| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 99.83948635634029, | |
| "eval_steps": 3110, | |
| "global_step": 31100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.983948635634029, | |
| "grad_norm": 4.509789943695068, | |
| "learning_rate": 9.001286173633441e-06, | |
| "loss": 2.5113, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 9.983948635634029, | |
| "eval_accuracy": 0.15658839274850955, | |
| "eval_loss": 2.851839542388916, | |
| "eval_runtime": 30.9634, | |
| "eval_samples_per_second": 265.442, | |
| "eval_steps_per_second": 13.274, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 19.967897271268058, | |
| "grad_norm": 3.6269993782043457, | |
| "learning_rate": 8.001607717041802e-06, | |
| "loss": 2.3003, | |
| "step": 6220 | |
| }, | |
| { | |
| "epoch": 19.967897271268058, | |
| "eval_accuracy": 0.14758486433872733, | |
| "eval_loss": 3.0242345333099365, | |
| "eval_runtime": 30.8398, | |
| "eval_samples_per_second": 266.506, | |
| "eval_steps_per_second": 13.327, | |
| "step": 6220 | |
| }, | |
| { | |
| "epoch": 29.951845906902086, | |
| "grad_norm": 4.074563980102539, | |
| "learning_rate": 7.0022508038585215e-06, | |
| "loss": 2.2124, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 29.951845906902086, | |
| "eval_accuracy": 0.14040637547146856, | |
| "eval_loss": 3.124624729156494, | |
| "eval_runtime": 30.966, | |
| "eval_samples_per_second": 265.42, | |
| "eval_steps_per_second": 13.273, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 39.935794542536115, | |
| "grad_norm": 4.5159077644348145, | |
| "learning_rate": 6.002572347266882e-06, | |
| "loss": 2.1682, | |
| "step": 12440 | |
| }, | |
| { | |
| "epoch": 39.935794542536115, | |
| "eval_accuracy": 0.13709088696921767, | |
| "eval_loss": 3.18056058883667, | |
| "eval_runtime": 30.9323, | |
| "eval_samples_per_second": 265.709, | |
| "eval_steps_per_second": 13.287, | |
| "step": 12440 | |
| }, | |
| { | |
| "epoch": 49.919743178170144, | |
| "grad_norm": 4.013918399810791, | |
| "learning_rate": 5.003215434083602e-06, | |
| "loss": 2.1384, | |
| "step": 15550 | |
| }, | |
| { | |
| "epoch": 49.919743178170144, | |
| "eval_accuracy": 0.1342255748874559, | |
| "eval_loss": 3.2024312019348145, | |
| "eval_runtime": 30.8117, | |
| "eval_samples_per_second": 266.749, | |
| "eval_steps_per_second": 13.339, | |
| "step": 15550 | |
| }, | |
| { | |
| "epoch": 59.90369181380417, | |
| "grad_norm": 3.466132640838623, | |
| "learning_rate": 4.003536977491962e-06, | |
| "loss": 2.1246, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.90369181380417, | |
| "eval_accuracy": 0.1322545321815306, | |
| "eval_loss": 3.2168166637420654, | |
| "eval_runtime": 30.7446, | |
| "eval_samples_per_second": 267.331, | |
| "eval_steps_per_second": 13.368, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 69.88764044943821, | |
| "grad_norm": 2.7771756649017334, | |
| "learning_rate": 3.004180064308682e-06, | |
| "loss": 2.1121, | |
| "step": 21770 | |
| }, | |
| { | |
| "epoch": 69.88764044943821, | |
| "eval_accuracy": 0.13129855908782787, | |
| "eval_loss": 3.2208762168884277, | |
| "eval_runtime": 30.8368, | |
| "eval_samples_per_second": 266.532, | |
| "eval_steps_per_second": 13.328, | |
| "step": 21770 | |
| }, | |
| { | |
| "epoch": 79.87158908507223, | |
| "grad_norm": 2.471256732940674, | |
| "learning_rate": 2.0045016077170422e-06, | |
| "loss": 2.0987, | |
| "step": 24880 | |
| }, | |
| { | |
| "epoch": 79.87158908507223, | |
| "eval_accuracy": 0.1307032485703857, | |
| "eval_loss": 3.213690996170044, | |
| "eval_runtime": 30.7206, | |
| "eval_samples_per_second": 267.54, | |
| "eval_steps_per_second": 13.379, | |
| "step": 24880 | |
| }, | |
| { | |
| "epoch": 89.85553772070627, | |
| "grad_norm": 2.8352723121643066, | |
| "learning_rate": 1.0048231511254019e-06, | |
| "loss": 2.0986, | |
| "step": 27990 | |
| }, | |
| { | |
| "epoch": 89.85553772070627, | |
| "eval_accuracy": 0.1300915223533547, | |
| "eval_loss": 3.2373690605163574, | |
| "eval_runtime": 30.864, | |
| "eval_samples_per_second": 266.297, | |
| "eval_steps_per_second": 13.316, | |
| "step": 27990 | |
| }, | |
| { | |
| "epoch": 99.83948635634029, | |
| "grad_norm": 2.9546291828155518, | |
| "learning_rate": 5.144694533762058e-09, | |
| "loss": 2.0962, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 99.83948635634029, | |
| "eval_accuracy": 0.12952913979802896, | |
| "eval_loss": 3.238151788711548, | |
| "eval_runtime": 31.0746, | |
| "eval_samples_per_second": 264.492, | |
| "eval_steps_per_second": 13.226, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 99.83948635634029, | |
| "step": 31100, | |
| "total_flos": 9.751242948802248e+17, | |
| "train_loss": 2.186087721061093, | |
| "train_runtime": 48974.7066, | |
| "train_samples_per_second": 127.117, | |
| "train_steps_per_second": 0.635 | |
| } | |
| ], | |
| "logging_steps": 3110, | |
| "max_steps": 31100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.751242948802248e+17, | |
| "train_batch_size": 50, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |