| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.2039775624681285, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002039775624681285, | |
| "eval_loss": 2.7500362396240234, | |
| "eval_runtime": 4.8128, | |
| "eval_samples_per_second": 171.624, | |
| "eval_steps_per_second": 21.609, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006119326874043855, | |
| "grad_norm": 119.72549438476562, | |
| "learning_rate": 3e-05, | |
| "loss": 10.7627, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01223865374808771, | |
| "grad_norm": 95.31661987304688, | |
| "learning_rate": 6e-05, | |
| "loss": 10.7585, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018357980622131564, | |
| "grad_norm": 111.39752960205078, | |
| "learning_rate": 9e-05, | |
| "loss": 10.8945, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.018357980622131564, | |
| "eval_loss": 2.742156505584717, | |
| "eval_runtime": 4.8217, | |
| "eval_samples_per_second": 171.311, | |
| "eval_steps_per_second": 21.569, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02447730749617542, | |
| "grad_norm": 87.65267944335938, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 11.0519, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.030596634370219276, | |
| "grad_norm": 95.65772247314453, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 10.8906, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03671596124426313, | |
| "grad_norm": 92.90438079833984, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 10.9398, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03671596124426313, | |
| "eval_loss": 2.678375720977783, | |
| "eval_runtime": 4.8305, | |
| "eval_samples_per_second": 170.996, | |
| "eval_steps_per_second": 21.53, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04283528811830699, | |
| "grad_norm": 93.3251724243164, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 10.3163, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04895461499235084, | |
| "grad_norm": 105.36555480957031, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 10.66, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.05507394186639469, | |
| "grad_norm": 104.30107116699219, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 10.3486, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.05507394186639469, | |
| "eval_loss": 2.6385228633880615, | |
| "eval_runtime": 4.8602, | |
| "eval_samples_per_second": 169.95, | |
| "eval_steps_per_second": 21.398, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06119326874043855, | |
| "grad_norm": 153.6848602294922, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 10.7138, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06731259561448241, | |
| "grad_norm": 126.95474243164062, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 10.0417, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.07343192248852626, | |
| "grad_norm": 170.2035369873047, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 10.5315, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.07343192248852626, | |
| "eval_loss": 2.6369853019714355, | |
| "eval_runtime": 4.8791, | |
| "eval_samples_per_second": 169.294, | |
| "eval_steps_per_second": 21.315, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.07955124936257012, | |
| "grad_norm": 132.4921875, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 10.5915, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.08567057623661398, | |
| "grad_norm": 109.94354248046875, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 10.4331, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.09178990311065782, | |
| "grad_norm": 100.11196899414062, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 10.891, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.09178990311065782, | |
| "eval_loss": 2.6252143383026123, | |
| "eval_runtime": 4.8716, | |
| "eval_samples_per_second": 169.555, | |
| "eval_steps_per_second": 21.348, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.09790922998470168, | |
| "grad_norm": 112.04855346679688, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 9.8393, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.10402855685874554, | |
| "grad_norm": 119.39227294921875, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 10.1454, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.11014788373278939, | |
| "grad_norm": 202.7808837890625, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 10.4947, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.11014788373278939, | |
| "eval_loss": 2.626685857772827, | |
| "eval_runtime": 4.8065, | |
| "eval_samples_per_second": 171.849, | |
| "eval_steps_per_second": 21.637, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.11626721060683325, | |
| "grad_norm": 114.66297149658203, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 10.7594, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.1223865374808771, | |
| "grad_norm": 120.82008361816406, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 10.5164, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12850586435492095, | |
| "grad_norm": 128.4127197265625, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 10.219, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.12850586435492095, | |
| "eval_loss": 2.6237473487854004, | |
| "eval_runtime": 4.8342, | |
| "eval_samples_per_second": 170.867, | |
| "eval_steps_per_second": 21.513, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.13462519122896482, | |
| "grad_norm": 99.31004333496094, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 10.3523, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.14074451810300867, | |
| "grad_norm": 106.18312072753906, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 10.6466, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.14686384497705252, | |
| "grad_norm": 143.47616577148438, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 10.473, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.14686384497705252, | |
| "eval_loss": 2.621034622192383, | |
| "eval_runtime": 4.8122, | |
| "eval_samples_per_second": 171.646, | |
| "eval_steps_per_second": 21.612, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1529831718510964, | |
| "grad_norm": 104.74443054199219, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 10.524, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.15910249872514023, | |
| "grad_norm": 114.83618927001953, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 10.5378, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.16522182559918408, | |
| "grad_norm": 140.494140625, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 10.4252, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.16522182559918408, | |
| "eval_loss": 2.618023157119751, | |
| "eval_runtime": 4.8413, | |
| "eval_samples_per_second": 170.617, | |
| "eval_steps_per_second": 21.482, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.17134115247322795, | |
| "grad_norm": 105.02456665039062, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 10.2756, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.1774604793472718, | |
| "grad_norm": 115.14602661132812, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 10.1504, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.18357980622131564, | |
| "grad_norm": 102.13778686523438, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 10.736, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.18357980622131564, | |
| "eval_loss": 2.619086742401123, | |
| "eval_runtime": 4.8863, | |
| "eval_samples_per_second": 169.043, | |
| "eval_steps_per_second": 21.284, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.18969913309535952, | |
| "grad_norm": 105.03943634033203, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 10.5, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.19581845996940336, | |
| "grad_norm": 116.5811767578125, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 10.7438, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2019377868434472, | |
| "grad_norm": 120.85993957519531, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 10.4505, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.2019377868434472, | |
| "eval_loss": 2.619232177734375, | |
| "eval_runtime": 4.8553, | |
| "eval_samples_per_second": 170.123, | |
| "eval_steps_per_second": 21.42, | |
| "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": 3154709430927360.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |