2025-08-17 00:49:01 - eval_util - INFO - Using naive RTN quantization.
2025-08-17 00:49:02 - 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 00:52:03 - 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 00:52:03 - eval_util - INFO - Evaluating wikitext...
2025-08-17 00:53:09 - eval_util - INFO - Wikitext completed: 6.401055335998535
2025-08-17 00:53:09 - eval_util - INFO - Starting evaluation of task: boolq
2025-08-17 00:53:09 - eval_util - INFO - GPU Memory before boolq: 3.81GB
2025-08-17 00:53:09 - 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 00:53:09 - 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 00:53:09 - 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 00:53:09 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 00:53:24 - lm_eval.api.task - WARNING - [Task: boolq] metric acc is defined, but aggregation is not. using default aggregation=mean
2025-08-17 00:53:24 - 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 00:53:28 - lm_eval.api.task - INFO - Building contexts for boolq on rank 0...
2025-08-17 00:53:29 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 00:56:47 - eval_util - INFO - ✅ Task boolq completed successfully
2025-08-17 00:56:47 - eval_util - INFO - GPU Memory after boolq: 3.81GB
2025-08-17 00:56:47 - eval_util - INFO - Starting evaluation of task: hellaswag
2025-08-17 00:56:47 - eval_util - INFO - GPU Memory before hellaswag: 3.81GB
2025-08-17 00:56:47 - 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 00:56:47 - 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 00:56:47 - 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 00:56:47 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 00:57:06 - lm_eval.api.task - INFO - Building contexts for hellaswag on rank 0...
2025-08-17 00:57:09 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 01:27:23 - eval_util - INFO - ✅ Task hellaswag completed successfully
2025-08-17 01:27:23 - eval_util - INFO - GPU Memory after hellaswag: 3.81GB
2025-08-17 01:27:23 - eval_util - INFO - Starting evaluation of task: winogrande
2025-08-17 01:27:23 - eval_util - INFO - GPU Memory before winogrande: 3.81GB
2025-08-17 01:27:23 - 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 01:27:23 - 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 01:27:23 - 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 01:27:23 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 01:27:40 - lm_eval.api.task - INFO - Building contexts for winogrande on rank 0...
2025-08-17 01:27:40 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 01:29:16 - eval_util - INFO - ✅ Task winogrande completed successfully
2025-08-17 01:29:16 - eval_util - INFO - GPU Memory after winogrande: 3.81GB
2025-08-17 01:29:16 - eval_util - INFO - Starting evaluation of task: arc_easy
2025-08-17 01:29:16 - eval_util - INFO - GPU Memory before arc_easy: 3.81GB
2025-08-17 01:29:16 - 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 01:29:16 - 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 01:29:16 - 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 01:29:16 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 01:29:30 - lm_eval.api.task - INFO - Building contexts for arc_easy on rank 0...
2025-08-17 01:29:32 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 01:35:17 - eval_util - INFO - ✅ Task arc_easy completed successfully
2025-08-17 01:35:17 - eval_util - INFO - GPU Memory after arc_easy: 3.81GB
2025-08-17 01:35:17 - eval_util - INFO - Starting evaluation of task: arc_challenge
2025-08-17 01:35:17 - eval_util - INFO - GPU Memory before arc_challenge: 3.81GB
2025-08-17 01:35:17 - 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 01:35:17 - 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 01:35:17 - 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 01:35:17 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 01:35:31 - lm_eval.api.task - INFO - Building contexts for arc_challenge on rank 0...
2025-08-17 01:35:32 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 01:38:32 - eval_util - INFO - ✅ Task arc_challenge completed successfully
2025-08-17 01:38:32 - eval_util - INFO - GPU Memory after arc_challenge: 3.81GB
2025-08-17 01:38:32 - eval_util - INFO - Evaluation Summary - Successful: 6, Failed: 0
2025-08-17 01:38:32 - eval_util - INFO - result:
2025-08-17 01:38:32 - eval_util - INFO - 	{'arc_challenge': {'acc,none': 0.42235494880546076,
2025-08-17 01:38:32 - eval_util - INFO - 	                   'acc_norm,none': 0.44197952218430037,
2025-08-17 01:38:32 - eval_util - INFO - 	                   'acc_norm_stderr,none': 0.014512682523128342,
2025-08-17 01:38:32 - eval_util - INFO - 	                   'acc_stderr,none': 0.01443413871337999,
2025-08-17 01:38:32 - eval_util - INFO - 	                   'alias': 'arc_challenge'},
2025-08-17 01:38:32 - eval_util - INFO - 	 'arc_easy': {'acc,none': 0.7550505050505051,
2025-08-17 01:38:32 - eval_util - INFO - 	              'acc_norm,none': 0.7255892255892256,
2025-08-17 01:38:32 - eval_util - INFO - 	              'acc_norm_stderr,none': 0.009156177122244527,
2025-08-17 01:38:32 - eval_util - INFO - 	              'acc_stderr,none': 0.008824588611219092,
2025-08-17 01:38:32 - eval_util - INFO - 	              'alias': 'arc_easy'},
2025-08-17 01:38:32 - eval_util - INFO - 	 'boolq': {'acc,none': 0.745565749235474,
2025-08-17 01:38:32 - eval_util - INFO - 	           'acc_stderr,none': 0.007617690099234393,
2025-08-17 01:38:32 - eval_util - INFO - 	           'alias': 'boolq'},
2025-08-17 01:38:32 - eval_util - INFO - 	 'hellaswag': {'acc,none': 0.5658235411272655,
2025-08-17 01:38:32 - eval_util - INFO - 	               'acc_norm,none': 0.758514240191197,
2025-08-17 01:38:32 - eval_util - INFO - 	               'acc_norm_stderr,none': 0.004271094187098005,
2025-08-17 01:38:32 - eval_util - INFO - 	               'acc_stderr,none': 0.004946353590937027,
2025-08-17 01:38:32 - eval_util - INFO - 	               'alias': 'hellaswag'},
2025-08-17 01:38:32 - eval_util - INFO - 	 'wikitext': 6.401055335998535,
2025-08-17 01:38:32 - eval_util - INFO - 	 'winogrande': {'acc,none': 0.6803472770323599,
2025-08-17 01:38:32 - eval_util - INFO - 	                'acc_stderr,none': 0.01310652851766514,
2025-08-17 01:38:32 - eval_util - INFO - 	                'alias': 'winogrande'}}
