| { |
| "best_metric": 1.652644395828247, |
| "best_model_checkpoint": "output/nirvana/checkpoint-144", |
| "epoch": 4.0, |
| "global_step": 144, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.14, |
| "learning_rate": 0.00013040646433810595, |
| "loss": 2.8877, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.29, |
| "learning_rate": 0.00011137140040750913, |
| "loss": 2.7784, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.43, |
| "learning_rate": 8.386493606940316e-05, |
| "loss": 2.2959, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.57, |
| "learning_rate": 5.3335063930596836e-05, |
| "loss": 2.5076, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.71, |
| "learning_rate": 2.5828599592490882e-05, |
| "loss": 2.4917, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.86, |
| "learning_rate": 6.793535661894054e-06, |
| "loss": 2.4772, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 0.0, |
| "loss": 2.6382, |
| "step": 35 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 2.5190556049346924, |
| "eval_runtime": 2.4884, |
| "eval_samples_per_second": 22.504, |
| "eval_steps_per_second": 2.813, |
| "step": 35 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 2.3264973163604736, |
| "eval_runtime": 1.1647, |
| "eval_samples_per_second": 45.506, |
| "eval_steps_per_second": 6.01, |
| "step": 36 |
| }, |
| { |
| "epoch": 1.11, |
| "learning_rate": 4.137086214086682e-06, |
| "loss": 2.6774, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.25, |
| "learning_rate": 2.009247481060283e-05, |
| "loss": 2.4577, |
| "step": 45 |
| }, |
| { |
| "epoch": 1.39, |
| "learning_rate": 4.513741816785908e-05, |
| "loss": 2.2823, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.53, |
| "learning_rate": 7.457888395248933e-05, |
| "loss": 2.298, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.67, |
| "learning_rate": 0.00010290000000000001, |
| "loss": 2.4088, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.81, |
| "learning_rate": 0.00012479383023822482, |
| "loss": 2.2944, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.94, |
| "learning_rate": 0.00013615781185663748, |
| "loss": 2.0032, |
| "step": 70 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 2.128573179244995, |
| "eval_runtime": 1.1938, |
| "eval_samples_per_second": 44.396, |
| "eval_steps_per_second": 5.864, |
| "step": 72 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 1.8826501369476318, |
| "eval_runtime": 1.8444, |
| "eval_samples_per_second": 22.772, |
| "eval_steps_per_second": 3.253, |
| "step": 74 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 0.0001369528677140173, |
| "loss": 2.0939, |
| "step": 75 |
| }, |
| { |
| "epoch": 2.16, |
| "learning_rate": 0.00012848876816285752, |
| "loss": 2.1654, |
| "step": 80 |
| }, |
| { |
| "epoch": 2.3, |
| "learning_rate": 0.00010939183589447406, |
| "loss": 2.0452, |
| "step": 85 |
| }, |
| { |
| "epoch": 2.43, |
| "learning_rate": 8.305259792170682e-05, |
| "loss": 1.7357, |
| "step": 90 |
| }, |
| { |
| "epoch": 2.57, |
| "learning_rate": 5.414740207829316e-05, |
| "loss": 2.0788, |
| "step": 95 |
| }, |
| { |
| "epoch": 2.7, |
| "learning_rate": 2.7808164105525978e-05, |
| "loss": 1.928, |
| "step": 100 |
| }, |
| { |
| "epoch": 2.84, |
| "learning_rate": 8.711231837142545e-06, |
| "loss": 1.9376, |
| "step": 105 |
| }, |
| { |
| "epoch": 2.97, |
| "learning_rate": 2.4713228598268823e-07, |
| "loss": 1.9652, |
| "step": 110 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 1.7358663082122803, |
| "eval_runtime": 1.8478, |
| "eval_samples_per_second": 22.729, |
| "eval_steps_per_second": 3.247, |
| "step": 111 |
| }, |
| { |
| "epoch": 3.19, |
| "learning_rate": 1.2406169761775193e-05, |
| "loss": 1.9469, |
| "step": 115 |
| }, |
| { |
| "epoch": 3.33, |
| "learning_rate": 3.4300000000000014e-05, |
| "loss": 1.7381, |
| "step": 120 |
| }, |
| { |
| "epoch": 3.47, |
| "learning_rate": 6.262111604751063e-05, |
| "loss": 1.8423, |
| "step": 125 |
| }, |
| { |
| "epoch": 3.61, |
| "learning_rate": 9.206258183214083e-05, |
| "loss": 1.6384, |
| "step": 130 |
| }, |
| { |
| "epoch": 3.75, |
| "learning_rate": 0.0001171075251893971, |
| "loss": 2.0094, |
| "step": 135 |
| }, |
| { |
| "epoch": 3.89, |
| "learning_rate": 0.00013306291378591332, |
| "loss": 1.8349, |
| "step": 140 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 1.652644395828247, |
| "eval_runtime": 0.6674, |
| "eval_samples_per_second": 74.92, |
| "eval_steps_per_second": 10.489, |
| "step": 144 |
| } |
| ], |
| "max_steps": 144, |
| "num_train_epochs": 4, |
| "total_flos": 146715475968000.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
