| { |
| "best_global_step": 6136, |
| "best_metric": 0.772878110408783, |
| "best_model_checkpoint": "distilbert_km_5_v1_mnli/checkpoint-6136", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 13806, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.0547219514846802, |
| "learning_rate": 4.9e-05, |
| "loss": 0.9818, |
| "step": 1534 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.5685175751400917, |
| "eval_loss": 0.9036023020744324, |
| "eval_runtime": 4.0479, |
| "eval_samples_per_second": 2424.699, |
| "eval_steps_per_second": 9.635, |
| "step": 1534 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.7676842212677002, |
| "learning_rate": 4.8e-05, |
| "loss": 0.8685, |
| "step": 3068 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6330106979113601, |
| "eval_loss": 0.8275231719017029, |
| "eval_runtime": 3.9772, |
| "eval_samples_per_second": 2467.801, |
| "eval_steps_per_second": 9.806, |
| "step": 3068 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.7819937467575073, |
| "learning_rate": 4.7e-05, |
| "loss": 0.7685, |
| "step": 4602 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6604177279673968, |
| "eval_loss": 0.7766311764717102, |
| "eval_runtime": 3.9789, |
| "eval_samples_per_second": 2466.793, |
| "eval_steps_per_second": 9.802, |
| "step": 4602 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.35194730758667, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.6808, |
| "step": 6136 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6723382577687214, |
| "eval_loss": 0.772878110408783, |
| "eval_runtime": 3.9521, |
| "eval_samples_per_second": 2483.486, |
| "eval_steps_per_second": 9.868, |
| "step": 6136 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.4090893268585205, |
| "learning_rate": 4.5e-05, |
| "loss": 0.602, |
| "step": 7670 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6697911360163016, |
| "eval_loss": 0.7900730967521667, |
| "eval_runtime": 3.9553, |
| "eval_samples_per_second": 2481.474, |
| "eval_steps_per_second": 9.86, |
| "step": 7670 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.6621053218841553, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.5232, |
| "step": 9204 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.674477840040754, |
| "eval_loss": 0.8526480197906494, |
| "eval_runtime": 3.9746, |
| "eval_samples_per_second": 2469.426, |
| "eval_steps_per_second": 9.812, |
| "step": 9204 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 3.343918800354004, |
| "learning_rate": 4.3e-05, |
| "loss": 0.4484, |
| "step": 10738 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6753948038716251, |
| "eval_loss": 0.9103152751922607, |
| "eval_runtime": 4.0023, |
| "eval_samples_per_second": 2452.35, |
| "eval_steps_per_second": 9.744, |
| "step": 10738 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 3.675567150115967, |
| "learning_rate": 4.2e-05, |
| "loss": 0.3773, |
| "step": 12272 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.6649006622516557, |
| "eval_loss": 1.0193079710006714, |
| "eval_runtime": 3.9743, |
| "eval_samples_per_second": 2469.638, |
| "eval_steps_per_second": 9.813, |
| "step": 12272 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 3.6849734783172607, |
| "learning_rate": 4.1e-05, |
| "loss": 0.3169, |
| "step": 13806 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.6694854814060112, |
| "eval_loss": 1.1595568656921387, |
| "eval_runtime": 3.9777, |
| "eval_samples_per_second": 2467.51, |
| "eval_steps_per_second": 9.805, |
| "step": 13806 |
| }, |
| { |
| "epoch": 9.0, |
| "step": 13806, |
| "total_flos": 2.340951304214108e+17, |
| "train_loss": 0.6185946463847806, |
| "train_runtime": 3246.9545, |
| "train_samples_per_second": 6047.236, |
| "train_steps_per_second": 23.622 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 76700, |
| "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": 2.340951304214108e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
