| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0547945205479454, | |
| "eval_steps": 30, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.410958904109589, | |
| "grad_norm": 10.242740631103516, | |
| "learning_rate": 9e-06, | |
| "loss": 3.1804, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.410958904109589, | |
| "eval_loss": 1.894700527191162, | |
| "eval_runtime": 202.5611, | |
| "eval_samples_per_second": 5.751, | |
| "eval_steps_per_second": 0.36, | |
| "eval_wer": 0.8896577289015198, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.821917808219178, | |
| "grad_norm": 6.609806537628174, | |
| "learning_rate": 7.75e-06, | |
| "loss": 1.5393, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.821917808219178, | |
| "eval_loss": 1.2656499147415161, | |
| "eval_runtime": 205.5663, | |
| "eval_samples_per_second": 5.667, | |
| "eval_steps_per_second": 0.355, | |
| "eval_wer": 0.7559619424193748, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2328767123287672, | |
| "grad_norm": 6.066283702850342, | |
| "learning_rate": 5.2500000000000006e-06, | |
| "loss": 1.1714, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2328767123287672, | |
| "eval_loss": 1.1013103723526, | |
| "eval_runtime": 197.662, | |
| "eval_samples_per_second": 5.894, | |
| "eval_steps_per_second": 0.369, | |
| "eval_wer": 0.7067836401828741, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.643835616438356, | |
| "grad_norm": 5.897896766662598, | |
| "learning_rate": 2.7500000000000004e-06, | |
| "loss": 1.0264, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.643835616438356, | |
| "eval_loss": 1.0345516204833984, | |
| "eval_runtime": 189.8865, | |
| "eval_samples_per_second": 6.135, | |
| "eval_steps_per_second": 0.384, | |
| "eval_wer": 0.6828123069319165, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0547945205479454, | |
| "grad_norm": 5.676390647888184, | |
| "learning_rate": 2.5000000000000004e-07, | |
| "loss": 0.9896, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0547945205479454, | |
| "eval_loss": 1.0129033327102661, | |
| "eval_runtime": 192.5239, | |
| "eval_samples_per_second": 6.051, | |
| "eval_steps_per_second": 0.379, | |
| "eval_wer": 0.6624243173112566, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0547945205479454, | |
| "step": 150, | |
| "total_flos": 4.7164860112896e+17, | |
| "train_loss": 1.5814088694254558, | |
| "train_runtime": 3714.6914, | |
| "train_samples_per_second": 5.169, | |
| "train_steps_per_second": 0.04 | |
| } | |
| ], | |
| "logging_steps": 30, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 30, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.7164860112896e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |