| { | |
| "best_metric": 0.8227848101265823, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-55-fold3/checkpoint-59", | |
| "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": 0.4050632911392405, | |
| "eval_loss": 0.8367149233818054, | |
| "eval_runtime": 1.1081, | |
| "eval_samples_per_second": 71.291, | |
| "eval_steps_per_second": 2.707, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.45569620253164556, | |
| "eval_loss": 0.7223008275032043, | |
| "eval_runtime": 1.1027, | |
| "eval_samples_per_second": 71.639, | |
| "eval_steps_per_second": 2.72, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 2.8950867652893066, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.7025, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_accuracy": 0.46835443037974683, | |
| "eval_loss": 0.7199137210845947, | |
| "eval_runtime": 1.1071, | |
| "eval_samples_per_second": 71.358, | |
| "eval_steps_per_second": 2.71, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.6096423864364624, | |
| "eval_runtime": 1.1237, | |
| "eval_samples_per_second": 70.301, | |
| "eval_steps_per_second": 2.67, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_accuracy": 0.5822784810126582, | |
| "eval_loss": 0.627830445766449, | |
| "eval_runtime": 1.1363, | |
| "eval_samples_per_second": 69.527, | |
| "eval_steps_per_second": 2.64, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 4.293572902679443, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6356, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.5629443526268005, | |
| "eval_runtime": 1.1192, | |
| "eval_samples_per_second": 70.587, | |
| "eval_steps_per_second": 2.681, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_accuracy": 0.6835443037974683, | |
| "eval_loss": 0.5924213528633118, | |
| "eval_runtime": 1.1352, | |
| "eval_samples_per_second": 69.592, | |
| "eval_steps_per_second": 2.643, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.536530077457428, | |
| "eval_runtime": 1.1272, | |
| "eval_samples_per_second": 70.087, | |
| "eval_steps_per_second": 2.662, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 3.6636528968811035, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5493, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.857142857142858, | |
| "eval_accuracy": 0.6329113924050633, | |
| "eval_loss": 0.6081864237785339, | |
| "eval_runtime": 1.1379, | |
| "eval_samples_per_second": 69.429, | |
| "eval_steps_per_second": 2.637, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5949367088607594, | |
| "eval_loss": 0.7238861918449402, | |
| "eval_runtime": 1.1332, | |
| "eval_samples_per_second": 69.714, | |
| "eval_steps_per_second": 2.647, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.5434973239898682, | |
| "eval_runtime": 1.1358, | |
| "eval_samples_per_second": 69.555, | |
| "eval_steps_per_second": 2.641, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 12.597495079040527, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.5205, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5569620253164557, | |
| "eval_loss": 0.8530091047286987, | |
| "eval_runtime": 1.137, | |
| "eval_samples_per_second": 69.479, | |
| "eval_steps_per_second": 2.638, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 12.857142857142858, | |
| "eval_accuracy": 0.6708860759493671, | |
| "eval_loss": 0.5530335307121277, | |
| "eval_runtime": 1.1759, | |
| "eval_samples_per_second": 67.183, | |
| "eval_steps_per_second": 2.551, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.4728139042854309, | |
| "eval_runtime": 1.1396, | |
| "eval_samples_per_second": 69.324, | |
| "eval_steps_per_second": 2.633, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 2.1170239448547363, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.4979, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 14.857142857142858, | |
| "eval_accuracy": 0.5569620253164557, | |
| "eval_loss": 0.9570804238319397, | |
| "eval_runtime": 1.1436, | |
| "eval_samples_per_second": 69.08, | |
| "eval_steps_per_second": 2.623, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.5192645788192749, | |
| "eval_runtime": 1.1502, | |
| "eval_samples_per_second": 68.682, | |
| "eval_steps_per_second": 2.608, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 16.857142857142858, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.45285508036613464, | |
| "eval_runtime": 1.144, | |
| "eval_samples_per_second": 69.058, | |
| "eval_steps_per_second": 2.622, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 5.5375895500183105, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.4957, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.46856963634490967, | |
| "eval_runtime": 1.1477, | |
| "eval_samples_per_second": 68.831, | |
| "eval_steps_per_second": 2.614, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 18.857142857142858, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.5060284733772278, | |
| "eval_runtime": 1.1584, | |
| "eval_samples_per_second": 68.196, | |
| "eval_steps_per_second": 2.59, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.345262050628662, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.3659, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.4820997416973114, | |
| "eval_runtime": 1.1568, | |
| "eval_samples_per_second": 68.291, | |
| "eval_steps_per_second": 2.593, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 20.857142857142858, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.6116397380828857, | |
| "eval_runtime": 1.1342, | |
| "eval_samples_per_second": 69.653, | |
| "eval_steps_per_second": 2.645, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.5860076546669006, | |
| "eval_runtime": 1.1445, | |
| "eval_samples_per_second": 69.025, | |
| "eval_steps_per_second": 2.621, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 3.9714572429656982, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.2973, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.7099510431289673, | |
| "eval_runtime": 1.1367, | |
| "eval_samples_per_second": 69.502, | |
| "eval_steps_per_second": 2.639, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.6445814371109009, | |
| "eval_runtime": 1.152, | |
| "eval_samples_per_second": 68.574, | |
| "eval_steps_per_second": 2.604, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 24.857142857142858, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.6294004321098328, | |
| "eval_runtime": 1.1524, | |
| "eval_samples_per_second": 68.553, | |
| "eval_steps_per_second": 2.603, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "grad_norm": 5.2691497802734375, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2647, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.5988398194313049, | |
| "eval_runtime": 1.1436, | |
| "eval_samples_per_second": 69.082, | |
| "eval_steps_per_second": 2.623, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 26.857142857142858, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.5255513191223145, | |
| "eval_runtime": 1.1562, | |
| "eval_samples_per_second": 68.329, | |
| "eval_steps_per_second": 2.595, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.6628075838088989, | |
| "eval_runtime": 1.1535, | |
| "eval_samples_per_second": 68.487, | |
| "eval_steps_per_second": 2.601, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 3.251382350921631, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2527, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 28.857142857142858, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.505411684513092, | |
| "eval_runtime": 1.1554, | |
| "eval_samples_per_second": 68.375, | |
| "eval_steps_per_second": 2.597, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7632080912590027, | |
| "eval_runtime": 1.1433, | |
| "eval_samples_per_second": 69.101, | |
| "eval_steps_per_second": 2.624, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 30.857142857142858, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5917405486106873, | |
| "eval_runtime": 1.1411, | |
| "eval_samples_per_second": 69.229, | |
| "eval_steps_per_second": 2.629, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 31.428571428571427, | |
| "grad_norm": 2.8656537532806396, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.2176, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.5292685031890869, | |
| "eval_runtime": 1.1391, | |
| "eval_samples_per_second": 69.351, | |
| "eval_steps_per_second": 2.634, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 32.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.6047636866569519, | |
| "eval_runtime": 1.1444, | |
| "eval_samples_per_second": 69.033, | |
| "eval_steps_per_second": 2.622, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.5710076689720154, | |
| "eval_runtime": 1.1506, | |
| "eval_samples_per_second": 68.662, | |
| "eval_steps_per_second": 2.607, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 3.1147148609161377, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1633, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 34.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.5901313424110413, | |
| "eval_runtime": 1.1453, | |
| "eval_samples_per_second": 68.979, | |
| "eval_steps_per_second": 2.619, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.8161200284957886, | |
| "eval_runtime": 1.1351, | |
| "eval_samples_per_second": 69.598, | |
| "eval_steps_per_second": 2.643, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 36.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.7201886773109436, | |
| "eval_runtime": 1.1562, | |
| "eval_samples_per_second": 68.325, | |
| "eval_steps_per_second": 2.595, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 37.142857142857146, | |
| "grad_norm": 4.740509033203125, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.1753, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.8239344358444214, | |
| "eval_runtime": 1.136, | |
| "eval_samples_per_second": 69.545, | |
| "eval_steps_per_second": 2.641, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 38.857142857142854, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 0.890842854976654, | |
| "eval_runtime": 1.1174, | |
| "eval_samples_per_second": 70.703, | |
| "eval_steps_per_second": 2.685, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 1.9374984502792358, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.1743, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.851948618888855, | |
| "eval_runtime": 1.1451, | |
| "eval_samples_per_second": 68.99, | |
| "eval_steps_per_second": 2.62, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 40.857142857142854, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 1.0070878267288208, | |
| "eval_runtime": 1.125, | |
| "eval_samples_per_second": 70.22, | |
| "eval_steps_per_second": 2.667, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.7842212915420532, | |
| "eval_runtime": 1.1609, | |
| "eval_samples_per_second": 68.052, | |
| "eval_steps_per_second": 2.584, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "grad_norm": 4.079115867614746, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1532, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.7827017307281494, | |
| "eval_runtime": 1.2227, | |
| "eval_samples_per_second": 64.612, | |
| "eval_steps_per_second": 2.454, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.7150283455848694, | |
| "eval_runtime": 1.2107, | |
| "eval_samples_per_second": 65.253, | |
| "eval_steps_per_second": 2.478, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 44.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.6905309557914734, | |
| "eval_runtime": 1.1211, | |
| "eval_samples_per_second": 70.469, | |
| "eval_steps_per_second": 2.676, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 2.595792293548584, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.1526, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.7088607594936709, | |
| "eval_loss": 0.9259989261627197, | |
| "eval_runtime": 1.1289, | |
| "eval_samples_per_second": 69.98, | |
| "eval_steps_per_second": 2.657, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 46.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7932636737823486, | |
| "eval_runtime": 1.1258, | |
| "eval_samples_per_second": 70.172, | |
| "eval_steps_per_second": 2.665, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.8579609394073486, | |
| "eval_runtime": 1.16, | |
| "eval_samples_per_second": 68.104, | |
| "eval_steps_per_second": 2.586, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "grad_norm": 3.7627644538879395, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.1519, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 48.857142857142854, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.689932644367218, | |
| "eval_runtime": 1.1366, | |
| "eval_samples_per_second": 69.506, | |
| "eval_steps_per_second": 2.639, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7069446444511414, | |
| "eval_runtime": 1.118, | |
| "eval_samples_per_second": 70.66, | |
| "eval_steps_per_second": 2.683, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 50.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.674085259437561, | |
| "eval_runtime": 1.144, | |
| "eval_samples_per_second": 69.058, | |
| "eval_steps_per_second": 2.622, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 51.42857142857143, | |
| "grad_norm": 1.8397268056869507, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1292, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.7182620763778687, | |
| "eval_runtime": 1.1524, | |
| "eval_samples_per_second": 68.553, | |
| "eval_steps_per_second": 2.603, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 52.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.8050850629806519, | |
| "eval_runtime": 1.1248, | |
| "eval_samples_per_second": 70.232, | |
| "eval_steps_per_second": 2.667, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6882942914962769, | |
| "eval_runtime": 1.1548, | |
| "eval_samples_per_second": 68.41, | |
| "eval_steps_per_second": 2.598, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 54.285714285714285, | |
| "grad_norm": 2.849743127822876, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.1305, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 54.857142857142854, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.826583206653595, | |
| "eval_runtime": 1.1097, | |
| "eval_samples_per_second": 71.193, | |
| "eval_steps_per_second": 2.704, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 1.0870975255966187, | |
| "eval_runtime": 1.1164, | |
| "eval_samples_per_second": 70.761, | |
| "eval_steps_per_second": 2.687, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 56.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7594748139381409, | |
| "eval_runtime": 1.1303, | |
| "eval_samples_per_second": 69.892, | |
| "eval_steps_per_second": 2.654, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 57.142857142857146, | |
| "grad_norm": 3.434027910232544, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1129, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.6880047917366028, | |
| "eval_runtime": 1.1278, | |
| "eval_samples_per_second": 70.046, | |
| "eval_steps_per_second": 2.66, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 58.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 1.0675818920135498, | |
| "eval_runtime": 1.1311, | |
| "eval_samples_per_second": 69.846, | |
| "eval_steps_per_second": 2.652, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 2.7458627223968506, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1369, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.8078165054321289, | |
| "eval_runtime": 1.1294, | |
| "eval_samples_per_second": 69.946, | |
| "eval_steps_per_second": 2.656, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 60.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7849780917167664, | |
| "eval_runtime": 1.1336, | |
| "eval_samples_per_second": 69.69, | |
| "eval_steps_per_second": 2.646, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.6975088715553284, | |
| "eval_runtime": 1.1508, | |
| "eval_samples_per_second": 68.647, | |
| "eval_steps_per_second": 2.607, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "grad_norm": 3.0175793170928955, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.127, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 62.857142857142854, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 0.7211695909500122, | |
| "eval_runtime": 1.1292, | |
| "eval_samples_per_second": 69.96, | |
| "eval_steps_per_second": 2.657, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 0.8966912627220154, | |
| "eval_runtime": 1.1532, | |
| "eval_samples_per_second": 68.505, | |
| "eval_steps_per_second": 2.601, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 64.85714285714286, | |
| "eval_accuracy": 0.759493670886076, | |
| "eval_loss": 1.004639983177185, | |
| "eval_runtime": 1.1649, | |
| "eval_samples_per_second": 67.82, | |
| "eval_steps_per_second": 2.575, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 65.71428571428571, | |
| "grad_norm": 2.0275487899780273, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.1238, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.8611046671867371, | |
| "eval_runtime": 1.1447, | |
| "eval_samples_per_second": 69.013, | |
| "eval_steps_per_second": 2.621, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 66.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.9676278829574585, | |
| "eval_runtime": 1.1478, | |
| "eval_samples_per_second": 68.828, | |
| "eval_steps_per_second": 2.614, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.7215189873417721, | |
| "eval_loss": 1.311516284942627, | |
| "eval_runtime": 1.1548, | |
| "eval_samples_per_second": 68.409, | |
| "eval_steps_per_second": 2.598, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 68.57142857142857, | |
| "grad_norm": 1.4194263219833374, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1068, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 68.85714285714286, | |
| "eval_accuracy": 0.7468354430379747, | |
| "eval_loss": 1.0992393493652344, | |
| "eval_runtime": 1.1324, | |
| "eval_samples_per_second": 69.761, | |
| "eval_steps_per_second": 2.649, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.8765135407447815, | |
| "eval_runtime": 1.1526, | |
| "eval_samples_per_second": 68.541, | |
| "eval_steps_per_second": 2.603, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 70.85714285714286, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.8509979248046875, | |
| "eval_runtime": 1.1503, | |
| "eval_samples_per_second": 68.676, | |
| "eval_steps_per_second": 2.608, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 71.42857142857143, | |
| "grad_norm": 2.365654945373535, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.1019, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7402685880661011, | |
| "eval_runtime": 1.1525, | |
| "eval_samples_per_second": 68.549, | |
| "eval_steps_per_second": 2.603, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 72.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7458766102790833, | |
| "eval_runtime": 1.1534, | |
| "eval_samples_per_second": 68.494, | |
| "eval_steps_per_second": 2.601, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7705420255661011, | |
| "eval_runtime": 1.1654, | |
| "eval_samples_per_second": 67.789, | |
| "eval_steps_per_second": 2.574, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 74.28571428571429, | |
| "grad_norm": 2.4320626258850098, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.1002, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 74.85714285714286, | |
| "eval_accuracy": 0.7974683544303798, | |
| "eval_loss": 0.7534676790237427, | |
| "eval_runtime": 1.1501, | |
| "eval_samples_per_second": 68.689, | |
| "eval_steps_per_second": 2.608, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.7124237418174744, | |
| "eval_runtime": 1.1477, | |
| "eval_samples_per_second": 68.833, | |
| "eval_steps_per_second": 2.614, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 76.85714285714286, | |
| "eval_accuracy": 0.7341772151898734, | |
| "eval_loss": 0.7013713121414185, | |
| "eval_runtime": 1.1413, | |
| "eval_samples_per_second": 69.219, | |
| "eval_steps_per_second": 2.629, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 77.14285714285714, | |
| "grad_norm": 3.408477306365967, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1222, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.8067706227302551, | |
| "eval_runtime": 1.1344, | |
| "eval_samples_per_second": 69.639, | |
| "eval_steps_per_second": 2.645, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 78.85714285714286, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.9450699090957642, | |
| "eval_runtime": 1.143, | |
| "eval_samples_per_second": 69.119, | |
| "eval_steps_per_second": 2.625, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 2.774991750717163, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.1091, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 1.0047645568847656, | |
| "eval_runtime": 1.181, | |
| "eval_samples_per_second": 66.89, | |
| "eval_steps_per_second": 2.54, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 80.85714285714286, | |
| "eval_accuracy": 0.7721518987341772, | |
| "eval_loss": 0.9518149495124817, | |
| "eval_runtime": 1.1372, | |
| "eval_samples_per_second": 69.469, | |
| "eval_steps_per_second": 2.638, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.8575441241264343, | |
| "eval_runtime": 1.1296, | |
| "eval_samples_per_second": 69.935, | |
| "eval_steps_per_second": 2.656, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 82.85714285714286, | |
| "grad_norm": 2.476370096206665, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.0957, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 82.85714285714286, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.8440907001495361, | |
| "eval_runtime": 1.1604, | |
| "eval_samples_per_second": 68.078, | |
| "eval_steps_per_second": 2.585, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.8601673245429993, | |
| "eval_runtime": 1.1451, | |
| "eval_samples_per_second": 68.988, | |
| "eval_steps_per_second": 2.62, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 84.85714285714286, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.8700512051582336, | |
| "eval_runtime": 1.1512, | |
| "eval_samples_per_second": 68.626, | |
| "eval_steps_per_second": 2.606, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "grad_norm": 2.8179068565368652, | |
| "learning_rate": 0.0, | |
| "loss": 0.1111, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "eval_accuracy": 0.7848101265822784, | |
| "eval_loss": 0.8731077909469604, | |
| "eval_runtime": 1.1363, | |
| "eval_samples_per_second": 69.522, | |
| "eval_steps_per_second": 2.64, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "step": 300, | |
| "total_flos": 2.9362240500074496e+18, | |
| "train_loss": 0.24259120881557464, | |
| "train_runtime": 1655.0776, | |
| "train_samples_per_second": 26.706, | |
| "train_steps_per_second": 0.181 | |
| }, | |
| { | |
| "epoch": 85.71428571428571, | |
| "eval_accuracy": 0.8227848101265823, | |
| "eval_loss": 0.45285508036613464, | |
| "eval_runtime": 1.1598, | |
| "eval_samples_per_second": 68.116, | |
| "eval_steps_per_second": 2.587, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "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": 2.9362240500074496e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |