| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.7084282460136673, | |
| "eval_steps": 200, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11389521640091116, | |
| "eval_loss": 0.7527692914009094, | |
| "eval_runtime": 198.1592, | |
| "eval_samples_per_second": 35.436, | |
| "eval_steps_per_second": 0.555, | |
| "eval_wer": 0.5048698494532216, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22779043280182232, | |
| "eval_loss": 0.6966450810432434, | |
| "eval_runtime": 197.214, | |
| "eval_samples_per_second": 35.606, | |
| "eval_steps_per_second": 0.558, | |
| "eval_wer": 0.5050160113740549, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2847380410022779, | |
| "grad_norm": 4.219395637512207, | |
| "learning_rate": 0.00027833333333333334, | |
| "loss": 2.117, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3416856492027335, | |
| "eval_loss": 0.6128434538841248, | |
| "eval_runtime": 200.109, | |
| "eval_samples_per_second": 35.091, | |
| "eval_steps_per_second": 0.55, | |
| "eval_wer": 0.47608923849639245, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.45558086560364464, | |
| "eval_loss": 0.6331803202629089, | |
| "eval_runtime": 200.832, | |
| "eval_samples_per_second": 34.965, | |
| "eval_steps_per_second": 0.548, | |
| "eval_wer": 0.5017472993263264, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5694760820045558, | |
| "grad_norm": 3.2030014991760254, | |
| "learning_rate": 0.00022288888888888887, | |
| "loss": 0.7606, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5694760820045558, | |
| "eval_loss": 0.5895215272903442, | |
| "eval_runtime": 203.9306, | |
| "eval_samples_per_second": 34.433, | |
| "eval_steps_per_second": 0.539, | |
| "eval_wer": 0.457659549024037, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.683371298405467, | |
| "eval_loss": 0.5552608370780945, | |
| "eval_runtime": 205.5709, | |
| "eval_samples_per_second": 34.159, | |
| "eval_steps_per_second": 0.535, | |
| "eval_wer": 0.4211057813683413, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7972665148063781, | |
| "eval_loss": 0.530360996723175, | |
| "eval_runtime": 215.1898, | |
| "eval_samples_per_second": 32.632, | |
| "eval_steps_per_second": 0.511, | |
| "eval_wer": 0.419604299817962, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.8542141230068337, | |
| "grad_norm": 2.892026424407959, | |
| "learning_rate": 0.00016766666666666666, | |
| "loss": 0.7049, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9111617312072893, | |
| "eval_loss": 0.5060806274414062, | |
| "eval_runtime": 208.7711, | |
| "eval_samples_per_second": 33.635, | |
| "eval_steps_per_second": 0.527, | |
| "eval_wer": 0.38730251531378307, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.0250569476082005, | |
| "eval_loss": 0.5090161561965942, | |
| "eval_runtime": 207.1232, | |
| "eval_samples_per_second": 33.903, | |
| "eval_steps_per_second": 0.531, | |
| "eval_wer": 0.3959127811956045, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "grad_norm": 2.3204939365386963, | |
| "learning_rate": 0.00011233333333333333, | |
| "loss": 0.6136, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "eval_loss": 0.4839297831058502, | |
| "eval_runtime": 206.3623, | |
| "eval_samples_per_second": 34.028, | |
| "eval_steps_per_second": 0.533, | |
| "eval_wer": 0.3758088733573393, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2528473804100229, | |
| "eval_loss": 0.46924272179603577, | |
| "eval_runtime": 205.266, | |
| "eval_samples_per_second": 34.209, | |
| "eval_steps_per_second": 0.536, | |
| "eval_wer": 0.3658565752933204, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.366742596810934, | |
| "eval_loss": 0.4569305181503296, | |
| "eval_runtime": 207.0588, | |
| "eval_samples_per_second": 33.913, | |
| "eval_steps_per_second": 0.531, | |
| "eval_wer": 0.35436293333687663, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.4236902050113895, | |
| "grad_norm": 3.092404365539551, | |
| "learning_rate": 5.688888888888888e-05, | |
| "loss": 0.5388, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4806378132118452, | |
| "eval_loss": 0.4487648606300354, | |
| "eval_runtime": 205.0091, | |
| "eval_samples_per_second": 34.252, | |
| "eval_steps_per_second": 0.537, | |
| "eval_wer": 0.3484765941614956, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.5945330296127562, | |
| "eval_loss": 0.441054105758667, | |
| "eval_runtime": 211.0185, | |
| "eval_samples_per_second": 33.277, | |
| "eval_steps_per_second": 0.521, | |
| "eval_wer": 0.3423112185917963, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "grad_norm": 0.7985823154449463, | |
| "learning_rate": 1.4444444444444445e-06, | |
| "loss": 0.5275, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "eval_loss": 0.4376124143600464, | |
| "eval_runtime": 210.3633, | |
| "eval_samples_per_second": 33.38, | |
| "eval_steps_per_second": 0.523, | |
| "eval_wer": 0.339454417411871, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "step": 3000, | |
| "total_flos": 5.591115044514249e+18, | |
| "train_loss": 0.8770465799967448, | |
| "train_runtime": 4793.5663, | |
| "train_samples_per_second": 10.013, | |
| "train_steps_per_second": 0.626 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3000, | |
| "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": 5.591115044514249e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |