| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 60.0, | |
| "eval_steps": 8874, | |
| "global_step": 88740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.476245641708374, | |
| "learning_rate": 9.0009015100293e-06, | |
| "loss": 2.4036, | |
| "step": 8874 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.011286497116089, | |
| "eval_runtime": 32.2427, | |
| "eval_samples_per_second": 301.308, | |
| "eval_steps_per_second": 6.854, | |
| "step": 8874 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 10.134785652160645, | |
| "learning_rate": 8.001239576290288e-06, | |
| "loss": 2.1954, | |
| "step": 17748 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 3.1597039699554443, | |
| "eval_runtime": 31.8518, | |
| "eval_samples_per_second": 305.006, | |
| "eval_steps_per_second": 6.938, | |
| "step": 17748 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 14.89818000793457, | |
| "learning_rate": 7.001690331304937e-06, | |
| "loss": 2.0709, | |
| "step": 26622 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 3.206042528152466, | |
| "eval_runtime": 33.1331, | |
| "eval_samples_per_second": 293.211, | |
| "eval_steps_per_second": 6.67, | |
| "step": 26622 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 20.7427921295166, | |
| "learning_rate": 6.002141086319586e-06, | |
| "loss": 2.0132, | |
| "step": 35496 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 3.307133913040161, | |
| "eval_runtime": 31.6331, | |
| "eval_samples_per_second": 307.115, | |
| "eval_steps_per_second": 6.986, | |
| "step": 35496 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 10.81413745880127, | |
| "learning_rate": 5.0025918413342355e-06, | |
| "loss": 1.9783, | |
| "step": 44370 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 3.3543457984924316, | |
| "eval_runtime": 31.6425, | |
| "eval_samples_per_second": 307.024, | |
| "eval_steps_per_second": 6.984, | |
| "step": 44370 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 13.179546356201172, | |
| "learning_rate": 4.003155285102547e-06, | |
| "loss": 1.9672, | |
| "step": 53244 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 3.45923113822937, | |
| "eval_runtime": 31.6946, | |
| "eval_samples_per_second": 306.519, | |
| "eval_steps_per_second": 6.973, | |
| "step": 53244 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "grad_norm": 35.83116149902344, | |
| "learning_rate": 3.003606040117197e-06, | |
| "loss": 1.9536, | |
| "step": 62118 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 3.473649740219116, | |
| "eval_runtime": 31.7269, | |
| "eval_samples_per_second": 306.207, | |
| "eval_steps_per_second": 6.966, | |
| "step": 62118 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 2.7362611293792725, | |
| "learning_rate": 2.0041694838855083e-06, | |
| "loss": 1.9473, | |
| "step": 70992 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 3.531722068786621, | |
| "eval_runtime": 31.6356, | |
| "eval_samples_per_second": 307.091, | |
| "eval_steps_per_second": 6.986, | |
| "step": 70992 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "grad_norm": 3.277851104736328, | |
| "learning_rate": 1.0046202389001578e-06, | |
| "loss": 1.9479, | |
| "step": 79866 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_loss": 3.5619990825653076, | |
| "eval_runtime": 31.6245, | |
| "eval_samples_per_second": 307.199, | |
| "eval_steps_per_second": 6.988, | |
| "step": 79866 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 2.524662733078003, | |
| "learning_rate": 5.070993914807302e-09, | |
| "loss": 1.9411, | |
| "step": 88740 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_loss": 3.5839552879333496, | |
| "eval_runtime": 31.4889, | |
| "eval_samples_per_second": 308.522, | |
| "eval_steps_per_second": 7.018, | |
| "step": 88740 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "step": 88740, | |
| "total_flos": 1.4015777943683174e+18, | |
| "train_loss": 2.0418326404735745, | |
| "train_runtime": 30807.4628, | |
| "train_samples_per_second": 115.158, | |
| "train_steps_per_second": 2.88 | |
| } | |
| ], | |
| "logging_steps": 8874, | |
| "max_steps": 88740, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4015777943683174e+18, | |
| "train_batch_size": 40, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |