| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0080645161290323, | |
| "eval_steps": 200, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06720430107526881, | |
| "eval_loss": 2.9829530715942383, | |
| "eval_runtime": 176.5397, | |
| "eval_samples_per_second": 39.776, | |
| "eval_steps_per_second": 2.487, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.13440860215053763, | |
| "eval_loss": 1.5613375902175903, | |
| "eval_runtime": 174.5758, | |
| "eval_samples_per_second": 40.223, | |
| "eval_steps_per_second": 2.515, | |
| "eval_wer": 0.9657582481829415, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.16801075268817203, | |
| "grad_norm": 2.138428211212158, | |
| "learning_rate": 0.00027833333333333334, | |
| "loss": 3.6118, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.20161290322580644, | |
| "eval_loss": 1.0701438188552856, | |
| "eval_runtime": 174.6767, | |
| "eval_samples_per_second": 40.2, | |
| "eval_steps_per_second": 2.513, | |
| "eval_wer": 0.7648786191684716, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.26881720430107525, | |
| "eval_loss": 0.8867517113685608, | |
| "eval_runtime": 174.8944, | |
| "eval_samples_per_second": 40.15, | |
| "eval_steps_per_second": 2.51, | |
| "eval_wer": 0.6947474720631419, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.33602150537634407, | |
| "grad_norm": 2.852606773376465, | |
| "learning_rate": 0.00022288888888888887, | |
| "loss": 0.9333, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.33602150537634407, | |
| "eval_loss": 0.7679557204246521, | |
| "eval_runtime": 175.5538, | |
| "eval_samples_per_second": 39.999, | |
| "eval_steps_per_second": 2.501, | |
| "eval_wer": 0.6070503195631087, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4032258064516129, | |
| "eval_loss": 0.7223904132843018, | |
| "eval_runtime": 175.6961, | |
| "eval_samples_per_second": 39.967, | |
| "eval_steps_per_second": 2.499, | |
| "eval_wer": 0.5453965638661157, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.47043010752688175, | |
| "eval_loss": 0.6732765436172485, | |
| "eval_runtime": 175.6064, | |
| "eval_samples_per_second": 39.987, | |
| "eval_steps_per_second": 2.5, | |
| "eval_wer": 0.5121646580475425, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.5040322580645161, | |
| "grad_norm": 4.027652740478516, | |
| "learning_rate": 0.00016744444444444443, | |
| "loss": 0.7446, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5376344086021505, | |
| "eval_loss": 0.6437448859214783, | |
| "eval_runtime": 175.5304, | |
| "eval_samples_per_second": 40.004, | |
| "eval_steps_per_second": 2.501, | |
| "eval_wer": 0.4966449195445063, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.6048387096774194, | |
| "eval_loss": 0.6063565015792847, | |
| "eval_runtime": 175.877, | |
| "eval_samples_per_second": 39.926, | |
| "eval_steps_per_second": 2.496, | |
| "eval_wer": 0.4774312706785899, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.6720430107526881, | |
| "grad_norm": 3.676745653152466, | |
| "learning_rate": 0.000112, | |
| "loss": 0.6579, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6720430107526881, | |
| "eval_loss": 0.5673760771751404, | |
| "eval_runtime": 176.4608, | |
| "eval_samples_per_second": 39.794, | |
| "eval_steps_per_second": 2.488, | |
| "eval_wer": 0.44613933217289625, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.739247311827957, | |
| "eval_loss": 0.5556111931800842, | |
| "eval_runtime": 175.9454, | |
| "eval_samples_per_second": 39.91, | |
| "eval_steps_per_second": 2.495, | |
| "eval_wer": 0.4325196986406941, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.8064516129032258, | |
| "eval_loss": 0.5264282822608948, | |
| "eval_runtime": 176.4778, | |
| "eval_samples_per_second": 39.79, | |
| "eval_steps_per_second": 2.488, | |
| "eval_wer": 0.4180363810308402, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.8400537634408602, | |
| "grad_norm": 2.7548441886901855, | |
| "learning_rate": 5.666666666666666e-05, | |
| "loss": 0.5823, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8736559139784946, | |
| "eval_loss": 0.5129852890968323, | |
| "eval_runtime": 176.2054, | |
| "eval_samples_per_second": 39.851, | |
| "eval_steps_per_second": 2.491, | |
| "eval_wer": 0.4022110312387887, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.9408602150537635, | |
| "eval_loss": 0.49821802973747253, | |
| "eval_runtime": 176.2887, | |
| "eval_samples_per_second": 39.832, | |
| "eval_steps_per_second": 2.49, | |
| "eval_wer": 0.39360076535696725, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.0080645161290323, | |
| "grad_norm": 0.7031016945838928, | |
| "learning_rate": 1.111111111111111e-06, | |
| "loss": 0.5426, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0080645161290323, | |
| "eval_loss": 0.4942198395729065, | |
| "eval_runtime": 177.2018, | |
| "eval_samples_per_second": 39.627, | |
| "eval_steps_per_second": 2.477, | |
| "eval_wer": 0.3917272352808302, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0080645161290323, | |
| "step": 3000, | |
| "total_flos": 6.709869296482936e+18, | |
| "train_loss": 1.1787635701497396, | |
| "train_runtime": 4443.9077, | |
| "train_samples_per_second": 10.801, | |
| "train_steps_per_second": 0.675 | |
| } | |
| ], | |
| "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": 6.709869296482936e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |