| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.3166023166023164, | |
| "eval_steps": 200, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07722007722007722, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 223.773, | |
| "eval_samples_per_second": 31.38, | |
| "eval_steps_per_second": 0.492, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15444015444015444, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 201.0458, | |
| "eval_samples_per_second": 34.927, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 0.8963291911755158, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.19305019305019305, | |
| "grad_norm": 4.3846588134765625, | |
| "learning_rate": 0.00024799999999999996, | |
| "loss": 3.9177, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.23166023166023167, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 194.7647, | |
| "eval_samples_per_second": 36.054, | |
| "eval_steps_per_second": 0.565, | |
| "eval_wer": 0.7594821653487074, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.3088803088803089, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 194.859, | |
| "eval_samples_per_second": 36.036, | |
| "eval_steps_per_second": 0.565, | |
| "eval_wer": 0.7512108345344293, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.3861003861003861, | |
| "grad_norm": 3.36423921585083, | |
| "learning_rate": 0.00027805555555555553, | |
| "loss": 0.9791, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3861003861003861, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 195.6019, | |
| "eval_samples_per_second": 35.899, | |
| "eval_steps_per_second": 0.562, | |
| "eval_wer": 0.598444418474803, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46332046332046334, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 197.8836, | |
| "eval_samples_per_second": 35.486, | |
| "eval_steps_per_second": 0.556, | |
| "eval_wer": 0.5867710646254528, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 203.8782, | |
| "eval_samples_per_second": 34.442, | |
| "eval_steps_per_second": 0.54, | |
| "eval_wer": 0.5255476348149014, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.5791505791505791, | |
| "grad_norm": 2.6850786209106445, | |
| "learning_rate": 0.0002503333333333333, | |
| "loss": 0.805, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6177606177606177, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 199.2388, | |
| "eval_samples_per_second": 35.244, | |
| "eval_steps_per_second": 0.552, | |
| "eval_wer": 0.5281575837845559, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.694980694980695, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 194.9574, | |
| "eval_samples_per_second": 36.018, | |
| "eval_steps_per_second": 0.564, | |
| "eval_wer": 0.4768805266643294, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.7722007722007722, | |
| "grad_norm": 2.9242658615112305, | |
| "learning_rate": 0.0002226111111111111, | |
| "loss": 0.7184, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7722007722007722, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 198.941, | |
| "eval_samples_per_second": 35.297, | |
| "eval_steps_per_second": 0.553, | |
| "eval_wer": 0.4743095321569086, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8494208494208494, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 207.8762, | |
| "eval_samples_per_second": 33.78, | |
| "eval_steps_per_second": 0.529, | |
| "eval_wer": 0.46802487891654654, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.9266409266409267, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 215.534, | |
| "eval_samples_per_second": 32.58, | |
| "eval_steps_per_second": 0.51, | |
| "eval_wer": 0.457026735745913, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.9652509652509652, | |
| "grad_norm": 3.865280866622925, | |
| "learning_rate": 0.00019483333333333332, | |
| "loss": 0.6704, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0038610038610039, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 212.4212, | |
| "eval_samples_per_second": 33.057, | |
| "eval_steps_per_second": 0.518, | |
| "eval_wer": 0.4252528793840001, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 211.2545, | |
| "eval_samples_per_second": 33.24, | |
| "eval_steps_per_second": 0.521, | |
| "eval_wer": 0.4163972316362173, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.1583011583011582, | |
| "grad_norm": 0.6646206974983215, | |
| "learning_rate": 0.00016716666666666665, | |
| "loss": 0.5664, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.1583011583011582, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 202.1583, | |
| "eval_samples_per_second": 34.735, | |
| "eval_steps_per_second": 0.544, | |
| "eval_wer": 0.41592977808941345, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.2355212355212355, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 198.8952, | |
| "eval_samples_per_second": 35.305, | |
| "eval_steps_per_second": 0.553, | |
| "eval_wer": 0.3995039798475582, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.3127413127413128, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 197.172, | |
| "eval_samples_per_second": 35.614, | |
| "eval_steps_per_second": 0.558, | |
| "eval_wer": 0.3940633399555919, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 1.06748366355896, | |
| "learning_rate": 0.00013944444444444442, | |
| "loss": 0.5359, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.3899613899613898, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 194.4658, | |
| "eval_samples_per_second": 36.109, | |
| "eval_steps_per_second": 0.566, | |
| "eval_wer": 0.38185760845571526, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.4671814671814671, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 201.7117, | |
| "eval_samples_per_second": 34.812, | |
| "eval_steps_per_second": 0.545, | |
| "eval_wer": 0.3810785192110423, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "grad_norm": 0.8601678013801575, | |
| "learning_rate": 0.00011166666666666667, | |
| "loss": 0.5172, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 196.5164, | |
| "eval_samples_per_second": 35.732, | |
| "eval_steps_per_second": 0.56, | |
| "eval_wer": 0.36905457520158935, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 193.7969, | |
| "eval_samples_per_second": 36.234, | |
| "eval_steps_per_second": 0.568, | |
| "eval_wer": 0.36086115331177854, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.698841698841699, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 196.3411, | |
| "eval_samples_per_second": 35.764, | |
| "eval_steps_per_second": 0.56, | |
| "eval_wer": 0.3599652006804046, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.7374517374517375, | |
| "grad_norm": 0.6527121663093567, | |
| "learning_rate": 8.394444444444443e-05, | |
| "loss": 0.4817, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.776061776061776, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 194.384, | |
| "eval_samples_per_second": 36.124, | |
| "eval_steps_per_second": 0.566, | |
| "eval_wer": 0.35086284133847534, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.8532818532818531, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 196.7828, | |
| "eval_samples_per_second": 35.684, | |
| "eval_steps_per_second": 0.559, | |
| "eval_wer": 0.3529663822990924, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.9305019305019306, | |
| "grad_norm": 0.7631692886352539, | |
| "learning_rate": 5.6166666666666665e-05, | |
| "loss": 0.4818, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.9305019305019306, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 194.6791, | |
| "eval_samples_per_second": 36.07, | |
| "eval_steps_per_second": 0.565, | |
| "eval_wer": 0.34340955423110386, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0077220077220077, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 200.454, | |
| "eval_samples_per_second": 35.03, | |
| "eval_steps_per_second": 0.549, | |
| "eval_wer": 0.336254918000857, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.0849420849420848, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 193.7233, | |
| "eval_samples_per_second": 36.248, | |
| "eval_steps_per_second": 0.568, | |
| "eval_wer": 0.33718982509446455, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.1235521235521237, | |
| "grad_norm": 1.005771279335022, | |
| "learning_rate": 2.844444444444444e-05, | |
| "loss": 0.4196, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 198.8274, | |
| "eval_samples_per_second": 35.317, | |
| "eval_steps_per_second": 0.553, | |
| "eval_wer": 0.3320348512588784, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.2393822393822393, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 197.3434, | |
| "eval_samples_per_second": 35.583, | |
| "eval_steps_per_second": 0.557, | |
| "eval_wer": 0.3292690844402893, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.3166023166023164, | |
| "grad_norm": 1.702697515487671, | |
| "learning_rate": 7.222222222222222e-07, | |
| "loss": 0.3743, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.3166023166023164, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 192.6595, | |
| "eval_samples_per_second": 36.448, | |
| "eval_steps_per_second": 0.571, | |
| "eval_wer": 0.3263604845935102, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.3166023166023164, | |
| "step": 6000, | |
| "total_flos": 1.2607274134194512e+19, | |
| "train_loss": 0.8722912038167318, | |
| "train_runtime": 9671.7075, | |
| "train_samples_per_second": 9.926, | |
| "train_steps_per_second": 0.62 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 1.2607274134194512e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |