| { | |
| "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": 11.533838272094727, | |
| "learning_rate": 0.0002465, | |
| "loss": 4.6618, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03206361421059382, | |
| "eval_loss": 1.5996026992797852, | |
| "eval_runtime": 185.1067, | |
| "eval_samples_per_second": 37.935, | |
| "eval_steps_per_second": 0.594, | |
| "eval_wer": 0.9161163448889834, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06412722842118763, | |
| "grad_norm": 4.801280975341797, | |
| "learning_rate": 0.0002874574468085106, | |
| "loss": 1.0278, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06412722842118763, | |
| "eval_loss": 1.1463252305984497, | |
| "eval_runtime": 184.8935, | |
| "eval_samples_per_second": 37.979, | |
| "eval_steps_per_second": 0.595, | |
| "eval_wer": 0.7792157748574922, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09619084263178146, | |
| "grad_norm": 4.746730327606201, | |
| "learning_rate": 0.0002715, | |
| "loss": 0.8164, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.09619084263178146, | |
| "eval_loss": 1.0590689182281494, | |
| "eval_runtime": 185.3865, | |
| "eval_samples_per_second": 37.878, | |
| "eval_steps_per_second": 0.593, | |
| "eval_wer": 0.7363238948165668, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.12825445684237527, | |
| "grad_norm": 21.847057342529297, | |
| "learning_rate": 0.00025554255319148935, | |
| "loss": 0.7124, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.12825445684237527, | |
| "eval_loss": 0.9373884797096252, | |
| "eval_runtime": 186.5741, | |
| "eval_samples_per_second": 37.637, | |
| "eval_steps_per_second": 0.59, | |
| "eval_wer": 0.6622596632960842, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16031807105296908, | |
| "grad_norm": 4.238761901855469, | |
| "learning_rate": 0.0002395851063829787, | |
| "loss": 0.6566, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.16031807105296908, | |
| "eval_loss": 0.8721805810928345, | |
| "eval_runtime": 187.0392, | |
| "eval_samples_per_second": 37.543, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.615168949893036, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.19238168526356292, | |
| "grad_norm": 5.65878963470459, | |
| "learning_rate": 0.00022362765957446805, | |
| "loss": 0.6101, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.19238168526356292, | |
| "eval_loss": 0.8194323182106018, | |
| "eval_runtime": 187.1888, | |
| "eval_samples_per_second": 37.513, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.5927264513214366, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22444529947415673, | |
| "grad_norm": 6.294471740722656, | |
| "learning_rate": 0.00020767021276595744, | |
| "loss": 0.5777, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.22444529947415673, | |
| "eval_loss": 0.7799355387687683, | |
| "eval_runtime": 187.6349, | |
| "eval_samples_per_second": 37.424, | |
| "eval_steps_per_second": 0.586, | |
| "eval_wer": 0.5707490134070343, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.25650891368475054, | |
| "grad_norm": 3.848400115966797, | |
| "learning_rate": 0.0001917446808510638, | |
| "loss": 0.5431, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.25650891368475054, | |
| "eval_loss": 0.7453157901763916, | |
| "eval_runtime": 187.5991, | |
| "eval_samples_per_second": 37.431, | |
| "eval_steps_per_second": 0.586, | |
| "eval_wer": 0.550233194700966, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.2885725278953444, | |
| "grad_norm": 9.198270797729492, | |
| "learning_rate": 0.00017585106382978722, | |
| "loss": 0.512, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.2885725278953444, | |
| "eval_loss": 0.7209311127662659, | |
| "eval_runtime": 188.2502, | |
| "eval_samples_per_second": 37.301, | |
| "eval_steps_per_second": 0.584, | |
| "eval_wer": 0.5346735938558843, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.32063614210593816, | |
| "grad_norm": 7.854990482330322, | |
| "learning_rate": 0.00015989361702127658, | |
| "loss": 0.4953, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32063614210593816, | |
| "eval_loss": 0.6942155957221985, | |
| "eval_runtime": 187.3133, | |
| "eval_samples_per_second": 37.488, | |
| "eval_steps_per_second": 0.587, | |
| "eval_wer": 0.5225022920846676, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.352699756316532, | |
| "grad_norm": 10.654298782348633, | |
| "learning_rate": 0.00014393617021276595, | |
| "loss": 0.4746, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.352699756316532, | |
| "eval_loss": 0.6680377721786499, | |
| "eval_runtime": 188.6546, | |
| "eval_samples_per_second": 37.221, | |
| "eval_steps_per_second": 0.583, | |
| "eval_wer": 0.495714798230112, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.38476337052712584, | |
| "grad_norm": 19.936861038208008, | |
| "learning_rate": 0.00012801063829787234, | |
| "loss": 0.4535, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38476337052712584, | |
| "eval_loss": 0.6434958577156067, | |
| "eval_runtime": 188.0277, | |
| "eval_samples_per_second": 37.346, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.47308627539563375, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.4168269847377196, | |
| "grad_norm": 6.950649261474609, | |
| "learning_rate": 0.0001120531914893617, | |
| "loss": 0.4249, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.4168269847377196, | |
| "eval_loss": 0.6293027400970459, | |
| "eval_runtime": 187.8772, | |
| "eval_samples_per_second": 37.375, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.4816832538301067, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.44889059894831346, | |
| "grad_norm": 5.574142932891846, | |
| "learning_rate": 9.612765957446806e-05, | |
| "loss": 0.4065, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.44889059894831346, | |
| "eval_loss": 0.5997486710548401, | |
| "eval_runtime": 188.7633, | |
| "eval_samples_per_second": 37.2, | |
| "eval_steps_per_second": 0.583, | |
| "eval_wer": 0.44947448145736724, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.4809542131589073, | |
| "grad_norm": 5.602737903594971, | |
| "learning_rate": 8.017021276595744e-05, | |
| "loss": 0.393, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.4809542131589073, | |
| "eval_loss": 0.5802670121192932, | |
| "eval_runtime": 189.0166, | |
| "eval_samples_per_second": 37.15, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.44260487117819797, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.5130178273695011, | |
| "grad_norm": 15.019088745117188, | |
| "learning_rate": 6.424468085106383e-05, | |
| "loss": 0.3808, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.5130178273695011, | |
| "eval_loss": 0.5661880970001221, | |
| "eval_runtime": 189.1322, | |
| "eval_samples_per_second": 37.127, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.4299153589603901, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.5450814415800949, | |
| "grad_norm": 3.79109525680542, | |
| "learning_rate": 4.8287234042553194e-05, | |
| "loss": 0.3722, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.5450814415800949, | |
| "eval_loss": 0.553141176700592, | |
| "eval_runtime": 188.8506, | |
| "eval_samples_per_second": 37.183, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.4194182755550831, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.5771450557906888, | |
| "grad_norm": 12.16059398651123, | |
| "learning_rate": 3.232978723404255e-05, | |
| "loss": 0.3622, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.5771450557906888, | |
| "eval_loss": 0.5399113297462463, | |
| "eval_runtime": 188.6804, | |
| "eval_samples_per_second": 37.216, | |
| "eval_steps_per_second": 0.583, | |
| "eval_wer": 0.40727354867856336, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.6092086700012825, | |
| "grad_norm": 7.863190650939941, | |
| "learning_rate": 1.6372340425531912e-05, | |
| "loss": 0.3526, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.6092086700012825, | |
| "eval_loss": 0.5277913808822632, | |
| "eval_runtime": 188.1946, | |
| "eval_samples_per_second": 37.312, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.40279567892212226, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.6412722842118763, | |
| "grad_norm": 5.124056339263916, | |
| "learning_rate": 4.1489361702127654e-07, | |
| "loss": 0.3337, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.6412722842118763, | |
| "eval_loss": 0.5226185917854309, | |
| "eval_runtime": 188.4518, | |
| "eval_samples_per_second": 37.262, | |
| "eval_steps_per_second": 0.584, | |
| "eval_wer": 0.40009832711037885, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.6412722842118763, | |
| "step": 10000, | |
| "total_flos": 1.1393778193380235e+19, | |
| "train_loss": 0.7283544036865235, | |
| "train_runtime": 7737.7643, | |
| "train_samples_per_second": 10.339, | |
| "train_steps_per_second": 1.292 | |
| } | |
| ], | |
| "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 | |
| } | |