| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.011701380762930026, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00011701380762930026, | |
| "eval_loss": 5.071834564208984, | |
| "eval_runtime": 261.8333, | |
| "eval_samples_per_second": 54.974, | |
| "eval_steps_per_second": 6.875, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00035104142288790077, | |
| "grad_norm": 6.64661169052124, | |
| "learning_rate": 3e-05, | |
| "loss": 4.9241, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0007020828457758015, | |
| "grad_norm": 3.513359785079956, | |
| "learning_rate": 6e-05, | |
| "loss": 4.6983, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0010531242686637023, | |
| "grad_norm": 3.402698278427124, | |
| "learning_rate": 9e-05, | |
| "loss": 4.7042, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0010531242686637023, | |
| "eval_loss": 4.4165263175964355, | |
| "eval_runtime": 262.1486, | |
| "eval_samples_per_second": 54.908, | |
| "eval_steps_per_second": 6.866, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.001404165691551603, | |
| "grad_norm": 3.1393704414367676, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 4.1249, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0017552071144395038, | |
| "grad_norm": 2.3505287170410156, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.8557, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0021062485373274046, | |
| "grad_norm": 2.0289347171783447, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.9179, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0021062485373274046, | |
| "eval_loss": 3.685943841934204, | |
| "eval_runtime": 261.9757, | |
| "eval_samples_per_second": 54.944, | |
| "eval_steps_per_second": 6.871, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0024572899602153054, | |
| "grad_norm": 2.3705809116363525, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 3.6618, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.002808331383103206, | |
| "grad_norm": 2.002850294113159, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.4756, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.003159372805991107, | |
| "grad_norm": 1.663278579711914, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 3.3452, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.003159372805991107, | |
| "eval_loss": 3.578341245651245, | |
| "eval_runtime": 262.0449, | |
| "eval_samples_per_second": 54.93, | |
| "eval_steps_per_second": 6.869, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0035104142288790077, | |
| "grad_norm": 1.877441167831421, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 3.4687, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0038614556517669084, | |
| "grad_norm": 1.7343926429748535, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.4485, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.004212497074654809, | |
| "grad_norm": 1.607003927230835, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 3.414, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.004212497074654809, | |
| "eval_loss": 3.5100536346435547, | |
| "eval_runtime": 261.9844, | |
| "eval_samples_per_second": 54.942, | |
| "eval_steps_per_second": 6.871, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.00456353849754271, | |
| "grad_norm": 2.106229305267334, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 3.4085, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.004914579920430611, | |
| "grad_norm": 1.4526017904281616, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 3.4435, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0052656213433185115, | |
| "grad_norm": 1.8056683540344238, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 3.6258, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0052656213433185115, | |
| "eval_loss": 3.472996234893799, | |
| "eval_runtime": 262.4211, | |
| "eval_samples_per_second": 54.851, | |
| "eval_steps_per_second": 6.859, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.005616662766206412, | |
| "grad_norm": 2.4865214824676514, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 3.4346, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.005967704189094313, | |
| "grad_norm": 1.5772249698638916, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 3.4463, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.006318745611982214, | |
| "grad_norm": 1.514796257019043, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 3.4054, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.006318745611982214, | |
| "eval_loss": 3.453535318374634, | |
| "eval_runtime": 262.1313, | |
| "eval_samples_per_second": 54.911, | |
| "eval_steps_per_second": 6.867, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0066697870348701145, | |
| "grad_norm": 1.575348973274231, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 3.4188, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.007020828457758015, | |
| "grad_norm": 1.889758586883545, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 3.2958, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.007371869880645916, | |
| "grad_norm": 1.7029871940612793, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 3.1771, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.007371869880645916, | |
| "eval_loss": 3.4409759044647217, | |
| "eval_runtime": 262.1868, | |
| "eval_samples_per_second": 54.9, | |
| "eval_steps_per_second": 6.865, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.007722911303533817, | |
| "grad_norm": 1.5269502401351929, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 3.4087, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.008073952726421718, | |
| "grad_norm": 2.0026628971099854, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 3.4728, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.008424994149309618, | |
| "grad_norm": 1.3194637298583984, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 3.1512, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.008424994149309618, | |
| "eval_loss": 3.4321975708007812, | |
| "eval_runtime": 262.2406, | |
| "eval_samples_per_second": 54.889, | |
| "eval_steps_per_second": 6.864, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.00877603557219752, | |
| "grad_norm": 1.8826606273651123, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 3.4841, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.00912707699508542, | |
| "grad_norm": 2.0330681800842285, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 3.3632, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.009478118417973322, | |
| "grad_norm": 1.4071763753890991, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 3.2883, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.009478118417973322, | |
| "eval_loss": 3.425398588180542, | |
| "eval_runtime": 262.4725, | |
| "eval_samples_per_second": 54.84, | |
| "eval_steps_per_second": 6.858, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.009829159840861221, | |
| "grad_norm": 1.6387964487075806, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 3.1512, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.010180201263749123, | |
| "grad_norm": 1.5741945505142212, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 3.2649, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.010531242686637023, | |
| "grad_norm": 1.4074112176895142, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 3.2711, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.010531242686637023, | |
| "eval_loss": 3.4211525917053223, | |
| "eval_runtime": 262.4203, | |
| "eval_samples_per_second": 54.851, | |
| "eval_steps_per_second": 6.859, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.010882284109524925, | |
| "grad_norm": 1.1304495334625244, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 3.0629, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.011233325532412824, | |
| "grad_norm": 1.7654685974121094, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 3.3242, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.011584366955300726, | |
| "grad_norm": 1.1728322505950928, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2.992, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.011584366955300726, | |
| "eval_loss": 3.4203412532806396, | |
| "eval_runtime": 262.2463, | |
| "eval_samples_per_second": 54.887, | |
| "eval_steps_per_second": 6.864, | |
| "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": 2.05873754406912e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |