2025-08-17 07:23:02 - eval_util - INFO - Using AWQ quantization.
2025-08-17 07:23:03 - 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 07:26:20 - 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 07:26:21 - eval_util - INFO - Evaluating wikitext...
2025-08-17 07:27:26 - eval_util - INFO - Wikitext completed: 5.722687721252441
2025-08-17 07:27:26 - eval_util - INFO - Starting evaluation of task: boolq
2025-08-17 07:27:26 - eval_util - INFO - GPU Memory before boolq: 3.81GB
2025-08-17 07:27:26 - 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 07:27:26 - 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 07:27:26 - 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 07:27:26 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 07:27:41 - lm_eval.api.task - WARNING - [Task: boolq] metric acc is defined, but aggregation is not. using default aggregation=mean
2025-08-17 07:27:41 - 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 07:27:45 - lm_eval.api.task - INFO - Building contexts for boolq on rank 0...
2025-08-17 07:27:46 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 07:31:02 - eval_util - INFO - ✅ Task boolq completed successfully
2025-08-17 07:31:02 - eval_util - INFO - GPU Memory after boolq: 3.81GB
2025-08-17 07:31:02 - eval_util - INFO - Starting evaluation of task: hellaswag
2025-08-17 07:31:02 - eval_util - INFO - GPU Memory before hellaswag: 3.81GB
2025-08-17 07:31:02 - 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 07:31:02 - 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 07:31:02 - 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 07:31:02 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 07:31:20 - lm_eval.api.task - INFO - Building contexts for hellaswag on rank 0...
2025-08-17 07:31:24 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 08:01:28 - eval_util - INFO - ✅ Task hellaswag completed successfully
2025-08-17 08:01:28 - eval_util - INFO - GPU Memory after hellaswag: 3.81GB
2025-08-17 08:01:28 - eval_util - INFO - Starting evaluation of task: winogrande
2025-08-17 08:01:28 - eval_util - INFO - GPU Memory before winogrande: 3.81GB
2025-08-17 08:01:28 - 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 08:01:28 - 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 08:01:28 - 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 08:01:28 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 08:01:45 - lm_eval.api.task - INFO - Building contexts for winogrande on rank 0...
2025-08-17 08:01:45 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 08:03:21 - eval_util - INFO - ✅ Task winogrande completed successfully
2025-08-17 08:03:21 - eval_util - INFO - GPU Memory after winogrande: 3.81GB
2025-08-17 08:03:21 - eval_util - INFO - Starting evaluation of task: arc_easy
2025-08-17 08:03:21 - eval_util - INFO - GPU Memory before arc_easy: 3.81GB
2025-08-17 08:03:21 - 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 08:03:21 - 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 08:03:21 - 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 08:03:21 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 08:03:35 - lm_eval.api.task - INFO - Building contexts for arc_easy on rank 0...
2025-08-17 08:03:37 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 08:09:20 - eval_util - INFO - ✅ Task arc_easy completed successfully
2025-08-17 08:09:20 - eval_util - INFO - GPU Memory after arc_easy: 3.81GB
2025-08-17 08:09:20 - eval_util - INFO - Starting evaluation of task: arc_challenge
2025-08-17 08:09:20 - eval_util - INFO - GPU Memory before arc_challenge: 3.81GB
2025-08-17 08:09:20 - 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 08:09:20 - 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 08:09:20 - 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 08:09:20 - lm_eval.evaluator - INFO - Using pre-initialized model
2025-08-17 08:09:33 - lm_eval.api.task - INFO - Building contexts for arc_challenge on rank 0...
2025-08-17 08:09:34 - lm_eval.evaluator - INFO - Running loglikelihood requests
2025-08-17 08:12:33 - eval_util - INFO - ✅ Task arc_challenge completed successfully
2025-08-17 08:12:33 - eval_util - INFO - GPU Memory after arc_challenge: 3.81GB
2025-08-17 08:12:33 - eval_util - INFO - Evaluation Summary - Successful: 6, Failed: 0
2025-08-17 08:12:33 - eval_util - INFO - result:
2025-08-17 08:12:33 - eval_util - INFO - 	{'arc_challenge': {'acc,none': 0.4129692832764505,
2025-08-17 08:12:33 - eval_util - INFO - 	                   'acc_norm,none': 0.43686006825938567,
2025-08-17 08:12:33 - eval_util - INFO - 	                   'acc_norm_stderr,none': 0.014494421584256527,
2025-08-17 08:12:33 - eval_util - INFO - 	                   'acc_stderr,none': 0.014388344935398324,
2025-08-17 08:12:33 - eval_util - INFO - 	                   'alias': 'arc_challenge'},
2025-08-17 08:12:33 - eval_util - INFO - 	 'arc_easy': {'acc,none': 0.7466329966329966,
2025-08-17 08:12:33 - eval_util - INFO - 	              'acc_norm,none': 0.7289562289562289,
2025-08-17 08:12:33 - eval_util - INFO - 	              'acc_norm_stderr,none': 0.009120919741760597,
2025-08-17 08:12:33 - eval_util - INFO - 	              'acc_stderr,none': 0.008924765424529264,
2025-08-17 08:12:33 - eval_util - INFO - 	              'alias': 'arc_easy'},
2025-08-17 08:12:33 - eval_util - INFO - 	 'boolq': {'acc,none': 0.7746177370030581,
2025-08-17 08:12:33 - eval_util - INFO - 	           'acc_stderr,none': 0.0073079593991191814,
2025-08-17 08:12:33 - eval_util - INFO - 	           'alias': 'boolq'},
2025-08-17 08:12:33 - eval_util - INFO - 	 'hellaswag': {'acc,none': 0.5577574188408684,
2025-08-17 08:12:33 - eval_util - INFO - 	               'acc_norm,none': 0.7457677753435571,
2025-08-17 08:12:33 - eval_util - INFO - 	               'acc_norm_stderr,none': 0.004345388614519977,
2025-08-17 08:12:33 - eval_util - INFO - 	               'acc_stderr,none': 0.004956378590571545,
2025-08-17 08:12:33 - eval_util - INFO - 	               'alias': 'hellaswag'},
2025-08-17 08:12:33 - eval_util - INFO - 	 'wikitext': 5.722687721252441,
2025-08-17 08:12:33 - eval_util - INFO - 	 'winogrande': {'acc,none': 0.6795580110497238,
2025-08-17 08:12:33 - eval_util - INFO - 	                'acc_stderr,none': 0.013115085457681714,
2025-08-17 08:12:33 - eval_util - INFO - 	                'alias': 'winogrande'}}
