| { | |
| "best_metric": 0.37990298867225647, | |
| "best_model_checkpoint": "../KevinKibe/nllb-200-distilled-1.3B-finetuned-finetuned-finetuned/checkpoint-600", | |
| "epoch": 1400.0, | |
| "eval_steps": 200, | |
| "global_step": 1400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 200.0, | |
| "grad_norm": 0.19941186904907227, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.0469, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "eval_gen_len": 20.0, | |
| "eval_loss": 0.39197173714637756, | |
| "eval_rouge": 0.0476, | |
| "eval_runtime": 9.9932, | |
| "eval_samples_per_second": 0.2, | |
| "eval_steps_per_second": 0.1, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "grad_norm": 0.08469633013010025, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.0069, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "eval_gen_len": 17.5, | |
| "eval_loss": 0.3805849552154541, | |
| "eval_rouge": 0.1364, | |
| "eval_runtime": 1.1672, | |
| "eval_samples_per_second": 1.713, | |
| "eval_steps_per_second": 0.857, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 600.0, | |
| "grad_norm": 0.05082901939749718, | |
| "learning_rate": 1.4e-05, | |
| "loss": 0.0034, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 600.0, | |
| "eval_gen_len": 17.5, | |
| "eval_loss": 0.37990298867225647, | |
| "eval_rouge": 0.1364, | |
| "eval_runtime": 1.2177, | |
| "eval_samples_per_second": 1.642, | |
| "eval_steps_per_second": 0.821, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 800.0, | |
| "grad_norm": 0.03562492877244949, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.0022, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 800.0, | |
| "eval_gen_len": 109.0, | |
| "eval_loss": 0.39079928398132324, | |
| "eval_rouge": 0.1364, | |
| "eval_runtime": 10.1532, | |
| "eval_samples_per_second": 0.197, | |
| "eval_steps_per_second": 0.098, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1000.0, | |
| "grad_norm": 0.027455538511276245, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0016, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1000.0, | |
| "eval_gen_len": 13.5, | |
| "eval_loss": 0.3873954713344574, | |
| "eval_rouge": 0.1765, | |
| "eval_runtime": 10.52, | |
| "eval_samples_per_second": 0.19, | |
| "eval_steps_per_second": 0.095, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1200.0, | |
| "grad_norm": 0.02252182736992836, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0013, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1200.0, | |
| "eval_gen_len": 13.5, | |
| "eval_loss": 0.39042866230010986, | |
| "eval_rouge": 0.1765, | |
| "eval_runtime": 10.4367, | |
| "eval_samples_per_second": 0.192, | |
| "eval_steps_per_second": 0.096, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1400.0, | |
| "grad_norm": 0.019408540800213814, | |
| "learning_rate": 6e-06, | |
| "loss": 0.0011, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1400.0, | |
| "eval_gen_len": 13.5, | |
| "eval_loss": 0.3919622004032135, | |
| "eval_rouge": 0.1765, | |
| "eval_runtime": 9.8993, | |
| "eval_samples_per_second": 0.202, | |
| "eval_steps_per_second": 0.101, | |
| "step": 1400 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2000, | |
| "save_steps": 200, | |
| "total_flos": 4096215416832000.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |