| { | |
| "best_metric": 0.3044010102748871, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-650", | |
| "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": 2.5466179847717285, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0208, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8045112781954887, | |
| "eval_loss": 0.8930914998054504, | |
| "eval_runtime": 0.9307, | |
| "eval_samples_per_second": 142.898, | |
| "eval_steps_per_second": 18.265, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.4668034315109253, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8683, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_loss": 0.6855143308639526, | |
| "eval_runtime": 0.9382, | |
| "eval_samples_per_second": 141.765, | |
| "eval_steps_per_second": 18.12, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 1.5885818004608154, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7297, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_loss": 0.5008501410484314, | |
| "eval_runtime": 0.9428, | |
| "eval_samples_per_second": 141.075, | |
| "eval_steps_per_second": 18.032, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.4792442321777344, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6084, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.0500737428665161, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4881, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9323308270676691, | |
| "eval_loss": 0.3685053288936615, | |
| "eval_runtime": 0.9444, | |
| "eval_samples_per_second": 140.829, | |
| "eval_steps_per_second": 18.001, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.9487807154655457, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3733, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_loss": 0.3044010102748871, | |
| "eval_runtime": 0.9975, | |
| "eval_samples_per_second": 133.327, | |
| "eval_steps_per_second": 17.042, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 2.811962604522705, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3229, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_loss": 0.3873802721500397, | |
| "eval_runtime": 0.9366, | |
| "eval_samples_per_second": 142.002, | |
| "eval_steps_per_second": 18.151, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 2.758183717727661, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4189, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 5.297445297241211, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5796, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5939849624060151, | |
| "eval_loss": 0.7966009974479675, | |
| "eval_runtime": 0.9577, | |
| "eval_samples_per_second": 138.872, | |
| "eval_steps_per_second": 17.75, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 28.03814697265625, | |
| "learning_rate": 5e-05, | |
| "loss": 0.765, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 0.9421801567077637, | |
| "eval_runtime": 0.9477, | |
| "eval_samples_per_second": 140.335, | |
| "eval_steps_per_second": 17.938, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 5.901334285736084, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0732, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.0903689861297607, | |
| "eval_runtime": 0.9358, | |
| "eval_samples_per_second": 142.119, | |
| "eval_steps_per_second": 18.166, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 1.3560993671417236, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0879, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 37.35905456542969, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1643, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.41353383458646614, | |
| "eval_loss": 1.0773515701293945, | |
| "eval_runtime": 0.9946, | |
| "eval_samples_per_second": 133.719, | |
| "eval_steps_per_second": 17.092, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.772706474360013e+17, | |
| "train_loss": 0.7308059575007512, | |
| "train_runtime": 175.3451, | |
| "train_samples_per_second": 58.969, | |
| "train_steps_per_second": 7.414 | |
| } | |
| ], | |
| "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 | |
| } | |