| { | |
| "best_metric": 0.9318181818181818, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-85-fold2/checkpoint-46", | |
| "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": 0.2727272727272727, | |
| "eval_loss": 1.2745391130447388, | |
| "eval_runtime": 0.6014, | |
| "eval_samples_per_second": 73.159, | |
| "eval_steps_per_second": 3.325, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4090909090909091, | |
| "eval_loss": 0.8028427362442017, | |
| "eval_runtime": 0.6032, | |
| "eval_samples_per_second": 72.94, | |
| "eval_steps_per_second": 3.315, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.7455913424491882, | |
| "eval_runtime": 0.6175, | |
| "eval_samples_per_second": 71.255, | |
| "eval_steps_per_second": 3.239, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.7981840372085571, | |
| "eval_runtime": 0.6096, | |
| "eval_samples_per_second": 72.178, | |
| "eval_steps_per_second": 3.281, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.3311023712158203, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7325, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6233007311820984, | |
| "eval_runtime": 0.6071, | |
| "eval_samples_per_second": 72.47, | |
| "eval_steps_per_second": 3.294, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.5093103051185608, | |
| "eval_runtime": 0.6339, | |
| "eval_samples_per_second": 69.412, | |
| "eval_steps_per_second": 3.155, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.5565758347511292, | |
| "eval_runtime": 0.6082, | |
| "eval_samples_per_second": 72.35, | |
| "eval_steps_per_second": 3.289, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.6838734745979309, | |
| "eval_runtime": 0.6019, | |
| "eval_samples_per_second": 73.107, | |
| "eval_steps_per_second": 3.323, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.48205462098121643, | |
| "eval_runtime": 0.6025, | |
| "eval_samples_per_second": 73.034, | |
| "eval_steps_per_second": 3.32, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.509533882141113, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4472, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7727272727272727, | |
| "eval_loss": 0.4365437626838684, | |
| "eval_runtime": 0.6039, | |
| "eval_samples_per_second": 72.86, | |
| "eval_steps_per_second": 3.312, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.5157524347305298, | |
| "eval_runtime": 0.6015, | |
| "eval_samples_per_second": 73.149, | |
| "eval_steps_per_second": 3.325, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.41960516571998596, | |
| "eval_runtime": 0.6153, | |
| "eval_samples_per_second": 71.513, | |
| "eval_steps_per_second": 3.251, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.35994166135787964, | |
| "eval_runtime": 0.6066, | |
| "eval_samples_per_second": 72.534, | |
| "eval_steps_per_second": 3.297, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.36041271686553955, | |
| "eval_runtime": 0.607, | |
| "eval_samples_per_second": 72.489, | |
| "eval_steps_per_second": 3.295, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.2131314277648926, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3483, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8181818181818182, | |
| "eval_loss": 0.36339208483695984, | |
| "eval_runtime": 0.5982, | |
| "eval_samples_per_second": 73.558, | |
| "eval_steps_per_second": 3.344, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2802547514438629, | |
| "eval_runtime": 0.6116, | |
| "eval_samples_per_second": 71.94, | |
| "eval_steps_per_second": 3.27, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.25918081402778625, | |
| "eval_runtime": 0.6122, | |
| "eval_samples_per_second": 71.876, | |
| "eval_steps_per_second": 3.267, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2654772102832794, | |
| "eval_runtime": 0.6337, | |
| "eval_samples_per_second": 69.432, | |
| "eval_steps_per_second": 3.156, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2333187609910965, | |
| "eval_runtime": 0.6516, | |
| "eval_samples_per_second": 67.523, | |
| "eval_steps_per_second": 3.069, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.3891751766204834, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.2514, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.25902560353279114, | |
| "eval_runtime": 0.6102, | |
| "eval_samples_per_second": 72.107, | |
| "eval_steps_per_second": 3.278, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.26424452662467957, | |
| "eval_runtime": 0.6031, | |
| "eval_samples_per_second": 72.952, | |
| "eval_steps_per_second": 3.316, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.26372984051704407, | |
| "eval_runtime": 0.6239, | |
| "eval_samples_per_second": 70.529, | |
| "eval_steps_per_second": 3.206, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19909273087978363, | |
| "eval_runtime": 0.6092, | |
| "eval_samples_per_second": 72.228, | |
| "eval_steps_per_second": 3.283, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.19407562911510468, | |
| "eval_runtime": 0.6232, | |
| "eval_samples_per_second": 70.602, | |
| "eval_steps_per_second": 3.209, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 3.262606382369995, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.1847, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.18681906163692474, | |
| "eval_runtime": 0.6022, | |
| "eval_samples_per_second": 73.069, | |
| "eval_steps_per_second": 3.321, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.1828092634677887, | |
| "eval_runtime": 0.6083, | |
| "eval_samples_per_second": 72.327, | |
| "eval_steps_per_second": 3.288, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.17112162709236145, | |
| "eval_runtime": 0.6062, | |
| "eval_samples_per_second": 72.577, | |
| "eval_steps_per_second": 3.299, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2423066645860672, | |
| "eval_runtime": 0.6086, | |
| "eval_samples_per_second": 72.296, | |
| "eval_steps_per_second": 3.286, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.21621073782444, | |
| "eval_runtime": 0.6093, | |
| "eval_samples_per_second": 72.213, | |
| "eval_steps_per_second": 3.282, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 2.5355308055877686, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1501, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.18539918959140778, | |
| "eval_runtime": 0.6182, | |
| "eval_samples_per_second": 71.169, | |
| "eval_steps_per_second": 3.235, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.30705034732818604, | |
| "eval_runtime": 0.6154, | |
| "eval_samples_per_second": 71.493, | |
| "eval_steps_per_second": 3.25, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2435438185930252, | |
| "eval_runtime": 0.622, | |
| "eval_samples_per_second": 70.736, | |
| "eval_steps_per_second": 3.215, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.17283663153648376, | |
| "eval_runtime": 0.6112, | |
| "eval_samples_per_second": 71.984, | |
| "eval_steps_per_second": 3.272, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.16438254714012146, | |
| "eval_runtime": 0.626, | |
| "eval_samples_per_second": 70.291, | |
| "eval_steps_per_second": 3.195, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 2.833869695663452, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.13, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8409090909090909, | |
| "eval_loss": 0.2768351137638092, | |
| "eval_runtime": 0.6028, | |
| "eval_samples_per_second": 72.987, | |
| "eval_steps_per_second": 3.318, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.15386301279067993, | |
| "eval_runtime": 0.6151, | |
| "eval_samples_per_second": 71.537, | |
| "eval_steps_per_second": 3.252, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.25800377130508423, | |
| "eval_runtime": 0.6195, | |
| "eval_samples_per_second": 71.023, | |
| "eval_steps_per_second": 3.228, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.17828179895877838, | |
| "eval_runtime": 0.6253, | |
| "eval_samples_per_second": 70.368, | |
| "eval_steps_per_second": 3.199, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8636363636363636, | |
| "eval_loss": 0.1782391220331192, | |
| "eval_runtime": 0.6139, | |
| "eval_samples_per_second": 71.678, | |
| "eval_steps_per_second": 3.258, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 2.5041236877441406, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1357, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.20351167023181915, | |
| "eval_runtime": 0.6115, | |
| "eval_samples_per_second": 71.96, | |
| "eval_steps_per_second": 3.271, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.21165017783641815, | |
| "eval_runtime": 0.6088, | |
| "eval_samples_per_second": 72.277, | |
| "eval_steps_per_second": 3.285, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.17929820716381073, | |
| "eval_runtime": 0.6063, | |
| "eval_samples_per_second": 72.573, | |
| "eval_steps_per_second": 3.299, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2002231925725937, | |
| "eval_runtime": 0.6078, | |
| "eval_samples_per_second": 72.392, | |
| "eval_steps_per_second": 3.291, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.23656944930553436, | |
| "eval_runtime": 0.6049, | |
| "eval_samples_per_second": 72.736, | |
| "eval_steps_per_second": 3.306, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 2.3605904579162598, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.105, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20083707571029663, | |
| "eval_runtime": 0.602, | |
| "eval_samples_per_second": 73.092, | |
| "eval_steps_per_second": 3.322, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.23675359785556793, | |
| "eval_runtime": 0.6125, | |
| "eval_samples_per_second": 71.834, | |
| "eval_steps_per_second": 3.265, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.21420449018478394, | |
| "eval_runtime": 0.6133, | |
| "eval_samples_per_second": 71.747, | |
| "eval_steps_per_second": 3.261, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.21165300905704498, | |
| "eval_runtime": 0.6093, | |
| "eval_samples_per_second": 72.213, | |
| "eval_steps_per_second": 3.282, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2621056139469147, | |
| "eval_runtime": 0.6066, | |
| "eval_samples_per_second": 72.53, | |
| "eval_steps_per_second": 3.297, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 3.734884023666382, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1091, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.22308549284934998, | |
| "eval_runtime": 0.6187, | |
| "eval_samples_per_second": 71.12, | |
| "eval_steps_per_second": 3.233, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19459280371665955, | |
| "eval_runtime": 0.6087, | |
| "eval_samples_per_second": 72.285, | |
| "eval_steps_per_second": 3.286, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.20005172491073608, | |
| "eval_runtime": 0.617, | |
| "eval_samples_per_second": 71.316, | |
| "eval_steps_per_second": 3.242, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20313407480716705, | |
| "eval_runtime": 0.606, | |
| "eval_samples_per_second": 72.607, | |
| "eval_steps_per_second": 3.3, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20779913663864136, | |
| "eval_runtime": 0.6127, | |
| "eval_samples_per_second": 71.816, | |
| "eval_steps_per_second": 3.264, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 1.503174066543579, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1054, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.22496341168880463, | |
| "eval_runtime": 0.5985, | |
| "eval_samples_per_second": 73.523, | |
| "eval_steps_per_second": 3.342, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2180013209581375, | |
| "eval_runtime": 0.6105, | |
| "eval_samples_per_second": 72.071, | |
| "eval_steps_per_second": 3.276, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1915305107831955, | |
| "eval_runtime": 0.6101, | |
| "eval_samples_per_second": 72.117, | |
| "eval_steps_per_second": 3.278, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.2227245271205902, | |
| "eval_runtime": 0.6127, | |
| "eval_samples_per_second": 71.814, | |
| "eval_steps_per_second": 3.264, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.23516863584518433, | |
| "eval_runtime": 0.6379, | |
| "eval_samples_per_second": 68.976, | |
| "eval_steps_per_second": 3.135, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 2.845384120941162, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0982, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.23286031186580658, | |
| "eval_runtime": 0.6246, | |
| "eval_samples_per_second": 70.446, | |
| "eval_steps_per_second": 3.202, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2135220468044281, | |
| "eval_runtime": 0.6138, | |
| "eval_samples_per_second": 71.681, | |
| "eval_steps_per_second": 3.258, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.19485782086849213, | |
| "eval_runtime": 0.6095, | |
| "eval_samples_per_second": 72.185, | |
| "eval_steps_per_second": 3.281, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.21486514806747437, | |
| "eval_runtime": 0.6086, | |
| "eval_samples_per_second": 72.296, | |
| "eval_steps_per_second": 3.286, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24350209534168243, | |
| "eval_runtime": 0.6151, | |
| "eval_samples_per_second": 71.537, | |
| "eval_steps_per_second": 3.252, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 1.9481189250946045, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0808, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2541444003582001, | |
| "eval_runtime": 0.6056, | |
| "eval_samples_per_second": 72.657, | |
| "eval_steps_per_second": 3.303, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24467478692531586, | |
| "eval_runtime": 0.6059, | |
| "eval_samples_per_second": 72.62, | |
| "eval_steps_per_second": 3.301, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19038386642932892, | |
| "eval_runtime": 0.6075, | |
| "eval_samples_per_second": 72.428, | |
| "eval_steps_per_second": 3.292, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24367865920066833, | |
| "eval_runtime": 0.6168, | |
| "eval_samples_per_second": 71.342, | |
| "eval_steps_per_second": 3.243, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.35932543873786926, | |
| "eval_runtime": 0.607, | |
| "eval_samples_per_second": 72.488, | |
| "eval_steps_per_second": 3.295, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 2.6166303157806396, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0843, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.4187169373035431, | |
| "eval_runtime": 0.6057, | |
| "eval_samples_per_second": 72.648, | |
| "eval_steps_per_second": 3.302, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.8863636363636364, | |
| "eval_loss": 0.35100072622299194, | |
| "eval_runtime": 0.605, | |
| "eval_samples_per_second": 72.726, | |
| "eval_steps_per_second": 3.306, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2315276712179184, | |
| "eval_runtime": 0.6151, | |
| "eval_samples_per_second": 71.531, | |
| "eval_steps_per_second": 3.251, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.20485752820968628, | |
| "eval_runtime": 0.6119, | |
| "eval_samples_per_second": 71.905, | |
| "eval_steps_per_second": 3.268, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.21503090858459473, | |
| "eval_runtime": 0.6275, | |
| "eval_samples_per_second": 70.124, | |
| "eval_steps_per_second": 3.187, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 2.6058237552642822, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0942, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.21164561808109283, | |
| "eval_runtime": 0.614, | |
| "eval_samples_per_second": 71.662, | |
| "eval_steps_per_second": 3.257, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.2013980746269226, | |
| "eval_runtime": 0.6105, | |
| "eval_samples_per_second": 72.068, | |
| "eval_steps_per_second": 3.276, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.21975825726985931, | |
| "eval_runtime": 0.6262, | |
| "eval_samples_per_second": 70.265, | |
| "eval_steps_per_second": 3.194, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2538079023361206, | |
| "eval_runtime": 0.6264, | |
| "eval_samples_per_second": 70.245, | |
| "eval_steps_per_second": 3.193, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.27548667788505554, | |
| "eval_runtime": 0.6345, | |
| "eval_samples_per_second": 69.35, | |
| "eval_steps_per_second": 3.152, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.9026525020599365, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0884, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.24910703301429749, | |
| "eval_runtime": 0.6525, | |
| "eval_samples_per_second": 67.433, | |
| "eval_steps_per_second": 3.065, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.21000614762306213, | |
| "eval_runtime": 0.607, | |
| "eval_samples_per_second": 72.491, | |
| "eval_steps_per_second": 3.295, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.1976689249277115, | |
| "eval_runtime": 0.6101, | |
| "eval_samples_per_second": 72.123, | |
| "eval_steps_per_second": 3.278, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.1979275345802307, | |
| "eval_runtime": 0.6116, | |
| "eval_samples_per_second": 71.94, | |
| "eval_steps_per_second": 3.27, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.21450987458229065, | |
| "eval_runtime": 0.612, | |
| "eval_samples_per_second": 71.895, | |
| "eval_steps_per_second": 3.268, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 2.5778989791870117, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0637, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.21915043890476227, | |
| "eval_runtime": 0.6078, | |
| "eval_samples_per_second": 72.395, | |
| "eval_steps_per_second": 3.291, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.2055453360080719, | |
| "eval_runtime": 0.607, | |
| "eval_samples_per_second": 72.493, | |
| "eval_steps_per_second": 3.295, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19937144219875336, | |
| "eval_runtime": 0.6058, | |
| "eval_samples_per_second": 72.632, | |
| "eval_steps_per_second": 3.301, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.19752758741378784, | |
| "eval_runtime": 0.613, | |
| "eval_samples_per_second": 71.774, | |
| "eval_steps_per_second": 3.262, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.19738946855068207, | |
| "eval_runtime": 0.6029, | |
| "eval_samples_per_second": 72.982, | |
| "eval_steps_per_second": 3.317, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 1.4332749843597412, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0923, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.1964712142944336, | |
| "eval_runtime": 0.6224, | |
| "eval_samples_per_second": 70.694, | |
| "eval_steps_per_second": 3.213, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.19246001541614532, | |
| "eval_runtime": 0.6066, | |
| "eval_samples_per_second": 72.537, | |
| "eval_steps_per_second": 3.297, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.19422116875648499, | |
| "eval_runtime": 0.6186, | |
| "eval_samples_per_second": 71.123, | |
| "eval_steps_per_second": 3.233, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19693025946617126, | |
| "eval_runtime": 0.6105, | |
| "eval_samples_per_second": 72.075, | |
| "eval_steps_per_second": 3.276, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1949141025543213, | |
| "eval_runtime": 0.613, | |
| "eval_samples_per_second": 71.78, | |
| "eval_steps_per_second": 3.263, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 1.9863300323486328, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0657, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19037796556949615, | |
| "eval_runtime": 0.6049, | |
| "eval_samples_per_second": 72.734, | |
| "eval_steps_per_second": 3.306, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.18766190111637115, | |
| "eval_runtime": 0.6224, | |
| "eval_samples_per_second": 70.689, | |
| "eval_steps_per_second": 3.213, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.1884867548942566, | |
| "eval_runtime": 0.6265, | |
| "eval_samples_per_second": 70.237, | |
| "eval_steps_per_second": 3.193, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.1901746690273285, | |
| "eval_runtime": 0.6104, | |
| "eval_samples_per_second": 72.087, | |
| "eval_steps_per_second": 3.277, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19220541417598724, | |
| "eval_runtime": 0.6116, | |
| "eval_samples_per_second": 71.943, | |
| "eval_steps_per_second": 3.27, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.5260493755340576, | |
| "learning_rate": 0.0, | |
| "loss": 0.0822, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19321714341640472, | |
| "eval_runtime": 0.6259, | |
| "eval_samples_per_second": 70.297, | |
| "eval_steps_per_second": 3.195, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.883089058199552e+18, | |
| "train_loss": 0.17245761930942535, | |
| "train_runtime": 1120.8578, | |
| "train_samples_per_second": 21.68, | |
| "train_steps_per_second": 0.178 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9318181818181818, | |
| "eval_loss": 0.19909273087978363, | |
| "eval_runtime": 0.6966, | |
| "eval_samples_per_second": 63.167, | |
| "eval_steps_per_second": 2.871, | |
| "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 | |
| } | |