| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "global_step": 15, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.07, |
| "learning_rate": 0.0005, |
| "loss": 10.5545, |
| "theoretical_loss": 20.841284642178216, |
| "tokens_seen": 65536 |
| }, |
| { |
| "epoch": 0.13, |
| "learning_rate": 0.00046428571428571433, |
| "loss": 10.5833, |
| "theoretical_loss": 17.59466794495971, |
| "tokens_seen": 131072 |
| }, |
| { |
| "epoch": 0.2, |
| "learning_rate": 0.00042857142857142855, |
| "loss": 9.0487, |
| "theoretical_loss": 15.967943933467634, |
| "tokens_seen": 196608 |
| }, |
| { |
| "epoch": 0.27, |
| "learning_rate": 0.0003928571428571429, |
| "loss": 8.6045, |
| "theoretical_loss": 14.920783596619636, |
| "tokens_seen": 262144 |
| }, |
| { |
| "epoch": 0.33, |
| "learning_rate": 0.00035714285714285714, |
| "loss": 8.612, |
| "theoretical_loss": 14.164683778394041, |
| "tokens_seen": 327680 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 0.00032142857142857147, |
| "loss": 8.0149, |
| "theoretical_loss": 13.581028313181289, |
| "tokens_seen": 393216 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 0.0002857142857142857, |
| "loss": 8.0364, |
| "theoretical_loss": 13.110275842032255, |
| "tokens_seen": 458752 |
| }, |
| { |
| "epoch": 0.53, |
| "learning_rate": 0.00025, |
| "loss": 7.6606, |
| "theoretical_loss": 12.71859646611439, |
| "tokens_seen": 524288 |
| }, |
| { |
| "epoch": 0.6, |
| "learning_rate": 0.00021428571428571427, |
| "loss": 7.5729, |
| "theoretical_loss": 12.38505930450469, |
| "tokens_seen": 589824 |
| }, |
| { |
| "epoch": 0.67, |
| "learning_rate": 0.00017857142857142857, |
| "loss": 7.5104, |
| "theoretical_loss": 12.095879447666144, |
| "tokens_seen": 655360 |
| }, |
| { |
| "epoch": 0.73, |
| "learning_rate": 0.00014285714285714284, |
| "loss": 7.3193, |
| "theoretical_loss": 11.841533071732742, |
| "tokens_seen": 720896 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 0.00010714285714285714, |
| "loss": 6.9659, |
| "theoretical_loss": 11.615186049337796, |
| "tokens_seen": 786432 |
| }, |
| { |
| "epoch": 0.87, |
| "learning_rate": 7.142857142857142e-05, |
| "loss": 6.9517, |
| "theoretical_loss": 11.411780980817854, |
| "tokens_seen": 851968 |
| }, |
| { |
| "epoch": 0.93, |
| "learning_rate": 3.571428571428571e-05, |
| "loss": 7.641, |
| "theoretical_loss": 11.227478542742938, |
| "tokens_seen": 917504 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 0.0, |
| "loss": 7.5328, |
| "theoretical_loss": 11.059300758608979, |
| "tokens_seen": 983040 |
| } |
| ], |
| "max_steps": 15, |
| "num_train_epochs": 9223372036854775807, |
| "total_flos": 501680701440000.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
