| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9546317788562986, | |
| "global_step": 7000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 6.0161, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 3.9784154891967773, | |
| "eval_rouge-1": 0.1824, | |
| "eval_rouge-2": 0.046, | |
| "eval_rouge-l": 0.1728, | |
| "eval_runtime": 265.003, | |
| "eval_samples_per_second": 15.751, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.8873366381252815e-05, | |
| "loss": 3.6793, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 2.9924702644348145, | |
| "eval_rouge-1": 0.3283, | |
| "eval_rouge-2": 0.1385, | |
| "eval_rouge-l": 0.3123, | |
| "eval_runtime": 236.629, | |
| "eval_samples_per_second": 17.639, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 4.662009914375845e-05, | |
| "loss": 2.7635, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 2.58980655670166, | |
| "eval_rouge-1": 0.3955, | |
| "eval_rouge-2": 0.2061, | |
| "eval_rouge-l": 0.3776, | |
| "eval_runtime": 233.9394, | |
| "eval_samples_per_second": 17.842, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.4366831906264086e-05, | |
| "loss": 2.2888, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "eval_loss": 2.4224066734313965, | |
| "eval_rouge-1": 0.42, | |
| "eval_rouge-2": 0.2296, | |
| "eval_rouge-l": 0.4011, | |
| "eval_runtime": 459.7714, | |
| "eval_samples_per_second": 9.078, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.211356466876972e-05, | |
| "loss": 2.1166, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 2.343660593032837, | |
| "eval_rouge-1": 0.4335, | |
| "eval_rouge-2": 0.2425, | |
| "eval_rouge-l": 0.413, | |
| "eval_runtime": 449.5335, | |
| "eval_samples_per_second": 9.285, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 3.986029743127535e-05, | |
| "loss": 1.8642, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 2.2839481830596924, | |
| "eval_rouge-1": 0.4426, | |
| "eval_rouge-2": 0.2535, | |
| "eval_rouge-l": 0.4222, | |
| "eval_runtime": 453.1272, | |
| "eval_samples_per_second": 9.212, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 3.760703019378098e-05, | |
| "loss": 1.8616, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_loss": 2.220813035964966, | |
| "eval_rouge-1": 0.4475, | |
| "eval_rouge-2": 0.262, | |
| "eval_rouge-l": 0.4285, | |
| "eval_runtime": 443.175, | |
| "eval_samples_per_second": 9.418, | |
| "step": 7000 | |
| } | |
| ], | |
| "max_steps": 23690, | |
| "num_train_epochs": 10, | |
| "total_flos": 3.3108670416384e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |