| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 37.57142857142857, | |
| "eval_steps": 500, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 18.408525466918945, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 2.259, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.691978454589844, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 1.6222, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 23.907058715820312, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 1.4985, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 15.894387245178223, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 1.2984, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.4949517250061035, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 1.199, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 14.757712364196777, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 1.1274, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.4316110610961914, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 1.0192, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 7.104844570159912, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 1.0137, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 45.069522857666016, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.927, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.8073647022247314, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.8765, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 26.81642723083496, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.8803, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 9.39847469329834, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.7795, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.370129346847534, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.6831, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.4130897521972656, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.7307, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 8.065000534057617, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.6719, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 48.87232208251953, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.6639, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.8513643741607666, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.6147, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 2.277904510498047, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.5719, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.250492572784424, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.5682, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.1722874641418457, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.4997, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 5.405400276184082, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.4997, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 14.250221252441406, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.437, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "grad_norm": 2.9079504013061523, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.457, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 1.7956665754318237, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.4715, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 5.249920845031738, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.4485, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "grad_norm": 2.336935520172119, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.4416, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 2.4214272499084473, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.4026, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 1.8111662864685059, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.351, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "grad_norm": 1.6475896835327148, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.3904, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 3.535529136657715, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.3094, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "grad_norm": 14.978922843933105, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.37, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 17.745275497436523, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.3506, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "grad_norm": 9.402239799499512, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.3409, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "grad_norm": 1.4599494934082031, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.3529, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 1.7524628639221191, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.285, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "grad_norm": 1.8424232006072998, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.27, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "grad_norm": 2.109724283218384, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.2549, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 37.57142857142857, | |
| "grad_norm": 5.070311069488525, | |
| "learning_rate": 9.971930719778446e-05, | |
| "loss": 0.3348, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 37.57142857142857, | |
| "step": 150, | |
| "total_flos": 1.0130277760185139e+18, | |
| "train_loss": 0.69614071448644, | |
| "train_runtime": 3435.6403, | |
| "train_samples_per_second": 0.393, | |
| "train_steps_per_second": 0.044 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "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": 1.0130277760185139e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |