| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 13.636363636363637, | |
| "eval_steps": 500, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 220.0, | |
| "learning_rate": 2.5e-05, | |
| "loss": 24.9691, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 35.25, | |
| "learning_rate": 0.000125, | |
| "loss": 21.906, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_loss": 7.653346538543701, | |
| "eval_runtime": 0.5608, | |
| "eval_samples_per_second": 3.567, | |
| "eval_steps_per_second": 1.783, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 9.3125, | |
| "learning_rate": 0.00019956059820218982, | |
| "loss": 13.5603, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 6.644189357757568, | |
| "eval_runtime": 0.5604, | |
| "eval_samples_per_second": 3.569, | |
| "eval_steps_per_second": 1.784, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 4.625, | |
| "learning_rate": 0.00019466156752904343, | |
| "loss": 10.2605, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 6.081549644470215, | |
| "eval_runtime": 0.5806, | |
| "eval_samples_per_second": 3.445, | |
| "eval_steps_per_second": 1.722, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 15.125, | |
| "learning_rate": 0.00018458320592590975, | |
| "loss": 9.9129, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.114753484725952, | |
| "eval_runtime": 0.5656, | |
| "eval_samples_per_second": 3.536, | |
| "eval_steps_per_second": 1.768, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 2.96875, | |
| "learning_rate": 0.00016987694277788417, | |
| "loss": 4.5895, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "eval_loss": 1.6582958698272705, | |
| "eval_runtime": 0.5842, | |
| "eval_samples_per_second": 3.424, | |
| "eval_steps_per_second": 1.712, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "grad_norm": 3.171875, | |
| "learning_rate": 0.0001513474193514842, | |
| "loss": 1.6316, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.415539264678955, | |
| "eval_runtime": 0.5662, | |
| "eval_samples_per_second": 3.532, | |
| "eval_steps_per_second": 1.766, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "grad_norm": 7.1875, | |
| "learning_rate": 0.0001300084635000341, | |
| "loss": 1.4115, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "eval_loss": 1.3542958498001099, | |
| "eval_runtime": 0.58, | |
| "eval_samples_per_second": 3.448, | |
| "eval_steps_per_second": 1.724, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "grad_norm": 1.296875, | |
| "learning_rate": 0.0001070276188945293, | |
| "loss": 1.2971, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.313336968421936, | |
| "eval_runtime": 0.5751, | |
| "eval_samples_per_second": 3.478, | |
| "eval_steps_per_second": 1.739, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "grad_norm": 1.6171875, | |
| "learning_rate": 8.366226381814697e-05, | |
| "loss": 1.1321, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "eval_loss": 1.2903474569320679, | |
| "eval_runtime": 0.5848, | |
| "eval_samples_per_second": 3.42, | |
| "eval_steps_per_second": 1.71, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 0.9375, | |
| "learning_rate": 6.119081473277501e-05, | |
| "loss": 1.062, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.90234375, | |
| "learning_rate": 4.084277875864776e-05, | |
| "loss": 0.9739, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.2820332050323486, | |
| "eval_runtime": 0.5646, | |
| "eval_samples_per_second": 3.542, | |
| "eval_steps_per_second": 1.771, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 2.3731482188961818e-05, | |
| "loss": 0.917, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "eval_loss": 1.2888375520706177, | |
| "eval_runtime": 0.5836, | |
| "eval_samples_per_second": 3.427, | |
| "eval_steps_per_second": 1.714, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 1.0793155744261351e-05, | |
| "loss": 0.8541, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.27806556224823, | |
| "eval_runtime": 0.5683, | |
| "eval_samples_per_second": 3.519, | |
| "eval_steps_per_second": 1.76, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "grad_norm": 0.6796875, | |
| "learning_rate": 2.735709467518699e-06, | |
| "loss": 0.8659, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "eval_loss": 1.2891546487808228, | |
| "eval_runtime": 0.7414, | |
| "eval_samples_per_second": 2.698, | |
| "eval_steps_per_second": 1.349, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 0.0, | |
| "loss": 0.8354, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "eval_loss": 1.2830009460449219, | |
| "eval_runtime": 0.5659, | |
| "eval_samples_per_second": 3.534, | |
| "eval_steps_per_second": 1.767, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "step": 75, | |
| "total_flos": 1.1503415671442637e+17, | |
| "train_loss": 4.788167775472005, | |
| "train_runtime": 283.7979, | |
| "train_samples_per_second": 4.651, | |
| "train_steps_per_second": 0.264 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 100, | |
| "total_flos": 1.1503415671442637e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |