| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 11223, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.331729484095162e-05, | |
| "loss": 1.7279, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3.663458968190324e-05, | |
| "loss": 1.5155, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.995188452285485e-05, | |
| "loss": 1.4301, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.326917936380647e-05, | |
| "loss": 1.3448, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.658647420475809e-05, | |
| "loss": 1.2982, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 9.903769045709705e-06, | |
| "loss": 1.2323, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.2210638866613206e-06, | |
| "loss": 1.2036, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3.217945290920431e-05, | |
| "loss": 1.1143, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 2.995188452285485e-05, | |
| "loss": 1.0045, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.772431613650539e-05, | |
| "loss": 0.9933, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 2.549674775015593e-05, | |
| "loss": 0.9654, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 2.326917936380647e-05, | |
| "loss": 0.9793, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 2.104161097745701e-05, | |
| "loss": 0.979, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 1.8814042591107548e-05, | |
| "loss": 0.9534, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.658647420475809e-05, | |
| "loss": 0.9262, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1.4358905818408625e-05, | |
| "loss": 0.6403, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.2131337432059165e-05, | |
| "loss": 0.6348, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 9.903769045709705e-06, | |
| "loss": 0.6605, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 7.676200659360243e-06, | |
| "loss": 0.6282, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 5.448632273010781e-06, | |
| "loss": 0.632, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3.2210638866613206e-06, | |
| "loss": 0.6298, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 9.934955003118596e-07, | |
| "loss": 0.6261, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 11223, | |
| "total_flos": 5.117130085328486e+16, | |
| "train_loss": 0.5631158361882754, | |
| "train_runtime": 5240.7351, | |
| "train_samples_per_second": 8.566, | |
| "train_steps_per_second": 2.141 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11223, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 5.117130085328486e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |