| { |
| "best_metric": 0.7026481628417969, |
| "best_model_checkpoint": "output/morgenshtern/checkpoint-100", |
| "epoch": 1.0, |
| "global_step": 100, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.05, |
| "learning_rate": 4.740143418587856e-05, |
| "loss": 0.8849, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1, |
| "learning_rate": 3.745625171786743e-05, |
| "loss": 0.6963, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.15, |
| "learning_rate": 2.827793169273699e-05, |
| "loss": 0.8572, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2, |
| "learning_rate": 2.009247481060302e-05, |
| "loss": 0.7158, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.25, |
| "learning_rate": 1.310143418587843e-05, |
| "loss": 0.8878, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3, |
| "learning_rate": 7.476952440678023e-06, |
| "loss": 0.9167, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.35, |
| "learning_rate": 3.3575229821526355e-06, |
| "loss": 0.7488, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 8.445798351735566e-07, |
| "loss": 0.832, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.45, |
| "learning_rate": 0.0, |
| "loss": 0.8448, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "learning_rate": 8.445798351735718e-07, |
| "loss": 0.9363, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.55, |
| "learning_rate": 3.3575229821523687e-06, |
| "loss": 0.8096, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6, |
| "learning_rate": 7.476952440677627e-06, |
| "loss": 0.8588, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.65, |
| "learning_rate": 1.3101434185878491e-05, |
| "loss": 0.7752, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7, |
| "learning_rate": 2.0092474810602402e-05, |
| "loss": 0.8502, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.75, |
| "learning_rate": 2.827793169273628e-05, |
| "loss": 0.773, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 3.745625171786751e-05, |
| "loss": 0.7943, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.85, |
| "learning_rate": 4.7401434185877724e-05, |
| "loss": 0.8335, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.9, |
| "learning_rate": 5.786859569823979e-05, |
| "loss": 0.7886, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.95, |
| "learning_rate": 6.860000000000016e-05, |
| "loss": 0.7602, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.0, |
| "learning_rate": 7.933140430175958e-05, |
| "loss": 0.6606, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.7026481628417969, |
| "eval_runtime": 3.0708, |
| "eval_samples_per_second": 43.311, |
| "eval_steps_per_second": 5.536, |
| "step": 100 |
| } |
| ], |
| "max_steps": 400, |
| "num_train_epochs": 4, |
| "total_flos": 103602290688000.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
