| { | |
| "best_metric": 0.5642260313034058, | |
| "best_model_checkpoint": "./beans_lora_outputs/checkpoint-650", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 1.4916294813156128, | |
| "learning_rate": 3.938461538461539e-05, | |
| "loss": 1.0725, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.1397809982299805, | |
| "learning_rate": 3.876923076923077e-05, | |
| "loss": 1.0828, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 1.600398063659668, | |
| "learning_rate": 3.815384615384616e-05, | |
| "loss": 1.0705, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.198091745376587, | |
| "learning_rate": 3.753846153846154e-05, | |
| "loss": 1.0374, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 1.3902909755706787, | |
| "learning_rate": 3.692307692307693e-05, | |
| "loss": 1.0509, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 1.3507941961288452, | |
| "learning_rate": 3.630769230769231e-05, | |
| "loss": 1.0385, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 1.9068666696548462, | |
| "learning_rate": 3.56923076923077e-05, | |
| "loss": 1.0204, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 1.7665146589279175, | |
| "learning_rate": 3.507692307692308e-05, | |
| "loss": 1.01, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 1.689713716506958, | |
| "learning_rate": 3.446153846153847e-05, | |
| "loss": 1.0112, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.324887275695801, | |
| "learning_rate": 3.384615384615385e-05, | |
| "loss": 1.0204, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 1.4830341339111328, | |
| "learning_rate": 3.3230769230769236e-05, | |
| "loss": 1.0078, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 1.819659948348999, | |
| "learning_rate": 3.261538461538462e-05, | |
| "loss": 0.9695, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.7536489963531494, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.9653, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_loss": 0.9567110538482666, | |
| "eval_runtime": 21.3884, | |
| "eval_samples_per_second": 6.218, | |
| "eval_steps_per_second": 0.795, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 1.4856749773025513, | |
| "learning_rate": 3.1384615384615386e-05, | |
| "loss": 0.9409, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 1.4876972436904907, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.9176, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.3868235349655151, | |
| "learning_rate": 3.0153846153846155e-05, | |
| "loss": 0.9248, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 1.7096399068832397, | |
| "learning_rate": 2.9538461538461543e-05, | |
| "loss": 0.9017, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 1.4847768545150757, | |
| "learning_rate": 2.8923076923076925e-05, | |
| "loss": 0.8809, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 1.456756353378296, | |
| "learning_rate": 2.8307692307692312e-05, | |
| "loss": 0.8762, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.316740870475769, | |
| "learning_rate": 2.7692307692307694e-05, | |
| "loss": 0.8704, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 1.3937312364578247, | |
| "learning_rate": 2.707692307692308e-05, | |
| "loss": 0.8389, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 2.5524656772613525, | |
| "learning_rate": 2.6461538461538463e-05, | |
| "loss": 0.8112, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 1.4633307456970215, | |
| "learning_rate": 2.584615384615385e-05, | |
| "loss": 0.8212, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 1.4309651851654053, | |
| "learning_rate": 2.523076923076923e-05, | |
| "loss": 0.8426, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 1.5278276205062866, | |
| "learning_rate": 2.461538461538462e-05, | |
| "loss": 0.7876, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.5827990770339966, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.7802, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_loss": 0.7600473761558533, | |
| "eval_runtime": 22.1213, | |
| "eval_samples_per_second": 6.012, | |
| "eval_steps_per_second": 0.768, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 1.1035858392715454, | |
| "learning_rate": 2.338461538461539e-05, | |
| "loss": 0.7393, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 1.7304904460906982, | |
| "learning_rate": 2.276923076923077e-05, | |
| "loss": 0.7731, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 1.8998841047286987, | |
| "learning_rate": 2.2153846153846158e-05, | |
| "loss": 0.7435, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 1.4001611471176147, | |
| "learning_rate": 2.153846153846154e-05, | |
| "loss": 0.7131, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 1.6339213848114014, | |
| "learning_rate": 2.0923076923076927e-05, | |
| "loss": 0.7559, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 1.0672645568847656, | |
| "learning_rate": 2.0307692307692308e-05, | |
| "loss": 0.7177, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 1.234423279762268, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 0.7056, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 1.1383863687515259, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 0.7015, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 1.238714575767517, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 0.7118, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 1.153242588043213, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 0.6856, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 1.6669644117355347, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 0.6866, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.3097412586212158, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 0.6598, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.921310544013977, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.6269, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_loss": 0.6367737054824829, | |
| "eval_runtime": 16.3183, | |
| "eval_samples_per_second": 8.15, | |
| "eval_steps_per_second": 1.042, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.251091718673706, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.6991, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 1.2219730615615845, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.6688, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 1.278702974319458, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.6816, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 1.672782301902771, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.634, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 1.4417272806167603, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.656, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 1.3969999551773071, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.6252, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 1.3986371755599976, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.636, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 1.205729603767395, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.6018, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 1.6087666749954224, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.6653, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 1.0829548835754395, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.5922, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.2072396278381348, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.5914, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 1.098717212677002, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.5882, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.4719703197479248, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.5764, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_loss": 0.5801168084144592, | |
| "eval_runtime": 11.2873, | |
| "eval_samples_per_second": 11.783, | |
| "eval_steps_per_second": 1.506, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 2.0378928184509277, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.588, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.6630451679229736, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.6869, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 2.175814151763916, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.6446, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.132444143295288, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.5948, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 1.2464724779129028, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.6202, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 1.079876184463501, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.5491, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 1.3481943607330322, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.5552, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 1.102620244026184, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.6199, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 2.219717502593994, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.6307, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 1.1284600496292114, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.5266, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 1.1227543354034424, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.5658, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 2.005279302597046, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.5571, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.8302509784698486, | |
| "learning_rate": 0.0, | |
| "loss": 0.5977, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_loss": 0.5642260313034058, | |
| "eval_runtime": 22.2136, | |
| "eval_samples_per_second": 5.987, | |
| "eval_steps_per_second": 0.765, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 4.0340206085566464e+17, | |
| "train_loss": 0.7649891574566181, | |
| "train_runtime": 920.2323, | |
| "train_samples_per_second": 5.618, | |
| "train_steps_per_second": 0.706 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.0340206085566464e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |