{ "best_metric": 0.9320158102766799, "best_model_checkpoint": "./sentiment_model\\checkpoint-92", "epoch": 4.0, "eval_steps": 500, "global_step": 92, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.43478260869565216, "grad_norm": 9.166400909423828, "learning_rate": 1.8550724637681162e-05, "loss": 1.0141, "step": 10 }, { "epoch": 0.8695652173913043, "grad_norm": 6.58314323425293, "learning_rate": 1.710144927536232e-05, "loss": 0.7957, "step": 20 }, { "epoch": 1.0, "eval_accuracy": 0.8, "eval_f1": 0.7970942760942761, "eval_loss": 0.6531133055686951, "eval_runtime": 1.1269, "eval_samples_per_second": 39.931, "eval_steps_per_second": 5.324, "step": 23 }, { "epoch": 1.3043478260869565, "grad_norm": 4.263016700744629, "learning_rate": 1.565217391304348e-05, "loss": 0.5277, "step": 30 }, { "epoch": 1.7391304347826086, "grad_norm": 11.197798728942871, "learning_rate": 1.420289855072464e-05, "loss": 0.4258, "step": 40 }, { "epoch": 2.0, "eval_accuracy": 0.8666666666666667, "eval_f1": 0.8631313131313132, "eval_loss": 0.37175920605659485, "eval_runtime": 1.0687, "eval_samples_per_second": 42.106, "eval_steps_per_second": 5.614, "step": 46 }, { "epoch": 2.1739130434782608, "grad_norm": 1.9997150897979736, "learning_rate": 1.2753623188405797e-05, "loss": 0.305, "step": 50 }, { "epoch": 2.608695652173913, "grad_norm": 4.433629989624023, "learning_rate": 1.1304347826086957e-05, "loss": 0.2419, "step": 60 }, { "epoch": 3.0, "eval_accuracy": 0.9111111111111111, "eval_f1": 0.9101010101010101, "eval_loss": 0.2871778607368469, "eval_runtime": 1.2144, "eval_samples_per_second": 37.057, "eval_steps_per_second": 4.941, "step": 69 }, { "epoch": 3.0434782608695654, "grad_norm": 4.2565693855285645, "learning_rate": 9.855072463768118e-06, "loss": 0.1732, "step": 70 }, { "epoch": 3.4782608695652173, "grad_norm": 5.562937259674072, "learning_rate": 8.405797101449275e-06, "loss": 0.1251, "step": 80 }, { "epoch": 3.9130434782608696, "grad_norm": 0.7796284556388855, "learning_rate": 6.956521739130435e-06, "loss": 0.1133, "step": 90 }, { "epoch": 4.0, "eval_accuracy": 0.9333333333333333, "eval_f1": 0.9320158102766799, "eval_loss": 0.2682146728038788, "eval_runtime": 1.2187, "eval_samples_per_second": 36.925, "eval_steps_per_second": 4.923, "step": 92 } ], "logging_steps": 10, "max_steps": 138, "num_input_tokens_seen": 0, "num_train_epochs": 6, "save_steps": 500, "total_flos": 32745287065680.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }