| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 625, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.8854656219482422, | |
| "learning_rate": 3.125e-06, | |
| "loss": 0.4597, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.9969694018363953, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.4563, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.8475977182388306, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 0.4264, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.5963095426559448, | |
| "learning_rate": 9.995510010438338e-06, | |
| "loss": 0.4412, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.5206061005592346, | |
| "learning_rate": 9.977283247594167e-06, | |
| "loss": 0.442, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.4866243302822113, | |
| "learning_rate": 9.945090192163202e-06, | |
| "loss": 0.4613, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.48311129212379456, | |
| "learning_rate": 9.8990211780905e-06, | |
| "loss": 0.4213, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.4780612587928772, | |
| "learning_rate": 9.83920547536085e-06, | |
| "loss": 0.4165, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.5186869502067566, | |
| "learning_rate": 9.765810927266281e-06, | |
| "loss": 0.4269, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.4753789007663727, | |
| "learning_rate": 9.679043479436558e-06, | |
| "loss": 0.4315, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.5088816285133362, | |
| "learning_rate": 9.579146601954276e-06, | |
| "loss": 0.422, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.611789882183075, | |
| "learning_rate": 9.466400606176062e-06, | |
| "loss": 0.4246, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.3785482347011566, | |
| "learning_rate": 9.341121858176877e-06, | |
| "loss": 0.3859, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.40997618436813354, | |
| "learning_rate": 9.203661891024547e-06, | |
| "loss": 0.3365, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.5316786766052246, | |
| "learning_rate": 9.054406418375443e-06, | |
| "loss": 0.3536, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.5229846835136414, | |
| "learning_rate": 8.893774252159186e-06, | |
| "loss": 0.3335, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.64065021276474, | |
| "learning_rate": 8.722216127389362e-06, | |
| "loss": 0.3572, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.4666633903980255, | |
| "learning_rate": 8.540213437397833e-06, | |
| "loss": 0.35, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.5635166764259338, | |
| "learning_rate": 8.348276883041584e-06, | |
| "loss": 0.3455, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.42854130268096924, | |
| "learning_rate": 8.146945039672446e-06, | |
| "loss": 0.3511, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.6392173171043396, | |
| "learning_rate": 7.93678284589079e-06, | |
| "loss": 0.3635, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.4988630712032318, | |
| "learning_rate": 7.718380018323743e-06, | |
| "loss": 0.3545, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.44719088077545166, | |
| "learning_rate": 7.492349396876097e-06, | |
| "loss": 0.3396, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.4074830412864685, | |
| "learning_rate": 7.2593252250971306e-06, | |
| "loss": 0.3659, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.4083346724510193, | |
| "learning_rate": 7.019961370488645e-06, | |
| "loss": 0.345, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.58194899559021, | |
| "learning_rate": 6.774929489748052e-06, | |
| "loss": 0.2951, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.41007938981056213, | |
| "learning_rate": 6.524917144094851e-06, | |
| "loss": 0.2788, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.5576044917106628, | |
| "learning_rate": 6.270625869968907e-06, | |
| "loss": 0.2732, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.46990445256233215, | |
| "learning_rate": 6.012769210514147e-06, | |
| "loss": 0.2657, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.49044930934906006, | |
| "learning_rate": 5.752070713371372e-06, | |
| "loss": 0.2864, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.38818588852882385, | |
| "learning_rate": 5.4892619003983735e-06, | |
| "loss": 0.2804, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.5136937499046326, | |
| "learning_rate": 5.2250802150143395e-06, | |
| "loss": 0.274, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.4540717601776123, | |
| "learning_rate": 4.960266952928317e-06, | |
| "loss": 0.2921, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.4595988094806671, | |
| "learning_rate": 4.695565182058113e-06, | |
| "loss": 0.2742, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.45735347270965576, | |
| "learning_rate": 4.431717657476394e-06, | |
| "loss": 0.2732, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.47884508967399597, | |
| "learning_rate": 4.169464737234588e-06, | |
| "loss": 0.283, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.39772123098373413, | |
| "learning_rate": 3.909542304912881e-06, | |
| "loss": 0.2834, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 0.45640942454338074, | |
| "learning_rate": 3.652679704725596e-06, | |
| "loss": 0.2619, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.43376386165618896, | |
| "learning_rate": 3.3995976949760813e-06, | |
| "loss": 0.2164, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.4014115035533905, | |
| "learning_rate": 3.1510064256037277e-06, | |
| "loss": 0.2228, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.2800000000000002, | |
| "grad_norm": 0.43393564224243164, | |
| "learning_rate": 2.9076034454981333e-06, | |
| "loss": 0.215, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.4281829595565796, | |
| "learning_rate": 2.6700717451719066e-06, | |
| "loss": 0.2377, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 0.3631593883037567, | |
| "learning_rate": 2.4390778402843606e-06, | |
| "loss": 0.2257, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 0.41745710372924805, | |
| "learning_rate": 2.215269901393805e-06, | |
| "loss": 0.2313, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.383023738861084, | |
| "learning_rate": 1.999275935186277e-06, | |
| "loss": 0.234, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 0.4755268096923828, | |
| "learning_rate": 1.7917020222842662e-06, | |
| "loss": 0.2363, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.40028688311576843, | |
| "learning_rate": 1.5931306165801453e-06, | |
| "loss": 0.2394, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.5058606863021851, | |
| "learning_rate": 1.4041189108663421e-06, | |
| "loss": 0.2113, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 0.40886273980140686, | |
| "learning_rate": 1.2251972733483615e-06, | |
| "loss": 0.2162, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5253835320472717, | |
| "learning_rate": 1.0568677594277898e-06, | |
| "loss": 0.233, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 0.35965532064437866, | |
| "learning_rate": 8.996027029312315e-07, | |
| "loss": 0.205, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 0.3577881157398224, | |
| "learning_rate": 7.538433907381814e-07, | |
| "loss": 0.2032, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 0.3823171555995941, | |
| "learning_rate": 6.199988245268779e-07, | |
| "loss": 0.2011, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.3407541811466217, | |
| "learning_rate": 4.984445731126542e-07, | |
| "loss": 0.1961, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.37284690141677856, | |
| "learning_rate": 3.8952171859913123e-07, | |
| "loss": 0.192, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 0.3782266080379486, | |
| "learning_rate": 2.9353589929939086e-07, | |
| "loss": 0.1935, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.5600000000000005, | |
| "grad_norm": 0.3706322908401489, | |
| "learning_rate": 2.1075645211266926e-07, | |
| "loss": 0.1994, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "grad_norm": 0.36876216530799866, | |
| "learning_rate": 1.4141565676307866e-07, | |
| "loss": 0.2027, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 0.40256306529045105, | |
| "learning_rate": 8.570808402103392e-08, | |
| "loss": 0.1975, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.3809185326099396, | |
| "learning_rate": 4.3790049736268106e-08, | |
| "loss": 0.1905, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "grad_norm": 0.3607443869113922, | |
| "learning_rate": 1.577917621443825e-08, | |
| "loss": 0.2059, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 0.40115296840667725, | |
| "learning_rate": 1.7540621680850467e-09, | |
| "loss": 0.2113, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 625, | |
| "total_flos": 123341386481664.0, | |
| "train_loss": 0.2985656312942505, | |
| "train_runtime": 10873.4333, | |
| "train_samples_per_second": 0.46, | |
| "train_steps_per_second": 0.057 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 625, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 123341386481664.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |