| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 36.8, | |
| "eval_steps": 20, | |
| "global_step": 92, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2e-05, | |
| "loss": 1.2516, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2581, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2684, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2529, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2441, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.00012, | |
| "loss": 1.2166, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.00014, | |
| "loss": 1.225, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.00016, | |
| "loss": 1.1785, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.00018, | |
| "loss": 1.1504, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1148, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 0.0001999390827019096, | |
| "loss": 1.0771, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.00019975640502598244, | |
| "loss": 1.0768, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 0.00019945218953682734, | |
| "loss": 1.0422, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.00019902680687415705, | |
| "loss": 1.0305, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 1.0148, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 1.0024, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 0.00019702957262759965, | |
| "loss": 0.9914, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.0001961261695938319, | |
| "loss": 0.9746, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.9764, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 0.9548, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9240104556083679, | |
| "eval_runtime": 1.9642, | |
| "eval_samples_per_second": 10.182, | |
| "eval_steps_per_second": 2.036, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 0.00019271838545667876, | |
| "loss": 0.9726, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 0.9381, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 0.0001898794046299167, | |
| "loss": 0.9339, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.00018829475928589271, | |
| "loss": 0.9256, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 0.9266, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 0.0001848048096156426, | |
| "loss": 0.9146, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "learning_rate": 0.00018290375725550417, | |
| "loss": 0.9155, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.8992, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "learning_rate": 0.00017880107536067218, | |
| "loss": 0.8951, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 0.8855, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "learning_rate": 0.00017431448254773944, | |
| "loss": 0.8901, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.0001719339800338651, | |
| "loss": 0.8704, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 13.2, | |
| "learning_rate": 0.00016946583704589973, | |
| "loss": 0.869, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.864, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 0.862, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 0.0001615661475325658, | |
| "loss": 0.8577, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 14.8, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.8544, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "learning_rate": 0.0001559192903470747, | |
| "loss": 0.8467, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 0.0001529919264233205, | |
| "loss": 0.8379, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.8514, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.8522964119911194, | |
| "eval_runtime": 1.9623, | |
| "eval_samples_per_second": 10.192, | |
| "eval_steps_per_second": 2.038, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 0.00014694715627858908, | |
| "loss": 0.8333, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "learning_rate": 0.00014383711467890774, | |
| "loss": 0.8379, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 17.2, | |
| "learning_rate": 0.00014067366430758004, | |
| "loss": 0.8278, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 0.00013746065934159123, | |
| "loss": 0.8273, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 0.8382, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.8208, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "learning_rate": 0.0001275637355816999, | |
| "loss": 0.8203, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 0.00012419218955996676, | |
| "loss": 0.8118, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 19.6, | |
| "learning_rate": 0.00012079116908177593, | |
| "loss": 0.8175, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.8059, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 20.4, | |
| "learning_rate": 0.00011391731009600654, | |
| "loss": 0.8114, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 0.8004, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 21.2, | |
| "learning_rate": 0.00010697564737441252, | |
| "loss": 0.8026, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 21.6, | |
| "learning_rate": 0.00010348994967025012, | |
| "loss": 0.7943, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.79, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "learning_rate": 9.651005032974994e-05, | |
| "loss": 0.7952, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 22.8, | |
| "learning_rate": 9.302435262558747e-05, | |
| "loss": 0.7898, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 23.2, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 0.7769, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 23.6, | |
| "learning_rate": 8.608268990399349e-05, | |
| "loss": 0.7812, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.7774, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.8498085141181946, | |
| "eval_runtime": 1.9653, | |
| "eval_samples_per_second": 10.177, | |
| "eval_steps_per_second": 2.035, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 24.4, | |
| "learning_rate": 7.920883091822408e-05, | |
| "loss": 0.7651, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 24.8, | |
| "learning_rate": 7.580781044003324e-05, | |
| "loss": 0.778, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 25.2, | |
| "learning_rate": 7.243626441830009e-05, | |
| "loss": 0.772, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.767, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 0.7569, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 26.4, | |
| "learning_rate": 6.25393406584088e-05, | |
| "loss": 0.7554, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 26.8, | |
| "learning_rate": 5.9326335692419995e-05, | |
| "loss": 0.76, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 27.2, | |
| "learning_rate": 5.616288532109225e-05, | |
| "loss": 0.7554, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 27.6, | |
| "learning_rate": 5.305284372141095e-05, | |
| "loss": 0.748, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.7527, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 28.4, | |
| "learning_rate": 4.700807357667952e-05, | |
| "loss": 0.7595, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "learning_rate": 4.4080709652925336e-05, | |
| "loss": 0.7394, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 29.2, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.7348, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 29.6, | |
| "learning_rate": 3.843385246743417e-05, | |
| "loss": 0.7497, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 0.7305, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 30.4, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.7317, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 30.8, | |
| "learning_rate": 3.053416295410026e-05, | |
| "loss": 0.7329, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 31.2, | |
| "learning_rate": 2.8066019966134904e-05, | |
| "loss": 0.7494, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 31.6, | |
| "learning_rate": 2.5685517452260567e-05, | |
| "loss": 0.7306, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.7178, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.8597297072410583, | |
| "eval_runtime": 1.9653, | |
| "eval_samples_per_second": 10.177, | |
| "eval_steps_per_second": 2.035, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 32.4, | |
| "learning_rate": 2.119892463932781e-05, | |
| "loss": 0.7243, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 32.8, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.7393, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 33.2, | |
| "learning_rate": 1.7096242744495837e-05, | |
| "loss": 0.7148, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 33.6, | |
| "learning_rate": 1.5195190384357404e-05, | |
| "loss": 0.7261, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 0.7228, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 34.4, | |
| "learning_rate": 1.1705240714107302e-05, | |
| "loss": 0.7194, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 34.8, | |
| "learning_rate": 1.0120595370083318e-05, | |
| "loss": 0.7252, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 35.2, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 0.7188, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 35.6, | |
| "learning_rate": 7.281614543321269e-06, | |
| "loss": 0.7199, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 0.7239, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 36.4, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.7099, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "learning_rate": 3.873830406168111e-06, | |
| "loss": 0.7265, | |
| "step": 92 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 7.297007876466278e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |