| { |
| "best_global_step": 4950, |
| "best_metric": 3.060495615005493, |
| "best_model_checkpoint": null, |
| "epoch": 10.1010101010101, |
| "eval_steps": 500, |
| "global_step": 5000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00202020202020202, |
| "grad_norm": 29.864723205566406, |
| "learning_rate": 0.0, |
| "loss": 10.4445, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.278876543045044, |
| "learning_rate": 9.88e-05, |
| "loss": 4.8246, |
| "step": 495 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 3.7829103469848633, |
| "eval_runtime": 19.668, |
| "eval_samples_per_second": 1616.332, |
| "eval_steps_per_second": 6.356, |
| "step": 495 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 1.0388227701187134, |
| "learning_rate": 9.478678038379531e-05, |
| "loss": 3.5358, |
| "step": 990 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 3.4372856616973877, |
| "eval_runtime": 19.5671, |
| "eval_samples_per_second": 1624.666, |
| "eval_steps_per_second": 6.388, |
| "step": 990 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.9951871037483215, |
| "learning_rate": 8.950959488272922e-05, |
| "loss": 3.2822, |
| "step": 1485 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 3.2929306030273438, |
| "eval_runtime": 19.6808, |
| "eval_samples_per_second": 1615.277, |
| "eval_steps_per_second": 6.351, |
| "step": 1485 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 1.0897934436798096, |
| "learning_rate": 8.42324093816631e-05, |
| "loss": 3.1379, |
| "step": 1980 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 3.211028575897217, |
| "eval_runtime": 19.92, |
| "eval_samples_per_second": 1595.88, |
| "eval_steps_per_second": 6.275, |
| "step": 1980 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 1.1176131963729858, |
| "learning_rate": 7.895522388059702e-05, |
| "loss": 3.0361, |
| "step": 2475 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 3.154623031616211, |
| "eval_runtime": 19.6905, |
| "eval_samples_per_second": 1614.486, |
| "eval_steps_per_second": 6.348, |
| "step": 2475 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 1.1129180192947388, |
| "learning_rate": 7.367803837953093e-05, |
| "loss": 2.9554, |
| "step": 2970 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 3.116886854171753, |
| "eval_runtime": 19.7127, |
| "eval_samples_per_second": 1612.665, |
| "eval_steps_per_second": 6.341, |
| "step": 2970 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 1.1529704332351685, |
| "learning_rate": 6.840085287846483e-05, |
| "loss": 2.8872, |
| "step": 3465 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 3.092864751815796, |
| "eval_runtime": 19.8571, |
| "eval_samples_per_second": 1600.942, |
| "eval_steps_per_second": 6.295, |
| "step": 3465 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 1.1491146087646484, |
| "learning_rate": 6.312366737739872e-05, |
| "loss": 2.8263, |
| "step": 3960 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 3.077211856842041, |
| "eval_runtime": 19.9288, |
| "eval_samples_per_second": 1595.183, |
| "eval_steps_per_second": 6.272, |
| "step": 3960 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 1.2301512956619263, |
| "learning_rate": 5.784648187633263e-05, |
| "loss": 2.7708, |
| "step": 4455 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 3.065768003463745, |
| "eval_runtime": 19.6931, |
| "eval_samples_per_second": 1614.269, |
| "eval_steps_per_second": 6.347, |
| "step": 4455 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 1.3216657638549805, |
| "learning_rate": 5.256929637526653e-05, |
| "loss": 2.719, |
| "step": 4950 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 3.060495615005493, |
| "eval_runtime": 19.7521, |
| "eval_samples_per_second": 1609.447, |
| "eval_steps_per_second": 6.328, |
| "step": 4950 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 9900, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 20, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8.351742541824e+16, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
