| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.046667185190946564, | |
| "eval_steps": 13, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00031111456793964375, | |
| "eval_loss": 3.795616626739502, | |
| "eval_runtime": 387.5587, | |
| "eval_samples_per_second": 13.969, | |
| "eval_steps_per_second": 1.747, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0009333437038189313, | |
| "grad_norm": 1.8201688528060913, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.6704, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0018666874076378626, | |
| "grad_norm": 1.9469057321548462, | |
| "learning_rate": 3e-05, | |
| "loss": 3.829, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002800031111456794, | |
| "grad_norm": 2.4936416149139404, | |
| "learning_rate": 4.5e-05, | |
| "loss": 3.6616, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003733374815275725, | |
| "grad_norm": 3.112192392349243, | |
| "learning_rate": 4.997482666353287e-05, | |
| "loss": 3.1786, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004044489383215369, | |
| "eval_loss": 2.5751025676727295, | |
| "eval_runtime": 389.7804, | |
| "eval_samples_per_second": 13.89, | |
| "eval_steps_per_second": 1.737, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.004666718519094656, | |
| "grad_norm": 2.7909445762634277, | |
| "learning_rate": 4.984280524733107e-05, | |
| "loss": 2.6525, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005600062222913588, | |
| "grad_norm": 2.6461470127105713, | |
| "learning_rate": 4.959823971496574e-05, | |
| "loss": 2.2006, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00653340592673252, | |
| "grad_norm": 2.1072444915771484, | |
| "learning_rate": 4.9242238009417175e-05, | |
| "loss": 1.9303, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00746674963055145, | |
| "grad_norm": 1.8727749586105347, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 1.8151, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.008088978766430738, | |
| "eval_loss": 1.7247991561889648, | |
| "eval_runtime": 389.9247, | |
| "eval_samples_per_second": 13.885, | |
| "eval_steps_per_second": 1.736, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.008400093334370381, | |
| "grad_norm": 1.5817272663116455, | |
| "learning_rate": 4.820287471297598e-05, | |
| "loss": 1.6959, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.009333437038189313, | |
| "grad_norm": 1.7775505781173706, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 1.7599, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.010266780742008244, | |
| "grad_norm": 1.512369155883789, | |
| "learning_rate": 4.674352832889239e-05, | |
| "loss": 1.6668, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.011200124445827176, | |
| "grad_norm": 1.427876353263855, | |
| "learning_rate": 4.586433134303257e-05, | |
| "loss": 1.6359, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.012133468149646108, | |
| "grad_norm": 1.557162880897522, | |
| "learning_rate": 4.489061372204453e-05, | |
| "loss": 1.5442, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.012133468149646108, | |
| "eval_loss": 1.5903956890106201, | |
| "eval_runtime": 389.6183, | |
| "eval_samples_per_second": 13.896, | |
| "eval_steps_per_second": 1.738, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01306681185346504, | |
| "grad_norm": 1.3834956884384155, | |
| "learning_rate": 4.382678665009028e-05, | |
| "loss": 1.5914, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01400015555728397, | |
| "grad_norm": 1.2795754671096802, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 1.512, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0149334992611029, | |
| "grad_norm": 1.383254885673523, | |
| "learning_rate": 4.144846814849282e-05, | |
| "loss": 1.5425, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.015866842964921832, | |
| "grad_norm": 1.4243319034576416, | |
| "learning_rate": 4.01447510960205e-05, | |
| "loss": 1.5153, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.016177957532861477, | |
| "eval_loss": 1.5446308851242065, | |
| "eval_runtime": 390.0683, | |
| "eval_samples_per_second": 13.88, | |
| "eval_steps_per_second": 1.736, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.016800186668740762, | |
| "grad_norm": 1.3810615539550781, | |
| "learning_rate": 3.8772424536302564e-05, | |
| "loss": 1.5528, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.017733530372559696, | |
| "grad_norm": 1.4023598432540894, | |
| "learning_rate": 3.7337705451608674e-05, | |
| "loss": 1.5652, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.018666874076378626, | |
| "grad_norm": 1.2947098016738892, | |
| "learning_rate": 3.5847093477938956e-05, | |
| "loss": 1.5906, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.01960021778019756, | |
| "grad_norm": 1.40787935256958, | |
| "learning_rate": 3.4307341460048633e-05, | |
| "loss": 1.4431, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.020222446916076844, | |
| "eval_loss": 1.5237524509429932, | |
| "eval_runtime": 389.7942, | |
| "eval_samples_per_second": 13.889, | |
| "eval_steps_per_second": 1.737, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.02053356148401649, | |
| "grad_norm": 1.2801828384399414, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 1.5227, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.02146690518783542, | |
| "grad_norm": 1.3553149700164795, | |
| "learning_rate": 3.110851015344735e-05, | |
| "loss": 1.5033, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.022400248891654352, | |
| "grad_norm": 1.339362621307373, | |
| "learning_rate": 2.9463922369965917e-05, | |
| "loss": 1.517, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.023333592595473282, | |
| "grad_norm": 1.307166337966919, | |
| "learning_rate": 2.7799111902582696e-05, | |
| "loss": 1.4759, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.024266936299292215, | |
| "grad_norm": 1.2301127910614014, | |
| "learning_rate": 2.6121620758762877e-05, | |
| "loss": 1.4794, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.024266936299292215, | |
| "eval_loss": 1.506866216659546, | |
| "eval_runtime": 389.7553, | |
| "eval_samples_per_second": 13.891, | |
| "eval_steps_per_second": 1.737, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.025200280003111145, | |
| "grad_norm": 1.2243834733963013, | |
| "learning_rate": 2.443904839260488e-05, | |
| "loss": 1.4497, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.02613362370693008, | |
| "grad_norm": 1.1867769956588745, | |
| "learning_rate": 2.2759017277414166e-05, | |
| "loss": 1.4718, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.02706696741074901, | |
| "grad_norm": 1.1967134475708008, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 1.5324, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.02800031111456794, | |
| "grad_norm": 1.2109935283660889, | |
| "learning_rate": 1.9436976651092144e-05, | |
| "loss": 1.5148, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.028311425682507583, | |
| "eval_loss": 1.4927889108657837, | |
| "eval_runtime": 389.8573, | |
| "eval_samples_per_second": 13.887, | |
| "eval_steps_per_second": 1.737, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.028933654818386872, | |
| "grad_norm": 1.2042676210403442, | |
| "learning_rate": 1.781001681419957e-05, | |
| "loss": 1.5146, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.0298669985222058, | |
| "grad_norm": 1.2717711925506592, | |
| "learning_rate": 1.621562939796643e-05, | |
| "loss": 1.4681, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.030800342226024735, | |
| "grad_norm": 1.1750036478042603, | |
| "learning_rate": 1.466103737583699e-05, | |
| "loss": 1.5074, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.031733685929843665, | |
| "grad_norm": 1.3590319156646729, | |
| "learning_rate": 1.3153283438175034e-05, | |
| "loss": 1.5168, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 0.032355915065722954, | |
| "eval_loss": 1.4859765768051147, | |
| "eval_runtime": 389.9351, | |
| "eval_samples_per_second": 13.884, | |
| "eval_steps_per_second": 1.736, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.032667029633662595, | |
| "grad_norm": 1.15152907371521, | |
| "learning_rate": 1.1699198087116589e-05, | |
| "loss": 1.4838, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.033600373337481525, | |
| "grad_norm": 1.2240478992462158, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 1.5484, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.03453371704130046, | |
| "grad_norm": 1.2042181491851807, | |
| "learning_rate": 8.978109650374397e-06, | |
| "loss": 1.5173, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.03546706074511939, | |
| "grad_norm": 1.205437183380127, | |
| "learning_rate": 7.723433775328384e-06, | |
| "loss": 1.472, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 0.03640040444893832, | |
| "grad_norm": 1.2650120258331299, | |
| "learning_rate": 6.547025062816486e-06, | |
| "loss": 1.5346, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.03640040444893832, | |
| "eval_loss": 1.4806684255599976, | |
| "eval_runtime": 390.1832, | |
| "eval_samples_per_second": 13.876, | |
| "eval_steps_per_second": 1.735, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.03733374815275725, | |
| "grad_norm": 1.1355212926864624, | |
| "learning_rate": 5.454212938299255e-06, | |
| "loss": 1.4929, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.03826709185657618, | |
| "grad_norm": 1.2554372549057007, | |
| "learning_rate": 4.4499481138022544e-06, | |
| "loss": 1.4357, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 0.03920043556039512, | |
| "grad_norm": 1.289537787437439, | |
| "learning_rate": 3.5387801599533475e-06, | |
| "loss": 1.4257, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.04013377926421405, | |
| "grad_norm": 1.439562201499939, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 1.4607, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 0.04044489383215369, | |
| "eval_loss": 1.4785248041152954, | |
| "eval_runtime": 390.0687, | |
| "eval_samples_per_second": 13.88, | |
| "eval_steps_per_second": 1.736, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.04106712296803298, | |
| "grad_norm": 1.2261285781860352, | |
| "learning_rate": 2.0118056862137357e-06, | |
| "loss": 1.5202, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.04200046667185191, | |
| "grad_norm": 1.3027156591415405, | |
| "learning_rate": 1.4029167422908107e-06, | |
| "loss": 1.5151, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.04293381037567084, | |
| "grad_norm": 1.1372510194778442, | |
| "learning_rate": 9.009284826036691e-07, | |
| "loss": 1.5352, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 0.043867154079489774, | |
| "grad_norm": 1.248124122619629, | |
| "learning_rate": 5.08115039419113e-07, | |
| "loss": 1.5216, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 0.044489383215369056, | |
| "eval_loss": 1.4776010513305664, | |
| "eval_runtime": 389.8727, | |
| "eval_samples_per_second": 13.887, | |
| "eval_steps_per_second": 1.736, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.044800497783308704, | |
| "grad_norm": 1.2946685552597046, | |
| "learning_rate": 2.262559558016325e-07, | |
| "loss": 1.5154, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.045733841487127634, | |
| "grad_norm": 1.1807682514190674, | |
| "learning_rate": 5.662812383859795e-08, | |
| "loss": 1.4998, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.046667185190946564, | |
| "grad_norm": 1.2081470489501953, | |
| "learning_rate": 0.0, | |
| "loss": 1.5239, | |
| "step": 150 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 150, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.225661850681344e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |