| { | |
| "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": 30.875, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.5389, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 6.346914291381836, | |
| "eval_runtime": 0.1184, | |
| "eval_samples_per_second": 8.443, | |
| "eval_steps_per_second": 8.443, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 30.75, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.5389, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 6.253310680389404, | |
| "eval_runtime": 0.1196, | |
| "eval_samples_per_second": 8.359, | |
| "eval_steps_per_second": 8.359, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 14.875, | |
| "learning_rate": 3e-06, | |
| "loss": 0.5017, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 6.210088729858398, | |
| "eval_runtime": 0.1203, | |
| "eval_samples_per_second": 8.313, | |
| "eval_steps_per_second": 8.313, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.40625, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.4689, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 6.316311359405518, | |
| "eval_runtime": 0.1184, | |
| "eval_samples_per_second": 8.446, | |
| "eval_steps_per_second": 8.446, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.578125, | |
| "learning_rate": 5e-06, | |
| "loss": 0.3604, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 6.414371967315674, | |
| "eval_runtime": 0.1193, | |
| "eval_samples_per_second": 8.381, | |
| "eval_steps_per_second": 8.381, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.9375, | |
| "learning_rate": 6e-06, | |
| "loss": 0.3107, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 6.412661075592041, | |
| "eval_runtime": 0.1183, | |
| "eval_samples_per_second": 8.452, | |
| "eval_steps_per_second": 8.452, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 23.625, | |
| "learning_rate": 7e-06, | |
| "loss": 0.2698, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 6.808929443359375, | |
| "eval_runtime": 0.1187, | |
| "eval_samples_per_second": 8.426, | |
| "eval_steps_per_second": 8.426, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 30464.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.317, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 7.338753700256348, | |
| "eval_runtime": 0.117, | |
| "eval_samples_per_second": 8.546, | |
| "eval_steps_per_second": 8.546, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 24.75, | |
| "learning_rate": 9e-06, | |
| "loss": 0.2228, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 6.506290435791016, | |
| "eval_runtime": 0.1168, | |
| "eval_samples_per_second": 8.564, | |
| "eval_steps_per_second": 8.564, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 12.0625, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1798, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 5.707322597503662, | |
| "eval_runtime": 0.1186, | |
| "eval_samples_per_second": 8.429, | |
| "eval_steps_per_second": 8.429, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 6.28125, | |
| "learning_rate": 9.755282581475769e-06, | |
| "loss": 0.1436, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 5.118520259857178, | |
| "eval_runtime": 0.1191, | |
| "eval_samples_per_second": 8.398, | |
| "eval_steps_per_second": 8.398, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.1183, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 4.899399280548096, | |
| "eval_runtime": 0.1192, | |
| "eval_samples_per_second": 8.386, | |
| "eval_steps_per_second": 8.386, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 7.938926261462366e-06, | |
| "loss": 0.1002, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 4.803285598754883, | |
| "eval_runtime": 0.1193, | |
| "eval_samples_per_second": 8.381, | |
| "eval_steps_per_second": 8.381, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.484375, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.0865, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 5.1707048416137695, | |
| "eval_runtime": 0.1192, | |
| "eval_samples_per_second": 8.389, | |
| "eval_steps_per_second": 8.389, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 2.8125, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0758, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 5.70893669128418, | |
| "eval_runtime": 0.1195, | |
| "eval_samples_per_second": 8.367, | |
| "eval_steps_per_second": 8.367, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.9375, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.0663, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 6.405214786529541, | |
| "eval_runtime": 0.117, | |
| "eval_samples_per_second": 8.55, | |
| "eval_steps_per_second": 8.55, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.859375, | |
| "learning_rate": 2.061073738537635e-06, | |
| "loss": 0.0601, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 6.781367778778076, | |
| "eval_runtime": 0.1172, | |
| "eval_samples_per_second": 8.534, | |
| "eval_steps_per_second": 8.534, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.28125, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 0.0545, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 6.95858097076416, | |
| "eval_runtime": 0.1189, | |
| "eval_samples_per_second": 8.409, | |
| "eval_steps_per_second": 8.409, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 2.421875, | |
| "learning_rate": 2.447174185242324e-07, | |
| "loss": 0.0505, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 6.976622581481934, | |
| "eval_runtime": 0.1189, | |
| "eval_samples_per_second": 8.413, | |
| "eval_steps_per_second": 8.413, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.296875, | |
| "learning_rate": 0.0, | |
| "loss": 0.0482, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 6.984273910522461, | |
| "eval_runtime": 0.1186, | |
| "eval_samples_per_second": 8.434, | |
| "eval_steps_per_second": 8.434, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 1, | |
| "total_flos": 9084979233423360.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |