| { | |
| "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": 0.7393712997436523, | |
| "eval_runtime": 197.0417, | |
| "eval_samples_per_second": 35.637, | |
| "eval_steps_per_second": 0.558, | |
| "eval_wer": 0.4965917697551123, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22779043280182232, | |
| "eval_loss": 0.6405971646308899, | |
| "eval_runtime": 190.1593, | |
| "eval_samples_per_second": 36.927, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.46169893301797793, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2847380410022779, | |
| "grad_norm": 11.500487327575684, | |
| "learning_rate": 0.00027833333333333334, | |
| "loss": 2.0443, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3416856492027335, | |
| "eval_loss": 0.6104918718338013, | |
| "eval_runtime": 188.7991, | |
| "eval_samples_per_second": 37.193, | |
| "eval_steps_per_second": 0.583, | |
| "eval_wer": 0.4496073559308521, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.45558086560364464, | |
| "eval_loss": 0.5864666700363159, | |
| "eval_runtime": 188.9909, | |
| "eval_samples_per_second": 37.155, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.44139571346948536, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5694760820045558, | |
| "grad_norm": 3.202120542526245, | |
| "learning_rate": 0.00022288888888888887, | |
| "loss": 0.7145, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5694760820045558, | |
| "eval_loss": 0.5439139008522034, | |
| "eval_runtime": 188.5161, | |
| "eval_samples_per_second": 37.249, | |
| "eval_steps_per_second": 0.584, | |
| "eval_wer": 0.41646846224371836, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.683371298405467, | |
| "eval_loss": 0.5428078174591064, | |
| "eval_runtime": 189.9544, | |
| "eval_samples_per_second": 36.967, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.41715940950583985, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.7972665148063781, | |
| "eval_loss": 0.5054725408554077, | |
| "eval_runtime": 189.6912, | |
| "eval_samples_per_second": 37.018, | |
| "eval_steps_per_second": 0.58, | |
| "eval_wer": 0.38716964084029815, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.8542141230068337, | |
| "grad_norm": 3.453420877456665, | |
| "learning_rate": 0.00016777777777777776, | |
| "loss": 0.6574, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9111617312072893, | |
| "eval_loss": 0.49158021807670593, | |
| "eval_runtime": 190.7112, | |
| "eval_samples_per_second": 36.82, | |
| "eval_steps_per_second": 0.577, | |
| "eval_wer": 0.3802601682190834, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.0250569476082005, | |
| "eval_loss": 0.4755226671695709, | |
| "eval_runtime": 189.3317, | |
| "eval_samples_per_second": 37.088, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.3658034255039264, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "grad_norm": 4.045414447784424, | |
| "learning_rate": 0.00011244444444444443, | |
| "loss": 0.577, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "eval_loss": 0.469484806060791, | |
| "eval_runtime": 189.7257, | |
| "eval_samples_per_second": 37.011, | |
| "eval_steps_per_second": 0.58, | |
| "eval_wer": 0.3594254507766513, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.2528473804100229, | |
| "eval_loss": 0.4655653238296509, | |
| "eval_runtime": 190.2356, | |
| "eval_samples_per_second": 36.912, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.35054943594786003, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.366742596810934, | |
| "eval_loss": 0.5080298185348511, | |
| "eval_runtime": 189.8354, | |
| "eval_samples_per_second": 36.99, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.3546951195205889, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.4236902050113895, | |
| "grad_norm": 3.1688034534454346, | |
| "learning_rate": 5.733333333333333e-05, | |
| "loss": 0.534, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4806378132118452, | |
| "eval_loss": 0.5481207370758057, | |
| "eval_runtime": 190.1317, | |
| "eval_samples_per_second": 36.932, | |
| "eval_steps_per_second": 0.579, | |
| "eval_wer": 0.36395647032248635, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.5945330296127562, | |
| "eval_loss": 0.6369026303291321, | |
| "eval_runtime": 191.1482, | |
| "eval_samples_per_second": 36.736, | |
| "eval_steps_per_second": 0.575, | |
| "eval_wer": 0.3902921909671933, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "grad_norm": 127.80626678466797, | |
| "learning_rate": 1.7777777777777775e-06, | |
| "loss": 0.7065, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "eval_loss": 0.6057178378105164, | |
| "eval_runtime": 190.1608, | |
| "eval_samples_per_second": 36.927, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.37381575625506586, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "step": 3000, | |
| "total_flos": 5.580592575460899e+18, | |
| "train_loss": 0.8723047383626302, | |
| "train_runtime": 4416.1126, | |
| "train_samples_per_second": 10.869, | |
| "train_steps_per_second": 0.679 | |
| } | |
| ], | |
| "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": 5.580592575460899e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |