| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 13872, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.819780853517878e-05, | |
| "loss": 1.0175, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.639561707035756e-05, | |
| "loss": 0.6937, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.459342560553633e-05, | |
| "loss": 0.6068, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.2791234140715114e-05, | |
| "loss": 0.55, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.098904267589389e-05, | |
| "loss": 0.5041, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 3.9186851211072664e-05, | |
| "loss": 0.4698, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3.7384659746251445e-05, | |
| "loss": 0.4439, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 3.558246828143022e-05, | |
| "loss": 0.4068, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 3.3780276816608994e-05, | |
| "loss": 0.3862, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 3.1978085351787776e-05, | |
| "loss": 0.332, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 3.0175893886966554e-05, | |
| "loss": 0.3044, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 2.8373702422145332e-05, | |
| "loss": 0.2945, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.657151095732411e-05, | |
| "loss": 0.2832, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.4769319492502884e-05, | |
| "loss": 0.2809, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 2.2967128027681662e-05, | |
| "loss": 0.263, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 2.116493656286044e-05, | |
| "loss": 0.2534, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 1.936274509803922e-05, | |
| "loss": 0.2419, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.7560553633217993e-05, | |
| "loss": 0.2397, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.575836216839677e-05, | |
| "loss": 0.2052, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.395617070357555e-05, | |
| "loss": 0.1783, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.2153979238754325e-05, | |
| "loss": 0.1752, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.0351787773933102e-05, | |
| "loss": 0.1696, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 8.54959630911188e-06, | |
| "loss": 0.1639, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 6.747404844290659e-06, | |
| "loss": 0.1594, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 4.945213379469435e-06, | |
| "loss": 0.1538, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3.143021914648212e-06, | |
| "loss": 0.154, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.3408304498269898e-06, | |
| "loss": 0.1492, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 13872, | |
| "total_flos": 2.370754172808069e+17, | |
| "train_loss": 0.33127041616516817, | |
| "train_runtime": 10514.6739, | |
| "train_samples_per_second": 168.836, | |
| "train_steps_per_second": 1.319 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 13872, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.370754172808069e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |