| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 3, | |
| "global_step": 36, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08333333333333333, | |
| "grad_norm": 0.8625234365463257, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4145, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08333333333333333, | |
| "eval_loss": 1.3638261556625366, | |
| "eval_runtime": 14.6873, | |
| "eval_samples_per_second": 0.34, | |
| "eval_steps_per_second": 0.204, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 0.554802656173706, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3786, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.6293703317642212, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4133, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.3478872776031494, | |
| "eval_runtime": 14.8158, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 1.862587571144104, | |
| "learning_rate": 8e-05, | |
| "loss": 1.3524, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 0.4966895878314972, | |
| "learning_rate": 0.0001, | |
| "loss": 1.247, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.4830384254455566, | |
| "learning_rate": 0.00012, | |
| "loss": 1.1718, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.0839580297470093, | |
| "eval_runtime": 14.8345, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.5833333333333334, | |
| "grad_norm": 0.8882748484611511, | |
| "learning_rate": 0.00014, | |
| "loss": 1.06, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.3599146902561188, | |
| "learning_rate": 0.00016, | |
| "loss": 0.9756, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.5338190793991089, | |
| "learning_rate": 0.00018, | |
| "loss": 0.8807, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.8535793423652649, | |
| "eval_runtime": 14.8372, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 0.8406828045845032, | |
| "learning_rate": 0.0002, | |
| "loss": 0.797, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.9166666666666666, | |
| "grad_norm": 0.4235399067401886, | |
| "learning_rate": 0.0001992708874098054, | |
| "loss": 0.7843, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.5914283990859985, | |
| "learning_rate": 0.0001970941817426052, | |
| "loss": 0.7696, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7616702914237976, | |
| "eval_runtime": 14.8238, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0833333333333333, | |
| "grad_norm": 0.2914833426475525, | |
| "learning_rate": 0.0001935016242685415, | |
| "loss": 0.6477, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.1666666666666667, | |
| "grad_norm": 0.297875314950943, | |
| "learning_rate": 0.000188545602565321, | |
| "loss": 0.6373, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.3945733904838562, | |
| "learning_rate": 0.00018229838658936564, | |
| "loss": 0.5582, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 0.707451343536377, | |
| "eval_runtime": 14.8217, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.32392948865890503, | |
| "learning_rate": 0.00017485107481711012, | |
| "loss": 0.6144, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.4166666666666667, | |
| "grad_norm": 0.3210783004760742, | |
| "learning_rate": 0.00016631226582407952, | |
| "loss": 0.6076, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.3447917401790619, | |
| "learning_rate": 0.00015680647467311557, | |
| "loss": 0.5734, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 0.6850055456161499, | |
| "eval_runtime": 14.8208, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.5833333333333335, | |
| "grad_norm": 0.3459540605545044, | |
| "learning_rate": 0.00014647231720437686, | |
| "loss": 0.5842, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.3534122407436371, | |
| "learning_rate": 0.00013546048870425356, | |
| "loss": 0.5423, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.36839067935943604, | |
| "learning_rate": 0.0001239315664287558, | |
| "loss": 0.5593, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 0.6518851518630981, | |
| "eval_runtime": 14.82, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.8333333333333335, | |
| "grad_norm": 0.31901416182518005, | |
| "learning_rate": 0.0001120536680255323, | |
| "loss": 0.5515, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.9166666666666665, | |
| "grad_norm": 0.41241687536239624, | |
| "learning_rate": 0.0001, | |
| "loss": 0.59, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.3511301279067993, | |
| "learning_rate": 8.79463319744677e-05, | |
| "loss": 0.5131, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6314570307731628, | |
| "eval_runtime": 14.8291, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "grad_norm": 0.2581290900707245, | |
| "learning_rate": 7.606843357124426e-05, | |
| "loss": 0.4988, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.1666666666666665, | |
| "grad_norm": 0.2543998658657074, | |
| "learning_rate": 6.453951129574644e-05, | |
| "loss": 0.4241, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.25313320755958557, | |
| "learning_rate": 5.3527682795623146e-05, | |
| "loss": 0.4138, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 0.6262981295585632, | |
| "eval_runtime": 14.8268, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "grad_norm": 0.26558130979537964, | |
| "learning_rate": 4.3193525326884435e-05, | |
| "loss": 0.3564, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.4166666666666665, | |
| "grad_norm": 0.27149203419685364, | |
| "learning_rate": 3.36877341759205e-05, | |
| "loss": 0.3952, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.2739770710468292, | |
| "learning_rate": 2.514892518288988e-05, | |
| "loss": 0.3607, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 0.6266031265258789, | |
| "eval_runtime": 14.8414, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.5833333333333335, | |
| "grad_norm": 0.24796062707901, | |
| "learning_rate": 1.7701613410634365e-05, | |
| "loss": 0.3474, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.30042949318885803, | |
| "learning_rate": 1.1454397434679021e-05, | |
| "loss": 0.3602, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.299396812915802, | |
| "learning_rate": 6.498375731458528e-06, | |
| "loss": 0.3951, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 0.627160906791687, | |
| "eval_runtime": 14.8411, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.8333333333333335, | |
| "grad_norm": 0.27229976654052734, | |
| "learning_rate": 2.905818257394799e-06, | |
| "loss": 0.3975, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.9166666666666665, | |
| "grad_norm": 0.2533663213253021, | |
| "learning_rate": 7.291125901946027e-07, | |
| "loss": 0.2966, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.29661864042282104, | |
| "learning_rate": 0.0, | |
| "loss": 0.345, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6269396543502808, | |
| "eval_runtime": 14.8316, | |
| "eval_samples_per_second": 0.337, | |
| "eval_steps_per_second": 0.202, | |
| "step": 36 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 36, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 12, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.8971140168679424e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |