| { | |
| "best_metric": 0.02520456165075302, | |
| "best_model_checkpoint": "./results_pnum/results_cvrev_pnum_f3_large_b4e15_5000/checkpoint-5000", | |
| "epoch": 7.704160246533128, | |
| "global_step": 15000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.914903088829532e-05, | |
| "loss": 0.5561, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.8292925142113556e-05, | |
| "loss": 0.0339, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.743681939593179e-05, | |
| "loss": 0.0311, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 4.658071364975002e-05, | |
| "loss": 0.0297, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.572460790356825e-05, | |
| "loss": 0.0287, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.4868502157386485e-05, | |
| "loss": 0.0285, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 4.4012396411204716e-05, | |
| "loss": 0.0278, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.3156290665022946e-05, | |
| "loss": 0.0272, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 4.2300184918841177e-05, | |
| "loss": 0.026, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.1444079172659414e-05, | |
| "loss": 0.0259, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_loss": 0.02520456165075302, | |
| "eval_runtime": 103.0087, | |
| "eval_samples_per_second": 25.192, | |
| "eval_steps_per_second": 6.3, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 4.058797342647764e-05, | |
| "loss": 0.0259, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.973186768029587e-05, | |
| "loss": 0.0251, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 3.8875761934114105e-05, | |
| "loss": 0.0247, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 3.8019656187932336e-05, | |
| "loss": 0.0243, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 3.7163550441750566e-05, | |
| "loss": 0.0242, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 3.63074446955688e-05, | |
| "loss": 0.023, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 3.545133894938703e-05, | |
| "loss": 0.0227, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 3.4595233203205265e-05, | |
| "loss": 0.0226, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 3.3739127457023495e-05, | |
| "loss": 0.0228, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 3.2883021710841726e-05, | |
| "loss": 0.021, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_loss": 0.026348939165472984, | |
| "eval_runtime": 102.8144, | |
| "eval_samples_per_second": 25.24, | |
| "eval_steps_per_second": 6.312, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 3.2026915964659956e-05, | |
| "loss": 0.0206, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 3.117081021847819e-05, | |
| "loss": 0.0208, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 3.0314704472296417e-05, | |
| "loss": 0.0204, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 2.945859872611465e-05, | |
| "loss": 0.0191, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "learning_rate": 2.860249297993288e-05, | |
| "loss": 0.019, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 2.7746387233751116e-05, | |
| "loss": 0.0191, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 2.6890281487569346e-05, | |
| "loss": 0.0179, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 2.603417574138758e-05, | |
| "loss": 0.0165, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 2.517806999520581e-05, | |
| "loss": 0.0172, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 2.432196424902404e-05, | |
| "loss": 0.0164, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "eval_loss": 0.029481180012226105, | |
| "eval_runtime": 102.7986, | |
| "eval_samples_per_second": 25.244, | |
| "eval_steps_per_second": 6.313, | |
| "step": 15000 | |
| } | |
| ], | |
| "max_steps": 29205, | |
| "num_train_epochs": 15, | |
| "total_flos": 1.29872527097856e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |