2025-08-16 23:59:37 - eval_util - INFO - Using naive RTN quantization.
2025-08-16 23:59:37 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
2025-08-17 00:02:39 - accelerate.utils.modeling - INFO - We will use 90% of the memory on device 0 for storing the model, and 10% for the buffer to avoid OOM. You can set `max_memory` in to a higher value to use more memory (at your own risk).
2025-08-17 00:02:40 - eval_util - INFO - Evaluating wikitext...
2025-08-17 00:03:45 - eval_util - INFO - Wikitext completed: 5.726220607757568
2025-08-17 00:03:45 - eval_util - INFO - Starting evaluation of task: boolq
2025-08-17 00:03:45 - eval_util - INFO - GPU Memory before boolq: 3.81GB
2025-08-17 00:03:45 - lm_eval.models.huggingface - WARNING - `pretrained` model kwarg is not of type `str`. Many other model arguments may be ignored. Please do not launch via accelerate or use `parallelize=True` if passing an existing model this way.
2025-08-17 00:03:45 - lm_eval.models.huggingface - WARNING - Passed an already-initialized model through `pretrained`, assuming single-process call to evaluate() or custom distributed integration
2025-08-17 00:03:45 - lm_eval.evaluator - INFO - Setting random seed to 0 | Setting numpy seed to 1234 | Setting torch manual seed to 1234 | Setting fewshot manual seed to 1234
2025-08-17 00:03:45 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 00:04:00 - lm_eval.api.task - WARNING - [Task: boolq] metric acc is defined, but aggregation is not. using default aggregation=mean
2025-08-17 00:04:00 - lm_eval.api.task - WARNING - [Task: boolq] metric acc is defined, but higher_is_better is not. using default higher_is_better=True
2025-08-17 00:04:04 - lm_eval.api.task - INFO - Building contexts for boolq on rank 0...
2025-08-17 00:04:05 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 00:07:24 - eval_util - INFO - ✅ Task boolq completed successfully
2025-08-17 00:07:24 - eval_util - INFO - GPU Memory after boolq: 3.81GB
2025-08-17 00:07:24 - eval_util - INFO - Starting evaluation of task: hellaswag
2025-08-17 00:07:24 - eval_util - INFO - GPU Memory before hellaswag: 3.81GB
2025-08-17 00:07:24 - lm_eval.models.huggingface - WARNING - `pretrained` model kwarg is not of type `str`. Many other model arguments may be ignored. Please do not launch via accelerate or use `parallelize=True` if passing an existing model this way.
2025-08-17 00:07:24 - lm_eval.models.huggingface - WARNING - Passed an already-initialized model through `pretrained`, assuming single-process call to evaluate() or custom distributed integration
2025-08-17 00:07:24 - lm_eval.evaluator - INFO - Setting random seed to 0 | Setting numpy seed to 1234 | Setting torch manual seed to 1234 | Setting fewshot manual seed to 1234
2025-08-17 00:07:24 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 00:07:43 - lm_eval.api.task - INFO - Building contexts for hellaswag on rank 0...
2025-08-17 00:07:46 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 00:37:46 - eval_util - INFO - ✅ Task hellaswag completed successfully
2025-08-17 00:37:46 - eval_util - INFO - GPU Memory after hellaswag: 3.81GB
2025-08-17 00:37:46 - eval_util - INFO - Starting evaluation of task: winogrande
2025-08-17 00:37:46 - eval_util - INFO - GPU Memory before winogrande: 3.81GB
2025-08-17 00:37:46 - lm_eval.models.huggingface - WARNING - `pretrained` model kwarg is not of type `str`. Many other model arguments may be ignored. Please do not launch via accelerate or use `parallelize=True` if passing an existing model this way.
2025-08-17 00:37:46 - lm_eval.models.huggingface - WARNING - Passed an already-initialized model through `pretrained`, assuming single-process call to evaluate() or custom distributed integration
2025-08-17 00:37:46 - lm_eval.evaluator - INFO - Setting random seed to 0 | Setting numpy seed to 1234 | Setting torch manual seed to 1234 | Setting fewshot manual seed to 1234
2025-08-17 00:37:46 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 00:38:02 - lm_eval.api.task - INFO - Building contexts for winogrande on rank 0...
2025-08-17 00:38:03 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 00:39:38 - eval_util - INFO - ✅ Task winogrande completed successfully
2025-08-17 00:39:38 - eval_util - INFO - GPU Memory after winogrande: 3.81GB
2025-08-17 00:39:38 - eval_util - INFO - Starting evaluation of task: arc_easy
2025-08-17 00:39:38 - eval_util - INFO - GPU Memory before arc_easy: 3.81GB
2025-08-17 00:39:38 - lm_eval.models.huggingface - WARNING - `pretrained` model kwarg is not of type `str`. Many other model arguments may be ignored. Please do not launch via accelerate or use `parallelize=True` if passing an existing model this way.
2025-08-17 00:39:38 - lm_eval.models.huggingface - WARNING - Passed an already-initialized model through `pretrained`, assuming single-process call to evaluate() or custom distributed integration
2025-08-17 00:39:38 - lm_eval.evaluator - INFO - Setting random seed to 0 | Setting numpy seed to 1234 | Setting torch manual seed to 1234 | Setting fewshot manual seed to 1234
2025-08-17 00:39:38 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 00:39:53 - lm_eval.api.task - INFO - Building contexts for arc_easy on rank 0...
2025-08-17 00:39:55 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 00:45:37 - eval_util - INFO - ✅ Task arc_easy completed successfully
2025-08-17 00:45:37 - eval_util - INFO - GPU Memory after arc_easy: 3.81GB
2025-08-17 00:45:37 - eval_util - INFO - Starting evaluation of task: arc_challenge
2025-08-17 00:45:37 - eval_util - INFO - GPU Memory before arc_challenge: 3.81GB
2025-08-17 00:45:37 - lm_eval.models.huggingface - WARNING - `pretrained` model kwarg is not of type `str`. Many other model arguments may be ignored. Please do not launch via accelerate or use `parallelize=True` if passing an existing model this way.
2025-08-17 00:45:37 - lm_eval.models.huggingface - WARNING - Passed an already-initialized model through `pretrained`, assuming single-process call to evaluate() or custom distributed integration
2025-08-17 00:45:37 - lm_eval.evaluator - INFO - Setting random seed to 0 | Setting numpy seed to 1234 | Setting torch manual seed to 1234 | Setting fewshot manual seed to 1234
2025-08-17 00:45:37 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 00:45:51 - lm_eval.api.task - INFO - Building contexts for arc_challenge on rank 0...
2025-08-17 00:45:52 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 00:48:51 - eval_util - INFO - ✅ Task arc_challenge completed successfully
2025-08-17 00:48:51 - eval_util - INFO - GPU Memory after arc_challenge: 3.81GB
2025-08-17 00:48:51 - eval_util - INFO - Evaluation Summary - Successful: 6, Failed: 0
2025-08-17 00:48:51 - eval_util - INFO - result:
2025-08-17 00:48:51 - eval_util - INFO - 	{'arc_challenge': {'acc,none': 0.4308873720136519,
2025-08-17 00:48:51 - eval_util - INFO - 	                   'acc_norm,none': 0.45051194539249145,
2025-08-17 00:48:51 - eval_util - INFO - 	                   'acc_norm_stderr,none': 0.014539646098471627,
2025-08-17 00:48:51 - eval_util - INFO - 	                   'acc_stderr,none': 0.014471133392642471,
2025-08-17 00:48:51 - eval_util - INFO - 	                   'alias': 'arc_challenge'},
2025-08-17 00:48:51 - eval_util - INFO - 	 'arc_easy': {'acc,none': 0.7622053872053872,
2025-08-17 00:48:51 - eval_util - INFO - 	              'acc_norm,none': 0.742003367003367,
2025-08-17 00:48:51 - eval_util - INFO - 	              'acc_norm_stderr,none': 0.008977970005203404,
2025-08-17 00:48:51 - eval_util - INFO - 	              'acc_stderr,none': 0.008735850753507987,
2025-08-17 00:48:51 - eval_util - INFO - 	              'alias': 'arc_easy'},
2025-08-17 00:48:51 - eval_util - INFO - 	 'boolq': {'acc,none': 0.7746177370030581,
2025-08-17 00:48:51 - eval_util - INFO - 	           'acc_stderr,none': 0.007307959399119174,
2025-08-17 00:48:51 - eval_util - INFO - 	           'alias': 'boolq'},
2025-08-17 00:48:51 - eval_util - INFO - 	 'hellaswag': {'acc,none': 0.5689105755825533,
2025-08-17 00:48:51 - eval_util - INFO - 	               'acc_norm,none': 0.7532364070902211,
2025-08-17 00:48:51 - eval_util - INFO - 	               'acc_norm_stderr,none': 0.004302468610485086,
2025-08-17 00:48:51 - eval_util - INFO - 	               'acc_stderr,none': 0.004942164585991471,
2025-08-17 00:48:51 - eval_util - INFO - 	               'alias': 'hellaswag'},
2025-08-17 00:48:51 - eval_util - INFO - 	 'wikitext': 5.726220607757568,
2025-08-17 00:48:51 - eval_util - INFO - 	 'winogrande': {'acc,none': 0.6874506708760852,
2025-08-17 00:48:51 - eval_util - INFO - 	                'acc_stderr,none': 0.013027563620748838,
2025-08-17 00:48:51 - eval_util - INFO - 	                'alias': 'winogrande'}}
