| { | |
| "best_metric": 0.9545454545454546, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-85-fold1/checkpoint-104", | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.25, | |
| "eval_loss": 1.1775949001312256, | |
| "eval_runtime": 0.6003, | |
| "eval_samples_per_second": 73.3, | |
| "eval_steps_per_second": 3.332, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3181818181818182, | |
| "eval_loss": 0.816531240940094, | |
| "eval_runtime": 0.5976, | |
| "eval_samples_per_second": 73.627, | |
| "eval_steps_per_second": 3.347, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.7525402307510376, | |
| "eval_runtime": 0.594, | |
| "eval_samples_per_second": 74.075, | |
| "eval_steps_per_second": 3.367, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.8073585033416748, | |
| "eval_runtime": 0.5987, | |
| "eval_samples_per_second": 73.492, | |
| "eval_steps_per_second": 3.341, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.2465157508850098, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7495, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.684214174747467, | |
| "eval_runtime": 0.6, | |
| "eval_samples_per_second": 73.337, | |
| "eval_steps_per_second": 3.333, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6818181818181818, | |
| "eval_loss": 0.5605855584144592, | |
| "eval_runtime": 0.6092, | |
| "eval_samples_per_second": 72.223, | |
| "eval_steps_per_second": 3.283, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6818181818181818, | |
| "eval_loss": 0.5376735329627991, | |
| "eval_runtime": 0.6076, | |
| "eval_samples_per_second": 72.414, | |
| "eval_steps_per_second": 3.292, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.5760537981987, | |
| "eval_runtime": 0.6158, | |
| "eval_samples_per_second": 71.446, | |
| "eval_steps_per_second": 3.248, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.5866416692733765, | |
| "eval_runtime": 0.6071, | |
| "eval_samples_per_second": 72.475, | |
| "eval_steps_per_second": 3.294, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.638932466506958, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4611, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.4894014596939087, | |
| "eval_runtime": 0.6083, | |
| "eval_samples_per_second": 72.333, | |
| "eval_steps_per_second": 3.288, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.6216031312942505, | |
| "eval_runtime": 0.603, | |
| "eval_samples_per_second": 72.968, | |
| "eval_steps_per_second": 3.317, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.6271905303001404, | |
| "eval_runtime": 0.6112, | |
| "eval_samples_per_second": 71.994, | |
| "eval_steps_per_second": 3.272, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.4475552439689636, | |
| "eval_runtime": 0.6178, | |
| "eval_samples_per_second": 71.224, | |
| "eval_steps_per_second": 3.237, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.4450555145740509, | |
| "eval_runtime": 0.6129, | |
| "eval_samples_per_second": 71.792, | |
| "eval_steps_per_second": 3.263, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 7.13038444519043, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3766, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.4369763433933258, | |
| "eval_runtime": 0.6167, | |
| "eval_samples_per_second": 71.353, | |
| "eval_steps_per_second": 3.243, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.49372145533561707, | |
| "eval_runtime": 0.6166, | |
| "eval_samples_per_second": 71.363, | |
| "eval_steps_per_second": 3.244, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.49774959683418274, | |
| "eval_runtime": 0.6137, | |
| "eval_samples_per_second": 71.701, | |
| "eval_steps_per_second": 3.259, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4550728499889374, | |
| "eval_runtime": 0.6181, | |
| "eval_samples_per_second": 71.186, | |
| "eval_steps_per_second": 3.236, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.3776405155658722, | |
| "eval_runtime": 0.6175, | |
| "eval_samples_per_second": 71.252, | |
| "eval_steps_per_second": 3.239, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 8.562040328979492, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3147, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3346579670906067, | |
| "eval_runtime": 0.6129, | |
| "eval_samples_per_second": 71.792, | |
| "eval_steps_per_second": 3.263, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.5178121328353882, | |
| "eval_runtime": 0.6164, | |
| "eval_samples_per_second": 71.384, | |
| "eval_steps_per_second": 3.245, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3430297076702118, | |
| "eval_runtime": 0.6347, | |
| "eval_samples_per_second": 69.322, | |
| "eval_steps_per_second": 3.151, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.30911171436309814, | |
| "eval_runtime": 0.6125, | |
| "eval_samples_per_second": 71.833, | |
| "eval_steps_per_second": 3.265, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3994681239128113, | |
| "eval_runtime": 0.6384, | |
| "eval_samples_per_second": 68.922, | |
| "eval_steps_per_second": 3.133, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 2.9600367546081543, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2176, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.2971316874027252, | |
| "eval_runtime": 0.6167, | |
| "eval_samples_per_second": 71.352, | |
| "eval_steps_per_second": 3.243, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.31372883915901184, | |
| "eval_runtime": 0.6415, | |
| "eval_samples_per_second": 68.59, | |
| "eval_steps_per_second": 3.118, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2693716585636139, | |
| "eval_runtime": 0.6209, | |
| "eval_samples_per_second": 70.861, | |
| "eval_steps_per_second": 3.221, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.25888314843177795, | |
| "eval_runtime": 0.6236, | |
| "eval_samples_per_second": 70.553, | |
| "eval_steps_per_second": 3.207, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3612332046031952, | |
| "eval_runtime": 0.6164, | |
| "eval_samples_per_second": 71.388, | |
| "eval_steps_per_second": 3.245, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 8.659514427185059, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1855, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.340602308511734, | |
| "eval_runtime": 0.6128, | |
| "eval_samples_per_second": 71.806, | |
| "eval_steps_per_second": 3.264, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.47381675243377686, | |
| "eval_runtime": 0.6119, | |
| "eval_samples_per_second": 71.912, | |
| "eval_steps_per_second": 3.269, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7954545454545454, | |
| "eval_loss": 0.7612176537513733, | |
| "eval_runtime": 0.6248, | |
| "eval_samples_per_second": 70.426, | |
| "eval_steps_per_second": 3.201, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.5307350158691406, | |
| "eval_runtime": 0.6178, | |
| "eval_samples_per_second": 71.226, | |
| "eval_steps_per_second": 3.238, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3345917761325836, | |
| "eval_runtime": 0.6139, | |
| "eval_samples_per_second": 71.679, | |
| "eval_steps_per_second": 3.258, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 4.022067070007324, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.2006, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.35621532797813416, | |
| "eval_runtime": 0.6167, | |
| "eval_samples_per_second": 71.349, | |
| "eval_steps_per_second": 3.243, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.5254650115966797, | |
| "eval_runtime": 0.6173, | |
| "eval_samples_per_second": 71.281, | |
| "eval_steps_per_second": 3.24, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.3794662356376648, | |
| "eval_runtime": 0.6176, | |
| "eval_samples_per_second": 71.243, | |
| "eval_steps_per_second": 3.238, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2924174964427948, | |
| "eval_runtime": 0.6167, | |
| "eval_samples_per_second": 71.348, | |
| "eval_steps_per_second": 3.243, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2921161949634552, | |
| "eval_runtime": 0.6222, | |
| "eval_samples_per_second": 70.72, | |
| "eval_steps_per_second": 3.215, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 1.585066795349121, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.161, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.38950470089912415, | |
| "eval_runtime": 0.6256, | |
| "eval_samples_per_second": 70.334, | |
| "eval_steps_per_second": 3.197, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.3420548439025879, | |
| "eval_runtime": 0.6243, | |
| "eval_samples_per_second": 70.482, | |
| "eval_steps_per_second": 3.204, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2674037218093872, | |
| "eval_runtime": 0.6201, | |
| "eval_samples_per_second": 70.959, | |
| "eval_steps_per_second": 3.225, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.25858479738235474, | |
| "eval_runtime": 0.6382, | |
| "eval_samples_per_second": 68.948, | |
| "eval_steps_per_second": 3.134, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4520129859447479, | |
| "eval_runtime": 0.6318, | |
| "eval_samples_per_second": 69.642, | |
| "eval_steps_per_second": 3.166, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 3.2059006690979004, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1588, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4299778938293457, | |
| "eval_runtime": 0.6213, | |
| "eval_samples_per_second": 70.818, | |
| "eval_steps_per_second": 3.219, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.24242053925991058, | |
| "eval_runtime": 0.6312, | |
| "eval_samples_per_second": 69.704, | |
| "eval_steps_per_second": 3.168, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.26448819041252136, | |
| "eval_runtime": 0.6377, | |
| "eval_samples_per_second": 69.001, | |
| "eval_steps_per_second": 3.136, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2531264126300812, | |
| "eval_runtime": 0.637, | |
| "eval_samples_per_second": 69.074, | |
| "eval_steps_per_second": 3.14, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2613573372364044, | |
| "eval_runtime": 0.6346, | |
| "eval_samples_per_second": 69.335, | |
| "eval_steps_per_second": 3.152, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 2.1700892448425293, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1103, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3023524284362793, | |
| "eval_runtime": 0.6154, | |
| "eval_samples_per_second": 71.493, | |
| "eval_steps_per_second": 3.25, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.27967941761016846, | |
| "eval_runtime": 0.6172, | |
| "eval_samples_per_second": 71.291, | |
| "eval_steps_per_second": 3.24, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.23065362870693207, | |
| "eval_runtime": 0.6263, | |
| "eval_samples_per_second": 70.252, | |
| "eval_steps_per_second": 3.193, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2635006010532379, | |
| "eval_runtime": 0.629, | |
| "eval_samples_per_second": 69.955, | |
| "eval_steps_per_second": 3.18, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.5111494064331055, | |
| "eval_runtime": 0.621, | |
| "eval_samples_per_second": 70.852, | |
| "eval_steps_per_second": 3.221, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 6.504031658172607, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1201, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.5371341109275818, | |
| "eval_runtime": 0.6372, | |
| "eval_samples_per_second": 69.049, | |
| "eval_steps_per_second": 3.139, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.29399573802948, | |
| "eval_runtime": 0.6191, | |
| "eval_samples_per_second": 71.069, | |
| "eval_steps_per_second": 3.23, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.30151382088661194, | |
| "eval_runtime": 0.6185, | |
| "eval_samples_per_second": 71.142, | |
| "eval_steps_per_second": 3.234, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.26309311389923096, | |
| "eval_runtime": 0.623, | |
| "eval_samples_per_second": 70.621, | |
| "eval_steps_per_second": 3.21, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.28302156925201416, | |
| "eval_runtime": 0.612, | |
| "eval_samples_per_second": 71.89, | |
| "eval_steps_per_second": 3.268, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 2.179642677307129, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1037, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.32017800211906433, | |
| "eval_runtime": 0.612, | |
| "eval_samples_per_second": 71.897, | |
| "eval_steps_per_second": 3.268, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.3525654077529907, | |
| "eval_runtime": 0.6201, | |
| "eval_samples_per_second": 70.958, | |
| "eval_steps_per_second": 3.225, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.39751797914505005, | |
| "eval_runtime": 0.6196, | |
| "eval_samples_per_second": 71.015, | |
| "eval_steps_per_second": 3.228, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.4785454571247101, | |
| "eval_runtime": 0.6207, | |
| "eval_samples_per_second": 70.889, | |
| "eval_steps_per_second": 3.222, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.43056878447532654, | |
| "eval_runtime": 0.6178, | |
| "eval_samples_per_second": 71.22, | |
| "eval_steps_per_second": 3.237, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 1.682313084602356, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.1, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.32298263907432556, | |
| "eval_runtime": 0.6363, | |
| "eval_samples_per_second": 69.146, | |
| "eval_steps_per_second": 3.143, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3006637990474701, | |
| "eval_runtime": 0.6204, | |
| "eval_samples_per_second": 70.927, | |
| "eval_steps_per_second": 3.224, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.266933798789978, | |
| "eval_runtime": 0.6293, | |
| "eval_samples_per_second": 69.917, | |
| "eval_steps_per_second": 3.178, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.23347026109695435, | |
| "eval_runtime": 0.6221, | |
| "eval_samples_per_second": 70.725, | |
| "eval_steps_per_second": 3.215, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.18453344702720642, | |
| "eval_runtime": 0.6177, | |
| "eval_samples_per_second": 71.237, | |
| "eval_steps_per_second": 3.238, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 1.9629833698272705, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0984, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2260511964559555, | |
| "eval_runtime": 0.6183, | |
| "eval_samples_per_second": 71.158, | |
| "eval_steps_per_second": 3.234, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.3014659881591797, | |
| "eval_runtime": 0.6221, | |
| "eval_samples_per_second": 70.73, | |
| "eval_steps_per_second": 3.215, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.31382787227630615, | |
| "eval_runtime": 0.6179, | |
| "eval_samples_per_second": 71.206, | |
| "eval_steps_per_second": 3.237, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.24435962736606598, | |
| "eval_runtime": 0.6487, | |
| "eval_samples_per_second": 67.831, | |
| "eval_steps_per_second": 3.083, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20601721107959747, | |
| "eval_runtime": 0.6342, | |
| "eval_samples_per_second": 69.375, | |
| "eval_steps_per_second": 3.153, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 4.01008939743042, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0826, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20242777466773987, | |
| "eval_runtime": 0.6238, | |
| "eval_samples_per_second": 70.54, | |
| "eval_steps_per_second": 3.206, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.25031936168670654, | |
| "eval_runtime": 0.6215, | |
| "eval_samples_per_second": 70.799, | |
| "eval_steps_per_second": 3.218, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.24986010789871216, | |
| "eval_runtime": 0.6255, | |
| "eval_samples_per_second": 70.346, | |
| "eval_steps_per_second": 3.198, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20987679064273834, | |
| "eval_runtime": 0.6173, | |
| "eval_samples_per_second": 71.276, | |
| "eval_steps_per_second": 3.24, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.22395505011081696, | |
| "eval_runtime": 0.6202, | |
| "eval_samples_per_second": 70.949, | |
| "eval_steps_per_second": 3.225, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.0807651281356812, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0701, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.22283396124839783, | |
| "eval_runtime": 0.6175, | |
| "eval_samples_per_second": 71.253, | |
| "eval_steps_per_second": 3.239, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2336987555027008, | |
| "eval_runtime": 0.6367, | |
| "eval_samples_per_second": 69.11, | |
| "eval_steps_per_second": 3.141, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.21134348213672638, | |
| "eval_runtime": 0.6161, | |
| "eval_samples_per_second": 71.417, | |
| "eval_steps_per_second": 3.246, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.19769272208213806, | |
| "eval_runtime": 0.617, | |
| "eval_samples_per_second": 71.314, | |
| "eval_steps_per_second": 3.242, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20209179818630219, | |
| "eval_runtime": 0.6172, | |
| "eval_samples_per_second": 71.289, | |
| "eval_steps_per_second": 3.24, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 2.9528820514678955, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0846, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.23302783071994781, | |
| "eval_runtime": 0.6169, | |
| "eval_samples_per_second": 71.325, | |
| "eval_steps_per_second": 3.242, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.23330368101596832, | |
| "eval_runtime": 0.619, | |
| "eval_samples_per_second": 71.078, | |
| "eval_steps_per_second": 3.231, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2130350023508072, | |
| "eval_runtime": 0.6495, | |
| "eval_samples_per_second": 67.74, | |
| "eval_steps_per_second": 3.079, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2089734971523285, | |
| "eval_runtime": 0.6253, | |
| "eval_samples_per_second": 70.368, | |
| "eval_steps_per_second": 3.199, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.21136285364627838, | |
| "eval_runtime": 0.6561, | |
| "eval_samples_per_second": 67.06, | |
| "eval_steps_per_second": 3.048, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 2.1979732513427734, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0932, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20612572133541107, | |
| "eval_runtime": 0.6471, | |
| "eval_samples_per_second": 67.991, | |
| "eval_steps_per_second": 3.091, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2173743098974228, | |
| "eval_runtime": 0.6284, | |
| "eval_samples_per_second": 70.02, | |
| "eval_steps_per_second": 3.183, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24294628202915192, | |
| "eval_runtime": 0.6351, | |
| "eval_samples_per_second": 69.28, | |
| "eval_steps_per_second": 3.149, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2458869367837906, | |
| "eval_runtime": 0.6213, | |
| "eval_samples_per_second": 70.815, | |
| "eval_steps_per_second": 3.219, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.22928108274936676, | |
| "eval_runtime": 0.6133, | |
| "eval_samples_per_second": 71.745, | |
| "eval_steps_per_second": 3.261, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 1.044729232788086, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0742, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.21271629631519318, | |
| "eval_runtime": 0.6153, | |
| "eval_samples_per_second": 71.511, | |
| "eval_steps_per_second": 3.251, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20136304199695587, | |
| "eval_runtime": 0.6155, | |
| "eval_samples_per_second": 71.482, | |
| "eval_steps_per_second": 3.249, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20151734352111816, | |
| "eval_runtime": 0.6209, | |
| "eval_samples_per_second": 70.861, | |
| "eval_steps_per_second": 3.221, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20625466108322144, | |
| "eval_runtime": 0.6261, | |
| "eval_samples_per_second": 70.282, | |
| "eval_steps_per_second": 3.195, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20877276360988617, | |
| "eval_runtime": 0.629, | |
| "eval_samples_per_second": 69.947, | |
| "eval_steps_per_second": 3.179, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.0264675617218018, | |
| "learning_rate": 0.0, | |
| "loss": 0.0701, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20956982672214508, | |
| "eval_runtime": 0.6231, | |
| "eval_samples_per_second": 70.611, | |
| "eval_steps_per_second": 3.21, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.883089058199552e+18, | |
| "train_loss": 0.19163743317127227, | |
| "train_runtime": 1125.642, | |
| "train_samples_per_second": 21.588, | |
| "train_steps_per_second": 0.178 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9545454545454546, | |
| "eval_loss": 0.23065362870693207, | |
| "eval_runtime": 0.6418, | |
| "eval_samples_per_second": 68.56, | |
| "eval_steps_per_second": 3.116, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.883089058199552e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |