| { | |
| "best_metric": 0.45234861969947815, | |
| "best_model_checkpoint": "qlora-extractor-checkpoints/checkpoint-598", | |
| "epoch": 2.392, | |
| "eval_steps": 299, | |
| "global_step": 598, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.10648680478334427, | |
| "learning_rate": 0.00019786096256684492, | |
| "loss": 1.2225, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.0971209928393364, | |
| "learning_rate": 0.00019518716577540107, | |
| "loss": 0.9559, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.120168536901474, | |
| "learning_rate": 0.00019251336898395722, | |
| "loss": 0.7891, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.13561870157718658, | |
| "learning_rate": 0.0001898395721925134, | |
| "loss": 0.6716, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.13140155375003815, | |
| "learning_rate": 0.0001871657754010695, | |
| "loss": 0.5196, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.12888653576374054, | |
| "learning_rate": 0.0001844919786096257, | |
| "loss": 0.8715, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.17288090288639069, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.7102, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.19013544917106628, | |
| "learning_rate": 0.00017914438502673798, | |
| "loss": 0.6303, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.16771665215492249, | |
| "learning_rate": 0.00017647058823529413, | |
| "loss": 0.5402, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.12465972453355789, | |
| "learning_rate": 0.00017379679144385028, | |
| "loss": 0.4714, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.1407775729894638, | |
| "learning_rate": 0.00017112299465240642, | |
| "loss": 0.7708, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.1658485233783722, | |
| "learning_rate": 0.00016844919786096257, | |
| "loss": 0.5489, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.18059569597244263, | |
| "learning_rate": 0.00016577540106951872, | |
| "loss": 0.6152, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.16564571857452393, | |
| "learning_rate": 0.0001631016042780749, | |
| "loss": 0.5372, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.15012599527835846, | |
| "learning_rate": 0.00016042780748663101, | |
| "loss": 0.4824, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.18662799894809723, | |
| "learning_rate": 0.0001577540106951872, | |
| "loss": 0.7024, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.26714515686035156, | |
| "learning_rate": 0.0001550802139037433, | |
| "loss": 0.6077, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.14682364463806152, | |
| "learning_rate": 0.00015240641711229948, | |
| "loss": 0.5559, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.20032133162021637, | |
| "learning_rate": 0.00014973262032085563, | |
| "loss": 0.5333, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.16041621565818787, | |
| "learning_rate": 0.00014705882352941178, | |
| "loss": 0.4873, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.15586546063423157, | |
| "learning_rate": 0.00014438502673796793, | |
| "loss": 0.6419, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.1745934784412384, | |
| "learning_rate": 0.00014171122994652405, | |
| "loss": 0.5824, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.15679331123828888, | |
| "learning_rate": 0.00013903743315508022, | |
| "loss": 0.5084, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.16343417763710022, | |
| "learning_rate": 0.00013636363636363637, | |
| "loss": 0.481, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.14305788278579712, | |
| "learning_rate": 0.00013368983957219252, | |
| "loss": 0.4472, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.20606954395771027, | |
| "learning_rate": 0.00013101604278074866, | |
| "loss": 0.5238, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.1933826059103012, | |
| "learning_rate": 0.0001283422459893048, | |
| "loss": 0.4699, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.1788238286972046, | |
| "learning_rate": 0.00012566844919786096, | |
| "loss": 0.4909, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.2144651561975479, | |
| "learning_rate": 0.00012299465240641713, | |
| "loss": 0.387, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.18218065798282623, | |
| "learning_rate": 0.00012005347593582887, | |
| "loss": 0.319, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.14094972610473633, | |
| "learning_rate": 0.00011737967914438503, | |
| "loss": 0.6182, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.20308974385261536, | |
| "learning_rate": 0.00011470588235294118, | |
| "loss": 0.4492, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.18815161287784576, | |
| "learning_rate": 0.00011203208556149734, | |
| "loss": 0.4472, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.1986108422279358, | |
| "learning_rate": 0.00010935828877005347, | |
| "loss": 0.3996, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.1628069281578064, | |
| "learning_rate": 0.00010668449197860964, | |
| "loss": 0.3645, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.14933669567108154, | |
| "learning_rate": 0.00010401069518716577, | |
| "loss": 0.5977, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.18734735250473022, | |
| "learning_rate": 0.00010133689839572193, | |
| "loss": 0.3784, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.2121153324842453, | |
| "learning_rate": 9.866310160427808e-05, | |
| "loss": 0.4581, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.19748178124427795, | |
| "learning_rate": 9.598930481283423e-05, | |
| "loss": 0.3858, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.18506589531898499, | |
| "learning_rate": 9.331550802139037e-05, | |
| "loss": 0.3706, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.15329745411872864, | |
| "learning_rate": 9.064171122994652e-05, | |
| "loss": 0.5347, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.19260501861572266, | |
| "learning_rate": 8.796791443850267e-05, | |
| "loss": 0.4812, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.1763940006494522, | |
| "learning_rate": 8.529411764705883e-05, | |
| "loss": 0.4317, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.20417028665542603, | |
| "learning_rate": 8.262032085561498e-05, | |
| "loss": 0.4145, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.2001723051071167, | |
| "learning_rate": 7.994652406417112e-05, | |
| "loss": 0.3884, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.2101861983537674, | |
| "learning_rate": 7.727272727272727e-05, | |
| "loss": 0.4758, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.2131994664669037, | |
| "learning_rate": 7.459893048128342e-05, | |
| "loss": 0.466, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.21261604130268097, | |
| "learning_rate": 7.192513368983958e-05, | |
| "loss": 0.4088, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.21039697527885437, | |
| "learning_rate": 6.925133689839573e-05, | |
| "loss": 0.3858, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.18674618005752563, | |
| "learning_rate": 6.657754010695188e-05, | |
| "loss": 0.3672, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.1681678146123886, | |
| "learning_rate": 6.390374331550802e-05, | |
| "loss": 0.5486, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.26280567049980164, | |
| "learning_rate": 6.122994652406417e-05, | |
| "loss": 0.3358, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 0.24697232246398926, | |
| "learning_rate": 5.8556149732620325e-05, | |
| "loss": 0.3871, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.22857435047626495, | |
| "learning_rate": 5.588235294117647e-05, | |
| "loss": 0.3822, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.2402905523777008, | |
| "learning_rate": 5.320855614973263e-05, | |
| "loss": 0.3009, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.19268670678138733, | |
| "learning_rate": 5.0534759358288774e-05, | |
| "loss": 0.3916, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.2800000000000002, | |
| "grad_norm": 0.2288196086883545, | |
| "learning_rate": 4.786096256684492e-05, | |
| "loss": 0.338, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.2444332093000412, | |
| "learning_rate": 4.518716577540107e-05, | |
| "loss": 0.3971, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 0.26502081751823425, | |
| "learning_rate": 4.251336898395722e-05, | |
| "loss": 0.2905, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.392, | |
| "eval_loss": 0.45234861969947815, | |
| "eval_runtime": 967.7709, | |
| "eval_samples_per_second": 0.207, | |
| "eval_steps_per_second": 0.207, | |
| "step": 598 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 299, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.783168801781924e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |