| { | |
| "best_metric": 2.166426420211792, | |
| "best_model_checkpoint": "./tomrp/checkpoint-15", | |
| "epoch": 5.0, | |
| "eval_steps": 3, | |
| "global_step": 15, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 5e-06, | |
| "loss": 2.779, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 2.6370952129364014, | |
| "eval_runtime": 2.2587, | |
| "eval_samples_per_second": 12.397, | |
| "eval_steps_per_second": 0.885, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1e-05, | |
| "loss": 2.8308, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.8055, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.6105611324310303, | |
| "eval_runtime": 2.2632, | |
| "eval_samples_per_second": 12.372, | |
| "eval_steps_per_second": 0.884, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2e-05, | |
| "loss": 2.7115, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.8004, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3e-05, | |
| "loss": 2.4321, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.45676589012146, | |
| "eval_runtime": 2.2658, | |
| "eval_samples_per_second": 12.358, | |
| "eval_steps_per_second": 0.883, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 3.5e-05, | |
| "loss": 2.5158, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 4e-05, | |
| "loss": 2.2594, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.0601, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.2703487873077393, | |
| "eval_runtime": 2.2646, | |
| "eval_samples_per_second": 12.364, | |
| "eval_steps_per_second": 0.883, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9706, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 1.7659, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 2.0104, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.1692585945129395, | |
| "eval_runtime": 2.2637, | |
| "eval_samples_per_second": 12.369, | |
| "eval_steps_per_second": 0.884, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 1.5315, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 1.4955, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 1.5786, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.166426420211792, | |
| "eval_runtime": 2.2642, | |
| "eval_samples_per_second": 12.367, | |
| "eval_steps_per_second": 0.883, | |
| "step": 15 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 3, | |
| "total_flos": 3399415706419200.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |