| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.366742596810934, | |
| "eval_steps": 200, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22779043280182232, | |
| "eval_loss": 2.9538023471832275, | |
| "eval_runtime": 176.5728, | |
| "eval_samples_per_second": 39.768, | |
| "eval_steps_per_second": 4.972, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45558086560364464, | |
| "eval_loss": 1.4731117486953735, | |
| "eval_runtime": 177.0417, | |
| "eval_samples_per_second": 39.663, | |
| "eval_steps_per_second": 4.959, | |
| "eval_wer": 0.8217488938200083, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5694760820045558, | |
| "grad_norm": 2.272386074066162, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 3.3429, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.683371298405467, | |
| "eval_loss": 0.852441132068634, | |
| "eval_runtime": 176.02, | |
| "eval_samples_per_second": 39.893, | |
| "eval_steps_per_second": 4.988, | |
| "eval_wer": 0.6397905898297878, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9111617312072893, | |
| "eval_loss": 0.7107498645782471, | |
| "eval_runtime": 176.239, | |
| "eval_samples_per_second": 39.844, | |
| "eval_steps_per_second": 4.982, | |
| "eval_wer": 0.5182237340384539, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "grad_norm": 0.8374969959259033, | |
| "learning_rate": 9.042857142857142e-05, | |
| "loss": 0.8743, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "eval_loss": 0.5667783617973328, | |
| "eval_runtime": 177.0088, | |
| "eval_samples_per_second": 39.67, | |
| "eval_steps_per_second": 4.96, | |
| "eval_wer": 0.4325462735353911, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.366742596810934, | |
| "eval_loss": 0.5374971628189087, | |
| "eval_runtime": 176.8891, | |
| "eval_samples_per_second": 39.697, | |
| "eval_steps_per_second": 4.964, | |
| "eval_wer": 0.4141431589577326, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.366742596810934, | |
| "step": 1200, | |
| "total_flos": 4.530371419942491e+18, | |
| "train_loss": 1.872582066853841, | |
| "train_runtime": 2251.6929, | |
| "train_samples_per_second": 17.054, | |
| "train_steps_per_second": 0.533 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 1200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.530371419942491e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |