| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 9.927797833935019, |
| "global_step": 22000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.23, |
| "learning_rate": 1.9548736462093864e-05, |
| "loss": 1.999, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.45, |
| "learning_rate": 1.9097472924187726e-05, |
| "loss": 1.6311, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.68, |
| "learning_rate": 1.8646209386281592e-05, |
| "loss": 1.4725, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.9, |
| "learning_rate": 1.8194945848375454e-05, |
| "loss": 1.4118, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 1.260017991065979, |
| "eval_runtime": 13.7559, |
| "eval_samples_per_second": 645.178, |
| "eval_steps_per_second": 80.693, |
| "step": 2216 |
| }, |
| { |
| "epoch": 1.13, |
| "learning_rate": 1.7743682310469313e-05, |
| "loss": 1.3416, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.35, |
| "learning_rate": 1.729241877256318e-05, |
| "loss": 1.2715, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.58, |
| "learning_rate": 1.684115523465704e-05, |
| "loss": 1.215, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.81, |
| "learning_rate": 1.6389891696750904e-05, |
| "loss": 1.1616, |
| "step": 4000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 1.0639451742172241, |
| "eval_runtime": 13.601, |
| "eval_samples_per_second": 652.527, |
| "eval_steps_per_second": 81.612, |
| "step": 4432 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 1.5938628158844766e-05, |
| "loss": 1.1381, |
| "step": 4500 |
| }, |
| { |
| "epoch": 2.26, |
| "learning_rate": 1.548736462093863e-05, |
| "loss": 1.1219, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.48, |
| "learning_rate": 1.5036101083032493e-05, |
| "loss": 1.0613, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.71, |
| "learning_rate": 1.4584837545126355e-05, |
| "loss": 1.078, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.93, |
| "learning_rate": 1.413357400722022e-05, |
| "loss": 1.0446, |
| "step": 6500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.9587522149085999, |
| "eval_runtime": 13.6249, |
| "eval_samples_per_second": 651.379, |
| "eval_steps_per_second": 81.468, |
| "step": 6648 |
| }, |
| { |
| "epoch": 3.16, |
| "learning_rate": 1.368231046931408e-05, |
| "loss": 1.0442, |
| "step": 7000 |
| }, |
| { |
| "epoch": 3.38, |
| "learning_rate": 1.3231046931407942e-05, |
| "loss": 1.0028, |
| "step": 7500 |
| }, |
| { |
| "epoch": 3.61, |
| "learning_rate": 1.2779783393501806e-05, |
| "loss": 0.9996, |
| "step": 8000 |
| }, |
| { |
| "epoch": 3.84, |
| "learning_rate": 1.2328519855595669e-05, |
| "loss": 0.974, |
| "step": 8500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.8713343739509583, |
| "eval_runtime": 13.5623, |
| "eval_samples_per_second": 654.389, |
| "eval_steps_per_second": 81.845, |
| "step": 8864 |
| }, |
| { |
| "epoch": 4.06, |
| "learning_rate": 1.1877256317689531e-05, |
| "loss": 0.9667, |
| "step": 9000 |
| }, |
| { |
| "epoch": 4.29, |
| "learning_rate": 1.1425992779783395e-05, |
| "loss": 0.9457, |
| "step": 9500 |
| }, |
| { |
| "epoch": 4.51, |
| "learning_rate": 1.0974729241877257e-05, |
| "loss": 0.9311, |
| "step": 10000 |
| }, |
| { |
| "epoch": 4.74, |
| "learning_rate": 1.0523465703971121e-05, |
| "loss": 0.9321, |
| "step": 10500 |
| }, |
| { |
| "epoch": 4.96, |
| "learning_rate": 1.0072202166064984e-05, |
| "loss": 0.9109, |
| "step": 11000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.8490720987319946, |
| "eval_runtime": 13.6964, |
| "eval_samples_per_second": 647.978, |
| "eval_steps_per_second": 81.043, |
| "step": 11080 |
| }, |
| { |
| "epoch": 5.19, |
| "learning_rate": 9.620938628158846e-06, |
| "loss": 0.8969, |
| "step": 11500 |
| }, |
| { |
| "epoch": 5.42, |
| "learning_rate": 9.169675090252709e-06, |
| "loss": 0.8924, |
| "step": 12000 |
| }, |
| { |
| "epoch": 5.64, |
| "learning_rate": 8.718411552346571e-06, |
| "loss": 0.8812, |
| "step": 12500 |
| }, |
| { |
| "epoch": 5.87, |
| "learning_rate": 8.267148014440433e-06, |
| "loss": 0.8651, |
| "step": 13000 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.7949955463409424, |
| "eval_runtime": 13.8508, |
| "eval_samples_per_second": 640.758, |
| "eval_steps_per_second": 80.14, |
| "step": 13296 |
| }, |
| { |
| "epoch": 6.09, |
| "learning_rate": 7.815884476534297e-06, |
| "loss": 0.8299, |
| "step": 13500 |
| }, |
| { |
| "epoch": 6.32, |
| "learning_rate": 7.36462093862816e-06, |
| "loss": 0.8698, |
| "step": 14000 |
| }, |
| { |
| "epoch": 6.54, |
| "learning_rate": 6.913357400722023e-06, |
| "loss": 0.8441, |
| "step": 14500 |
| }, |
| { |
| "epoch": 6.77, |
| "learning_rate": 6.4620938628158845e-06, |
| "loss": 0.8028, |
| "step": 15000 |
| }, |
| { |
| "epoch": 6.99, |
| "learning_rate": 6.010830324909748e-06, |
| "loss": 0.8449, |
| "step": 15500 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.7730704545974731, |
| "eval_runtime": 13.7429, |
| "eval_samples_per_second": 645.79, |
| "eval_steps_per_second": 80.769, |
| "step": 15512 |
| }, |
| { |
| "epoch": 7.22, |
| "learning_rate": 5.559566787003611e-06, |
| "loss": 0.8188, |
| "step": 16000 |
| }, |
| { |
| "epoch": 7.45, |
| "learning_rate": 5.108303249097474e-06, |
| "loss": 0.8166, |
| "step": 16500 |
| }, |
| { |
| "epoch": 7.67, |
| "learning_rate": 4.6570397111913365e-06, |
| "loss": 0.8159, |
| "step": 17000 |
| }, |
| { |
| "epoch": 7.9, |
| "learning_rate": 4.205776173285199e-06, |
| "loss": 0.8304, |
| "step": 17500 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.7481271028518677, |
| "eval_runtime": 7.0241, |
| "eval_samples_per_second": 1263.503, |
| "eval_steps_per_second": 158.027, |
| "step": 17728 |
| }, |
| { |
| "epoch": 8.12, |
| "learning_rate": 3.7545126353790616e-06, |
| "loss": 0.815, |
| "step": 18000 |
| }, |
| { |
| "epoch": 8.35, |
| "learning_rate": 3.303249097472924e-06, |
| "loss": 0.7999, |
| "step": 18500 |
| }, |
| { |
| "epoch": 8.57, |
| "learning_rate": 2.8519855595667872e-06, |
| "loss": 0.7977, |
| "step": 19000 |
| }, |
| { |
| "epoch": 8.8, |
| "learning_rate": 2.40072202166065e-06, |
| "loss": 0.7856, |
| "step": 19500 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.7300745248794556, |
| "eval_runtime": 7.4235, |
| "eval_samples_per_second": 1195.525, |
| "eval_steps_per_second": 149.525, |
| "step": 19944 |
| }, |
| { |
| "epoch": 9.03, |
| "learning_rate": 1.949458483754513e-06, |
| "loss": 0.7805, |
| "step": 20000 |
| }, |
| { |
| "epoch": 9.25, |
| "learning_rate": 1.4981949458483756e-06, |
| "loss": 0.7801, |
| "step": 20500 |
| }, |
| { |
| "epoch": 9.48, |
| "learning_rate": 1.0469314079422384e-06, |
| "loss": 0.7528, |
| "step": 21000 |
| }, |
| { |
| "epoch": 9.7, |
| "learning_rate": 5.956678700361012e-07, |
| "loss": 0.7698, |
| "step": 21500 |
| }, |
| { |
| "epoch": 9.93, |
| "learning_rate": 1.444043321299639e-07, |
| "loss": 0.7997, |
| "step": 22000 |
| } |
| ], |
| "max_steps": 22160, |
| "num_train_epochs": 10, |
| "total_flos": 1139660369332650.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
