|
{ |
|
"best_metric": 0.8227848101265823, |
|
"best_model_checkpoint": "beit-base-patch16-224-fold4/checkpoint-248", |
|
"epoch": 85.71428571428571, |
|
"eval_steps": 500, |
|
"global_step": 300, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.8571428571428571, |
|
"eval_accuracy": 0.379746835443038, |
|
"eval_loss": 0.8664439916610718, |
|
"eval_runtime": 1.3597, |
|
"eval_samples_per_second": 58.1, |
|
"eval_steps_per_second": 2.206, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.4810126582278481, |
|
"eval_loss": 0.8144369125366211, |
|
"eval_runtime": 1.2392, |
|
"eval_samples_per_second": 63.752, |
|
"eval_steps_per_second": 2.421, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 2.857142857142857, |
|
"grad_norm": 5.919072151184082, |
|
"learning_rate": 1.6666666666666667e-05, |
|
"loss": 0.7345, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 2.857142857142857, |
|
"eval_accuracy": 0.5822784810126582, |
|
"eval_loss": 0.6451208591461182, |
|
"eval_runtime": 1.2159, |
|
"eval_samples_per_second": 64.971, |
|
"eval_steps_per_second": 2.467, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.5189873417721519, |
|
"eval_loss": 0.7268054485321045, |
|
"eval_runtime": 1.2597, |
|
"eval_samples_per_second": 62.712, |
|
"eval_steps_per_second": 2.381, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 4.857142857142857, |
|
"eval_accuracy": 0.7341772151898734, |
|
"eval_loss": 0.5730710029602051, |
|
"eval_runtime": 1.2594, |
|
"eval_samples_per_second": 62.726, |
|
"eval_steps_per_second": 2.382, |
|
"step": 17 |
|
}, |
|
{ |
|
"epoch": 5.714285714285714, |
|
"grad_norm": 5.122115135192871, |
|
"learning_rate": 3.3333333333333335e-05, |
|
"loss": 0.6447, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.5316455696202531, |
|
"eval_loss": 0.6948190331459045, |
|
"eval_runtime": 1.2989, |
|
"eval_samples_per_second": 60.819, |
|
"eval_steps_per_second": 2.31, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 6.857142857142857, |
|
"eval_accuracy": 0.6962025316455697, |
|
"eval_loss": 0.5315247774124146, |
|
"eval_runtime": 1.2437, |
|
"eval_samples_per_second": 63.519, |
|
"eval_steps_per_second": 2.412, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.6835443037974683, |
|
"eval_loss": 0.5292339324951172, |
|
"eval_runtime": 1.2386, |
|
"eval_samples_per_second": 63.78, |
|
"eval_steps_per_second": 2.422, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 8.571428571428571, |
|
"grad_norm": 5.269715309143066, |
|
"learning_rate": 5e-05, |
|
"loss": 0.5582, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 8.857142857142858, |
|
"eval_accuracy": 0.5822784810126582, |
|
"eval_loss": 0.6226171851158142, |
|
"eval_runtime": 1.2341, |
|
"eval_samples_per_second": 64.013, |
|
"eval_steps_per_second": 2.431, |
|
"step": 31 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.6329113924050633, |
|
"eval_loss": 0.5478971004486084, |
|
"eval_runtime": 1.2666, |
|
"eval_samples_per_second": 62.371, |
|
"eval_steps_per_second": 2.369, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 10.857142857142858, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.5230492949485779, |
|
"eval_runtime": 1.2346, |
|
"eval_samples_per_second": 63.987, |
|
"eval_steps_per_second": 2.43, |
|
"step": 38 |
|
}, |
|
{ |
|
"epoch": 11.428571428571429, |
|
"grad_norm": 7.602634906768799, |
|
"learning_rate": 4.814814814814815e-05, |
|
"loss": 0.4807, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.6708860759493671, |
|
"eval_loss": 0.6146434545516968, |
|
"eval_runtime": 1.2409, |
|
"eval_samples_per_second": 63.662, |
|
"eval_steps_per_second": 2.418, |
|
"step": 42 |
|
}, |
|
{ |
|
"epoch": 12.857142857142858, |
|
"eval_accuracy": 0.6708860759493671, |
|
"eval_loss": 0.6345846652984619, |
|
"eval_runtime": 1.2574, |
|
"eval_samples_per_second": 62.829, |
|
"eval_steps_per_second": 2.386, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.5822784810126582, |
|
"eval_loss": 1.0584036111831665, |
|
"eval_runtime": 1.2461, |
|
"eval_samples_per_second": 63.397, |
|
"eval_steps_per_second": 2.407, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 14.285714285714286, |
|
"grad_norm": 17.224761962890625, |
|
"learning_rate": 4.62962962962963e-05, |
|
"loss": 0.4441, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 14.857142857142858, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.47162705659866333, |
|
"eval_runtime": 1.239, |
|
"eval_samples_per_second": 63.763, |
|
"eval_steps_per_second": 2.421, |
|
"step": 52 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.759493670886076, |
|
"eval_loss": 0.44540753960609436, |
|
"eval_runtime": 1.2481, |
|
"eval_samples_per_second": 63.298, |
|
"eval_steps_per_second": 2.404, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 16.857142857142858, |
|
"eval_accuracy": 0.759493670886076, |
|
"eval_loss": 0.5688183903694153, |
|
"eval_runtime": 1.239, |
|
"eval_samples_per_second": 63.76, |
|
"eval_steps_per_second": 2.421, |
|
"step": 59 |
|
}, |
|
{ |
|
"epoch": 17.142857142857142, |
|
"grad_norm": 6.364123344421387, |
|
"learning_rate": 4.4444444444444447e-05, |
|
"loss": 0.3829, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.5232765078544617, |
|
"eval_runtime": 1.252, |
|
"eval_samples_per_second": 63.097, |
|
"eval_steps_per_second": 2.396, |
|
"step": 63 |
|
}, |
|
{ |
|
"epoch": 18.857142857142858, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.5291085839271545, |
|
"eval_runtime": 1.2688, |
|
"eval_samples_per_second": 62.265, |
|
"eval_steps_per_second": 2.365, |
|
"step": 66 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"grad_norm": 10.538686752319336, |
|
"learning_rate": 4.259259259259259e-05, |
|
"loss": 0.3352, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.4980542063713074, |
|
"eval_runtime": 1.3123, |
|
"eval_samples_per_second": 60.199, |
|
"eval_steps_per_second": 2.286, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 20.857142857142858, |
|
"eval_accuracy": 0.7468354430379747, |
|
"eval_loss": 0.8590209484100342, |
|
"eval_runtime": 1.252, |
|
"eval_samples_per_second": 63.098, |
|
"eval_steps_per_second": 2.396, |
|
"step": 73 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.5201254487037659, |
|
"eval_runtime": 1.2196, |
|
"eval_samples_per_second": 64.777, |
|
"eval_steps_per_second": 2.46, |
|
"step": 77 |
|
}, |
|
{ |
|
"epoch": 22.857142857142858, |
|
"grad_norm": 9.761859893798828, |
|
"learning_rate": 4.074074074074074e-05, |
|
"loss": 0.341, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 22.857142857142858, |
|
"eval_accuracy": 0.7468354430379747, |
|
"eval_loss": 0.6529780626296997, |
|
"eval_runtime": 1.2528, |
|
"eval_samples_per_second": 63.056, |
|
"eval_steps_per_second": 2.395, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.5731784701347351, |
|
"eval_runtime": 1.2777, |
|
"eval_samples_per_second": 61.831, |
|
"eval_steps_per_second": 2.348, |
|
"step": 84 |
|
}, |
|
{ |
|
"epoch": 24.857142857142858, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.6532183885574341, |
|
"eval_runtime": 1.2546, |
|
"eval_samples_per_second": 62.966, |
|
"eval_steps_per_second": 2.391, |
|
"step": 87 |
|
}, |
|
{ |
|
"epoch": 25.714285714285715, |
|
"grad_norm": 6.66892147064209, |
|
"learning_rate": 3.888888888888889e-05, |
|
"loss": 0.2554, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.5655595064163208, |
|
"eval_runtime": 1.2494, |
|
"eval_samples_per_second": 63.232, |
|
"eval_steps_per_second": 2.401, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 26.857142857142858, |
|
"eval_accuracy": 0.759493670886076, |
|
"eval_loss": 0.8085242509841919, |
|
"eval_runtime": 1.2356, |
|
"eval_samples_per_second": 63.936, |
|
"eval_steps_per_second": 2.428, |
|
"step": 94 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.4826371371746063, |
|
"eval_runtime": 1.2427, |
|
"eval_samples_per_second": 63.57, |
|
"eval_steps_per_second": 2.414, |
|
"step": 98 |
|
}, |
|
{ |
|
"epoch": 28.571428571428573, |
|
"grad_norm": 6.952823162078857, |
|
"learning_rate": 3.7037037037037037e-05, |
|
"loss": 0.325, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 28.857142857142858, |
|
"eval_accuracy": 0.7468354430379747, |
|
"eval_loss": 0.8227420449256897, |
|
"eval_runtime": 1.2487, |
|
"eval_samples_per_second": 63.264, |
|
"eval_steps_per_second": 2.402, |
|
"step": 101 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.5349090695381165, |
|
"eval_runtime": 1.2487, |
|
"eval_samples_per_second": 63.268, |
|
"eval_steps_per_second": 2.403, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 30.857142857142858, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.7055901288986206, |
|
"eval_runtime": 1.2375, |
|
"eval_samples_per_second": 63.84, |
|
"eval_steps_per_second": 2.424, |
|
"step": 108 |
|
}, |
|
{ |
|
"epoch": 31.428571428571427, |
|
"grad_norm": 3.615732192993164, |
|
"learning_rate": 3.518518518518519e-05, |
|
"loss": 0.2572, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 32.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.5414595007896423, |
|
"eval_runtime": 1.2397, |
|
"eval_samples_per_second": 63.727, |
|
"eval_steps_per_second": 2.42, |
|
"step": 112 |
|
}, |
|
{ |
|
"epoch": 32.857142857142854, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.595321536064148, |
|
"eval_runtime": 1.2534, |
|
"eval_samples_per_second": 63.029, |
|
"eval_steps_per_second": 2.393, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 34.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.5883573889732361, |
|
"eval_runtime": 1.2422, |
|
"eval_samples_per_second": 63.599, |
|
"eval_steps_per_second": 2.415, |
|
"step": 119 |
|
}, |
|
{ |
|
"epoch": 34.285714285714285, |
|
"grad_norm": 5.199797630310059, |
|
"learning_rate": 3.3333333333333335e-05, |
|
"loss": 0.1994, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 34.857142857142854, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.5971169471740723, |
|
"eval_runtime": 1.2584, |
|
"eval_samples_per_second": 62.78, |
|
"eval_steps_per_second": 2.384, |
|
"step": 122 |
|
}, |
|
{ |
|
"epoch": 36.0, |
|
"eval_accuracy": 0.7468354430379747, |
|
"eval_loss": 0.8480885028839111, |
|
"eval_runtime": 1.2599, |
|
"eval_samples_per_second": 62.705, |
|
"eval_steps_per_second": 2.381, |
|
"step": 126 |
|
}, |
|
{ |
|
"epoch": 36.857142857142854, |
|
"eval_accuracy": 0.759493670886076, |
|
"eval_loss": 0.6106647849082947, |
|
"eval_runtime": 1.2386, |
|
"eval_samples_per_second": 63.782, |
|
"eval_steps_per_second": 2.422, |
|
"step": 129 |
|
}, |
|
{ |
|
"epoch": 37.142857142857146, |
|
"grad_norm": 7.35693883895874, |
|
"learning_rate": 3.148148148148148e-05, |
|
"loss": 0.209, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 38.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.9241662621498108, |
|
"eval_runtime": 1.2404, |
|
"eval_samples_per_second": 63.691, |
|
"eval_steps_per_second": 2.419, |
|
"step": 133 |
|
}, |
|
{ |
|
"epoch": 38.857142857142854, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.7105510234832764, |
|
"eval_runtime": 1.249, |
|
"eval_samples_per_second": 63.253, |
|
"eval_steps_per_second": 2.402, |
|
"step": 136 |
|
}, |
|
{ |
|
"epoch": 40.0, |
|
"grad_norm": 4.4857497215271, |
|
"learning_rate": 2.962962962962963e-05, |
|
"loss": 0.188, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 40.0, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.9580705761909485, |
|
"eval_runtime": 1.2736, |
|
"eval_samples_per_second": 62.028, |
|
"eval_steps_per_second": 2.355, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 40.857142857142854, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.9281572699546814, |
|
"eval_runtime": 1.2512, |
|
"eval_samples_per_second": 63.139, |
|
"eval_steps_per_second": 2.398, |
|
"step": 143 |
|
}, |
|
{ |
|
"epoch": 42.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0447419881820679, |
|
"eval_runtime": 1.263, |
|
"eval_samples_per_second": 62.549, |
|
"eval_steps_per_second": 2.375, |
|
"step": 147 |
|
}, |
|
{ |
|
"epoch": 42.857142857142854, |
|
"grad_norm": 4.7659502029418945, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.1982, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 42.857142857142854, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.974770724773407, |
|
"eval_runtime": 1.242, |
|
"eval_samples_per_second": 63.606, |
|
"eval_steps_per_second": 2.415, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 44.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.7131812572479248, |
|
"eval_runtime": 1.2552, |
|
"eval_samples_per_second": 62.94, |
|
"eval_steps_per_second": 2.39, |
|
"step": 154 |
|
}, |
|
{ |
|
"epoch": 44.857142857142854, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.695663332939148, |
|
"eval_runtime": 1.2512, |
|
"eval_samples_per_second": 63.138, |
|
"eval_steps_per_second": 2.398, |
|
"step": 157 |
|
}, |
|
{ |
|
"epoch": 45.714285714285715, |
|
"grad_norm": 5.4431376457214355, |
|
"learning_rate": 2.5925925925925925e-05, |
|
"loss": 0.1799, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 46.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.8673846125602722, |
|
"eval_runtime": 1.2584, |
|
"eval_samples_per_second": 62.78, |
|
"eval_steps_per_second": 2.384, |
|
"step": 161 |
|
}, |
|
{ |
|
"epoch": 46.857142857142854, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.9036404490470886, |
|
"eval_runtime": 1.2321, |
|
"eval_samples_per_second": 64.12, |
|
"eval_steps_per_second": 2.435, |
|
"step": 164 |
|
}, |
|
{ |
|
"epoch": 48.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.9909658432006836, |
|
"eval_runtime": 1.2617, |
|
"eval_samples_per_second": 62.612, |
|
"eval_steps_per_second": 2.378, |
|
"step": 168 |
|
}, |
|
{ |
|
"epoch": 48.57142857142857, |
|
"grad_norm": 3.8231964111328125, |
|
"learning_rate": 2.4074074074074074e-05, |
|
"loss": 0.1502, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 48.857142857142854, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 1.0820796489715576, |
|
"eval_runtime": 1.2453, |
|
"eval_samples_per_second": 63.438, |
|
"eval_steps_per_second": 2.409, |
|
"step": 171 |
|
}, |
|
{ |
|
"epoch": 50.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.8116889595985413, |
|
"eval_runtime": 1.2355, |
|
"eval_samples_per_second": 63.943, |
|
"eval_steps_per_second": 2.428, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 50.857142857142854, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.903709888458252, |
|
"eval_runtime": 1.2502, |
|
"eval_samples_per_second": 63.19, |
|
"eval_steps_per_second": 2.4, |
|
"step": 178 |
|
}, |
|
{ |
|
"epoch": 51.42857142857143, |
|
"grad_norm": 4.519005298614502, |
|
"learning_rate": 2.2222222222222223e-05, |
|
"loss": 0.1591, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 52.0, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.8272249102592468, |
|
"eval_runtime": 1.2424, |
|
"eval_samples_per_second": 63.585, |
|
"eval_steps_per_second": 2.415, |
|
"step": 182 |
|
}, |
|
{ |
|
"epoch": 52.857142857142854, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0108956098556519, |
|
"eval_runtime": 1.2435, |
|
"eval_samples_per_second": 63.529, |
|
"eval_steps_per_second": 2.412, |
|
"step": 185 |
|
}, |
|
{ |
|
"epoch": 54.0, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.8490244150161743, |
|
"eval_runtime": 1.2485, |
|
"eval_samples_per_second": 63.274, |
|
"eval_steps_per_second": 2.403, |
|
"step": 189 |
|
}, |
|
{ |
|
"epoch": 54.285714285714285, |
|
"grad_norm": 4.413188457489014, |
|
"learning_rate": 2.037037037037037e-05, |
|
"loss": 0.1588, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 54.857142857142854, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.897348165512085, |
|
"eval_runtime": 1.2596, |
|
"eval_samples_per_second": 62.716, |
|
"eval_steps_per_second": 2.382, |
|
"step": 192 |
|
}, |
|
{ |
|
"epoch": 56.0, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.9084045886993408, |
|
"eval_runtime": 1.2452, |
|
"eval_samples_per_second": 63.443, |
|
"eval_steps_per_second": 2.409, |
|
"step": 196 |
|
}, |
|
{ |
|
"epoch": 56.857142857142854, |
|
"eval_accuracy": 0.759493670886076, |
|
"eval_loss": 0.8033677935600281, |
|
"eval_runtime": 1.2471, |
|
"eval_samples_per_second": 63.346, |
|
"eval_steps_per_second": 2.406, |
|
"step": 199 |
|
}, |
|
{ |
|
"epoch": 57.142857142857146, |
|
"grad_norm": 2.8926455974578857, |
|
"learning_rate": 1.8518518518518518e-05, |
|
"loss": 0.1353, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 58.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.8884690403938293, |
|
"eval_runtime": 1.2311, |
|
"eval_samples_per_second": 64.169, |
|
"eval_steps_per_second": 2.437, |
|
"step": 203 |
|
}, |
|
{ |
|
"epoch": 58.857142857142854, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0836836099624634, |
|
"eval_runtime": 1.2508, |
|
"eval_samples_per_second": 63.158, |
|
"eval_steps_per_second": 2.398, |
|
"step": 206 |
|
}, |
|
{ |
|
"epoch": 60.0, |
|
"grad_norm": 6.144521713256836, |
|
"learning_rate": 1.6666666666666667e-05, |
|
"loss": 0.1308, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 60.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.8627098202705383, |
|
"eval_runtime": 1.2285, |
|
"eval_samples_per_second": 64.307, |
|
"eval_steps_per_second": 2.442, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 60.857142857142854, |
|
"eval_accuracy": 0.810126582278481, |
|
"eval_loss": 0.9842409491539001, |
|
"eval_runtime": 1.2516, |
|
"eval_samples_per_second": 63.117, |
|
"eval_steps_per_second": 2.397, |
|
"step": 213 |
|
}, |
|
{ |
|
"epoch": 62.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.9965667724609375, |
|
"eval_runtime": 1.2496, |
|
"eval_samples_per_second": 63.222, |
|
"eval_steps_per_second": 2.401, |
|
"step": 217 |
|
}, |
|
{ |
|
"epoch": 62.857142857142854, |
|
"grad_norm": 3.8410747051239014, |
|
"learning_rate": 1.4814814814814815e-05, |
|
"loss": 0.1402, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 62.857142857142854, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.9638678431510925, |
|
"eval_runtime": 1.2493, |
|
"eval_samples_per_second": 63.235, |
|
"eval_steps_per_second": 2.401, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 64.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.010369062423706, |
|
"eval_runtime": 1.2619, |
|
"eval_samples_per_second": 62.603, |
|
"eval_steps_per_second": 2.377, |
|
"step": 224 |
|
}, |
|
{ |
|
"epoch": 64.85714285714286, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 0.8939818143844604, |
|
"eval_runtime": 1.2535, |
|
"eval_samples_per_second": 63.023, |
|
"eval_steps_per_second": 2.393, |
|
"step": 227 |
|
}, |
|
{ |
|
"epoch": 65.71428571428571, |
|
"grad_norm": 6.720986366271973, |
|
"learning_rate": 1.2962962962962962e-05, |
|
"loss": 0.1301, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 66.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.9348751306533813, |
|
"eval_runtime": 1.2555, |
|
"eval_samples_per_second": 62.921, |
|
"eval_steps_per_second": 2.389, |
|
"step": 231 |
|
}, |
|
{ |
|
"epoch": 66.85714285714286, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.1485346555709839, |
|
"eval_runtime": 1.2412, |
|
"eval_samples_per_second": 63.646, |
|
"eval_steps_per_second": 2.417, |
|
"step": 234 |
|
}, |
|
{ |
|
"epoch": 68.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0020633935928345, |
|
"eval_runtime": 1.2514, |
|
"eval_samples_per_second": 63.128, |
|
"eval_steps_per_second": 2.397, |
|
"step": 238 |
|
}, |
|
{ |
|
"epoch": 68.57142857142857, |
|
"grad_norm": 6.518858909606934, |
|
"learning_rate": 1.1111111111111112e-05, |
|
"loss": 0.1399, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 68.85714285714286, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.8970956206321716, |
|
"eval_runtime": 1.2438, |
|
"eval_samples_per_second": 63.517, |
|
"eval_steps_per_second": 2.412, |
|
"step": 241 |
|
}, |
|
{ |
|
"epoch": 70.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.8603967428207397, |
|
"eval_runtime": 1.2534, |
|
"eval_samples_per_second": 63.028, |
|
"eval_steps_per_second": 2.393, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 70.85714285714286, |
|
"eval_accuracy": 0.8227848101265823, |
|
"eval_loss": 0.8625026941299438, |
|
"eval_runtime": 1.2393, |
|
"eval_samples_per_second": 63.747, |
|
"eval_steps_per_second": 2.421, |
|
"step": 248 |
|
}, |
|
{ |
|
"epoch": 71.42857142857143, |
|
"grad_norm": 4.293798446655273, |
|
"learning_rate": 9.259259259259259e-06, |
|
"loss": 0.1203, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 72.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 1.030073642730713, |
|
"eval_runtime": 1.2476, |
|
"eval_samples_per_second": 63.32, |
|
"eval_steps_per_second": 2.405, |
|
"step": 252 |
|
}, |
|
{ |
|
"epoch": 72.85714285714286, |
|
"eval_accuracy": 0.7721518987341772, |
|
"eval_loss": 1.23548424243927, |
|
"eval_runtime": 1.2525, |
|
"eval_samples_per_second": 63.073, |
|
"eval_steps_per_second": 2.395, |
|
"step": 255 |
|
}, |
|
{ |
|
"epoch": 74.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0715705156326294, |
|
"eval_runtime": 1.2512, |
|
"eval_samples_per_second": 63.142, |
|
"eval_steps_per_second": 2.398, |
|
"step": 259 |
|
}, |
|
{ |
|
"epoch": 74.28571428571429, |
|
"grad_norm": 5.128173828125, |
|
"learning_rate": 7.4074074074074075e-06, |
|
"loss": 0.1385, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 74.85714285714286, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.9597594738006592, |
|
"eval_runtime": 1.2314, |
|
"eval_samples_per_second": 64.153, |
|
"eval_steps_per_second": 2.436, |
|
"step": 262 |
|
}, |
|
{ |
|
"epoch": 76.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.9965071678161621, |
|
"eval_runtime": 1.2328, |
|
"eval_samples_per_second": 64.082, |
|
"eval_steps_per_second": 2.433, |
|
"step": 266 |
|
}, |
|
{ |
|
"epoch": 76.85714285714286, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 1.1017982959747314, |
|
"eval_runtime": 1.2424, |
|
"eval_samples_per_second": 63.588, |
|
"eval_steps_per_second": 2.415, |
|
"step": 269 |
|
}, |
|
{ |
|
"epoch": 77.14285714285714, |
|
"grad_norm": 3.9869062900543213, |
|
"learning_rate": 5.555555555555556e-06, |
|
"loss": 0.115, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 78.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 1.0894912481307983, |
|
"eval_runtime": 1.2436, |
|
"eval_samples_per_second": 63.527, |
|
"eval_steps_per_second": 2.412, |
|
"step": 273 |
|
}, |
|
{ |
|
"epoch": 78.85714285714286, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0261164903640747, |
|
"eval_runtime": 1.2491, |
|
"eval_samples_per_second": 63.244, |
|
"eval_steps_per_second": 2.402, |
|
"step": 276 |
|
}, |
|
{ |
|
"epoch": 80.0, |
|
"grad_norm": 3.888028860092163, |
|
"learning_rate": 3.7037037037037037e-06, |
|
"loss": 0.1227, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 80.0, |
|
"eval_accuracy": 0.7974683544303798, |
|
"eval_loss": 0.980012059211731, |
|
"eval_runtime": 1.2662, |
|
"eval_samples_per_second": 62.39, |
|
"eval_steps_per_second": 2.369, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 80.85714285714286, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 0.9959664940834045, |
|
"eval_runtime": 1.2589, |
|
"eval_samples_per_second": 62.755, |
|
"eval_steps_per_second": 2.383, |
|
"step": 283 |
|
}, |
|
{ |
|
"epoch": 82.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0250811576843262, |
|
"eval_runtime": 1.2496, |
|
"eval_samples_per_second": 63.219, |
|
"eval_steps_per_second": 2.401, |
|
"step": 287 |
|
}, |
|
{ |
|
"epoch": 82.85714285714286, |
|
"grad_norm": 2.959679126739502, |
|
"learning_rate": 1.8518518518518519e-06, |
|
"loss": 0.1044, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 82.85714285714286, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.023586630821228, |
|
"eval_runtime": 1.2687, |
|
"eval_samples_per_second": 62.269, |
|
"eval_steps_per_second": 2.365, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 84.0, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.013852596282959, |
|
"eval_runtime": 1.2435, |
|
"eval_samples_per_second": 63.532, |
|
"eval_steps_per_second": 2.413, |
|
"step": 294 |
|
}, |
|
{ |
|
"epoch": 84.85714285714286, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0072976350784302, |
|
"eval_runtime": 1.2357, |
|
"eval_samples_per_second": 63.932, |
|
"eval_steps_per_second": 2.428, |
|
"step": 297 |
|
}, |
|
{ |
|
"epoch": 85.71428571428571, |
|
"grad_norm": 3.944120168685913, |
|
"learning_rate": 0.0, |
|
"loss": 0.1246, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 85.71428571428571, |
|
"eval_accuracy": 0.7848101265822784, |
|
"eval_loss": 1.0065951347351074, |
|
"eval_runtime": 1.2292, |
|
"eval_samples_per_second": 64.27, |
|
"eval_steps_per_second": 2.441, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 85.71428571428571, |
|
"step": 300, |
|
"total_flos": 2.9349165326823014e+18, |
|
"train_loss": 0.2534414354960124, |
|
"train_runtime": 1834.4014, |
|
"train_samples_per_second": 24.095, |
|
"train_steps_per_second": 0.164 |
|
}, |
|
{ |
|
"epoch": 85.71428571428571, |
|
"eval_accuracy": 0.8227848101265823, |
|
"eval_loss": 0.8625026941299438, |
|
"eval_runtime": 1.2267, |
|
"eval_samples_per_second": 64.4, |
|
"eval_steps_per_second": 2.446, |
|
"step": 300 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 300, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 100, |
|
"save_steps": 500, |
|
"total_flos": 2.9349165326823014e+18, |
|
"train_batch_size": 32, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|