| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06538084341288003, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013076168682576005, | |
| "eval_loss": 4.681145668029785, | |
| "eval_runtime": 37.3046, | |
| "eval_samples_per_second": 8.632, | |
| "eval_steps_per_second": 4.316, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006538084341288003, | |
| "grad_norm": 15.046794891357422, | |
| "learning_rate": 5e-05, | |
| "loss": 4.7171, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.013076168682576005, | |
| "grad_norm": 4.514803409576416, | |
| "learning_rate": 0.0001, | |
| "loss": 3.624, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.013076168682576005, | |
| "eval_loss": 2.933568239212036, | |
| "eval_runtime": 37.6526, | |
| "eval_samples_per_second": 8.552, | |
| "eval_steps_per_second": 4.276, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.019614253023864006, | |
| "grad_norm": 4.907962799072266, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.8585, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02615233736515201, | |
| "grad_norm": 2.664832830429077, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.4406, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02615233736515201, | |
| "eval_loss": 2.2351837158203125, | |
| "eval_runtime": 37.7323, | |
| "eval_samples_per_second": 8.534, | |
| "eval_steps_per_second": 4.267, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.032690421706440015, | |
| "grad_norm": 2.24057936668396, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.2194, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03922850604772801, | |
| "grad_norm": 2.625696897506714, | |
| "learning_rate": 5e-05, | |
| "loss": 1.9821, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03922850604772801, | |
| "eval_loss": 2.027791738510132, | |
| "eval_runtime": 37.9951, | |
| "eval_samples_per_second": 8.475, | |
| "eval_steps_per_second": 4.237, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04576659038901602, | |
| "grad_norm": 2.2278287410736084, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 2.0424, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.05230467473030402, | |
| "grad_norm": 2.1391711235046387, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 2.0098, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05230467473030402, | |
| "eval_loss": 1.9652129411697388, | |
| "eval_runtime": 37.8932, | |
| "eval_samples_per_second": 8.498, | |
| "eval_steps_per_second": 4.249, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.058842759071592025, | |
| "grad_norm": 2.061873435974121, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.9192, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06538084341288003, | |
| "grad_norm": 2.319734573364258, | |
| "learning_rate": 0.0, | |
| "loss": 1.9173, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06538084341288003, | |
| "eval_loss": 1.9535859823226929, | |
| "eval_runtime": 37.7406, | |
| "eval_samples_per_second": 8.532, | |
| "eval_steps_per_second": 4.266, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.533650861555712e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |