| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.999384615384617, | |
| "global_step": 8120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9440000057220459, | |
| "eval_loss": 0.185967355966568, | |
| "eval_runtime": 3.2155, | |
| "eval_samples_per_second": 155.495, | |
| "eval_steps_per_second": 19.592, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 2.2079363472639895e-05, | |
| "loss": 0.3271, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9599999785423279, | |
| "eval_loss": 0.1548820436000824, | |
| "eval_runtime": 3.2405, | |
| "eval_samples_per_second": 154.299, | |
| "eval_steps_per_second": 19.442, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 2.0630586341889244e-05, | |
| "loss": 0.1813, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9520000219345093, | |
| "eval_loss": 0.18529100716114044, | |
| "eval_runtime": 3.2782, | |
| "eval_samples_per_second": 152.524, | |
| "eval_steps_per_second": 19.218, | |
| "step": 1218 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.9181809211138593e-05, | |
| "loss": 0.1331, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9580000042915344, | |
| "eval_loss": 0.21402151882648468, | |
| "eval_runtime": 3.387, | |
| "eval_samples_per_second": 147.622, | |
| "eval_steps_per_second": 18.6, | |
| "step": 1624 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 1.7733032080387945e-05, | |
| "loss": 0.09, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9520000219345093, | |
| "eval_loss": 0.21573218703269958, | |
| "eval_runtime": 3.3463, | |
| "eval_samples_per_second": 149.417, | |
| "eval_steps_per_second": 18.827, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9639999866485596, | |
| "eval_loss": 0.19065596163272858, | |
| "eval_runtime": 3.2501, | |
| "eval_samples_per_second": 153.842, | |
| "eval_steps_per_second": 19.384, | |
| "step": 2436 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 1.6284254949637297e-05, | |
| "loss": 0.0675, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9459999799728394, | |
| "eval_loss": 0.3131004273891449, | |
| "eval_runtime": 3.2761, | |
| "eval_samples_per_second": 152.622, | |
| "eval_steps_per_second": 19.23, | |
| "step": 2842 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 1.4835477818886648e-05, | |
| "loss": 0.0497, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9440000057220459, | |
| "eval_loss": 0.3736623525619507, | |
| "eval_runtime": 3.2136, | |
| "eval_samples_per_second": 155.587, | |
| "eval_steps_per_second": 19.604, | |
| "step": 3248 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 1.3386700688135999e-05, | |
| "loss": 0.0423, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9520000219345093, | |
| "eval_loss": 0.31935977935791016, | |
| "eval_runtime": 3.2821, | |
| "eval_samples_per_second": 152.343, | |
| "eval_steps_per_second": 19.195, | |
| "step": 3654 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 1.193792355738535e-05, | |
| "loss": 0.0332, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9580000042915344, | |
| "eval_loss": 0.3113400340080261, | |
| "eval_runtime": 3.3622, | |
| "eval_samples_per_second": 148.714, | |
| "eval_steps_per_second": 18.738, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.29622161388397217, | |
| "eval_runtime": 3.2804, | |
| "eval_samples_per_second": 152.418, | |
| "eval_steps_per_second": 19.205, | |
| "step": 4466 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 1.04891464266347e-05, | |
| "loss": 0.0218, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9580000042915344, | |
| "eval_loss": 0.337551087141037, | |
| "eval_runtime": 3.3285, | |
| "eval_samples_per_second": 150.217, | |
| "eval_steps_per_second": 18.927, | |
| "step": 4872 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 9.04036929588405e-06, | |
| "loss": 0.0207, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.338717520236969, | |
| "eval_runtime": 3.1749, | |
| "eval_samples_per_second": 157.484, | |
| "eval_steps_per_second": 19.843, | |
| "step": 5278 | |
| }, | |
| { | |
| "epoch": 13.55, | |
| "learning_rate": 7.5915921651334015e-06, | |
| "loss": 0.0157, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9559999704360962, | |
| "eval_loss": 0.3549070358276367, | |
| "eval_runtime": 3.2845, | |
| "eval_samples_per_second": 152.231, | |
| "eval_steps_per_second": 19.181, | |
| "step": 5684 | |
| }, | |
| { | |
| "epoch": 14.78, | |
| "learning_rate": 6.142815034382753e-06, | |
| "loss": 0.0101, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.3411114811897278, | |
| "eval_runtime": 3.247, | |
| "eval_samples_per_second": 153.988, | |
| "eval_steps_per_second": 19.402, | |
| "step": 6090 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9580000042915344, | |
| "eval_loss": 0.40595561265945435, | |
| "eval_runtime": 3.2347, | |
| "eval_samples_per_second": 154.576, | |
| "eval_steps_per_second": 19.477, | |
| "step": 6496 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 4.694037903632104e-06, | |
| "loss": 0.0112, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.3584845960140228, | |
| "eval_runtime": 3.4061, | |
| "eval_samples_per_second": 146.793, | |
| "eval_steps_per_second": 18.496, | |
| "step": 6902 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 3.245260772881454e-06, | |
| "loss": 0.0058, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9620000123977661, | |
| "eval_loss": 0.3812974989414215, | |
| "eval_runtime": 3.194, | |
| "eval_samples_per_second": 156.545, | |
| "eval_steps_per_second": 19.725, | |
| "step": 7308 | |
| }, | |
| { | |
| "epoch": 18.47, | |
| "learning_rate": 1.796483642130805e-06, | |
| "loss": 0.0043, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9660000205039978, | |
| "eval_loss": 0.3624423146247864, | |
| "eval_runtime": 3.2008, | |
| "eval_samples_per_second": 156.209, | |
| "eval_steps_per_second": 19.682, | |
| "step": 7714 | |
| }, | |
| { | |
| "epoch": 19.7, | |
| "learning_rate": 3.477065113801558e-07, | |
| "loss": 0.0019, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9660000205039978, | |
| "eval_loss": 0.3639107644557953, | |
| "eval_runtime": 3.2121, | |
| "eval_samples_per_second": 155.662, | |
| "eval_steps_per_second": 19.613, | |
| "step": 8120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 8120, | |
| "total_flos": 8550999537666048.0, | |
| "train_loss": 0.06258585786290943, | |
| "train_runtime": 5261.4565, | |
| "train_samples_per_second": 24.708, | |
| "train_steps_per_second": 1.543 | |
| } | |
| ], | |
| "max_steps": 8120, | |
| "num_train_epochs": 20, | |
| "total_flos": 8550999537666048.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |