| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.20512820512820512, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0020512820512820513, | |
| "eval_loss": 3.666609048843384, | |
| "eval_runtime": 74.3332, | |
| "eval_samples_per_second": 11.058, | |
| "eval_steps_per_second": 1.386, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006153846153846154, | |
| "grad_norm": 55.59685516357422, | |
| "learning_rate": 1.5e-05, | |
| "loss": 14.9637, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.012307692307692308, | |
| "grad_norm": 49.33617401123047, | |
| "learning_rate": 3e-05, | |
| "loss": 14.2659, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018461538461538463, | |
| "grad_norm": 40.35097885131836, | |
| "learning_rate": 4.5e-05, | |
| "loss": 12.0429, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.018461538461538463, | |
| "eval_loss": 2.3609907627105713, | |
| "eval_runtime": 75.0844, | |
| "eval_samples_per_second": 10.948, | |
| "eval_steps_per_second": 1.372, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.024615384615384615, | |
| "grad_norm": 30.74170684814453, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 8.4852, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03076923076923077, | |
| "grad_norm": 30.664113998413086, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 6.9901, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.036923076923076927, | |
| "grad_norm": 37.1199951171875, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 5.2257, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.036923076923076927, | |
| "eval_loss": 1.283078670501709, | |
| "eval_runtime": 75.1182, | |
| "eval_samples_per_second": 10.943, | |
| "eval_steps_per_second": 1.371, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.043076923076923075, | |
| "grad_norm": 25.036869049072266, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 5.1287, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04923076923076923, | |
| "grad_norm": 19.620019912719727, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 4.4763, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.055384615384615386, | |
| "grad_norm": 24.405311584472656, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 3.8667, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.055384615384615386, | |
| "eval_loss": 0.9862188100814819, | |
| "eval_runtime": 75.1057, | |
| "eval_samples_per_second": 10.945, | |
| "eval_steps_per_second": 1.371, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06153846153846154, | |
| "grad_norm": 22.108592987060547, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 4.0031, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06769230769230769, | |
| "grad_norm": 19.9141788482666, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 3.8066, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.07384615384615385, | |
| "grad_norm": 22.954469680786133, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 3.7173, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.07384615384615385, | |
| "eval_loss": 0.9269228577613831, | |
| "eval_runtime": 75.1147, | |
| "eval_samples_per_second": 10.943, | |
| "eval_steps_per_second": 1.371, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 24.711549758911133, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 3.9268, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.08615384615384615, | |
| "grad_norm": 16.412107467651367, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 3.9029, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.09230769230769231, | |
| "grad_norm": 19.484609603881836, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 3.7293, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.09230769230769231, | |
| "eval_loss": 0.888292133808136, | |
| "eval_runtime": 75.1111, | |
| "eval_samples_per_second": 10.944, | |
| "eval_steps_per_second": 1.371, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.09846153846153846, | |
| "grad_norm": 17.762903213500977, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 3.8463, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.10461538461538461, | |
| "grad_norm": 16.857402801513672, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 3.2797, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.11076923076923077, | |
| "grad_norm": 17.598472595214844, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 3.4031, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.11076923076923077, | |
| "eval_loss": 0.8580312132835388, | |
| "eval_runtime": 75.1055, | |
| "eval_samples_per_second": 10.945, | |
| "eval_steps_per_second": 1.371, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.11692307692307692, | |
| "grad_norm": 19.14990234375, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 3.5648, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.12307692307692308, | |
| "grad_norm": 15.191015243530273, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 3.2272, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12923076923076923, | |
| "grad_norm": 16.499048233032227, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 3.1709, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.12923076923076923, | |
| "eval_loss": 0.8416498303413391, | |
| "eval_runtime": 75.1294, | |
| "eval_samples_per_second": 10.941, | |
| "eval_steps_per_second": 1.371, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.13538461538461538, | |
| "grad_norm": 16.03173828125, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 3.1598, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.14153846153846153, | |
| "grad_norm": 15.822704315185547, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 3.1743, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.1476923076923077, | |
| "grad_norm": 22.87495994567871, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 3.7806, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1476923076923077, | |
| "eval_loss": 0.8304129838943481, | |
| "eval_runtime": 75.1064, | |
| "eval_samples_per_second": 10.944, | |
| "eval_steps_per_second": 1.371, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 17.428983688354492, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 3.6015, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 19.788145065307617, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 3.2009, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.16615384615384615, | |
| "grad_norm": 18.700437545776367, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 3.3891, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.16615384615384615, | |
| "eval_loss": 0.8223448991775513, | |
| "eval_runtime": 75.1008, | |
| "eval_samples_per_second": 10.945, | |
| "eval_steps_per_second": 1.371, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.1723076923076923, | |
| "grad_norm": 16.810592651367188, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 3.3512, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.17846153846153845, | |
| "grad_norm": 17.362895965576172, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 3.5883, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.18461538461538463, | |
| "grad_norm": 22.981599807739258, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 3.2509, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.18461538461538463, | |
| "eval_loss": 0.8187195062637329, | |
| "eval_runtime": 75.1228, | |
| "eval_samples_per_second": 10.942, | |
| "eval_steps_per_second": 1.371, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.19076923076923077, | |
| "grad_norm": 16.67460823059082, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 3.4548, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.19692307692307692, | |
| "grad_norm": 21.03778839111328, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 3.7234, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.20307692307692307, | |
| "grad_norm": 17.65099334716797, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 3.2285, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.20307692307692307, | |
| "eval_loss": 0.8180755376815796, | |
| "eval_runtime": 75.1032, | |
| "eval_samples_per_second": 10.945, | |
| "eval_steps_per_second": 1.371, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.406258997362688e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |