| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2416626389560174, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 75.8236083984375, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 3.7148, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 52.23726272583008, | |
| "learning_rate": 9e-07, | |
| "loss": 3.2721, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 35.391998291015625, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 2.4262, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 15.225934028625488, | |
| "learning_rate": 1.9e-06, | |
| "loss": 1.8828, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 13.488654136657715, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.5398, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 8.773133277893066, | |
| "learning_rate": 2.8500000000000002e-06, | |
| "loss": 1.2772, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 6.736795902252197, | |
| "learning_rate": 3.3500000000000005e-06, | |
| "loss": 1.1159, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.9463019371032715, | |
| "learning_rate": 3.85e-06, | |
| "loss": 0.9883, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.3851969242095947, | |
| "learning_rate": 4.35e-06, | |
| "loss": 0.902, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.4642181396484375, | |
| "learning_rate": 4.85e-06, | |
| "loss": 0.8701, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.6514687538146973, | |
| "learning_rate": 5.3500000000000004e-06, | |
| "loss": 0.8548, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.4448617994785309, | |
| "learning_rate": 5.850000000000001e-06, | |
| "loss": 0.8038, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5425733327865601, | |
| "learning_rate": 6.35e-06, | |
| "loss": 0.825, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.47352057695388794, | |
| "learning_rate": 6.8500000000000005e-06, | |
| "loss": 0.812, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.40401744842529297, | |
| "learning_rate": 7.35e-06, | |
| "loss": 0.7912, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3112718164920807, | |
| "learning_rate": 7.850000000000001e-06, | |
| "loss": 0.8057, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.45351892709732056, | |
| "learning_rate": 8.350000000000001e-06, | |
| "loss": 0.7912, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.38080641627311707, | |
| "learning_rate": 8.85e-06, | |
| "loss": 0.7956, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.22949431836605072, | |
| "learning_rate": 9.35e-06, | |
| "loss": 0.7785, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.8331744074821472, | |
| "learning_rate": 9.85e-06, | |
| "loss": 0.7769, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.30213871598243713, | |
| "learning_rate": 1.035e-05, | |
| "loss": 0.7797, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.32015591859817505, | |
| "learning_rate": 1.0850000000000001e-05, | |
| "loss": 0.7941, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.22447015345096588, | |
| "learning_rate": 1.1350000000000001e-05, | |
| "loss": 0.8014, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.2546006143093109, | |
| "learning_rate": 1.185e-05, | |
| "loss": 0.771, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.23833389580249786, | |
| "learning_rate": 1.235e-05, | |
| "loss": 0.7956, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.1831669956445694, | |
| "learning_rate": 1.285e-05, | |
| "loss": 0.7833, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.19860319793224335, | |
| "learning_rate": 1.3350000000000001e-05, | |
| "loss": 0.7956, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.21220380067825317, | |
| "learning_rate": 1.3850000000000001e-05, | |
| "loss": 0.7942, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.26025038957595825, | |
| "learning_rate": 1.435e-05, | |
| "loss": 0.7877, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.2029470056295395, | |
| "learning_rate": 1.485e-05, | |
| "loss": 0.7637, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.21639183163642883, | |
| "learning_rate": 1.535e-05, | |
| "loss": 0.7885, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.23670387268066406, | |
| "learning_rate": 1.5850000000000002e-05, | |
| "loss": 0.7558, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2855335772037506, | |
| "learning_rate": 1.635e-05, | |
| "loss": 0.7854, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.21515759825706482, | |
| "learning_rate": 1.6850000000000003e-05, | |
| "loss": 0.7577, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.18981383740901947, | |
| "learning_rate": 1.7349999999999998e-05, | |
| "loss": 0.7802, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.2674611508846283, | |
| "learning_rate": 1.785e-05, | |
| "loss": 0.7809, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.24821774661540985, | |
| "learning_rate": 1.8350000000000002e-05, | |
| "loss": 0.7965, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.3644929528236389, | |
| "learning_rate": 1.885e-05, | |
| "loss": 0.7957, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.3568972647190094, | |
| "learning_rate": 1.9350000000000003e-05, | |
| "loss": 0.7882, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.3236319422721863, | |
| "learning_rate": 1.985e-05, | |
| "loss": 0.7841, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.24553723633289337, | |
| "learning_rate": 2.035e-05, | |
| "loss": 0.7824, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.1855400949716568, | |
| "learning_rate": 2.085e-05, | |
| "loss": 0.7645, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.26212435960769653, | |
| "learning_rate": 2.135e-05, | |
| "loss": 0.7824, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.3167509138584137, | |
| "learning_rate": 2.1850000000000003e-05, | |
| "loss": 0.7908, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.21845023334026337, | |
| "learning_rate": 2.235e-05, | |
| "loss": 0.7888, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.30318892002105713, | |
| "learning_rate": 2.2850000000000003e-05, | |
| "loss": 0.7682, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.28828638792037964, | |
| "learning_rate": 2.3350000000000002e-05, | |
| "loss": 0.7588, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.2422240674495697, | |
| "learning_rate": 2.385e-05, | |
| "loss": 0.7787, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.3579326272010803, | |
| "learning_rate": 2.435e-05, | |
| "loss": 0.7932, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.3549540042877197, | |
| "learning_rate": 2.485e-05, | |
| "loss": 0.7634, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 0.7744565606117249, | |
| "eval_runtime": 43.9626, | |
| "eval_samples_per_second": 45.493, | |
| "eval_steps_per_second": 0.364, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 10345, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.2648108400962437e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |