| { | |
| "best_metric": 0.6878666281700134, | |
| "best_model_checkpoint": "distilbert_lda_20_v1_book_rte/checkpoint-30", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.5146413445472717, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.7084, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4620938628158845, | |
| "eval_loss": 0.6956509351730347, | |
| "eval_runtime": 0.1331, | |
| "eval_samples_per_second": 2080.925, | |
| "eval_steps_per_second": 15.025, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.37892234325408936, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6931, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5379061371841155, | |
| "eval_loss": 0.6898268461227417, | |
| "eval_runtime": 0.1382, | |
| "eval_samples_per_second": 2004.196, | |
| "eval_steps_per_second": 14.471, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.28291600942611694, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6875, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5631768953068592, | |
| "eval_loss": 0.6878666281700134, | |
| "eval_runtime": 0.1347, | |
| "eval_samples_per_second": 2055.944, | |
| "eval_steps_per_second": 14.844, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6326797008514404, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6709, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5306859205776173, | |
| "eval_loss": 0.7002270221710205, | |
| "eval_runtime": 0.1819, | |
| "eval_samples_per_second": 1522.567, | |
| "eval_steps_per_second": 10.993, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.6822352409362793, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.633, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5342960288808665, | |
| "eval_loss": 0.7305744886398315, | |
| "eval_runtime": 0.1343, | |
| "eval_samples_per_second": 2062.462, | |
| "eval_steps_per_second": 14.891, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.8245058059692383, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5346, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5667870036101083, | |
| "eval_loss": 0.8086202144622803, | |
| "eval_runtime": 0.1443, | |
| "eval_samples_per_second": 1919.669, | |
| "eval_steps_per_second": 13.86, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.110736846923828, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.4039, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5451263537906137, | |
| "eval_loss": 0.9744732975959778, | |
| "eval_runtime": 0.1588, | |
| "eval_samples_per_second": 1743.965, | |
| "eval_steps_per_second": 12.592, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.084683418273926, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2679, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.516245487364621, | |
| "eval_loss": 1.1202261447906494, | |
| "eval_runtime": 0.1386, | |
| "eval_samples_per_second": 1997.954, | |
| "eval_steps_per_second": 14.426, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 80, | |
| "total_flos": 1319375290613760.0, | |
| "train_loss": 0.5749205261468887, | |
| "train_runtime": 36.2687, | |
| "train_samples_per_second": 3432.71, | |
| "train_steps_per_second": 13.786 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1319375290613760.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |