| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.036281837312241494, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007256367462448298, | |
| "grad_norm": 45.75, | |
| "learning_rate": 5e-06, | |
| "loss": 1.9641, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0014512734924896596, | |
| "grad_norm": 46.5, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7641, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0021769102387344894, | |
| "grad_norm": 35.75, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.4723, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0029025469849793192, | |
| "grad_norm": 31.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0985, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.003628183731224149, | |
| "grad_norm": 28.875, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.0381, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.004353820477468979, | |
| "grad_norm": 26.625, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6977, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005079457223713809, | |
| "grad_norm": 23.75, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.6392, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.0058050939699586385, | |
| "grad_norm": 26.5, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7459, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.006530730716203468, | |
| "grad_norm": 29.875, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6155, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.007256367462448298, | |
| "grad_norm": 21.5, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3941, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.007982004208693128, | |
| "grad_norm": 17.0, | |
| "learning_rate": 4.9999934086574596e-05, | |
| "loss": 0.4513, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.008707640954937958, | |
| "grad_norm": 23.875, | |
| "learning_rate": 4.9999736346645943e-05, | |
| "loss": 0.5292, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.009433277701182788, | |
| "grad_norm": 17.25, | |
| "learning_rate": 4.999940678125673e-05, | |
| "loss": 0.3869, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.010158914447427617, | |
| "grad_norm": 24.75, | |
| "learning_rate": 4.9998945392144796e-05, | |
| "loss": 0.4121, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.010884551193672447, | |
| "grad_norm": 16.625, | |
| "learning_rate": 4.999835218174307e-05, | |
| "loss": 0.4199, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.011610187939917277, | |
| "grad_norm": 11.8125, | |
| "learning_rate": 4.99976271531796e-05, | |
| "loss": 0.4694, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.012335824686162107, | |
| "grad_norm": 15.9375, | |
| "learning_rate": 4.9996770310277506e-05, | |
| "loss": 0.3736, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.013061461432406937, | |
| "grad_norm": 18.0, | |
| "learning_rate": 4.9995781657555e-05, | |
| "loss": 0.4224, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.013787098178651766, | |
| "grad_norm": 23.0, | |
| "learning_rate": 4.99946612002253e-05, | |
| "loss": 0.4307, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.014512734924896596, | |
| "grad_norm": 19.375, | |
| "learning_rate": 4.9993408944196676e-05, | |
| "loss": 0.3449, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.015238371671141426, | |
| "grad_norm": 11.0, | |
| "learning_rate": 4.9992024896072364e-05, | |
| "loss": 0.3243, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.015964008417386256, | |
| "grad_norm": 17.875, | |
| "learning_rate": 4.999050906315055e-05, | |
| "loss": 0.3002, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.016689645163631087, | |
| "grad_norm": 6.875, | |
| "learning_rate": 4.998886145342434e-05, | |
| "loss": 0.3882, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.017415281909875915, | |
| "grad_norm": 12.125, | |
| "learning_rate": 4.9987082075581684e-05, | |
| "loss": 0.4211, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.018140918656120747, | |
| "grad_norm": 16.875, | |
| "learning_rate": 4.9985170939005386e-05, | |
| "loss": 0.2967, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.018866555402365575, | |
| "grad_norm": 26.125, | |
| "learning_rate": 4.998312805377302e-05, | |
| "loss": 0.3283, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.019592192148610407, | |
| "grad_norm": 18.0, | |
| "learning_rate": 4.998095343065685e-05, | |
| "loss": 0.3455, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.020317828894855235, | |
| "grad_norm": 15.4375, | |
| "learning_rate": 4.997864708112384e-05, | |
| "loss": 0.2891, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.021043465641100066, | |
| "grad_norm": 30.25, | |
| "learning_rate": 4.997620901733554e-05, | |
| "loss": 0.3645, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.021769102387344894, | |
| "grad_norm": 9.875, | |
| "learning_rate": 4.997363925214803e-05, | |
| "loss": 0.3047, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.022494739133589726, | |
| "grad_norm": 18.625, | |
| "learning_rate": 4.9970937799111896e-05, | |
| "loss": 0.2943, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.023220375879834554, | |
| "grad_norm": 30.625, | |
| "learning_rate": 4.996810467247207e-05, | |
| "loss": 0.3315, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.023946012626079385, | |
| "grad_norm": 14.3125, | |
| "learning_rate": 4.9965139887167856e-05, | |
| "loss": 0.2473, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.024671649372324213, | |
| "grad_norm": 21.5, | |
| "learning_rate": 4.996204345883278e-05, | |
| "loss": 0.3173, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.025397286118569045, | |
| "grad_norm": 16.125, | |
| "learning_rate": 4.9958815403794546e-05, | |
| "loss": 0.3813, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.026122922864813873, | |
| "grad_norm": 15.0625, | |
| "learning_rate": 4.995545573907492e-05, | |
| "loss": 0.3366, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.026848559611058705, | |
| "grad_norm": 19.5, | |
| "learning_rate": 4.995196448238966e-05, | |
| "loss": 0.3188, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.027574196357303533, | |
| "grad_norm": 20.125, | |
| "learning_rate": 4.9948341652148436e-05, | |
| "loss": 0.2776, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.028299833103548364, | |
| "grad_norm": 17.25, | |
| "learning_rate": 4.994458726745468e-05, | |
| "loss": 0.3574, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.029025469849793192, | |
| "grad_norm": 5.375, | |
| "learning_rate": 4.9940701348105554e-05, | |
| "loss": 0.2566, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.029751106596038024, | |
| "grad_norm": 31.0, | |
| "learning_rate": 4.99366839145918e-05, | |
| "loss": 0.275, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.030476743342282852, | |
| "grad_norm": 21.75, | |
| "learning_rate": 4.993253498809762e-05, | |
| "loss": 0.2928, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.031202380088527683, | |
| "grad_norm": 21.25, | |
| "learning_rate": 4.9928254590500646e-05, | |
| "loss": 0.3466, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.03192801683477251, | |
| "grad_norm": 15.25, | |
| "learning_rate": 4.9923842744371707e-05, | |
| "loss": 0.2911, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.03265365358101734, | |
| "grad_norm": 8.9375, | |
| "learning_rate": 4.99192994729748e-05, | |
| "loss": 0.2153, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.033379290327262175, | |
| "grad_norm": 20.5, | |
| "learning_rate": 4.991462480026693e-05, | |
| "loss": 0.383, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.034104927073507, | |
| "grad_norm": 6.46875, | |
| "learning_rate": 4.9909818750898e-05, | |
| "loss": 0.3099, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.03483056381975183, | |
| "grad_norm": 34.0, | |
| "learning_rate": 4.990488135021065e-05, | |
| "loss": 0.2897, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.03555620056599666, | |
| "grad_norm": 7.78125, | |
| "learning_rate": 4.989981262424017e-05, | |
| "loss": 0.3098, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.036281837312241494, | |
| "grad_norm": 41.0, | |
| "learning_rate": 4.989461259971432e-05, | |
| "loss": 0.2428, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 13781, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 0.0, | |
| "train_batch_size": 5, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |