 19.43it/s] Running loglikelihood requests: 25%|β–ˆβ–ˆβ– | 849/3420 [01:18<02:12, 19.43it/s] Running loglikelihood requests: 25%|β–ˆβ–ˆβ–Œ | 857/3420 [01:19<02:11, 19.47it/s] Running loglikelihood requests: 25%|β–ˆβ–ˆβ–Œ | 865/3420 [01:19<02:11, 19.48it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–Œ | 873/3420 [01:19<02:10, 19.51it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–Œ | 881/3420 [01:20<02:10, 19.52it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–Œ | 889/3420 [01:20<02:09, 19.54it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–Œ | 897/3420 [01:21<02:08, 19.57it/s] Running loglikelihood requests: 26%|β–ˆβ–ˆβ–‹ | 905/3420 [01:21<02:08, 19.57it/s] Running loglikelihood requests: 27%|β–ˆβ–ˆβ–‹ | 913/3420 [01:21<02:07, 19.61it/s] Running loglikelihood requests: 27%|β–ˆβ–ˆβ–‹ | 921/3420 [01:22<02:07, 19.62it/s] Running loglikelihood requests: 27%|β–ˆβ–ˆβ–‹ | 929/3420 [01:22<02:06, 19.63it/s] Running loglikelihood requests: 27%|β–ˆβ–ˆβ–‹ | 937/3420 [01:23<02:06, 19.65it/s] Running loglikelihood requests: 28%|β–ˆβ–ˆβ–Š | 945/3420 [01:23<02:05, 19.66it/s] Running loglikelihood requests: 28%|β–ˆβ–ˆβ–Š | 953/3420 [01:23<02:05, 19.68it/s] Running loglikelihood requests: 28%|β–ˆβ–ˆβ–Š | 961/3420 [01:24<02:04, 19.68it/s] Running loglikelihood requests: 28%|β–ˆβ–ˆβ–Š | 969/3420 [01:24<02:04, 19.70it/s] Running loglikelihood requests: 29%|β–ˆβ–ˆβ–Š | 977/3420 [01:25<02:03, 19.71it/s] Running loglikelihood requests: 29%|β–ˆβ–ˆβ–‰ | 985/3420 [01:25<02:03, 19.71it/s] Running loglikelihood requests: 29%|β–ˆβ–ˆβ–‰ | 993/3420 [01:25<02:03, 19.72it/s] Running loglikelihood requests: 29%|β–ˆβ–ˆβ–‰ | 1001/3420 [01:26<02:02, 19.73it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–‰ | 1009/3420 [01:26<02:02, 19.74it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–‰ | 1017/3420 [01:27<02:00, 19.87it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–‰ | 1025/3420 [01:27<01:59, 19.99it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–ˆ | 1033/3420 [01:27<01:58, 20.08it/s] Running loglikelihood requests: 30%|β–ˆβ–ˆβ–ˆ | 1041/3420 [01:28<01:58, 20.12it/s] Running loglikelihood requests: 31%|β–ˆβ–ˆβ–ˆ | 1049/3420 [01:28<01:57, 20.14it/s] Running loglikelihood requests: 31%|β–ˆβ–ˆβ–ˆ | 1057/3420 [01:29<01:57, 20.16it/s] Running loglikelihood requests: 31%|β–ˆβ–ˆβ–ˆ | 1065/3420 [01:29<01:56, 20.21it/s] Running loglikelihood requests: 31%|β–ˆβ–ˆβ–ˆβ– | 1073/3420 [01:29<01:56, 20.22it/s] Running loglikelihood requests: 32%|β–ˆβ–ˆβ–ˆβ– | 1081/3420 [01:30<01:55, 20.23it/s] Running loglikelihood requests: 32%|β–ˆβ–ˆβ–ˆβ– | 1089/3420 [01:30<01:55, 20.25it/s] Running loglikelihood requests: 32%|β–ˆβ–ˆβ–ˆβ– | 1097/3420 [01:31<01:54, 20.27it/s] Running loglikelihood requests: 32%|β–ˆβ–ˆβ–ˆβ– | 1105/3420 [01:31<01:54, 20.28it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1113/3420 [01:31<01:53, 20.29it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1121/3420 [01:32<01:53, 20.33it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1129/3420 [01:32<01:52, 20.33it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1137/3420 [01:33<01:52, 20.33it/s] Running loglikelihood requests: 33%|β–ˆβ–ˆβ–ˆβ–Ž | 1145/3420 [01:33<01:51, 20.33it/s] Running loglikelihood requests: 34%|β–ˆβ–ˆβ–ˆβ–Ž | 1153/3420 [01:33<01:51, 20.34it/s] Running loglikelihood requests: 34%|β–ˆβ–ˆβ–ˆβ– | 1161/3420 [01:34<01:51, 20.34it/s] Running loglikelihood requests: 34%|β–ˆβ–ˆβ–ˆβ– | 1169/3420 [01:34<01:50, 20.35it/s] Running loglikelihood requests: 34%|β–ˆβ–ˆβ–ˆβ– | 1177/3420 [01:35<01:50, 20.35it/s] Running loglikelihood requests: 35%|β–ˆβ–ˆβ–ˆβ– | 1185/3420 [01:35<01:49, 20.35it/s] Running loglikelihood requests: 35%|β–ˆβ–ˆβ–ˆβ– | 1193/3420 [01:35<01:49, 20.37it/s] Running loglikelihood requests: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 1201/3420 [01:36<01:48, 20.37it/s] Running loglikelihood requests: 35%|β–ˆβ–ˆβ–ˆβ–Œ | 1209/3420 [01:36<01:48, 20.37it/s] Running loglikelihood requests: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 1217/3420 [01:37<01:48, 20.39it/s] Running loglikelihood requests: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 1225/3420 [01:37<01:47, 20.39it/s] Running loglikelihood requests: 36%|β–ˆβ–ˆβ–ˆβ–Œ | 1233/3420 [01:37<01:47, 20.43it/s] Running loglikelihood requests: 36%|β–ˆβ–ˆβ–ˆβ–‹ | 1241/3420 [01:38<01:46, 20.43it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1249/3420 [01:38<01:46, 20.45it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1257/3420 [01:38<01:45, 20.46it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1265/3420 [01:39<01:45, 20.48it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1273/3420 [01:39<01:44, 20.48it/s] Running loglikelihood requests: 37%|β–ˆβ–ˆβ–ˆβ–‹ | 1281/3420 [01:40<01:44, 20.51it/s] Running loglikelihood requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1289/3420 [01:40<01:43, 20.50it/s] Running loglikelihood requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1297/3420 [01:40<01:43, 20.50it/s] Running loglikelihood requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1305/3420 [01:41<01:43, 20.49it/s] Running loglikelihood requests: 38%|β–ˆβ–ˆβ–ˆβ–Š | 1313/3420 [01:41<01:42, 20.50it/s] Running loglikelihood requests: 39%|β–ˆβ–ˆβ–ˆβ–Š | 1321/3420 [01:42<01:42, 20.50it/s] Running loglikelihood requests: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 1329/3420 [01:42<01:41, 20.51it/s] Running loglikelihood requests: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 1337/3420 [01:42<01:41, 20.54it/s] Running loglikelihood requests: 39%|β–ˆβ–ˆβ–ˆβ–‰ | 1345/3420 [01:43<01:41, 20.54it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 1353/3420 [01:43<01:40, 20.54it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–‰ | 1361/3420 [01:44<01:40, 20.54it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 1369/3420 [01:44<01:39, 20.54it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 1377/3420 [01:44<01:39, 20.58it/s] Running loglikelihood requests: 40%|β–ˆβ–ˆβ–ˆβ–ˆ | 1385/3420 [01:45<01:38, 20.58it/s] Running loglikelihood requests: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 1393/3420 [01:45<01:38, 20.59it/s] Running loglikelihood requests: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 1401/3420 [01:45<01:38, 20.59it/s] Running loglikelihood requests: 41%|β–ˆβ–ˆβ–ˆβ–ˆ | 1409/3420 [01:46<01:37, 20.59it/s] Running loglikelihood requests: 41%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1417/3420 [01:46<01:37, 20.59it/s] Running loglikelihood requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1425/3420 [01:47<01:36, 20.60it/s] Running loglikelihood requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1433/3420 [01:47<01:36, 20.61it/s] Running loglikelihood requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1441/3420 [01:47<01:36, 20.61it/s] Running loglikelihood requests: 42%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1449/3420 [01:48<01:35, 20.63it/s] Running loglikelihood requests: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1457/3420 [01:48<01:35, 20.63it/s] Running loglikelihood requests: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1465/3420 [01:49<01:34, 20.63it/s] Running loglikelihood requests: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1473/3420 [01:49<01:34, 20.65it/s] Running loglikelihood requests: 43%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1481/3420 [01:49<01:33, 20.65it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1489/3420 [01:50<01:33, 20.68it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1497/3420 [01:50<01:32, 20.69it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1505/3420 [01:50<01:31, 20.98it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1513/3420 [01:51<01:29, 21.22it/s] Running loglikelihood requests: 44%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1521/3420 [01:51<01:28, 21.40it/s] Running loglikelihood requests: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1529/3420 [01:52<01:27, 21.53it/s] Running loglikelihood requests: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ– | 1537/3420 [01:52<01:27, 21.63it/s] Running loglikelihood requests: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1545/3420 [01:52<01:26, 21.71it/s] Running loglikelihood requests: 45%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1553/3420 [01:53<01:25, 21.75it/s] Running loglikelihood requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1561/3420 [01:53<01:25, 21.81it/s] Running loglikelihood requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1569/3420 [01:53<01:24, 21.85it/s] Running loglikelihood requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1577/3420 [01:54<01:24, 21.93it/s] Running loglikelihood requests: 46%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1585/3420 [01:54<01:23, 21.97it/s] Running loglikelihood requests: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1593/3420 [01:55<01:23, 21.99it/s] Running loglikelihood requests: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1601/3420 [01:55<01:22, 22.02it/s] Running loglikelihood requests: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1609/3420 [01:55<01:22, 22.04it/s] Running loglikelihood requests: 47%|β–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1617/3420 [01:56<01:21, 22.07it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1625/3420 [01:56<01:21, 22.09it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1633/3420 [01:56<01:20, 22.10it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1641/3420 [01:57<01:20, 22.12it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1649/3420 [01:57<01:19, 22.14it/s] Running loglikelihood requests: 48%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1657/3420 [01:57<01:19, 22.15it/s] Running loglikelihood requests: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1665/3420 [01:58<01:19, 22.17it/s] Running loglikelihood requests: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1673/3420 [01:58<01:18, 22.18it/s] Running loglikelihood requests: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1681/3420 [01:58<01:18, 22.19it/s] Running loglikelihood requests: 49%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1689/3420 [01:59<01:17, 22.23it/s] Running loglikelihood requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1697/3420 [01:59<01:17, 22.23it/s] Running loglikelihood requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 1705/3420 [02:00<01:17, 22.25it/s] Running loglikelihood requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1713/3420 [02:00<01:16, 22.27it/s] Running loglikelihood requests: 50%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1721/3420 [02:00<01:16, 22.28it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1729/3420 [02:01<01:15, 22.30it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1737/3420 [02:01<01:15, 22.32it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 1745/3420 [02:01<01:15, 22.33it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1753/3420 [02:02<01:14, 22.34it/s] Running loglikelihood requests: 51%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1761/3420 [02:02<01:14, 22.33it/s] Running loglikelihood requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1769/3420 [02:02<01:13, 22.34it/s] Running loglikelihood requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1777/3420 [02:03<01:13, 22.34it/s] Running loglikelihood requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1785/3420 [02:03<01:13, 22.35it/s] Running loglikelihood requests: 52%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1793/3420 [02:03<01:12, 22.37it/s] Running loglikelihood requests: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1801/3420 [02:04<01:12, 22.40it/s] Running loglikelihood requests: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1809/3420 [02:04<01:11, 22.42it/s] Running loglikelihood requests: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1817/3420 [02:05<01:11, 22.42it/s] Running loglikelihood requests: 53%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1825/3420 [02:05<01:11, 22.42it/s] Running loglikelihood requests: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 1833/3420 [02:05<01:10, 22.41it/s] Running loglikelihood requests: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1841/3420 [02:06<01:10, 22.42it/s] Running loglikelihood requests: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1849/3420 [02:06<01:10, 22.44it/s] Running loglikelihood requests: 54%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1857/3420 [02:06<01:09, 22.45it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1865/3420 [02:07<01:09, 22.44it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 1873/3420 [02:07<01:08, 22.44it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1881/3420 [02:07<01:06, 23.20it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1889/3420 [02:08<01:04, 23.82it/s] Running loglikelihood requests: 55%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1897/3420 [02:08<01:02, 24.28it/s] Running loglikelihood requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1905/3420 [02:08<01:01, 24.56it/s] Running loglikelihood requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1913/3420 [02:09<01:00, 24.78it/s] Running loglikelihood requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 1921/3420 [02:09<00:59, 24.98it/s] Running loglikelihood requests: 56%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1929/3420 [02:09<00:59, 25.13it/s] Running loglikelihood requests: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1937/3420 [02:10<00:58, 25.23it/s] Running loglikelihood requests: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1945/3420 [02:10<00:58, 25.27it/s] Running loglikelihood requests: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1953/3420 [02:10<00:57, 25.31it/s] Running loglikelihood requests: 57%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 1961/3420 [02:11<00:57, 25.38it/s] Running loglikelihood requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1969/3420 [02:11<00:57, 25.40it/s] Running loglikelihood requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1977/3420 [02:11<00:56, 25.41it/s] Running loglikelihood requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1985/3420 [02:11<00:56, 25.47it/s] Running loglikelihood requests: 58%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 1993/3420 [02:12<00:56, 25.47it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2001/3420 [02:12<00:55, 25.47it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2009/3420 [02:12<00:55, 25.52it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2017/3420 [02:13<00:54, 25.52it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2025/3420 [02:13<00:54, 25.51it/s] Running loglikelihood requests: 59%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2033/3420 [02:13<00:54, 25.56it/s] Running loglikelihood requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2041/3420 [02:14<00:53, 25.56it/s] Running loglikelihood requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2049/3420 [02:14<00:53, 25.59it/s] Running loglikelihood requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2057/3420 [02:14<00:53, 25.63it/s] Running loglikelihood requests: 60%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2065/3420 [02:15<00:52, 25.60it/s] Running loglikelihood requests: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2073/3420 [02:15<00:52, 25.64it/s] Running loglikelihood requests: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2081/3420 [02:15<00:52, 25.65it/s] Running loglikelihood requests: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2089/3420 [02:16<00:51, 25.66it/s] Running loglikelihood requests: 61%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2097/3420 [02:16<00:51, 25.66it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2105/3420 [02:16<00:51, 25.64it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2113/3420 [02:16<00:50, 25.66it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2121/3420 [02:17<00:50, 25.65it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2129/3420 [02:17<00:50, 25.68it/s] Running loglikelihood requests: 62%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2137/3420 [02:17<00:49, 25.73it/s] Running loglikelihood requests: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2145/3420 [02:18<00:49, 25.69it/s] Running loglikelihood requests: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2153/3420 [02:18<00:49, 25.71it/s] Running loglikelihood requests: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2161/3420 [02:18<00:49, 25.69it/s] Running loglikelihood requests: 63%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2169/3420 [02:19<00:48, 25.78it/s] Running loglikelihood requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2177/3420 [02:19<00:48, 25.80it/s] Running loglikelihood requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2185/3420 [02:19<00:47, 25.87it/s] Running loglikelihood requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2193/3420 [02:20<00:47, 25.87it/s] Running loglikelihood requests: 64%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2201/3420 [02:20<00:47, 25.91it/s] Running loglikelihood requests: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2209/3420 [02:20<00:46, 25.91it/s] Running loglikelihood requests: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2217/3420 [02:20<00:46, 25.92it/s] Running loglikelihood requests: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2225/3420 [02:21<00:46, 25.98it/s] Running loglikelihood requests: 65%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2233/3420 [02:21<00:45, 25.97it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2241/3420 [02:21<00:45, 26.02it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2249/3420 [02:22<00:44, 26.06it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2257/3420 [02:22<00:44, 26.05it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2265/3420 [02:22<00:44, 26.07it/s] Running loglikelihood requests: 66%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2273/3420 [02:23<00:43, 26.13it/s] Running loglikelihood requests: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2281/3420 [02:23<00:43, 26.11it/s] Running loglikelihood requests: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2289/3420 [02:23<00:43, 26.11it/s] Running loglikelihood requests: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2297/3420 [02:24<00:42, 26.16it/s] Running loglikelihood requests: 67%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2305/3420 [02:24<00:42, 26.15it/s] Running loglikelihood requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2313/3420 [02:24<00:42, 26.14it/s] Running loglikelihood requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2321/3420 [02:24<00:42, 26.16it/s] Running loglikelihood requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2329/3420 [02:25<00:41, 26.17it/s] Running loglikelihood requests: 68%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2337/3420 [02:25<00:41, 26.17it/s] Running loglikelihood requests: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2345/3420 [02:25<00:41, 26.18it/s] Running loglikelihood requests: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2353/3420 [02:26<00:40, 26.21it/s] Running loglikelihood requests: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2361/3420 [02:26<00:40, 26.22it/s] Running loglikelihood requests: 69%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2369/3420 [02:26<00:40, 26.22it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2377/3420 [02:27<00:38, 26.92it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2385/3420 [02:27<00:37, 27.43it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2393/3420 [02:27<00:36, 27.81it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2401/3420 [02:27<00:36, 28.08it/s] Running loglikelihood requests: 70%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2409/3420 [02:28<00:35, 28.27it/s] Running loglikelihood requests: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2417/3420 [02:28<00:35, 28.42it/s] Running loglikelihood requests: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2425/3420 [02:28<00:34, 28.52it/s] Running loglikelihood requests: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2433/3420 [02:29<00:34, 28.65it/s] Running loglikelihood requests: 71%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2441/3420 [02:29<00:34, 28.70it/s] Running loglikelihood requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2449/3420 [02:29<00:33, 28.75it/s] Running loglikelihood requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2457/3420 [02:29<00:33, 28.79it/s] Running loglikelihood requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2465/3420 [02:30<00:33, 28.82it/s] Running loglikelihood requests: 72%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2473/3420 [02:30<00:32, 28.84it/s] Running loglikelihood requests: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2481/3420 [02:30<00:32, 28.86it/s] Running loglikelihood requests: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2489/3420 [02:30<00:32, 28.93it/s] Running loglikelihood requests: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2501/3420 [02:31<00:27, 33.25it/s] Running loglikelihood requests: 73%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2509/3420 [02:31<00:28, 31.97it/s] Running loglikelihood requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2517/3420 [02:31<00:29, 31.07it/s] Running loglikelihood requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2525/3420 [02:32<00:29, 30.44it/s] Running loglikelihood requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2533/3420 [02:32<00:29, 30.01it/s] Running loglikelihood requests: 74%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2541/3420 [02:32<00:29, 29.69it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2549/3420 [02:32<00:29, 29.50it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2557/3420 [02:33<00:29, 29.37it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2565/3420 [02:33<00:29, 29.27it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2573/3420 [02:33<00:28, 29.21it/s] Running loglikelihood requests: 75%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2581/3420 [02:33<00:28, 29.28it/s] Running loglikelihood requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2589/3420 [02:34<00:28, 29.32it/s] Running loglikelihood requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2597/3420 [02:34<00:28, 29.39it/s] Running loglikelihood requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2605/3420 [02:34<00:27, 29.42it/s] Running loglikelihood requests: 76%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2613/3420 [02:35<00:27, 29.45it/s] Running loglikelihood requests: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2621/3420 [02:35<00:27, 29.46it/s] Running loglikelihood requests: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2629/3420 [02:35<00:26, 29.50it/s] Running loglikelihood requests: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2637/3420 [02:35<00:26, 29.49it/s] Running loglikelihood requests: 77%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2645/3420 [02:36<00:26, 29.50it/s] Running loglikelihood requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2653/3420 [02:36<00:25, 29.58it/s] Running loglikelihood requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2661/3420 [02:36<00:25, 29.55it/s] Running loglikelihood requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2669/3420 [02:36<00:25, 29.54it/s] Running loglikelihood requests: 78%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2677/3420 [02:37<00:25, 29.53it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2685/3420 [02:37<00:24, 29.57it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2693/3420 [02:37<00:24, 29.56it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2701/3420 [02:38<00:24, 29.58it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2709/3420 [02:38<00:24, 29.57it/s] Running loglikelihood requests: 79%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2717/3420 [02:38<00:23, 29.57it/s] Running loglikelihood requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2725/3420 [02:38<00:23, 29.57it/s] Running loglikelihood requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 2733/3420 [02:39<00:23, 29.57it/s] Running loglikelihood requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2741/3420 [02:39<00:22, 29.64it/s] Running loglikelihood requests: 80%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2749/3420 [02:39<00:22, 29.64it/s] Running loglikelihood requests: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2757/3420 [02:39<00:22, 29.69it/s] Running loglikelihood requests: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2765/3420 [02:40<00:22, 29.68it/s] Running loglikelihood requests: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 2773/3420 [02:40<00:21, 29.69it/s] Running loglikelihood requests: 81%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2781/3420 [02:40<00:21, 29.74it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2789/3420 [02:41<00:21, 29.72it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2797/3420 [02:41<00:20, 29.77it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2805/3420 [02:41<00:20, 29.76it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2813/3420 [02:41<00:20, 29.81it/s] Running loglikelihood requests: 82%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2821/3420 [02:42<00:20, 29.79it/s] Running loglikelihood requests: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2829/3420 [02:42<00:19, 29.83it/s] Running loglikelihood requests: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2837/3420 [02:42<00:19, 29.81it/s] Running loglikelihood requests: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2845/3420 [02:42<00:19, 29.80it/s] Running loglikelihood requests: 83%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2853/3420 [02:43<00:19, 29.81it/s] Running loglikelihood requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž | 2861/3420 [02:43<00:18, 29.81it/s] Running loglikelihood requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2869/3420 [02:43<00:18, 29.86it/s] Running loglikelihood requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2877/3420 [02:43<00:18, 29.90it/s] Running loglikelihood requests: 84%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2885/3420 [02:44<00:17, 29.88it/s] Running loglikelihood requests: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2893/3420 [02:44<00:17, 29.91it/s] Running loglikelihood requests: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ– | 2901/3420 [02:44<00:17, 29.93it/s] Running loglikelihood requests: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2909/3420 [02:45<00:17, 29.93it/s] Running loglikelihood requests: 85%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2917/3420 [02:45<00:16, 29.96it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2925/3420 [02:45<00:16, 30.52it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2933/3420 [02:45<00:15, 30.94it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2941/3420 [02:46<00:15, 31.21it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ | 2949/3420 [02:46<00:14, 31.46it/s] Running loglikelihood requests: 86%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2957/3420 [02:46<00:14, 31.65it/s] Running loglikelihood requests: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2965/3420 [02:46<00:14, 31.81it/s] Running loglikelihood requests: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2973/3420 [02:47<00:14, 31.92it/s] Running loglikelihood requests: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2981/3420 [02:47<00:13, 32.07it/s] Running loglikelihood requests: 87%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹ | 2989/3420 [02:47<00:13, 32.13it/s] Running loglikelihood requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 2997/3420 [02:47<00:13, 32.16it/s] Running loglikelihood requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3005/3420 [02:48<00:12, 32.20it/s] Running loglikelihood requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3013/3420 [02:48<00:12, 32.28it/s] Running loglikelihood requests: 88%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3021/3420 [02:48<00:12, 32.29it/s] Running loglikelihood requests: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š | 3029/3420 [02:48<00:12, 32.31it/s] Running loglikelihood requests: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3037/3420 [02:49<00:11, 32.37it/s] Running loglikelihood requests: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3045/3420 [02:49<00:11, 32.37it/s] Running loglikelihood requests: 89%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3053/3420 [02:49<00:11, 32.39it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3061/3420 [02:49<00:11, 32.39it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3069/3420 [02:50<00:10, 32.47it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰ | 3077/3420 [02:50<00:10, 32.47it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3085/3420 [02:50<00:10, 32.52it/s] Running loglikelihood requests: 90%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3093/3420 [02:50<00:10, 32.50it/s] Running loglikelihood requests: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3101/3420 [02:50<00:09, 32.50it/s] Running loglikelihood requests: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3109/3420 [02:51<00:09, 32.55it/s] Running loglikelihood requests: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ | 3117/3420 [02:51<00:09, 32.54it/s] Running loglikelihood requests: 91%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3125/3420 [02:51<00:09, 32.61it/s] Running loglikelihood requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3133/3420 [02:51<00:08, 32.60it/s] Running loglikelihood requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3141/3420 [02:52<00:08, 32.66it/s] Running loglikelihood requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3149/3420 [02:52<00:08, 32.64it/s] Running loglikelihood requests: 92%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3157/3420 [02:52<00:08, 32.71it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3165/3420 [02:52<00:07, 32.70it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3173/3420 [02:53<00:07, 32.77it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3181/3420 [02:53<00:07, 32.84it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3189/3420 [02:53<00:07, 32.81it/s] Running loglikelihood requests: 93%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3197/3420 [02:53<00:06, 32.88it/s] Running loglikelihood requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Ž| 3205/3420 [02:54<00:06, 32.93it/s] Running loglikelihood requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3213/3420 [02:54<00:06, 32.92it/s] Running loglikelihood requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3221/3420 [02:54<00:06, 32.96it/s] Running loglikelihood requests: 94%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3229/3420 [02:54<00:05, 33.02it/s] Running loglikelihood requests: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3237/3420 [02:55<00:05, 33.10it/s] Running loglikelihood requests: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–| 3245/3420 [02:55<00:05, 33.14it/s] Running loglikelihood requests: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3253/3420 [02:55<00:05, 33.17it/s] Running loglikelihood requests: 95%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3261/3420 [02:55<00:04, 33.25it/s] Running loglikelihood requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3269/3420 [02:56<00:04, 33.30it/s] Running loglikelihood requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3277/3420 [02:56<00:04, 33.34it/s] Running loglikelihood requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Œ| 3285/3420 [02:56<00:04, 33.36it/s] Running loglikelihood requests: 96%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3293/3420 [02:56<00:03, 33.40it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3301/3420 [02:57<00:03, 33.46it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3309/3420 [02:57<00:03, 33.53it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3317/3420 [02:57<00:03, 33.53it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3325/3420 [02:57<00:02, 33.55it/s] Running loglikelihood requests: 97%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‹| 3333/3420 [02:58<00:02, 33.59it/s] Running loglikelihood requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3341/3420 [02:58<00:02, 35.18it/s] Running loglikelihood requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3349/3420 [02:58<00:01, 36.40it/s] Running loglikelihood requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3357/3420 [02:58<00:01, 37.37it/s] Running loglikelihood requests: 98%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3365/3420 [02:58<00:01, 38.08it/s] Running loglikelihood requests: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–Š| 3373/3420 [02:59<00:01, 38.61it/s] Running loglikelihood requests: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3381/3420 [02:59<00:01, 39.00it/s] Running loglikelihood requests: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3389/3420 [02:59<00:00, 39.29it/s] Running loglikelihood requests: 99%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3397/3420 [02:59<00:00, 39.51it/s] Running loglikelihood requests: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3405/3420 [02:59<00:00, 39.71it/s] Running loglikelihood requests: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–‰| 3413/3420 [03:00<00:00, 39.90it/s] Running loglikelihood requests: 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 3420/3420 [03:00<00:00, 19.00it/s]
fatal: detected dubious ownership in repository at '/root/steering-resistance'
To add an exception for this directory, call:
git config --global --add safe.directory /root/steering-resistance
2026-07-24:07:33:33 INFO [loggers.evaluation_tracker:247] Saving results aggregated
hf ({'pretrained': 'Qwen/Qwen2.5-7B-Instruct', 'dtype': 'bfloat16', 'peft': '/root/steering-resistance/results/full_7b/m1_resist_adapter'}), gen_kwargs: ({}), limit: 15.0, num_fewshot: 5, batch_size: 2
| Tasks |Version|Filter|n-shot|Metric| |Value | |Stderr|
|---------------------------------------|------:|------|-----:|------|---|-----:|---|-----:|
|mmlu | 2|none | |acc | |0.7649|Β± |0.0140|
| - humanities | 2|none | 5|acc |↑ |0.8000|Β± |0.0276|
| - formal_logic | 1|none | 5|acc |↑ |0.6000|Β± |0.1309|
| - high_school_european_history | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - high_school_us_history | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - high_school_world_history | 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - international_law | 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - jurisprudence | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - logical_fallacies | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - moral_disputes | 1|none | 5|acc |↑ |0.6000|Β± |0.1309|
| - moral_scenarios | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - philosophy | 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - prehistory | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - professional_law | 1|none | 5|acc |↑ |0.6000|Β± |0.1309|
| - world_religions | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - other | 2|none | 5|acc |↑ |0.7487|Β± |0.0303|
| - business_ethics | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - clinical_knowledge | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - college_medicine | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - global_facts | 1|none | 5|acc |↑ |0.4667|Β± |0.1333|
| - human_aging | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - management | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - marketing | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - medical_genetics | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - miscellaneous | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - nutrition | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - professional_accounting | 1|none | 5|acc |↑ |0.4667|Β± |0.1333|
| - professional_medicine | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - virology | 1|none | 5|acc |↑ |0.6000|Β± |0.1309|
| - social sciences | 2|none | 5|acc |↑ |0.8278|Β± |0.0279|
| - econometrics | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - high_school_geography | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - high_school_government_and_politics| 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - high_school_macroeconomics | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - high_school_microeconomics | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - high_school_psychology | 1|none | 5|acc |↑ |1.0000|Β± |0.0000|
| - human_sexuality | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - professional_psychology | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - public_relations | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - security_studies | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - sociology | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - us_foreign_policy | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - stem | 2|none | 5|acc |↑ |0.7123|Β± |0.0259|
| - abstract_algebra | 1|none | 5|acc |↑ |0.5333|Β± |0.1333|
| - anatomy | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - astronomy | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - college_biology | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - college_chemistry | 1|none | 5|acc |↑ |0.7333|Β± |0.1182|
| - college_computer_science | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - college_mathematics | 1|none | 5|acc |↑ |0.4000|Β± |0.1309|
| - college_physics | 1|none | 5|acc |↑ |0.4667|Β± |0.1333|
| - computer_security | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - conceptual_physics | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - electrical_engineering | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - elementary_mathematics | 1|none | 5|acc |↑ |0.8000|Β± |0.1069|
| - high_school_biology | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - high_school_chemistry | 1|none | 5|acc |↑ |0.5333|Β± |0.1333|
| - high_school_computer_science | 1|none | 5|acc |↑ |0.9333|Β± |0.0667|
| - high_school_mathematics | 1|none | 5|acc |↑ |0.5333|Β± |0.1333|
| - high_school_physics | 1|none | 5|acc |↑ |0.6667|Β± |0.1260|
| - high_school_statistics | 1|none | 5|acc |↑ |0.8667|Β± |0.0909|
| - machine_learning | 1|none | 5|acc |↑ |0.6000|Β± |0.1309|
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|------------------|------:|------|-----:|------|---|-----:|---|-----:|
|mmlu | 2|none | |acc | |0.7649|Β± |0.0140|
| - humanities | 2|none | 5|acc |↑ |0.8000|Β± |0.0276|
| - other | 2|none | 5|acc |↑ |0.7487|Β± |0.0303|
| - social sciences| 2|none | 5|acc |↑ |0.8278|Β± |0.0279|
| - stem | 2|none | 5|acc |↑ |0.7123|Β± |0.0259|