| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.893964110929853, |
| "global_step": 6000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.41, |
| "learning_rate": 1.8368678629690052e-05, |
| "loss": 1.8033, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.82, |
| "learning_rate": 1.67373572593801e-05, |
| "loss": 1.3748, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_bleu": 56.2549, |
| "eval_gen_len": 46.4902, |
| "eval_loss": 1.0630319118499756, |
| "eval_runtime": 198.2655, |
| "eval_samples_per_second": 4.363, |
| "eval_steps_per_second": 0.55, |
| "step": 1226 |
| }, |
| { |
| "epoch": 1.22, |
| "learning_rate": 1.5106035889070147e-05, |
| "loss": 1.2066, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.63, |
| "learning_rate": 1.3474714518760197e-05, |
| "loss": 1.0606, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_bleu": 59.9691, |
| "eval_gen_len": 41.7179, |
| "eval_loss": 0.8989508748054504, |
| "eval_runtime": 167.0597, |
| "eval_samples_per_second": 5.178, |
| "eval_steps_per_second": 0.652, |
| "step": 2452 |
| }, |
| { |
| "epoch": 2.04, |
| "learning_rate": 1.1843393148450246e-05, |
| "loss": 1.0004, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.45, |
| "learning_rate": 1.0212071778140294e-05, |
| "loss": 0.9003, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.85, |
| "learning_rate": 8.580750407830342e-06, |
| "loss": 0.8476, |
| "step": 3500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_bleu": 62.2787, |
| "eval_gen_len": 43.8925, |
| "eval_loss": 0.8091979622840881, |
| "eval_runtime": 172.38, |
| "eval_samples_per_second": 5.018, |
| "eval_steps_per_second": 0.632, |
| "step": 3678 |
| }, |
| { |
| "epoch": 3.26, |
| "learning_rate": 6.949429037520392e-06, |
| "loss": 0.8063, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.67, |
| "learning_rate": 5.31810766721044e-06, |
| "loss": 0.7727, |
| "step": 4500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_bleu": 63.4386, |
| "eval_gen_len": 43.7595, |
| "eval_loss": 0.7679935693740845, |
| "eval_runtime": 172.1652, |
| "eval_samples_per_second": 5.024, |
| "eval_steps_per_second": 0.633, |
| "step": 4904 |
| }, |
| { |
| "epoch": 4.08, |
| "learning_rate": 3.68678629690049e-06, |
| "loss": 0.7669, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.49, |
| "learning_rate": 2.0554649265905383e-06, |
| "loss": 0.7119, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.89, |
| "learning_rate": 4.2414355628058736e-07, |
| "loss": 0.7228, |
| "step": 6000 |
| } |
| ], |
| "max_steps": 6130, |
| "num_train_epochs": 5, |
| "total_flos": 922220642893824.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
