| { | |
| "best_metric": 0.751246988773346, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-1300", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 1300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.51617431640625, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1019, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.41353383458646614, | |
| "eval_loss": 1.094665288925171, | |
| "eval_runtime": 0.9302, | |
| "eval_samples_per_second": 142.976, | |
| "eval_steps_per_second": 18.275, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.6315650343894958, | |
| "learning_rate": 5e-05, | |
| "loss": 1.091, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.518796992481203, | |
| "eval_loss": 1.002743124961853, | |
| "eval_runtime": 0.9515, | |
| "eval_samples_per_second": 139.773, | |
| "eval_steps_per_second": 17.866, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 1.4020596742630005, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0433, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.518796992481203, | |
| "eval_loss": 0.9840062856674194, | |
| "eval_runtime": 0.9456, | |
| "eval_samples_per_second": 140.656, | |
| "eval_steps_per_second": 17.979, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.387670636177063, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9915, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.7062584161758423, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9823, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5639097744360902, | |
| "eval_loss": 0.9470232129096985, | |
| "eval_runtime": 0.9624, | |
| "eval_samples_per_second": 138.194, | |
| "eval_steps_per_second": 17.664, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.5388019680976868, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9523, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5789473684210527, | |
| "eval_loss": 0.924637496471405, | |
| "eval_runtime": 0.9596, | |
| "eval_samples_per_second": 138.602, | |
| "eval_steps_per_second": 17.716, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 1.5363554954528809, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9431, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.556390977443609, | |
| "eval_loss": 0.890453577041626, | |
| "eval_runtime": 0.9858, | |
| "eval_samples_per_second": 134.909, | |
| "eval_steps_per_second": 17.244, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 1.7380107641220093, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8877, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 2.0791220664978027, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8587, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5939849624060151, | |
| "eval_loss": 0.8433966636657715, | |
| "eval_runtime": 0.9653, | |
| "eval_samples_per_second": 137.788, | |
| "eval_steps_per_second": 17.612, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 1.4432713985443115, | |
| "learning_rate": 5e-05, | |
| "loss": 0.834, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5714285714285714, | |
| "eval_loss": 0.8318859934806824, | |
| "eval_runtime": 0.9617, | |
| "eval_samples_per_second": 138.302, | |
| "eval_steps_per_second": 17.678, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 2.059706449508667, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8149, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6766917293233082, | |
| "eval_loss": 0.7831493616104126, | |
| "eval_runtime": 0.9673, | |
| "eval_samples_per_second": 137.496, | |
| "eval_steps_per_second": 17.575, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 1.22969651222229, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7628, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.152574300765991, | |
| "learning_rate": 5e-05, | |
| "loss": 0.729, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6541353383458647, | |
| "eval_loss": 0.751246988773346, | |
| "eval_runtime": 0.9753, | |
| "eval_samples_per_second": 136.369, | |
| "eval_steps_per_second": 17.431, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.772706474360013e+17, | |
| "train_loss": 0.9225140674297626, | |
| "train_runtime": 175.6098, | |
| "train_samples_per_second": 58.881, | |
| "train_steps_per_second": 7.403 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 8.772706474360013e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |