| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.907652942544778, | |
| "global_step": 12500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.9224625881988063e-05, | |
| "loss": 0.9542, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.844925176397612e-05, | |
| "loss": 0.9606, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1.767387764596418e-05, | |
| "loss": 0.9752, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.689850352795224e-05, | |
| "loss": 0.9758, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.6123129409940298e-05, | |
| "loss": 0.9911, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.5347755291928356e-05, | |
| "loss": 1.0023, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.4572381173916417e-05, | |
| "loss": 1.0072, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.3797007055904474e-05, | |
| "loss": 1.0212, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 38.6916, | |
| "eval_gen_len": 28.7748, | |
| "eval_loss": 1.050680160522461, | |
| "eval_runtime": 181.503, | |
| "eval_samples_per_second": 7.779, | |
| "eval_steps_per_second": 0.127, | |
| "step": 4299 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.3021632937892535e-05, | |
| "loss": 1.0013, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 1.2246258819880595e-05, | |
| "loss": 0.9579, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1.1470884701868653e-05, | |
| "loss": 0.9527, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.0695510583856712e-05, | |
| "loss": 0.959, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 9.92013646584477e-06, | |
| "loss": 0.9606, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 9.14476234783283e-06, | |
| "loss": 0.9579, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 8.36938822982089e-06, | |
| "loss": 0.9563, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 7.594014111808948e-06, | |
| "loss": 0.957, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 6.818639993797007e-06, | |
| "loss": 0.9558, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 38.8229, | |
| "eval_gen_len": 28.7649, | |
| "eval_loss": 1.0432215929031372, | |
| "eval_runtime": 185.3749, | |
| "eval_samples_per_second": 7.617, | |
| "eval_steps_per_second": 0.124, | |
| "step": 8598 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 6.043265875785067e-06, | |
| "loss": 0.9358, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 5.2678917577731256e-06, | |
| "loss": 0.9268, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.492517639761185e-06, | |
| "loss": 0.9238, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 3.717143521749244e-06, | |
| "loss": 0.9267, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 2.9417694037373037e-06, | |
| "loss": 0.9219, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 2.1663952857253625e-06, | |
| "loss": 0.9285, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.3910211677134219e-06, | |
| "loss": 0.925, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 6.15647049701481e-07, | |
| "loss": 0.9215, | |
| "step": 12500 | |
| } | |
| ], | |
| "max_steps": 12897, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.868489262563328e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |