| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6412722842118763, | |
| "eval_steps": 500, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03206361421059382, | |
| "grad_norm": 8.037480354309082, | |
| "learning_rate": 0.0002465, | |
| "loss": 4.7051, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03206361421059382, | |
| "eval_loss": 1.7504417896270752, | |
| "eval_runtime": 184.6482, | |
| "eval_samples_per_second": 38.029, | |
| "eval_steps_per_second": 0.596, | |
| "eval_wer": 0.9570416827223323, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06412722842118763, | |
| "grad_norm": 7.763198375701904, | |
| "learning_rate": 0.0002874574468085106, | |
| "loss": 1.0409, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06412722842118763, | |
| "eval_loss": 1.1511156558990479, | |
| "eval_runtime": 184.729, | |
| "eval_samples_per_second": 38.012, | |
| "eval_steps_per_second": 0.595, | |
| "eval_wer": 0.776093224730597, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09619084263178146, | |
| "grad_norm": 3.2267072200775146, | |
| "learning_rate": 0.0002715, | |
| "loss": 0.8183, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.09619084263178146, | |
| "eval_loss": 1.0506497621536255, | |
| "eval_runtime": 186.2316, | |
| "eval_samples_per_second": 37.706, | |
| "eval_steps_per_second": 0.591, | |
| "eval_wer": 0.70972242522489, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.12825445684237527, | |
| "grad_norm": 6.372620582580566, | |
| "learning_rate": 0.00025554255319148935, | |
| "loss": 0.7091, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.12825445684237527, | |
| "eval_loss": 0.9421387314796448, | |
| "eval_runtime": 186.5946, | |
| "eval_samples_per_second": 37.632, | |
| "eval_steps_per_second": 0.59, | |
| "eval_wer": 0.6609707809032807, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16031807105296908, | |
| "grad_norm": 5.675894260406494, | |
| "learning_rate": 0.0002395851063829787, | |
| "loss": 0.6547, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.16031807105296908, | |
| "eval_loss": 0.8725515007972717, | |
| "eval_runtime": 187.2013, | |
| "eval_samples_per_second": 37.51, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.6127639219229594, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.19238168526356292, | |
| "grad_norm": 6.913870334625244, | |
| "learning_rate": 0.00022362765957446805, | |
| "loss": 0.6088, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.19238168526356292, | |
| "eval_loss": 0.8246235847473145, | |
| "eval_runtime": 188.3497, | |
| "eval_samples_per_second": 37.282, | |
| "eval_steps_per_second": 0.584, | |
| "eval_wer": 0.5989582641278784, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22444529947415673, | |
| "grad_norm": 4.30249547958374, | |
| "learning_rate": 0.00020767021276595744, | |
| "loss": 0.5781, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.22444529947415673, | |
| "eval_loss": 0.802536129951477, | |
| "eval_runtime": 187.0791, | |
| "eval_samples_per_second": 37.535, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.5747352476115813, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.25650891368475054, | |
| "grad_norm": 3.4820008277893066, | |
| "learning_rate": 0.0001917446808510638, | |
| "loss": 0.5429, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.25650891368475054, | |
| "eval_loss": 0.7359501123428345, | |
| "eval_runtime": 186.7747, | |
| "eval_samples_per_second": 37.596, | |
| "eval_steps_per_second": 0.589, | |
| "eval_wer": 0.53048804794111, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.2885725278953444, | |
| "grad_norm": 11.696717262268066, | |
| "learning_rate": 0.00017585106382978722, | |
| "loss": 0.5104, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.2885725278953444, | |
| "eval_loss": 0.7335178852081299, | |
| "eval_runtime": 187.3685, | |
| "eval_samples_per_second": 37.477, | |
| "eval_steps_per_second": 0.587, | |
| "eval_wer": 0.5394039251119468, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.32063614210593816, | |
| "grad_norm": 7.053103446960449, | |
| "learning_rate": 0.00015989361702127658, | |
| "loss": 0.501, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32063614210593816, | |
| "eval_loss": 0.6932825446128845, | |
| "eval_runtime": 186.2726, | |
| "eval_samples_per_second": 37.697, | |
| "eval_steps_per_second": 0.591, | |
| "eval_wer": 0.5087763589736776, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.352699756316532, | |
| "grad_norm": 6.128586769104004, | |
| "learning_rate": 0.00014393617021276595, | |
| "loss": 0.4708, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.352699756316532, | |
| "eval_loss": 0.6770374774932861, | |
| "eval_runtime": 188.2655, | |
| "eval_samples_per_second": 37.298, | |
| "eval_steps_per_second": 0.584, | |
| "eval_wer": 0.5112743990751937, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.38476337052712584, | |
| "grad_norm": 7.154539108276367, | |
| "learning_rate": 0.00012801063829787234, | |
| "loss": 0.4526, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38476337052712584, | |
| "eval_loss": 0.6608560681343079, | |
| "eval_runtime": 187.3283, | |
| "eval_samples_per_second": 37.485, | |
| "eval_steps_per_second": 0.587, | |
| "eval_wer": 0.48059368314753054, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.4168269847377196, | |
| "grad_norm": 5.313536643981934, | |
| "learning_rate": 0.0001120531914893617, | |
| "loss": 0.4235, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.4168269847377196, | |
| "eval_loss": 0.637322187423706, | |
| "eval_runtime": 186.315, | |
| "eval_samples_per_second": 37.689, | |
| "eval_steps_per_second": 0.59, | |
| "eval_wer": 0.485842224850184, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.44889059894831346, | |
| "grad_norm": 6.399425983428955, | |
| "learning_rate": 9.612765957446806e-05, | |
| "loss": 0.4032, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.44889059894831346, | |
| "eval_loss": 0.6047533750534058, | |
| "eval_runtime": 186.8155, | |
| "eval_samples_per_second": 37.588, | |
| "eval_steps_per_second": 0.589, | |
| "eval_wer": 0.4466176802774419, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.4809542131589073, | |
| "grad_norm": 11.48141098022461, | |
| "learning_rate": 8.017021276595744e-05, | |
| "loss": 0.3863, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.4809542131589073, | |
| "eval_loss": 0.5946004390716553, | |
| "eval_runtime": 186.2938, | |
| "eval_samples_per_second": 37.693, | |
| "eval_steps_per_second": 0.59, | |
| "eval_wer": 0.4432160937562285, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.5130178273695011, | |
| "grad_norm": 32.89252471923828, | |
| "learning_rate": 6.424468085106383e-05, | |
| "loss": 0.3766, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.5130178273695011, | |
| "eval_loss": 0.5737225413322449, | |
| "eval_runtime": 186.9085, | |
| "eval_samples_per_second": 37.569, | |
| "eval_steps_per_second": 0.589, | |
| "eval_wer": 0.4298489217236477, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.5450814415800949, | |
| "grad_norm": 4.741519451141357, | |
| "learning_rate": 4.8287234042553194e-05, | |
| "loss": 0.3746, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.5450814415800949, | |
| "eval_loss": 0.5668203234672546, | |
| "eval_runtime": 186.8619, | |
| "eval_samples_per_second": 37.579, | |
| "eval_steps_per_second": 0.589, | |
| "eval_wer": 0.4247731168365245, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.5771450557906888, | |
| "grad_norm": 13.890504837036133, | |
| "learning_rate": 3.232978723404255e-05, | |
| "loss": 0.3586, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.5771450557906888, | |
| "eval_loss": 0.5485312342643738, | |
| "eval_runtime": 187.9252, | |
| "eval_samples_per_second": 37.366, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.4100772000690947, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.6092086700012825, | |
| "grad_norm": 8.073569297790527, | |
| "learning_rate": 1.6372340425531912e-05, | |
| "loss": 0.3552, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.6092086700012825, | |
| "eval_loss": 0.5377594828605652, | |
| "eval_runtime": 187.0305, | |
| "eval_samples_per_second": 37.545, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.40320758978992544, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.6412722842118763, | |
| "grad_norm": 6.519000053405762, | |
| "learning_rate": 4.1489361702127654e-07, | |
| "loss": 0.3326, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.6412722842118763, | |
| "eval_loss": 0.5324302911758423, | |
| "eval_runtime": 186.1276, | |
| "eval_samples_per_second": 37.727, | |
| "eval_steps_per_second": 0.591, | |
| "eval_wer": 0.40138720950318235, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.6412722842118763, | |
| "step": 10000, | |
| "total_flos": 1.1393778193380235e+19, | |
| "train_loss": 0.73015986328125, | |
| "train_runtime": 7697.7754, | |
| "train_samples_per_second": 10.393, | |
| "train_steps_per_second": 1.299 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 1.1393778193380235e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |