| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04697091006138244, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00047445363698366103, | |
| "eval_loss": 1.2206629514694214, | |
| "eval_runtime": 2455.084, | |
| "eval_samples_per_second": 2.892, | |
| "eval_steps_per_second": 1.446, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001423360910950983, | |
| "grad_norm": 42.169185638427734, | |
| "learning_rate": 3e-05, | |
| "loss": 38.8115, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002846721821901966, | |
| "grad_norm": 30.064516067504883, | |
| "learning_rate": 6e-05, | |
| "loss": 36.5887, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004270082732852949, | |
| "grad_norm": 24.0466365814209, | |
| "learning_rate": 9e-05, | |
| "loss": 31.7486, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004270082732852949, | |
| "eval_loss": 0.872442364692688, | |
| "eval_runtime": 2457.1575, | |
| "eval_samples_per_second": 2.89, | |
| "eval_steps_per_second": 1.445, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005693443643803932, | |
| "grad_norm": 26.121597290039062, | |
| "learning_rate": 0.00012, | |
| "loss": 28.0893, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.007116804554754915, | |
| "grad_norm": 22.931711196899414, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 26.3657, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.008540165465705898, | |
| "grad_norm": 15.313858032226562, | |
| "learning_rate": 0.00018, | |
| "loss": 22.876, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008540165465705898, | |
| "eval_loss": 0.7377527952194214, | |
| "eval_runtime": 730.2515, | |
| "eval_samples_per_second": 9.723, | |
| "eval_steps_per_second": 4.861, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.009963526376656881, | |
| "grad_norm": 14.844147682189941, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 23.8388, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.011386887287607865, | |
| "grad_norm": 13.040925025939941, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 23.2438, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.012810248198558846, | |
| "grad_norm": 13.732440948486328, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 23.504, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.012810248198558846, | |
| "eval_loss": 0.708083987236023, | |
| "eval_runtime": 2490.0558, | |
| "eval_samples_per_second": 2.851, | |
| "eval_steps_per_second": 1.426, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01423360910950983, | |
| "grad_norm": 15.219266891479492, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 23.009, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.015656970020460813, | |
| "grad_norm": 11.847440719604492, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 21.94, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.017080330931411796, | |
| "grad_norm": 11.252641677856445, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 22.467, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.017080330931411796, | |
| "eval_loss": 0.6968169808387756, | |
| "eval_runtime": 1821.4615, | |
| "eval_samples_per_second": 3.898, | |
| "eval_steps_per_second": 1.949, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01850369184236278, | |
| "grad_norm": 11.163553237915039, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 22.5554, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.019927052753313763, | |
| "grad_norm": 13.20541763305664, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 22.9775, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.021350413664264746, | |
| "grad_norm": 10.816298484802246, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 22.2379, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.021350413664264746, | |
| "eval_loss": 0.6865644454956055, | |
| "eval_runtime": 673.7961, | |
| "eval_samples_per_second": 10.537, | |
| "eval_steps_per_second": 5.269, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.02277377457521573, | |
| "grad_norm": 10.934300422668457, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 21.4568, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.024197135486166713, | |
| "grad_norm": 10.668505668640137, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 21.2971, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.025620496397117692, | |
| "grad_norm": 10.522425651550293, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 22.4085, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.025620496397117692, | |
| "eval_loss": 0.6808061003684998, | |
| "eval_runtime": 674.5537, | |
| "eval_samples_per_second": 10.525, | |
| "eval_steps_per_second": 5.263, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.027043857308068676, | |
| "grad_norm": 10.834729194641113, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 22.0063, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.02846721821901966, | |
| "grad_norm": 10.45215892791748, | |
| "learning_rate": 0.0001, | |
| "loss": 21.7114, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.029890579129970642, | |
| "grad_norm": 10.509422302246094, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 22.147, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.029890579129970642, | |
| "eval_loss": 0.6736528873443604, | |
| "eval_runtime": 674.4225, | |
| "eval_samples_per_second": 10.528, | |
| "eval_steps_per_second": 5.264, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.031313940040921626, | |
| "grad_norm": 10.096047401428223, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 21.0243, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.03273730095187261, | |
| "grad_norm": 10.614026069641113, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 22.6279, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.03416066186282359, | |
| "grad_norm": 9.915692329406738, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 21.7931, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03416066186282359, | |
| "eval_loss": 0.6665549278259277, | |
| "eval_runtime": 673.4954, | |
| "eval_samples_per_second": 10.542, | |
| "eval_steps_per_second": 5.271, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03558402277377457, | |
| "grad_norm": 10.606143951416016, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 21.1355, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.03700738368472556, | |
| "grad_norm": 9.829690933227539, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 20.704, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.03843074459567654, | |
| "grad_norm": 9.532461166381836, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 20.9497, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.03843074459567654, | |
| "eval_loss": 0.6634002327919006, | |
| "eval_runtime": 673.9541, | |
| "eval_samples_per_second": 10.535, | |
| "eval_steps_per_second": 5.267, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.039854105506627525, | |
| "grad_norm": 10.550149917602539, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 21.4862, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.041277466417578505, | |
| "grad_norm": 9.829084396362305, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 21.2587, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.04270082732852949, | |
| "grad_norm": 10.305970191955566, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 21.4128, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04270082732852949, | |
| "eval_loss": 0.6616738438606262, | |
| "eval_runtime": 674.758, | |
| "eval_samples_per_second": 10.522, | |
| "eval_steps_per_second": 5.261, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04412418823948047, | |
| "grad_norm": 10.540351867675781, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 22.0735, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.04554754915043146, | |
| "grad_norm": 10.391834259033203, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 22.2369, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.04697091006138244, | |
| "grad_norm": 10.5779390335083, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 21.495, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.04697091006138244, | |
| "eval_loss": 0.6612978577613831, | |
| "eval_runtime": 674.8148, | |
| "eval_samples_per_second": 10.521, | |
| "eval_steps_per_second": 5.261, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.2196614900128154e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |