| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "eval_steps": 780, | |
| "global_step": 7800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 19.43649673461914, | |
| "learning_rate": 9e-06, | |
| "loss": 2.648, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.4339828491210938, | |
| "eval_runtime": 9.7661, | |
| "eval_samples_per_second": 254.862, | |
| "eval_steps_per_second": 2.56, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.434560775756836, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.797, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.5685608386993408, | |
| "eval_runtime": 9.4627, | |
| "eval_samples_per_second": 263.033, | |
| "eval_steps_per_second": 2.642, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 8.09458065032959, | |
| "learning_rate": 7e-06, | |
| "loss": 0.3112, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 1.2473446130752563, | |
| "eval_runtime": 9.4652, | |
| "eval_samples_per_second": 262.964, | |
| "eval_steps_per_second": 2.641, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 3.432204008102417, | |
| "learning_rate": 6e-06, | |
| "loss": 0.1758, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 1.0605839490890503, | |
| "eval_runtime": 9.5192, | |
| "eval_samples_per_second": 261.472, | |
| "eval_steps_per_second": 2.626, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 2.8126320838928223, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1176, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 0.946902871131897, | |
| "eval_runtime": 9.5137, | |
| "eval_samples_per_second": 261.623, | |
| "eval_steps_per_second": 2.628, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 2.959984064102173, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.09, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 0.8606055974960327, | |
| "eval_runtime": 9.4719, | |
| "eval_samples_per_second": 262.778, | |
| "eval_steps_per_second": 2.639, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 1.7250601053237915, | |
| "learning_rate": 3e-06, | |
| "loss": 0.0733, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_loss": 0.8264588713645935, | |
| "eval_runtime": 9.5045, | |
| "eval_samples_per_second": 261.876, | |
| "eval_steps_per_second": 2.63, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 2.693885326385498, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.0631, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_loss": 0.7703806161880493, | |
| "eval_runtime": 9.4961, | |
| "eval_samples_per_second": 262.107, | |
| "eval_steps_per_second": 2.633, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 0.8978404998779297, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.0576, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_loss": 0.7507320046424866, | |
| "eval_runtime": 9.5204, | |
| "eval_samples_per_second": 261.439, | |
| "eval_steps_per_second": 2.626, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 2.5377581119537354, | |
| "learning_rate": 0.0, | |
| "loss": 0.0545, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_loss": 0.743861734867096, | |
| "eval_runtime": 9.509, | |
| "eval_samples_per_second": 261.751, | |
| "eval_steps_per_second": 2.629, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 7800, | |
| "total_flos": 1.31699559685632e+17, | |
| "train_loss": 0.43882344857240335, | |
| "train_runtime": 4515.1061, | |
| "train_samples_per_second": 110.252, | |
| "train_steps_per_second": 1.728 | |
| } | |
| ], | |
| "logging_steps": 780, | |
| "max_steps": 7800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "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": 1.31699559685632e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |