| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.844827586206897, | |
| "global_step": 16500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.942873563218391e-05, | |
| "loss": 5.5518, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.885402298850575e-05, | |
| "loss": 0.3979, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.8279310344827587e-05, | |
| "loss": 0.1115, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.7704597701149427e-05, | |
| "loss": 0.1044, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.7129885057471266e-05, | |
| "loss": 0.0986, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.6555172413793105e-05, | |
| "loss": 0.0899, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.5980459770114944e-05, | |
| "loss": 0.0872, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.5405747126436783e-05, | |
| "loss": 0.085, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.4831034482758623e-05, | |
| "loss": 0.0869, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.425632183908046e-05, | |
| "loss": 0.0879, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.36816091954023e-05, | |
| "loss": 0.0797, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.07041613757610321, | |
| "eval_runtime": 64.4372, | |
| "eval_samples_per_second": 40.008, | |
| "eval_steps_per_second": 10.01, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.310689655172414e-05, | |
| "loss": 0.079, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.253218390804598e-05, | |
| "loss": 0.071, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 1.1957471264367817e-05, | |
| "loss": 0.0701, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.1382758620689656e-05, | |
| "loss": 0.0739, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.0808045977011495e-05, | |
| "loss": 0.0723, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.0233333333333336e-05, | |
| "loss": 0.0682, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 9.658620689655173e-06, | |
| "loss": 0.068, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 9.083908045977013e-06, | |
| "loss": 0.0686, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 8.509195402298852e-06, | |
| "loss": 0.0662, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 7.934482758620691e-06, | |
| "loss": 0.0671, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 7.359770114942529e-06, | |
| "loss": 0.0684, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 6.785057471264368e-06, | |
| "loss": 0.0669, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.06483030319213867, | |
| "eval_runtime": 64.5114, | |
| "eval_samples_per_second": 39.962, | |
| "eval_steps_per_second": 9.998, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 6.211494252873563e-06, | |
| "loss": 0.062, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5.636781609195403e-06, | |
| "loss": 0.061, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 5.062068965517242e-06, | |
| "loss": 0.0581, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.487356321839081e-06, | |
| "loss": 0.0575, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.913793103448276e-06, | |
| "loss": 0.0643, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 3.3390804597701155e-06, | |
| "loss": 0.0623, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 2.764367816091954e-06, | |
| "loss": 0.0565, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2.1896551724137934e-06, | |
| "loss": 0.0592, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 1.6149425287356323e-06, | |
| "loss": 0.0627, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.0402298850574713e-06, | |
| "loss": 0.0565, | |
| "step": 16500 | |
| } | |
| ], | |
| "max_steps": 17400, | |
| "num_train_epochs": 3, | |
| "total_flos": 3.575505883024589e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |