| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 18.823529411764707, | |
| "global_step": 5600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 0.8018701076507568, | |
| "eval_runtime": 123.0686, | |
| "eval_samples_per_second": 17.892, | |
| "eval_steps_per_second": 2.243, | |
| "eval_wer": 0.8813849519642347, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0007940000000000001, | |
| "loss": 1.228, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_loss": 1.0962814092636108, | |
| "eval_runtime": 123.3569, | |
| "eval_samples_per_second": 17.851, | |
| "eval_steps_per_second": 2.237, | |
| "eval_wer": 0.9425473223627889, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_loss": 1.104669451713562, | |
| "eval_runtime": 123.8837, | |
| "eval_samples_per_second": 17.775, | |
| "eval_steps_per_second": 2.228, | |
| "eval_wer": 0.9653762008941311, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0009454044117647059, | |
| "loss": 0.7814, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_loss": 0.9275535345077515, | |
| "eval_runtime": 124.6162, | |
| "eval_samples_per_second": 17.67, | |
| "eval_steps_per_second": 2.215, | |
| "eval_wer": 0.9294207172072672, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_loss": 0.8205996751785278, | |
| "eval_runtime": 124.5979, | |
| "eval_samples_per_second": 17.673, | |
| "eval_steps_per_second": 2.215, | |
| "eval_wer": 0.9016455816608009, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.0008718749999999999, | |
| "loss": 0.617, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_loss": 0.7664089798927307, | |
| "eval_runtime": 124.9893, | |
| "eval_samples_per_second": 17.618, | |
| "eval_steps_per_second": 2.208, | |
| "eval_wer": 0.9076381622752783, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "eval_loss": 0.8046385049819946, | |
| "eval_runtime": 124.9226, | |
| "eval_samples_per_second": 17.627, | |
| "eval_steps_per_second": 2.209, | |
| "eval_wer": 0.902787025587368, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.0007983455882352941, | |
| "loss": 0.5312, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "eval_loss": 0.7806461453437805, | |
| "eval_runtime": 125.1179, | |
| "eval_samples_per_second": 17.599, | |
| "eval_steps_per_second": 2.206, | |
| "eval_wer": 0.9512983924664701, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_loss": 0.7088819146156311, | |
| "eval_runtime": 127.0024, | |
| "eval_samples_per_second": 17.338, | |
| "eval_steps_per_second": 2.173, | |
| "eval_wer": 0.8616950442309521, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.0007248161764705882, | |
| "loss": 0.4404, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_loss": 0.7011399269104004, | |
| "eval_runtime": 125.0855, | |
| "eval_samples_per_second": 17.604, | |
| "eval_steps_per_second": 2.206, | |
| "eval_wer": 0.8952725197374679, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "eval_loss": 0.6890342235565186, | |
| "eval_runtime": 125.964, | |
| "eval_samples_per_second": 17.481, | |
| "eval_steps_per_second": 2.191, | |
| "eval_wer": 0.8790069437838867, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.0006512867647058823, | |
| "loss": 0.3961, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "eval_loss": 0.6665045619010925, | |
| "eval_runtime": 125.5126, | |
| "eval_samples_per_second": 17.544, | |
| "eval_steps_per_second": 2.199, | |
| "eval_wer": 0.8834775991629411, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "eval_loss": 0.6596990823745728, | |
| "eval_runtime": 125.1836, | |
| "eval_samples_per_second": 17.59, | |
| "eval_steps_per_second": 2.205, | |
| "eval_wer": 0.8537049367449824, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 0.0005777573529411765, | |
| "loss": 0.3402, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "eval_loss": 0.6772456765174866, | |
| "eval_runtime": 125.8045, | |
| "eval_samples_per_second": 17.503, | |
| "eval_steps_per_second": 2.194, | |
| "eval_wer": 0.8308760582136403, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.08, | |
| "eval_loss": 0.6295859813690186, | |
| "eval_runtime": 125.3009, | |
| "eval_samples_per_second": 17.574, | |
| "eval_steps_per_second": 2.203, | |
| "eval_wer": 0.8315419005041378, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 0.0005042279411764706, | |
| "loss": 0.2939, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "eval_loss": 0.6442817449569702, | |
| "eval_runtime": 125.5719, | |
| "eval_samples_per_second": 17.536, | |
| "eval_steps_per_second": 2.198, | |
| "eval_wer": 0.8326833444307048, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "eval_loss": 0.5965331196784973, | |
| "eval_runtime": 126.3146, | |
| "eval_samples_per_second": 17.433, | |
| "eval_steps_per_second": 2.185, | |
| "eval_wer": 0.8219347474555313, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.00043069852941176473, | |
| "loss": 0.2325, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "eval_loss": 0.613932728767395, | |
| "eval_runtime": 125.4233, | |
| "eval_samples_per_second": 17.557, | |
| "eval_steps_per_second": 2.201, | |
| "eval_wer": 0.8033862836488157, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "eval_loss": 0.5964781641960144, | |
| "eval_runtime": 125.9783, | |
| "eval_samples_per_second": 17.479, | |
| "eval_steps_per_second": 2.191, | |
| "eval_wer": 0.8020545990678208, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 0.0003571691176470588, | |
| "loss": 0.1823, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "eval_loss": 0.640591561794281, | |
| "eval_runtime": 125.8961, | |
| "eval_samples_per_second": 17.491, | |
| "eval_steps_per_second": 2.192, | |
| "eval_wer": 0.8089032626272235, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "eval_loss": 0.6114814877510071, | |
| "eval_runtime": 126.0685, | |
| "eval_samples_per_second": 17.467, | |
| "eval_steps_per_second": 2.189, | |
| "eval_wer": 0.797964424997622, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 0.0002836397058823529, | |
| "loss": 0.138, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "eval_loss": 0.6256727576255798, | |
| "eval_runtime": 126.3311, | |
| "eval_samples_per_second": 17.43, | |
| "eval_steps_per_second": 2.185, | |
| "eval_wer": 0.7777989156282697, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 15.46, | |
| "eval_loss": 0.6141476035118103, | |
| "eval_runtime": 126.0957, | |
| "eval_samples_per_second": 17.463, | |
| "eval_steps_per_second": 2.189, | |
| "eval_wer": 0.7941596119090649, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 0.00021011029411764707, | |
| "loss": 0.0784, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "eval_loss": 0.6295885443687439, | |
| "eval_runtime": 126.3665, | |
| "eval_samples_per_second": 17.425, | |
| "eval_steps_per_second": 2.184, | |
| "eval_wer": 0.7922572053647865, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.81, | |
| "eval_loss": 0.5918190479278564, | |
| "eval_runtime": 126.3879, | |
| "eval_samples_per_second": 17.423, | |
| "eval_steps_per_second": 2.184, | |
| "eval_wer": 0.7728526586131457, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "learning_rate": 0.00013658088235294117, | |
| "loss": 0.0344, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "eval_loss": 0.5919037461280823, | |
| "eval_runtime": 126.2976, | |
| "eval_samples_per_second": 17.435, | |
| "eval_steps_per_second": 2.185, | |
| "eval_wer": 0.7636259868733949, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 18.15, | |
| "eval_loss": 0.576080858707428, | |
| "eval_runtime": 126.7739, | |
| "eval_samples_per_second": 17.37, | |
| "eval_steps_per_second": 2.177, | |
| "eval_wer": 0.7495481784457338, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "learning_rate": 6.30514705882353e-05, | |
| "loss": -0.0038, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "eval_loss": 0.5868940949440002, | |
| "eval_runtime": 126.4596, | |
| "eval_samples_per_second": 17.413, | |
| "eval_steps_per_second": 2.183, | |
| "eval_wer": 0.7588699705126986, | |
| "step": 5600 | |
| } | |
| ], | |
| "max_steps": 5940, | |
| "num_train_epochs": 20, | |
| "total_flos": 4.6199251281753915e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |