| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1662510390689942, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008312551953449709, | |
| "eval_loss": 1.7031567096710205, | |
| "eval_runtime": 38.9167, | |
| "eval_samples_per_second": 13.028, | |
| "eval_steps_per_second": 6.527, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004156275976724855, | |
| "grad_norm": 8.313850402832031, | |
| "learning_rate": 5e-05, | |
| "loss": 5.5336, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00831255195344971, | |
| "grad_norm": 11.550032615661621, | |
| "learning_rate": 0.0001, | |
| "loss": 4.8425, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.012468827930174564, | |
| "grad_norm": 15.0049467086792, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 4.3692, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01662510390689942, | |
| "grad_norm": 13.11844253540039, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 4.9362, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.020781379883624274, | |
| "grad_norm": 12.807136535644531, | |
| "learning_rate": 9.847001329696653e-05, | |
| "loss": 4.7307, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.02493765586034913, | |
| "grad_norm": 13.963868141174316, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 4.3436, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.029093931837073983, | |
| "grad_norm": 15.964492797851562, | |
| "learning_rate": 9.578866633275288e-05, | |
| "loss": 5.1, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.03325020781379884, | |
| "grad_norm": 14.075353622436523, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 3.9034, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03740648379052369, | |
| "grad_norm": 19.31031608581543, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 4.8119, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.04156275976724855, | |
| "grad_norm": 16.785146713256836, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 5.285, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04156275976724855, | |
| "eval_loss": 1.1468664407730103, | |
| "eval_runtime": 39.3179, | |
| "eval_samples_per_second": 12.895, | |
| "eval_steps_per_second": 6.46, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0457190357439734, | |
| "grad_norm": 8.89350700378418, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 4.4179, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.04987531172069826, | |
| "grad_norm": 9.902567863464355, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 4.1408, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05403158769742311, | |
| "grad_norm": 9.637911796569824, | |
| "learning_rate": 8.07106356344834e-05, | |
| "loss": 4.3677, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.058187863674147966, | |
| "grad_norm": 11.88183307647705, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 4.5824, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.06234413965087282, | |
| "grad_norm": 13.11361026763916, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 4.5274, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.06650041562759768, | |
| "grad_norm": 10.361695289611816, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 4.5816, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.07065669160432253, | |
| "grad_norm": 13.685104370117188, | |
| "learning_rate": 6.623497346023418e-05, | |
| "loss": 4.2433, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.07481296758104738, | |
| "grad_norm": 14.716741561889648, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 5.1209, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07896924355777224, | |
| "grad_norm": 16.79210662841797, | |
| "learning_rate": 5.8229729514036705e-05, | |
| "loss": 5.0024, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.0831255195344971, | |
| "grad_norm": 15.1632080078125, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 5.2588, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0831255195344971, | |
| "eval_loss": 1.1230735778808594, | |
| "eval_runtime": 39.4665, | |
| "eval_samples_per_second": 12.846, | |
| "eval_steps_per_second": 6.436, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08728179551122195, | |
| "grad_norm": 9.77318286895752, | |
| "learning_rate": 5e-05, | |
| "loss": 4.0232, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.0914380714879468, | |
| "grad_norm": 10.034825325012207, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 4.1443, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.09559434746467166, | |
| "grad_norm": 9.069893836975098, | |
| "learning_rate": 4.17702704859633e-05, | |
| "loss": 4.2779, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.09975062344139651, | |
| "grad_norm": 13.375626564025879, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 4.5112, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.10390689941812137, | |
| "grad_norm": 10.194912910461426, | |
| "learning_rate": 3.3765026539765834e-05, | |
| "loss": 4.295, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.10806317539484622, | |
| "grad_norm": 11.304728507995605, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 4.1605, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.11221945137157108, | |
| "grad_norm": 9.311614990234375, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 4.3502, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.11637572734829593, | |
| "grad_norm": 21.666893005371094, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 4.6441, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.12053200332502079, | |
| "grad_norm": 13.352088928222656, | |
| "learning_rate": 1.928936436551661e-05, | |
| "loss": 4.7608, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.12468827930174564, | |
| "grad_norm": 14.888147354125977, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 4.6675, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.12468827930174564, | |
| "eval_loss": 1.1012681722640991, | |
| "eval_runtime": 39.5131, | |
| "eval_samples_per_second": 12.831, | |
| "eval_steps_per_second": 6.428, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.1288445552784705, | |
| "grad_norm": 10.447840690612793, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 4.186, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.13300083125519535, | |
| "grad_norm": 13.874579429626465, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 4.3065, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.1371571072319202, | |
| "grad_norm": 10.623826026916504, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 3.7722, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.14131338320864506, | |
| "grad_norm": 11.924473762512207, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 4.3879, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.14546965918536992, | |
| "grad_norm": 16.149940490722656, | |
| "learning_rate": 4.2113336672471245e-06, | |
| "loss": 4.9389, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.14962593516209477, | |
| "grad_norm": 11.651994705200195, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 4.0122, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.15378221113881962, | |
| "grad_norm": 10.336478233337402, | |
| "learning_rate": 1.5299867030334814e-06, | |
| "loss": 4.6848, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.15793848711554448, | |
| "grad_norm": 12.979530334472656, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 4.2951, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.16209476309226933, | |
| "grad_norm": 14.301395416259766, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 4.8291, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.1662510390689942, | |
| "grad_norm": 17.349889755249023, | |
| "learning_rate": 0.0, | |
| "loss": 5.4328, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1662510390689942, | |
| "eval_loss": 1.1004496812820435, | |
| "eval_runtime": 39.5297, | |
| "eval_samples_per_second": 12.826, | |
| "eval_steps_per_second": 6.426, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 7.01377152024576e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |