| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0014008545212579673, | |
| "eval_steps": 4, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.669515070859891e-05, | |
| "grad_norm": 7.04739236831665, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 3.7065, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.669515070859891e-05, | |
| "eval_loss": 3.7707154750823975, | |
| "eval_runtime": 1173.2407, | |
| "eval_samples_per_second": 7.686, | |
| "eval_steps_per_second": 1.922, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.339030141719783e-05, | |
| "grad_norm": 6.620449542999268, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 3.6812, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00014008545212579673, | |
| "grad_norm": 6.225563049316406, | |
| "learning_rate": 6e-06, | |
| "loss": 3.5994, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00018678060283439565, | |
| "grad_norm": 8.202012062072754, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 3.7328, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00018678060283439565, | |
| "eval_loss": 3.7528493404388428, | |
| "eval_runtime": 1171.428, | |
| "eval_samples_per_second": 7.698, | |
| "eval_steps_per_second": 1.925, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00023347575354299455, | |
| "grad_norm": 6.334644317626953, | |
| "learning_rate": 1e-05, | |
| "loss": 3.1678, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00028017090425159347, | |
| "grad_norm": 6.9361701011657715, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 3.7147, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00032686605496019236, | |
| "grad_norm": 7.059310436248779, | |
| "learning_rate": 9.842915805643156e-06, | |
| "loss": 3.9362, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0003735612056687913, | |
| "grad_norm": 7.780526161193848, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 4.0172, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0003735612056687913, | |
| "eval_loss": 3.5856099128723145, | |
| "eval_runtime": 1169.9387, | |
| "eval_samples_per_second": 7.708, | |
| "eval_steps_per_second": 1.927, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0004202563563773902, | |
| "grad_norm": 6.989567756652832, | |
| "learning_rate": 9.381533400219319e-06, | |
| "loss": 3.4218, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0004669515070859891, | |
| "grad_norm": 7.0737690925598145, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 3.5655, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.000513646657794588, | |
| "grad_norm": 7.442623138427734, | |
| "learning_rate": 8.644843137107058e-06, | |
| "loss": 3.648, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0005603418085031869, | |
| "grad_norm": 7.113268852233887, | |
| "learning_rate": 8.18711994874345e-06, | |
| "loss": 3.2079, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0005603418085031869, | |
| "eval_loss": 3.3224551677703857, | |
| "eval_runtime": 1172.781, | |
| "eval_samples_per_second": 7.689, | |
| "eval_steps_per_second": 1.923, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0006070369592117859, | |
| "grad_norm": 7.907409191131592, | |
| "learning_rate": 7.679133974894984e-06, | |
| "loss": 3.2207, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0006537321099203847, | |
| "grad_norm": 6.5973124504089355, | |
| "learning_rate": 7.128896457825364e-06, | |
| "loss": 3.2725, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0007004272606289837, | |
| "grad_norm": 6.867315769195557, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 3.1614, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007471224113375826, | |
| "grad_norm": 5.5230865478515625, | |
| "learning_rate": 5.936906572928625e-06, | |
| "loss": 2.6292, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0007471224113375826, | |
| "eval_loss": 3.058130979537964, | |
| "eval_runtime": 1170.8364, | |
| "eval_samples_per_second": 7.702, | |
| "eval_steps_per_second": 1.926, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0007938175620461815, | |
| "grad_norm": 7.0929179191589355, | |
| "learning_rate": 5.3139525976465675e-06, | |
| "loss": 3.1603, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0008405127127547804, | |
| "grad_norm": 6.769837856292725, | |
| "learning_rate": 4.686047402353433e-06, | |
| "loss": 2.8255, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0008872078634633793, | |
| "grad_norm": 5.799893379211426, | |
| "learning_rate": 4.063093427071376e-06, | |
| "loss": 2.664, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0009339030141719782, | |
| "grad_norm": 7.688692569732666, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 3.2977, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0009339030141719782, | |
| "eval_loss": 2.881070852279663, | |
| "eval_runtime": 1170.4506, | |
| "eval_samples_per_second": 7.705, | |
| "eval_steps_per_second": 1.927, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0009805981648805771, | |
| "grad_norm": 7.305593013763428, | |
| "learning_rate": 2.871103542174637e-06, | |
| "loss": 2.8879, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.001027293315589176, | |
| "grad_norm": 6.604673385620117, | |
| "learning_rate": 2.320866025105016e-06, | |
| "loss": 2.6268, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.001073988466297775, | |
| "grad_norm": 7.184069633483887, | |
| "learning_rate": 1.8128800512565514e-06, | |
| "loss": 2.789, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0011206836170063739, | |
| "grad_norm": 7.4371538162231445, | |
| "learning_rate": 1.3551568628929434e-06, | |
| "loss": 2.9913, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0011206836170063739, | |
| "eval_loss": 2.791896343231201, | |
| "eval_runtime": 1170.987, | |
| "eval_samples_per_second": 7.701, | |
| "eval_steps_per_second": 1.926, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0011673787677149727, | |
| "grad_norm": 7.216193675994873, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 3.0909, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0012140739184235718, | |
| "grad_norm": 6.531351089477539, | |
| "learning_rate": 6.184665997806832e-07, | |
| "loss": 2.8438, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0012607690691321706, | |
| "grad_norm": 6.790028095245361, | |
| "learning_rate": 3.511175705587433e-07, | |
| "loss": 2.5153, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0013074642198407694, | |
| "grad_norm": 6.867988586425781, | |
| "learning_rate": 1.5708419435684463e-07, | |
| "loss": 2.7938, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0013074642198407694, | |
| "eval_loss": 2.758043050765991, | |
| "eval_runtime": 1170.5173, | |
| "eval_samples_per_second": 7.704, | |
| "eval_steps_per_second": 1.926, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0013541593705493685, | |
| "grad_norm": 7.407482147216797, | |
| "learning_rate": 3.9426493427611177e-08, | |
| "loss": 2.8072, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0014008545212579673, | |
| "grad_norm": 8.236345291137695, | |
| "learning_rate": 0.0, | |
| "loss": 2.8171, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.150341542313984e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |