| { | |
| "best_metric": 2.5737111568450928, | |
| "best_model_checkpoint": "output/aaron-watson/checkpoint-100", | |
| "epoch": 4.0, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0001240985658141214, | |
| "loss": 3.6755, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 8.97985658141214e-05, | |
| "loss": 3.521, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.740143418587861e-05, | |
| "loss": 3.4758, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 1.3101434185878613e-05, | |
| "loss": 3.2857, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0, | |
| "loss": 3.2955, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.0358455181121826, | |
| "eval_runtime": 1.5968, | |
| "eval_samples_per_second": 21.293, | |
| "eval_steps_per_second": 3.131, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.3101434185878598e-05, | |
| "loss": 3.1772, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 4.74014341858786e-05, | |
| "loss": 3.1081, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 8.979856581412138e-05, | |
| "loss": 3.3239, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.0001240985658141214, | |
| "loss": 3.0381, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001372, | |
| "loss": 3.0911, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.996838331222534, | |
| "eval_runtime": 1.5805, | |
| "eval_samples_per_second": 21.512, | |
| "eval_steps_per_second": 3.163, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 0.00012409856581412136, | |
| "loss": 3.0342, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 8.979856581412141e-05, | |
| "loss": 2.8495, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 4.7401434185878625e-05, | |
| "loss": 2.895, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 1.310143418587862e-05, | |
| "loss": 2.8209, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.7285, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.9697048664093018, | |
| "eval_runtime": 1.7446, | |
| "eval_samples_per_second": 21.208, | |
| "eval_steps_per_second": 2.866, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.310143418587859e-05, | |
| "loss": 2.8552, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 4.740143418587858e-05, | |
| "loss": 2.8477, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 8.979856581412137e-05, | |
| "loss": 2.825, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 0.00012409856581412136, | |
| "loss": 2.7383, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0001372, | |
| "loss": 2.839, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.5737111568450928, | |
| "eval_runtime": 1.5782, | |
| "eval_samples_per_second": 20.911, | |
| "eval_steps_per_second": 3.168, | |
| "step": 100 | |
| } | |
| ], | |
| "max_steps": 325, | |
| "num_train_epochs": 13, | |
| "total_flos": 102557122560000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |