| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9822485207100593, | |
| "eval_steps": 500, | |
| "global_step": 63, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23668639053254437, | |
| "grad_norm": 1.5945452451705933, | |
| "learning_rate": 4.922693215572695e-05, | |
| "loss": 2.9767, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.47337278106508873, | |
| "grad_norm": 1.7332061529159546, | |
| "learning_rate": 4.6955539334255716e-05, | |
| "loss": 2.9754, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7100591715976331, | |
| "grad_norm": 2.9891419410705566, | |
| "learning_rate": 4.332629679574566e-05, | |
| "loss": 2.9723, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.9467455621301775, | |
| "grad_norm": 2.947935104370117, | |
| "learning_rate": 3.856365659664399e-05, | |
| "loss": 2.7976, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.183431952662722, | |
| "grad_norm": 2.278214931488037, | |
| "learning_rate": 3.2962166256292113e-05, | |
| "loss": 2.7409, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.4201183431952662, | |
| "grad_norm": 2.263528823852539, | |
| "learning_rate": 2.686825233966061e-05, | |
| "loss": 2.4753, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.6568047337278107, | |
| "grad_norm": 2.759286403656006, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.54, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.893491124260355, | |
| "grad_norm": 2.5087263584136963, | |
| "learning_rate": 1.4717822421734718e-05, | |
| "loss": 2.4686, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.1301775147928996, | |
| "grad_norm": 2.494452476501465, | |
| "learning_rate": 9.412754953531663e-06, | |
| "loss": 2.4011, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.366863905325444, | |
| "grad_norm": 2.437476634979248, | |
| "learning_rate": 5.071687319426946e-06, | |
| "loss": 2.4184, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.603550295857988, | |
| "grad_norm": 3.0760037899017334, | |
| "learning_rate": 1.9630947032398067e-06, | |
| "loss": 2.3458, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.8402366863905324, | |
| "grad_norm": 3.3869755268096924, | |
| "learning_rate": 2.7922934437178695e-07, | |
| "loss": 2.3665, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.9822485207100593, | |
| "step": 63, | |
| "total_flos": 2011752588509184.0, | |
| "train_loss": 2.6121579503256176, | |
| "train_runtime": 3118.9448, | |
| "train_samples_per_second": 0.325, | |
| "train_steps_per_second": 0.02 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 63, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2011752588509184.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |