| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.0, | |
| "eval_steps": 500, | |
| "global_step": 750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 3.1039228439331055, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.2535, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.9736298322677612, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.1389, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.9868311285972595, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.1137, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.7625114321708679, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.0243, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.8334870338439941, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.0505, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.7253231406211853, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9779, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.6524099707603455, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.9797, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.68406742811203, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.9878, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.6620015501976013, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.9352, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.7781227231025696, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.9205, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.6981615424156189, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.9168, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.5809632539749146, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.9345, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.6306660771369934, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.8842, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.645531952381134, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.8496, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.7515069842338562, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.8134, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.8647974133491516, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 0.8645, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.5082105398178101, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 0.8339, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.6905492544174194, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.8524, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.7012932896614075, | |
| "learning_rate": 9.30074573947683e-06, | |
| "loss": 0.8566, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.6087114810943604, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.8107, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.7151578664779663, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.9001, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.746478796005249, | |
| "learning_rate": 8.903970133383297e-06, | |
| "loss": 0.8522, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.5843214392662048, | |
| "learning_rate": 8.754400164907496e-06, | |
| "loss": 0.8164, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6262736916542053, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 0.8739, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.6539023518562317, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.8353, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.7221904397010803, | |
| "learning_rate": 8.258286144107277e-06, | |
| "loss": 0.7376, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.6120831966400146, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 0.7247, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.5770366787910461, | |
| "learning_rate": 7.891661680839932e-06, | |
| "loss": 0.6993, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.4881764352321625, | |
| "learning_rate": 7.698753289757565e-06, | |
| "loss": 0.6871, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.6157920360565186, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.7247, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.5322688817977905, | |
| "learning_rate": 7.295832266935059e-06, | |
| "loss": 0.7549, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.7124238014221191, | |
| "learning_rate": 7.08669227240909e-06, | |
| "loss": 0.6858, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.5642361044883728, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 0.6737, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.7349158525466919, | |
| "learning_rate": 6.655317089424791e-06, | |
| "loss": 0.7092, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.6204760670661926, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.6878, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.4934922754764557, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 0.6883, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.5973473787307739, | |
| "learning_rate": 5.982583047664151e-06, | |
| "loss": 0.7312, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 0.9578109979629517, | |
| "learning_rate": 5.753428561247416e-06, | |
| "loss": 0.6466, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.6420013904571533, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 0.5745, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.6083827614784241, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.6074, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.2800000000000002, | |
| "grad_norm": 0.5411021709442139, | |
| "learning_rate": 5.0581763290069865e-06, | |
| "loss": 0.5958, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.542936384677887, | |
| "learning_rate": 4.825502516487497e-06, | |
| "loss": 0.517, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 0.643763542175293, | |
| "learning_rate": 4.59320662666071e-06, | |
| "loss": 0.6125, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 0.5902148485183716, | |
| "learning_rate": 4.361791760697027e-06, | |
| "loss": 0.5574, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.5409561991691589, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.5619, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 0.5303182601928711, | |
| "learning_rate": 3.903606879060483e-06, | |
| "loss": 0.5723, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.5597081184387207, | |
| "learning_rate": 3.6778291898139907e-06, | |
| "loss": 0.57, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.557049572467804, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.558, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 0.5532881021499634, | |
| "learning_rate": 3.2353471764306567e-06, | |
| "loss": 0.5642, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6130328178405762, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.582, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 0.6685742139816284, | |
| "learning_rate": 2.8081442660546126e-06, | |
| "loss": 0.4826, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 0.7372143268585205, | |
| "learning_rate": 2.601434433748771e-06, | |
| "loss": 0.4799, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 0.5380725860595703, | |
| "learning_rate": 2.3999193603539234e-06, | |
| "loss": 0.4595, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.611395537853241, | |
| "learning_rate": 2.204035482646267e-06, | |
| "loss": 0.4414, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.6131682991981506, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.42, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 0.5948655605316162, | |
| "learning_rate": 1.8308451630064484e-06, | |
| "loss": 0.4228, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.5600000000000005, | |
| "grad_norm": 0.5260858535766602, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 0.4486, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "grad_norm": 0.6490086913108826, | |
| "learning_rate": 1.4850947128716914e-06, | |
| "loss": 0.4064, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 0.512779951095581, | |
| "learning_rate": 1.3234549597008572e-06, | |
| "loss": 0.4806, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.6591514945030212, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.4537, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "grad_norm": 0.5841071605682373, | |
| "learning_rate": 1.0243960175257605e-06, | |
| "loss": 0.425, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 0.5767258405685425, | |
| "learning_rate": 8.876245235966884e-07, | |
| "loss": 0.4903, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 0.681662380695343, | |
| "learning_rate": 7.597595192178702e-07, | |
| "loss": 0.4259, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "grad_norm": 0.6221340894699097, | |
| "learning_rate": 6.410779315161885e-07, | |
| "loss": 0.4243, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "grad_norm": 0.5294339656829834, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.3387, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.28, | |
| "grad_norm": 0.5295357704162598, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 0.3378, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "grad_norm": 0.5497756004333496, | |
| "learning_rate": 3.426013053692878e-07, | |
| "loss": 0.3857, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "grad_norm": 0.5334697961807251, | |
| "learning_rate": 2.63016787428354e-07, | |
| "loss": 0.3996, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "grad_norm": 0.48104792833328247, | |
| "learning_rate": 1.9369152030840553e-07, | |
| "loss": 0.3743, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 0.53028404712677, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.3542, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "grad_norm": 0.5596076846122742, | |
| "learning_rate": 8.639676646793382e-08, | |
| "loss": 0.3587, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "grad_norm": 0.5660611391067505, | |
| "learning_rate": 4.865965629214819e-08, | |
| "loss": 0.414, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "grad_norm": 0.4768368899822235, | |
| "learning_rate": 2.1646046750978255e-08, | |
| "loss": 0.3887, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "grad_norm": 0.5675653219223022, | |
| "learning_rate": 5.414443307377171e-09, | |
| "loss": 0.3811, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.5585565567016602, | |
| "learning_rate": 0.0, | |
| "loss": 0.3763, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 750, | |
| "total_flos": 118427603697664.0, | |
| "train_loss": 0.662273271560669, | |
| "train_runtime": 26488.1763, | |
| "train_samples_per_second": 0.227, | |
| "train_steps_per_second": 0.028 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 100000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 118427603697664.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |