| { | |
| "best_metric": 0.946372926235199, | |
| "best_model_checkpoint": "finetune-tiktok-brat7\\checkpoint-24", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 24, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 48.439212799072266, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.298, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 17.51978874206543, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.3288, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 36.70003128051758, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5146, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 36.54575729370117, | |
| "learning_rate": 9.523809523809525e-06, | |
| "loss": 1.3186, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 38.397705078125, | |
| "learning_rate": 9.047619047619049e-06, | |
| "loss": 1.0083, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 52.007572174072266, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.9936, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 57.42391586303711, | |
| "learning_rate": 8.095238095238097e-06, | |
| "loss": 1.1921, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 39.129276275634766, | |
| "learning_rate": 7.61904761904762e-06, | |
| "loss": 1.2205, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6521739130434783, | |
| "eval_f1_macro": 0.4761904761904762, | |
| "eval_f1_micro": 0.6521739130434783, | |
| "eval_f1_weighted": 0.6521739130434783, | |
| "eval_loss": 1.1324578523635864, | |
| "eval_precision_macro": 0.4761904761904762, | |
| "eval_precision_micro": 0.6521739130434783, | |
| "eval_precision_weighted": 0.6521739130434783, | |
| "eval_recall_macro": 0.4761904761904762, | |
| "eval_recall_micro": 0.6521739130434783, | |
| "eval_recall_weighted": 0.6521739130434783, | |
| "eval_runtime": 27.369, | |
| "eval_samples_per_second": 0.84, | |
| "eval_steps_per_second": 0.073, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 30.69573211669922, | |
| "learning_rate": 7.1428571428571436e-06, | |
| "loss": 0.8532, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 19.239377975463867, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.718, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 36.98495101928711, | |
| "learning_rate": 6.1904761904761914e-06, | |
| "loss": 0.7192, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 29.336002349853516, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 1.0733, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 8.844223022460938, | |
| "learning_rate": 5.2380952380952384e-06, | |
| "loss": 0.2654, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 17.313766479492188, | |
| "learning_rate": 4.761904761904762e-06, | |
| "loss": 0.643, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 12.483030319213867, | |
| "learning_rate": 4.2857142857142855e-06, | |
| "loss": 0.3099, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 8.137201309204102, | |
| "learning_rate": 3.80952380952381e-06, | |
| "loss": 0.0809, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6956521739130435, | |
| "eval_f1_macro": 0.49096880131362886, | |
| "eval_f1_micro": 0.6956521739130435, | |
| "eval_f1_weighted": 0.6791604197901049, | |
| "eval_loss": 0.9708660840988159, | |
| "eval_precision_macro": 0.48253968253968255, | |
| "eval_precision_micro": 0.6956521739130435, | |
| "eval_precision_weighted": 0.6637681159420289, | |
| "eval_recall_macro": 0.5, | |
| "eval_recall_micro": 0.6956521739130435, | |
| "eval_recall_weighted": 0.6956521739130435, | |
| "eval_runtime": 27.3832, | |
| "eval_samples_per_second": 0.84, | |
| "eval_steps_per_second": 0.073, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.125, | |
| "grad_norm": 15.641371726989746, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.4351, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 23.520936965942383, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.4472, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.375, | |
| "grad_norm": 19.283954620361328, | |
| "learning_rate": 2.380952380952381e-06, | |
| "loss": 0.216, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 35.63965606689453, | |
| "learning_rate": 1.904761904761905e-06, | |
| "loss": 0.5644, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.625, | |
| "grad_norm": 14.39991283416748, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 0.4427, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 22.22780418395996, | |
| "learning_rate": 9.523809523809525e-07, | |
| "loss": 0.4644, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.875, | |
| "grad_norm": 21.318103790283203, | |
| "learning_rate": 4.7619047619047623e-07, | |
| "loss": 0.5123, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 20.380773544311523, | |
| "learning_rate": 0.0, | |
| "loss": 0.8922, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7391304347826086, | |
| "eval_f1_macro": 0.5230769230769231, | |
| "eval_f1_micro": 0.7391304347826086, | |
| "eval_f1_weighted": 0.7210702341137124, | |
| "eval_loss": 0.946372926235199, | |
| "eval_precision_macro": 0.5277777777777778, | |
| "eval_precision_micro": 0.7391304347826086, | |
| "eval_precision_weighted": 0.710144927536232, | |
| "eval_recall_macro": 0.5238095238095238, | |
| "eval_recall_micro": 0.7391304347826086, | |
| "eval_recall_weighted": 0.7391304347826086, | |
| "eval_runtime": 27.6232, | |
| "eval_samples_per_second": 0.833, | |
| "eval_steps_per_second": 0.072, | |
| "step": 24 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 24, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.01 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 46571074937856.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |