| { |
| "best_global_step": 7670, |
| "best_metric": 0.8416083455085754, |
| "best_model_checkpoint": "distilbert_rand_10_v1_mnli/checkpoint-7670", |
| "epoch": 10.0, |
| "eval_steps": 500, |
| "global_step": 15340, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.0366369485855103, |
| "learning_rate": 4.9e-05, |
| "loss": 0.982, |
| "step": 1534 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.5616912888436068, |
| "eval_loss": 0.9146104454994202, |
| "eval_runtime": 4.0819, |
| "eval_samples_per_second": 2404.539, |
| "eval_steps_per_second": 9.554, |
| "step": 1534 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.764380931854248, |
| "learning_rate": 4.8e-05, |
| "loss": 0.884, |
| "step": 3068 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.6010188487009679, |
| "eval_loss": 0.8732452988624573, |
| "eval_runtime": 3.9446, |
| "eval_samples_per_second": 2488.19, |
| "eval_steps_per_second": 9.887, |
| "step": 3068 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.8879926204681396, |
| "learning_rate": 4.7e-05, |
| "loss": 0.8126, |
| "step": 4602 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.6172185430463576, |
| "eval_loss": 0.8506290912628174, |
| "eval_runtime": 3.9806, |
| "eval_samples_per_second": 2465.721, |
| "eval_steps_per_second": 9.798, |
| "step": 4602 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 2.1562070846557617, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.749, |
| "step": 6136 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6308711156393275, |
| "eval_loss": 0.8454623222351074, |
| "eval_runtime": 4.0037, |
| "eval_samples_per_second": 2451.474, |
| "eval_steps_per_second": 9.741, |
| "step": 6136 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.052584171295166, |
| "learning_rate": 4.5e-05, |
| "loss": 0.6871, |
| "step": 7670 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6388181355068773, |
| "eval_loss": 0.8416083455085754, |
| "eval_runtime": 3.9697, |
| "eval_samples_per_second": 2472.488, |
| "eval_steps_per_second": 9.824, |
| "step": 7670 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 2.593092441558838, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.6248, |
| "step": 9204 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.636984207845135, |
| "eval_loss": 0.8793249130249023, |
| "eval_runtime": 3.9938, |
| "eval_samples_per_second": 2457.556, |
| "eval_steps_per_second": 9.765, |
| "step": 9204 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 2.703984022140503, |
| "learning_rate": 4.3e-05, |
| "loss": 0.5603, |
| "step": 10738 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.6335201222618441, |
| "eval_loss": 0.9110007882118225, |
| "eval_runtime": 3.9818, |
| "eval_samples_per_second": 2464.94, |
| "eval_steps_per_second": 9.794, |
| "step": 10738 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 2.898940324783325, |
| "learning_rate": 4.2e-05, |
| "loss": 0.4983, |
| "step": 12272 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.6288334182373917, |
| "eval_loss": 1.0409351587295532, |
| "eval_runtime": 4.0092, |
| "eval_samples_per_second": 2448.093, |
| "eval_steps_per_second": 9.728, |
| "step": 12272 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 3.5689072608947754, |
| "learning_rate": 4.1e-05, |
| "loss": 0.4385, |
| "step": 13806 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.6254712175241977, |
| "eval_loss": 1.1441553831100464, |
| "eval_runtime": 3.9974, |
| "eval_samples_per_second": 2455.361, |
| "eval_steps_per_second": 9.756, |
| "step": 13806 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 4.5675368309021, |
| "learning_rate": 4e-05, |
| "loss": 0.3844, |
| "step": 15340 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.6274070300560367, |
| "eval_loss": 1.2257394790649414, |
| "eval_runtime": 3.9922, |
| "eval_samples_per_second": 2458.545, |
| "eval_steps_per_second": 9.769, |
| "step": 15340 |
| }, |
| { |
| "epoch": 10.0, |
| "step": 15340, |
| "total_flos": 2.6010570046823424e+17, |
| "train_loss": 0.6621012184113225, |
| "train_runtime": 3607.1482, |
| "train_samples_per_second": 5443.386, |
| "train_steps_per_second": 21.263 |
| } |
| ], |
| "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": 2.6010570046823424e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
