| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 1, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 12.5, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.7672, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.0074498653411865, | |
| "eval_runtime": 0.1009, | |
| "eval_samples_per_second": 9.911, | |
| "eval_steps_per_second": 9.911, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 12.5625, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.7672, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.6057260036468506, | |
| "eval_runtime": 0.1016, | |
| "eval_samples_per_second": 9.839, | |
| "eval_steps_per_second": 9.839, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 6.84375, | |
| "learning_rate": 3e-06, | |
| "loss": 0.7422, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.2898356914520264, | |
| "eval_runtime": 0.1036, | |
| "eval_samples_per_second": 9.649, | |
| "eval_steps_per_second": 9.649, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 10.0, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.7211, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.14534854888916, | |
| "eval_runtime": 0.1037, | |
| "eval_samples_per_second": 9.647, | |
| "eval_steps_per_second": 9.647, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 72.0, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6591, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.636004090309143, | |
| "eval_runtime": 0.1032, | |
| "eval_samples_per_second": 9.688, | |
| "eval_steps_per_second": 9.688, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 53.25, | |
| "learning_rate": 6e-06, | |
| "loss": 0.4514, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.7588775753974915, | |
| "eval_runtime": 0.1037, | |
| "eval_samples_per_second": 9.646, | |
| "eval_steps_per_second": 9.646, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 26.625, | |
| "learning_rate": 7e-06, | |
| "loss": 0.24, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.6620572805404663, | |
| "eval_runtime": 0.1031, | |
| "eval_samples_per_second": 9.703, | |
| "eval_steps_per_second": 9.703, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 23.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1584, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8120734691619873, | |
| "eval_runtime": 0.1013, | |
| "eval_samples_per_second": 9.872, | |
| "eval_steps_per_second": 9.872, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 21.25, | |
| "learning_rate": 9e-06, | |
| "loss": 0.1235, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.7537858486175537, | |
| "eval_runtime": 0.102, | |
| "eval_samples_per_second": 9.8, | |
| "eval_steps_per_second": 9.8, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 9.75, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0998, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.7743023037910461, | |
| "eval_runtime": 0.103, | |
| "eval_samples_per_second": 9.708, | |
| "eval_steps_per_second": 9.708, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 7.3125, | |
| "learning_rate": 9.755282581475769e-06, | |
| "loss": 0.0869, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.7770701050758362, | |
| "eval_runtime": 0.1025, | |
| "eval_samples_per_second": 9.756, | |
| "eval_steps_per_second": 9.756, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 4.625, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.1692, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.8292599320411682, | |
| "eval_runtime": 0.1014, | |
| "eval_samples_per_second": 9.862, | |
| "eval_steps_per_second": 9.862, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 7.938926261462366e-06, | |
| "loss": 0.0702, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.8938868045806885, | |
| "eval_runtime": 0.1028, | |
| "eval_samples_per_second": 9.729, | |
| "eval_steps_per_second": 9.729, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.984375, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.063, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.9581525325775146, | |
| "eval_runtime": 0.1015, | |
| "eval_samples_per_second": 9.856, | |
| "eval_steps_per_second": 9.856, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0567, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.9825366139411926, | |
| "eval_runtime": 0.1012, | |
| "eval_samples_per_second": 9.886, | |
| "eval_steps_per_second": 9.886, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.640625, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.052, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.9959811568260193, | |
| "eval_runtime": 0.1016, | |
| "eval_samples_per_second": 9.843, | |
| "eval_steps_per_second": 9.843, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 2.061073738537635e-06, | |
| "loss": 0.0488, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.9883414506912231, | |
| "eval_runtime": 0.1016, | |
| "eval_samples_per_second": 9.844, | |
| "eval_steps_per_second": 9.844, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 0.0457, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.0003694295883179, | |
| "eval_runtime": 0.1045, | |
| "eval_samples_per_second": 9.568, | |
| "eval_steps_per_second": 9.568, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 1.3984375, | |
| "learning_rate": 2.447174185242324e-07, | |
| "loss": 0.0436, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 1.0056235790252686, | |
| "eval_runtime": 0.1034, | |
| "eval_samples_per_second": 9.667, | |
| "eval_steps_per_second": 9.667, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.5703125, | |
| "learning_rate": 0.0, | |
| "loss": 0.0427, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.0059328079223633, | |
| "eval_runtime": 0.1034, | |
| "eval_samples_per_second": 9.671, | |
| "eval_steps_per_second": 9.671, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 1, | |
| "total_flos": 9895376479846400.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |