| { |
| "best_global_step": 3, |
| "best_metric": 0.733714759349823, |
| "best_model_checkpoint": "distilbert_km_50_v2_wnli/checkpoint-3", |
| "epoch": 6.0, |
| "eval_steps": 500, |
| "global_step": 18, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.4733049869537354, |
| "learning_rate": 4.9e-05, |
| "loss": 0.7297, |
| "step": 3 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.38028169014084506, |
| "eval_loss": 0.733714759349823, |
| "eval_runtime": 0.041, |
| "eval_samples_per_second": 1732.256, |
| "eval_steps_per_second": 24.398, |
| "step": 3 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.1206154823303223, |
| "learning_rate": 4.8e-05, |
| "loss": 0.702, |
| "step": 6 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.352112676056338, |
| "eval_loss": 0.7760233879089355, |
| "eval_runtime": 0.0485, |
| "eval_samples_per_second": 1463.01, |
| "eval_steps_per_second": 20.606, |
| "step": 6 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.1792210340499878, |
| "learning_rate": 4.7e-05, |
| "loss": 0.6908, |
| "step": 9 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.2535211267605634, |
| "eval_loss": 0.783203125, |
| "eval_runtime": 0.0452, |
| "eval_samples_per_second": 1571.945, |
| "eval_steps_per_second": 22.14, |
| "step": 9 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 1.597712516784668, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.6823, |
| "step": 12 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.28169014084507044, |
| "eval_loss": 0.794729471206665, |
| "eval_runtime": 0.0404, |
| "eval_samples_per_second": 1756.15, |
| "eval_steps_per_second": 24.735, |
| "step": 12 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.416773796081543, |
| "learning_rate": 4.5e-05, |
| "loss": 0.6838, |
| "step": 15 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.14084507042253522, |
| "eval_loss": 0.8307656645774841, |
| "eval_runtime": 0.0491, |
| "eval_samples_per_second": 1446.361, |
| "eval_steps_per_second": 20.371, |
| "step": 15 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 1.5318540334701538, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.6831, |
| "step": 18 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.16901408450704225, |
| "eval_loss": 0.875907838344574, |
| "eval_runtime": 0.0471, |
| "eval_samples_per_second": 1507.246, |
| "eval_steps_per_second": 21.229, |
| "step": 18 |
| }, |
| { |
| "epoch": 6.0, |
| "step": 18, |
| "total_flos": 252350394439680.0, |
| "train_loss": 0.695286922984653, |
| "train_runtime": 11.8363, |
| "train_samples_per_second": 2682.435, |
| "train_steps_per_second": 12.673 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 150, |
| "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": 252350394439680.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
