| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.536067892503536, | |
| "eval_steps": 100, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14144271570014144, | |
| "eval_loss": 1.2105046510696411, | |
| "eval_runtime": 283.0228, | |
| "eval_samples_per_second": 19.984, | |
| "eval_steps_per_second": 2.498, | |
| "eval_wer": 0.8709347154900737, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2828854314002829, | |
| "eval_loss": 0.9787197113037109, | |
| "eval_runtime": 281.4976, | |
| "eval_samples_per_second": 20.093, | |
| "eval_steps_per_second": 2.512, | |
| "eval_wer": 0.6986290412887332, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4243281471004243, | |
| "eval_loss": 1.190703272819519, | |
| "eval_runtime": 281.4959, | |
| "eval_samples_per_second": 20.093, | |
| "eval_steps_per_second": 2.512, | |
| "eval_wer": 0.7126745692758074, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5657708628005658, | |
| "eval_loss": 1.0559463500976562, | |
| "eval_runtime": 280.2535, | |
| "eval_samples_per_second": 20.182, | |
| "eval_steps_per_second": 2.523, | |
| "eval_wer": 0.7169298203516181, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7072135785007072, | |
| "grad_norm": 5.1558146476745605, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 1.4456, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7072135785007072, | |
| "eval_loss": 1.2105939388275146, | |
| "eval_runtime": 281.5867, | |
| "eval_samples_per_second": 20.086, | |
| "eval_steps_per_second": 2.511, | |
| "eval_wer": 0.794404184863464, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8486562942008486, | |
| "eval_loss": 1.0231719017028809, | |
| "eval_runtime": 280.7447, | |
| "eval_samples_per_second": 20.146, | |
| "eval_steps_per_second": 2.518, | |
| "eval_wer": 0.7033002191614276, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9900990099009901, | |
| "eval_loss": 1.038698673248291, | |
| "eval_runtime": 281.8275, | |
| "eval_samples_per_second": 20.069, | |
| "eval_steps_per_second": 2.509, | |
| "eval_wer": 0.7335988865959591, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1315417256011315, | |
| "eval_loss": 0.7234079837799072, | |
| "eval_runtime": 280.5616, | |
| "eval_samples_per_second": 20.16, | |
| "eval_steps_per_second": 2.52, | |
| "eval_wer": 0.5223240709635104, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.272984441301273, | |
| "eval_loss": 0.7242198586463928, | |
| "eval_runtime": 286.25, | |
| "eval_samples_per_second": 19.759, | |
| "eval_steps_per_second": 2.47, | |
| "eval_wer": 0.5566220345219242, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4144271570014144, | |
| "grad_norm": 1.7350859642028809, | |
| "learning_rate": 0.0002259, | |
| "loss": 0.9155, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4144271570014144, | |
| "eval_loss": 0.7096899151802063, | |
| "eval_runtime": 280.6062, | |
| "eval_samples_per_second": 20.156, | |
| "eval_steps_per_second": 2.52, | |
| "eval_wer": 0.5258754459215178, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5558698727015559, | |
| "eval_loss": 0.6368164420127869, | |
| "eval_runtime": 282.6165, | |
| "eval_samples_per_second": 20.013, | |
| "eval_steps_per_second": 2.502, | |
| "eval_wer": 0.47965957991393515, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6973125884016973, | |
| "eval_loss": 0.6065136194229126, | |
| "eval_runtime": 280.4494, | |
| "eval_samples_per_second": 20.168, | |
| "eval_steps_per_second": 2.521, | |
| "eval_wer": 0.4652941082369503, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8387553041018387, | |
| "eval_loss": 0.6206967830657959, | |
| "eval_runtime": 282.4382, | |
| "eval_samples_per_second": 20.026, | |
| "eval_steps_per_second": 2.503, | |
| "eval_wer": 0.4716929820351618, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.9801980198019802, | |
| "eval_loss": 0.5924867987632751, | |
| "eval_runtime": 281.7785, | |
| "eval_samples_per_second": 20.073, | |
| "eval_steps_per_second": 2.509, | |
| "eval_wer": 0.470669162227448, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.1216407355021216, | |
| "grad_norm": 1.3663442134857178, | |
| "learning_rate": 0.0001512, | |
| "loss": 0.7436, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.1216407355021216, | |
| "eval_loss": 0.5382007360458374, | |
| "eval_runtime": 284.0884, | |
| "eval_samples_per_second": 19.909, | |
| "eval_steps_per_second": 2.489, | |
| "eval_wer": 0.40460079026091406, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.263083451202263, | |
| "eval_loss": 0.5200654864311218, | |
| "eval_runtime": 282.3736, | |
| "eval_samples_per_second": 20.03, | |
| "eval_steps_per_second": 2.504, | |
| "eval_wer": 0.3995616771448225, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.4045261669024045, | |
| "eval_loss": 0.4883446991443634, | |
| "eval_runtime": 283.5276, | |
| "eval_samples_per_second": 19.949, | |
| "eval_steps_per_second": 2.494, | |
| "eval_wer": 0.36979091679864345, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.545968882602546, | |
| "eval_loss": 0.47043663263320923, | |
| "eval_runtime": 282.1263, | |
| "eval_samples_per_second": 20.048, | |
| "eval_steps_per_second": 2.506, | |
| "eval_wer": 0.36585560941274337, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.6874115983026874, | |
| "eval_loss": 0.4443446099758148, | |
| "eval_runtime": 282.7688, | |
| "eval_samples_per_second": 20.002, | |
| "eval_steps_per_second": 2.5, | |
| "eval_wer": 0.3521460223000752, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.828854314002829, | |
| "grad_norm": 1.0020660161972046, | |
| "learning_rate": 7.635e-05, | |
| "loss": 0.5645, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.828854314002829, | |
| "eval_loss": 0.4469930827617645, | |
| "eval_runtime": 282.9465, | |
| "eval_samples_per_second": 19.99, | |
| "eval_steps_per_second": 2.499, | |
| "eval_wer": 0.34761881908784054, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.9702970297029703, | |
| "eval_loss": 0.41922062635421753, | |
| "eval_runtime": 281.1649, | |
| "eval_samples_per_second": 20.116, | |
| "eval_steps_per_second": 2.515, | |
| "eval_wer": 0.3241669466173953, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.1117397454031117, | |
| "eval_loss": 0.41775766015052795, | |
| "eval_runtime": 282.3459, | |
| "eval_samples_per_second": 20.032, | |
| "eval_steps_per_second": 2.504, | |
| "eval_wer": 0.3160883684471533, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.253182461103253, | |
| "eval_loss": 0.4122001826763153, | |
| "eval_runtime": 282.9107, | |
| "eval_samples_per_second": 19.992, | |
| "eval_steps_per_second": 2.499, | |
| "eval_wer": 0.305370254835149, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.3946251768033946, | |
| "eval_loss": 0.396011620759964, | |
| "eval_runtime": 280.8684, | |
| "eval_samples_per_second": 20.138, | |
| "eval_steps_per_second": 2.517, | |
| "eval_wer": 0.2990353697749196, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "grad_norm": 0.5122537612915039, | |
| "learning_rate": 1.9499999999999995e-06, | |
| "loss": 0.4232, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "eval_loss": 0.3903259038925171, | |
| "eval_runtime": 282.302, | |
| "eval_samples_per_second": 20.035, | |
| "eval_steps_per_second": 2.504, | |
| "eval_wer": 0.2956919582153541, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "step": 2500, | |
| "total_flos": 6.538015641955614e+19, | |
| "train_loss": 0.8184805297851563, | |
| "train_runtime": 14471.5924, | |
| "train_samples_per_second": 5.528, | |
| "train_steps_per_second": 0.173 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 400, | |
| "total_flos": 6.538015641955614e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |