| { | |
| "best_metric": 0.6874601244926453, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-2600", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 2.136049747467041, | |
| "learning_rate": 1.98974358974359e-05, | |
| "loss": 1.1239, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.8187670707702637, | |
| "learning_rate": 1.9794871794871798e-05, | |
| "loss": 1.1221, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 2.0219993591308594, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.1164, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.4619803428649902, | |
| "learning_rate": 1.958974358974359e-05, | |
| "loss": 1.1044, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 1.6733014583587646, | |
| "learning_rate": 1.9487179487179488e-05, | |
| "loss": 1.1082, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 1.4969494342803955, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 1.1043, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 2.6347556114196777, | |
| "learning_rate": 1.9282051282051284e-05, | |
| "loss": 1.1028, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 2.5843420028686523, | |
| "learning_rate": 1.9179487179487182e-05, | |
| "loss": 1.0908, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 2.0522916316986084, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 1.094, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.4885082244873047, | |
| "learning_rate": 1.8974358974358975e-05, | |
| "loss": 1.0912, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 1.7014166116714478, | |
| "learning_rate": 1.8871794871794873e-05, | |
| "loss": 1.0949, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 2.294283866882324, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 1.0992, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.129885673522949, | |
| "learning_rate": 1.866666666666667e-05, | |
| "loss": 1.0864, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.42857142857142855, | |
| "eval_loss": 1.0877832174301147, | |
| "eval_runtime": 0.7833, | |
| "eval_samples_per_second": 169.792, | |
| "eval_steps_per_second": 21.703, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 2.331717014312744, | |
| "learning_rate": 1.8564102564102567e-05, | |
| "loss": 1.0774, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 2.5262138843536377, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 1.0719, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.5971320867538452, | |
| "learning_rate": 1.835897435897436e-05, | |
| "loss": 1.0781, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 2.383288860321045, | |
| "learning_rate": 1.8256410256410257e-05, | |
| "loss": 1.0929, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 2.169706106185913, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 1.0805, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 2.1174418926239014, | |
| "learning_rate": 1.8051282051282053e-05, | |
| "loss": 1.08, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.7236179113388062, | |
| "learning_rate": 1.794871794871795e-05, | |
| "loss": 1.0766, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 1.7772722244262695, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 1.0676, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 3.53834867477417, | |
| "learning_rate": 1.7743589743589744e-05, | |
| "loss": 1.0695, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 2.0417070388793945, | |
| "learning_rate": 1.7641025641025642e-05, | |
| "loss": 1.0706, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 1.9734611511230469, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 1.0863, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 1.9997600317001343, | |
| "learning_rate": 1.7435897435897438e-05, | |
| "loss": 1.068, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.0024373531341553, | |
| "learning_rate": 1.7333333333333336e-05, | |
| "loss": 1.0629, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5488721804511278, | |
| "eval_loss": 1.0593525171279907, | |
| "eval_runtime": 0.7442, | |
| "eval_samples_per_second": 178.706, | |
| "eval_steps_per_second": 22.842, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 1.977807641029358, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 1.0711, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 2.3906102180480957, | |
| "learning_rate": 1.7128205128205128e-05, | |
| "loss": 1.0597, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 2.3670897483825684, | |
| "learning_rate": 1.7025641025641026e-05, | |
| "loss": 1.0576, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 3.026155948638916, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 1.0434, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 1.9082350730895996, | |
| "learning_rate": 1.6820512820512822e-05, | |
| "loss": 1.0566, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 1.9187153577804565, | |
| "learning_rate": 1.671794871794872e-05, | |
| "loss": 1.0476, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 1.4435549974441528, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 1.032, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 2.1457245349884033, | |
| "learning_rate": 1.6512820512820513e-05, | |
| "loss": 1.0475, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 1.9391709566116333, | |
| "learning_rate": 1.641025641025641e-05, | |
| "loss": 1.0486, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 1.8148127794265747, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 1.0407, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 2.444157123565674, | |
| "learning_rate": 1.6205128205128207e-05, | |
| "loss": 1.0356, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.9061695337295532, | |
| "learning_rate": 1.6102564102564105e-05, | |
| "loss": 1.0239, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.859686851501465, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.0434, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6766917293233082, | |
| "eval_loss": 1.0230107307434082, | |
| "eval_runtime": 0.7471, | |
| "eval_samples_per_second": 178.027, | |
| "eval_steps_per_second": 22.755, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.2021689414978027, | |
| "learning_rate": 1.5897435897435897e-05, | |
| "loss": 1.0424, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 1.8670283555984497, | |
| "learning_rate": 1.5794871794871795e-05, | |
| "loss": 1.0299, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 2.193986654281616, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 1.0369, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 2.26470685005188, | |
| "learning_rate": 1.558974358974359e-05, | |
| "loss": 1.0159, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 2.18507981300354, | |
| "learning_rate": 1.548717948717949e-05, | |
| "loss": 1.0282, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 1.8047341108322144, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.0381, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 2.6463704109191895, | |
| "learning_rate": 1.5282051282051282e-05, | |
| "loss": 1.0322, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 1.6456215381622314, | |
| "learning_rate": 1.517948717948718e-05, | |
| "loss": 1.0049, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 2.774256706237793, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 1.0091, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 1.572251319885254, | |
| "learning_rate": 1.4974358974358976e-05, | |
| "loss": 0.998, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.5640805959701538, | |
| "learning_rate": 1.4871794871794874e-05, | |
| "loss": 1.0222, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 1.9231537580490112, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 0.9979, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.481942176818848, | |
| "learning_rate": 1.4666666666666666e-05, | |
| "loss": 1.0214, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6766917293233082, | |
| "eval_loss": 0.9964542388916016, | |
| "eval_runtime": 0.7616, | |
| "eval_samples_per_second": 174.638, | |
| "eval_steps_per_second": 22.322, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 2.7514402866363525, | |
| "learning_rate": 1.4564102564102564e-05, | |
| "loss": 1.0128, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.8411396741867065, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 1.0145, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 2.670154571533203, | |
| "learning_rate": 1.435897435897436e-05, | |
| "loss": 1.0227, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 2.1951498985290527, | |
| "learning_rate": 1.4256410256410258e-05, | |
| "loss": 1.0321, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 1.9692825078964233, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.9829, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 2.611340284347534, | |
| "learning_rate": 1.405128205128205e-05, | |
| "loss": 0.9918, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 2.4288899898529053, | |
| "learning_rate": 1.3948717948717949e-05, | |
| "loss": 0.9879, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 1.7537823915481567, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.9793, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 3.1588003635406494, | |
| "learning_rate": 1.3743589743589745e-05, | |
| "loss": 1.0002, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 2.2472622394561768, | |
| "learning_rate": 1.3641025641025643e-05, | |
| "loss": 1.0094, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 1.7958937883377075, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.9703, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 2.415766477584839, | |
| "learning_rate": 1.3435897435897435e-05, | |
| "loss": 0.9703, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.948933124542236, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 1.0026, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7443609022556391, | |
| "eval_loss": 0.9569369554519653, | |
| "eval_runtime": 0.7647, | |
| "eval_samples_per_second": 173.928, | |
| "eval_steps_per_second": 22.231, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.076923076923077, | |
| "grad_norm": 2.1397032737731934, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.9645, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.153846153846154, | |
| "grad_norm": 2.7277321815490723, | |
| "learning_rate": 1.312820512820513e-05, | |
| "loss": 1.0063, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.230769230769231, | |
| "grad_norm": 2.391350030899048, | |
| "learning_rate": 1.3025641025641027e-05, | |
| "loss": 0.9918, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.3076923076923075, | |
| "grad_norm": 2.751174211502075, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.9849, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 2.77424693107605, | |
| "learning_rate": 1.2820512820512823e-05, | |
| "loss": 0.9745, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.461538461538462, | |
| "grad_norm": 1.9156702756881714, | |
| "learning_rate": 1.2717948717948718e-05, | |
| "loss": 0.9684, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 5.538461538461538, | |
| "grad_norm": 1.9521454572677612, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.9503, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.615384615384615, | |
| "grad_norm": 2.468419313430786, | |
| "learning_rate": 1.2512820512820514e-05, | |
| "loss": 0.9641, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.6923076923076925, | |
| "grad_norm": 2.520923614501953, | |
| "learning_rate": 1.2410256410256412e-05, | |
| "loss": 0.9471, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.769230769230769, | |
| "grad_norm": 2.1003758907318115, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.9513, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.846153846153846, | |
| "grad_norm": 2.192279100418091, | |
| "learning_rate": 1.2205128205128208e-05, | |
| "loss": 0.9527, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.923076923076923, | |
| "grad_norm": 3.8428618907928467, | |
| "learning_rate": 1.2102564102564102e-05, | |
| "loss": 0.938, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.9151530265808105, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.9753, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7819548872180451, | |
| "eval_loss": 0.9288201332092285, | |
| "eval_runtime": 0.7499, | |
| "eval_samples_per_second": 177.349, | |
| "eval_steps_per_second": 22.669, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.076923076923077, | |
| "grad_norm": 2.7967398166656494, | |
| "learning_rate": 1.1897435897435898e-05, | |
| "loss": 0.9428, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 2.5342345237731934, | |
| "learning_rate": 1.1794871794871796e-05, | |
| "loss": 0.9406, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.230769230769231, | |
| "grad_norm": 1.877543330192566, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.9319, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 6.3076923076923075, | |
| "grad_norm": 2.4524621963500977, | |
| "learning_rate": 1.1589743589743592e-05, | |
| "loss": 0.9332, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 6.384615384615385, | |
| "grad_norm": 2.4967362880706787, | |
| "learning_rate": 1.1487179487179487e-05, | |
| "loss": 0.9367, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 6.461538461538462, | |
| "grad_norm": 3.2078776359558105, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.9339, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 6.538461538461538, | |
| "grad_norm": 2.926706075668335, | |
| "learning_rate": 1.1282051282051283e-05, | |
| "loss": 0.9416, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 6.615384615384615, | |
| "grad_norm": 1.8625017404556274, | |
| "learning_rate": 1.117948717948718e-05, | |
| "loss": 0.9111, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.6923076923076925, | |
| "grad_norm": 2.7141189575195312, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.9574, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.769230769230769, | |
| "grad_norm": 2.307347536087036, | |
| "learning_rate": 1.0974358974358977e-05, | |
| "loss": 0.9259, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.846153846153846, | |
| "grad_norm": 2.3937132358551025, | |
| "learning_rate": 1.0871794871794871e-05, | |
| "loss": 0.9207, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 3.0794668197631836, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.9418, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 4.111669063568115, | |
| "learning_rate": 1.0666666666666667e-05, | |
| "loss": 0.9252, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7969924812030075, | |
| "eval_loss": 0.8874692916870117, | |
| "eval_runtime": 0.7823, | |
| "eval_samples_per_second": 170.013, | |
| "eval_steps_per_second": 21.731, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.076923076923077, | |
| "grad_norm": 2.7561662197113037, | |
| "learning_rate": 1.0564102564102565e-05, | |
| "loss": 0.911, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 7.153846153846154, | |
| "grad_norm": 3.2020223140716553, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.912, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 7.230769230769231, | |
| "grad_norm": 3.459304094314575, | |
| "learning_rate": 1.0358974358974361e-05, | |
| "loss": 0.8994, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 7.3076923076923075, | |
| "grad_norm": 2.774078369140625, | |
| "learning_rate": 1.0256410256410256e-05, | |
| "loss": 0.9079, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 7.384615384615385, | |
| "grad_norm": 2.7169668674468994, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.9256, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 7.461538461538462, | |
| "grad_norm": 2.171323299407959, | |
| "learning_rate": 1.0051282051282052e-05, | |
| "loss": 0.8898, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 7.538461538461538, | |
| "grad_norm": 2.7350351810455322, | |
| "learning_rate": 9.94871794871795e-06, | |
| "loss": 0.9243, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 7.615384615384615, | |
| "grad_norm": 2.3926539421081543, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.8868, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 2.0602715015411377, | |
| "learning_rate": 9.743589743589744e-06, | |
| "loss": 0.8837, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.769230769230769, | |
| "grad_norm": 2.885303497314453, | |
| "learning_rate": 9.641025641025642e-06, | |
| "loss": 0.8827, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.846153846153846, | |
| "grad_norm": 2.261361837387085, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.9047, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.923076923076923, | |
| "grad_norm": 2.6180179119110107, | |
| "learning_rate": 9.435897435897436e-06, | |
| "loss": 0.861, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.225304126739502, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.9192, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8120300751879699, | |
| "eval_loss": 0.850643515586853, | |
| "eval_runtime": 0.756, | |
| "eval_samples_per_second": 175.92, | |
| "eval_steps_per_second": 22.486, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.076923076923077, | |
| "grad_norm": 2.1875813007354736, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.8953, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 8.153846153846153, | |
| "grad_norm": 2.1640567779541016, | |
| "learning_rate": 9.128205128205129e-06, | |
| "loss": 0.8658, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 8.23076923076923, | |
| "grad_norm": 2.660614490509033, | |
| "learning_rate": 9.025641025641027e-06, | |
| "loss": 0.8995, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 8.307692307692308, | |
| "grad_norm": 2.104029417037964, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.8569, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 8.384615384615385, | |
| "grad_norm": 2.2643303871154785, | |
| "learning_rate": 8.820512820512821e-06, | |
| "loss": 0.8972, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 2.632410764694214, | |
| "learning_rate": 8.717948717948719e-06, | |
| "loss": 0.8715, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.538461538461538, | |
| "grad_norm": 1.6500084400177002, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.8716, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 8.615384615384615, | |
| "grad_norm": 6.204855442047119, | |
| "learning_rate": 8.512820512820513e-06, | |
| "loss": 0.8985, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 8.692307692307692, | |
| "grad_norm": 3.729611873626709, | |
| "learning_rate": 8.410256410256411e-06, | |
| "loss": 0.8837, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 8.76923076923077, | |
| "grad_norm": 3.685739278793335, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.8865, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 8.846153846153847, | |
| "grad_norm": 2.7028560638427734, | |
| "learning_rate": 8.205128205128205e-06, | |
| "loss": 0.875, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.923076923076923, | |
| "grad_norm": 2.7692482471466064, | |
| "learning_rate": 8.102564102564103e-06, | |
| "loss": 0.8867, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.9854462146759033, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9008, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8045112781954887, | |
| "eval_loss": 0.8337866067886353, | |
| "eval_runtime": 0.7963, | |
| "eval_samples_per_second": 167.03, | |
| "eval_steps_per_second": 21.35, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.076923076923077, | |
| "grad_norm": 1.9381572008132935, | |
| "learning_rate": 7.897435897435898e-06, | |
| "loss": 0.8969, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 9.153846153846153, | |
| "grad_norm": 2.219219446182251, | |
| "learning_rate": 7.794871794871796e-06, | |
| "loss": 0.8412, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 2.1302294731140137, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.8483, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.307692307692308, | |
| "grad_norm": 2.541210174560547, | |
| "learning_rate": 7.58974358974359e-06, | |
| "loss": 0.8536, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 9.384615384615385, | |
| "grad_norm": 1.952871322631836, | |
| "learning_rate": 7.487179487179488e-06, | |
| "loss": 0.8707, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 9.461538461538462, | |
| "grad_norm": 3.273028612136841, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.8547, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 9.538461538461538, | |
| "grad_norm": 2.6495628356933594, | |
| "learning_rate": 7.282051282051282e-06, | |
| "loss": 0.8709, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 9.615384615384615, | |
| "grad_norm": 1.998024582862854, | |
| "learning_rate": 7.17948717948718e-06, | |
| "loss": 0.8278, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 9.692307692307692, | |
| "grad_norm": 2.7621707916259766, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.8544, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 9.76923076923077, | |
| "grad_norm": 1.844375491142273, | |
| "learning_rate": 6.974358974358974e-06, | |
| "loss": 0.8324, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 9.846153846153847, | |
| "grad_norm": 2.149479866027832, | |
| "learning_rate": 6.871794871794872e-06, | |
| "loss": 0.8146, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 9.923076923076923, | |
| "grad_norm": 2.2224795818328857, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.8367, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.8497843742370605, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.8079, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_loss": 0.8103837370872498, | |
| "eval_runtime": 0.7593, | |
| "eval_samples_per_second": 175.164, | |
| "eval_steps_per_second": 22.389, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.076923076923077, | |
| "grad_norm": 2.0343823432922363, | |
| "learning_rate": 6.564102564102565e-06, | |
| "loss": 0.8408, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 10.153846153846153, | |
| "grad_norm": 2.4245193004608154, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.899, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 10.23076923076923, | |
| "grad_norm": 2.3912925720214844, | |
| "learning_rate": 6.358974358974359e-06, | |
| "loss": 0.8758, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 10.307692307692308, | |
| "grad_norm": 2.1387076377868652, | |
| "learning_rate": 6.256410256410257e-06, | |
| "loss": 0.8295, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 10.384615384615385, | |
| "grad_norm": 2.142160415649414, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.8075, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 10.461538461538462, | |
| "grad_norm": 2.6838831901550293, | |
| "learning_rate": 6.051282051282051e-06, | |
| "loss": 0.8448, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 10.538461538461538, | |
| "grad_norm": 2.476369857788086, | |
| "learning_rate": 5.948717948717949e-06, | |
| "loss": 0.817, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 10.615384615384615, | |
| "grad_norm": 3.031463861465454, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.8177, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 10.692307692307692, | |
| "grad_norm": 2.2818636894226074, | |
| "learning_rate": 5.743589743589743e-06, | |
| "loss": 0.8124, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 10.76923076923077, | |
| "grad_norm": 3.245805263519287, | |
| "learning_rate": 5.641025641025641e-06, | |
| "loss": 0.8674, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.846153846153847, | |
| "grad_norm": 2.194627046585083, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.831, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 10.923076923076923, | |
| "grad_norm": 1.8149436712265015, | |
| "learning_rate": 5.435897435897436e-06, | |
| "loss": 0.8391, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.0584821701049805, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 0.8332, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_loss": 0.7806060314178467, | |
| "eval_runtime": 0.742, | |
| "eval_samples_per_second": 179.256, | |
| "eval_steps_per_second": 22.912, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 11.076923076923077, | |
| "grad_norm": 1.9833248853683472, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.8484, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 11.153846153846153, | |
| "grad_norm": 5.478232383728027, | |
| "learning_rate": 5.128205128205128e-06, | |
| "loss": 0.8308, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 11.23076923076923, | |
| "grad_norm": 2.5792922973632812, | |
| "learning_rate": 5.025641025641026e-06, | |
| "loss": 0.802, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 11.307692307692308, | |
| "grad_norm": 2.730989694595337, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.8225, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 11.384615384615385, | |
| "grad_norm": 2.7447853088378906, | |
| "learning_rate": 4.820512820512821e-06, | |
| "loss": 0.8176, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 11.461538461538462, | |
| "grad_norm": 2.6465837955474854, | |
| "learning_rate": 4.717948717948718e-06, | |
| "loss": 0.8471, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "grad_norm": 2.4876015186309814, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.8349, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.615384615384615, | |
| "grad_norm": 3.2605788707733154, | |
| "learning_rate": 4.512820512820513e-06, | |
| "loss": 0.8285, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 11.692307692307692, | |
| "grad_norm": 3.278341293334961, | |
| "learning_rate": 4.4102564102564104e-06, | |
| "loss": 0.8546, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 11.76923076923077, | |
| "grad_norm": 2.0945637226104736, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.8096, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 11.846153846153847, | |
| "grad_norm": 2.161726474761963, | |
| "learning_rate": 4.2051282051282055e-06, | |
| "loss": 0.7938, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 11.923076923076923, | |
| "grad_norm": 2.1052703857421875, | |
| "learning_rate": 4.102564102564103e-06, | |
| "loss": 0.8295, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 3.460094451904297, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.8103, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_loss": 0.7585543990135193, | |
| "eval_runtime": 0.7508, | |
| "eval_samples_per_second": 177.133, | |
| "eval_steps_per_second": 22.641, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 12.076923076923077, | |
| "grad_norm": 2.943866014480591, | |
| "learning_rate": 3.897435897435898e-06, | |
| "loss": 0.7903, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 12.153846153846153, | |
| "grad_norm": 2.6185402870178223, | |
| "learning_rate": 3.794871794871795e-06, | |
| "loss": 0.8229, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 12.23076923076923, | |
| "grad_norm": 1.6378310918807983, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.8246, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 12.307692307692308, | |
| "grad_norm": 2.3109569549560547, | |
| "learning_rate": 3.58974358974359e-06, | |
| "loss": 0.8363, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.384615384615385, | |
| "grad_norm": 2.3602941036224365, | |
| "learning_rate": 3.487179487179487e-06, | |
| "loss": 0.8078, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 12.461538461538462, | |
| "grad_norm": 3.0623390674591064, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.794, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 12.538461538461538, | |
| "grad_norm": 2.947983741760254, | |
| "learning_rate": 3.2820512820512823e-06, | |
| "loss": 0.8033, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 12.615384615384615, | |
| "grad_norm": 1.8083330392837524, | |
| "learning_rate": 3.1794871794871795e-06, | |
| "loss": 0.8158, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 12.692307692307692, | |
| "grad_norm": 3.2873637676239014, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.7651, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 12.76923076923077, | |
| "grad_norm": 2.3777670860290527, | |
| "learning_rate": 2.9743589743589746e-06, | |
| "loss": 0.8566, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 12.846153846153847, | |
| "grad_norm": 1.8692084550857544, | |
| "learning_rate": 2.8717948717948717e-06, | |
| "loss": 0.8218, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 12.923076923076923, | |
| "grad_norm": 2.2379138469696045, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.7984, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 4.131476879119873, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 0.8149, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_loss": 0.757113516330719, | |
| "eval_runtime": 0.7762, | |
| "eval_samples_per_second": 171.337, | |
| "eval_steps_per_second": 21.9, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 13.076923076923077, | |
| "grad_norm": 2.9936656951904297, | |
| "learning_rate": 2.564102564102564e-06, | |
| "loss": 0.7917, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 13.153846153846153, | |
| "grad_norm": 2.5392699241638184, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.8241, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 13.23076923076923, | |
| "grad_norm": 3.0166265964508057, | |
| "learning_rate": 2.358974358974359e-06, | |
| "loss": 0.8117, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 13.307692307692308, | |
| "grad_norm": 1.8728867769241333, | |
| "learning_rate": 2.2564102564102566e-06, | |
| "loss": 0.8155, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 13.384615384615385, | |
| "grad_norm": 2.50715708732605, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.7814, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 13.461538461538462, | |
| "grad_norm": 5.447348594665527, | |
| "learning_rate": 2.0512820512820513e-06, | |
| "loss": 0.8253, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 13.538461538461538, | |
| "grad_norm": 2.6522035598754883, | |
| "learning_rate": 1.948717948717949e-06, | |
| "loss": 0.8486, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 13.615384615384615, | |
| "grad_norm": 2.1300199031829834, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.8027, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 13.692307692307692, | |
| "grad_norm": 2.1135923862457275, | |
| "learning_rate": 1.7435897435897436e-06, | |
| "loss": 0.7852, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 13.76923076923077, | |
| "grad_norm": 1.871300220489502, | |
| "learning_rate": 1.6410256410256412e-06, | |
| "loss": 0.8224, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 13.846153846153847, | |
| "grad_norm": 3.240356206893921, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.7895, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 13.923076923076923, | |
| "grad_norm": 2.5182340145111084, | |
| "learning_rate": 1.4358974358974359e-06, | |
| "loss": 0.7316, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 4.281803607940674, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 0.8186, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_loss": 0.7540305852890015, | |
| "eval_runtime": 0.7703, | |
| "eval_samples_per_second": 172.654, | |
| "eval_steps_per_second": 22.069, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 14.076923076923077, | |
| "grad_norm": 2.050518751144409, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.8222, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 14.153846153846153, | |
| "grad_norm": 2.051259994506836, | |
| "learning_rate": 1.1282051282051283e-06, | |
| "loss": 0.7878, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 14.23076923076923, | |
| "grad_norm": 2.8861193656921387, | |
| "learning_rate": 1.0256410256410257e-06, | |
| "loss": 0.78, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 14.307692307692308, | |
| "grad_norm": 4.159270763397217, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.774, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 14.384615384615385, | |
| "grad_norm": 2.8624985218048096, | |
| "learning_rate": 8.205128205128206e-07, | |
| "loss": 0.7882, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 14.461538461538462, | |
| "grad_norm": 2.5051703453063965, | |
| "learning_rate": 7.179487179487179e-07, | |
| "loss": 0.7883, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 14.538461538461538, | |
| "grad_norm": 3.003545045852661, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.7817, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 14.615384615384615, | |
| "grad_norm": 2.8403878211975098, | |
| "learning_rate": 5.128205128205128e-07, | |
| "loss": 0.8294, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 14.692307692307692, | |
| "grad_norm": 2.124030590057373, | |
| "learning_rate": 4.102564102564103e-07, | |
| "loss": 0.7978, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 14.76923076923077, | |
| "grad_norm": 4.762181758880615, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.8038, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 14.846153846153847, | |
| "grad_norm": 3.256133794784546, | |
| "learning_rate": 2.0512820512820514e-07, | |
| "loss": 0.8535, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 14.923076923076923, | |
| "grad_norm": 2.355344772338867, | |
| "learning_rate": 1.0256410256410257e-07, | |
| "loss": 0.7587, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 4.202574729919434, | |
| "learning_rate": 0.0, | |
| "loss": 0.7929, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8120300751879699, | |
| "eval_loss": 0.7412300109863281, | |
| "eval_runtime": 0.8087, | |
| "eval_samples_per_second": 164.47, | |
| "eval_steps_per_second": 21.022, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 15.076923076923077, | |
| "grad_norm": 3.3559834957122803, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.7535, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 15.153846153846153, | |
| "grad_norm": 1.8613739013671875, | |
| "learning_rate": 4.8461538461538465e-06, | |
| "loss": 0.7581, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 15.23076923076923, | |
| "grad_norm": 2.3707966804504395, | |
| "learning_rate": 4.76923076923077e-06, | |
| "loss": 0.7836, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 15.307692307692308, | |
| "grad_norm": 2.6265199184417725, | |
| "learning_rate": 4.692307692307693e-06, | |
| "loss": 0.8334, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 2.078848123550415, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.7772, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.461538461538462, | |
| "grad_norm": 2.6433162689208984, | |
| "learning_rate": 4.538461538461539e-06, | |
| "loss": 0.7955, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 15.538461538461538, | |
| "grad_norm": 3.458962917327881, | |
| "learning_rate": 4.461538461538462e-06, | |
| "loss": 0.787, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 15.615384615384615, | |
| "grad_norm": 5.090147495269775, | |
| "learning_rate": 4.384615384615385e-06, | |
| "loss": 0.7875, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 15.692307692307692, | |
| "grad_norm": 1.9066407680511475, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.7764, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 15.76923076923077, | |
| "grad_norm": 3.097341299057007, | |
| "learning_rate": 4.230769230769231e-06, | |
| "loss": 0.7335, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 15.846153846153847, | |
| "grad_norm": 2.7201600074768066, | |
| "learning_rate": 4.1538461538461545e-06, | |
| "loss": 0.7747, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 15.923076923076923, | |
| "grad_norm": 2.303032398223877, | |
| "learning_rate": 4.076923076923077e-06, | |
| "loss": 0.7738, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 4.420492172241211, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.774, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_loss": 0.7370420694351196, | |
| "eval_runtime": 0.776, | |
| "eval_samples_per_second": 171.384, | |
| "eval_steps_per_second": 21.906, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 16.076923076923077, | |
| "grad_norm": 3.9969003200531006, | |
| "learning_rate": 3.923076923076923e-06, | |
| "loss": 0.8316, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 16.153846153846153, | |
| "grad_norm": 2.3731822967529297, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.8162, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 16.23076923076923, | |
| "grad_norm": 2.232074737548828, | |
| "learning_rate": 3.7692307692307694e-06, | |
| "loss": 0.8138, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 16.307692307692307, | |
| "grad_norm": 2.8799118995666504, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.8434, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 16.384615384615383, | |
| "grad_norm": 2.2093818187713623, | |
| "learning_rate": 3.6153846153846156e-06, | |
| "loss": 0.7886, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 16.46153846153846, | |
| "grad_norm": 1.984840750694275, | |
| "learning_rate": 3.538461538461539e-06, | |
| "loss": 0.7682, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 16.53846153846154, | |
| "grad_norm": 2.711601495742798, | |
| "learning_rate": 3.4615384615384617e-06, | |
| "loss": 0.7471, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 16.615384615384617, | |
| "grad_norm": 2.130311965942383, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.7535, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 16.692307692307693, | |
| "grad_norm": 2.327207565307617, | |
| "learning_rate": 3.307692307692308e-06, | |
| "loss": 0.718, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 16.76923076923077, | |
| "grad_norm": 2.198944091796875, | |
| "learning_rate": 3.2307692307692313e-06, | |
| "loss": 0.8146, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 16.846153846153847, | |
| "grad_norm": 2.388453483581543, | |
| "learning_rate": 3.153846153846154e-06, | |
| "loss": 0.8368, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 16.923076923076923, | |
| "grad_norm": 2.2575690746307373, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.749, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 6.020498275756836, | |
| "learning_rate": 3e-06, | |
| "loss": 0.7613, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_loss": 0.7059224247932434, | |
| "eval_runtime": 0.7496, | |
| "eval_samples_per_second": 177.439, | |
| "eval_steps_per_second": 22.68, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 17.076923076923077, | |
| "grad_norm": 3.134481430053711, | |
| "learning_rate": 2.9230769230769236e-06, | |
| "loss": 0.7609, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 17.153846153846153, | |
| "grad_norm": 2.0070559978485107, | |
| "learning_rate": 2.846153846153846e-06, | |
| "loss": 0.7483, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 17.23076923076923, | |
| "grad_norm": 3.491682291030884, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.7696, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 17.307692307692307, | |
| "grad_norm": 1.9866397380828857, | |
| "learning_rate": 2.6923076923076923e-06, | |
| "loss": 0.7609, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 17.384615384615383, | |
| "grad_norm": 3.458582878112793, | |
| "learning_rate": 2.615384615384616e-06, | |
| "loss": 0.7813, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 17.46153846153846, | |
| "grad_norm": 2.1126835346221924, | |
| "learning_rate": 2.5384615384615385e-06, | |
| "loss": 0.7003, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 17.53846153846154, | |
| "grad_norm": 3.5276880264282227, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.8305, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 17.615384615384617, | |
| "grad_norm": 2.3967173099517822, | |
| "learning_rate": 2.384615384615385e-06, | |
| "loss": 0.7627, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 17.692307692307693, | |
| "grad_norm": 4.473978042602539, | |
| "learning_rate": 2.307692307692308e-06, | |
| "loss": 0.7332, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 17.76923076923077, | |
| "grad_norm": 2.1642568111419678, | |
| "learning_rate": 2.230769230769231e-06, | |
| "loss": 0.7678, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 17.846153846153847, | |
| "grad_norm": 3.03192138671875, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.7565, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 17.923076923076923, | |
| "grad_norm": 2.9610419273376465, | |
| "learning_rate": 2.0769230769230773e-06, | |
| "loss": 0.7651, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.160178184509277, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.7778, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_loss": 0.6930322647094727, | |
| "eval_runtime": 0.7854, | |
| "eval_samples_per_second": 169.332, | |
| "eval_steps_per_second": 21.644, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 18.076923076923077, | |
| "grad_norm": 2.168921947479248, | |
| "learning_rate": 1.9230769230769234e-06, | |
| "loss": 0.7234, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 18.153846153846153, | |
| "grad_norm": 3.935608386993408, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.8192, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 18.23076923076923, | |
| "grad_norm": 1.8215328454971313, | |
| "learning_rate": 1.7692307692307695e-06, | |
| "loss": 0.7271, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 18.307692307692307, | |
| "grad_norm": 2.687016010284424, | |
| "learning_rate": 1.6923076923076926e-06, | |
| "loss": 0.8063, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 18.384615384615383, | |
| "grad_norm": 2.3364577293395996, | |
| "learning_rate": 1.6153846153846157e-06, | |
| "loss": 0.7699, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 18.46153846153846, | |
| "grad_norm": 2.7465319633483887, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.8214, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.53846153846154, | |
| "grad_norm": 3.3499436378479004, | |
| "learning_rate": 1.4615384615384618e-06, | |
| "loss": 0.7432, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 18.615384615384617, | |
| "grad_norm": 3.7266149520874023, | |
| "learning_rate": 1.3846153846153848e-06, | |
| "loss": 0.797, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 18.692307692307693, | |
| "grad_norm": 2.661741256713867, | |
| "learning_rate": 1.307692307692308e-06, | |
| "loss": 0.7404, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 18.76923076923077, | |
| "grad_norm": 3.166747808456421, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.8197, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 18.846153846153847, | |
| "grad_norm": 3.200448989868164, | |
| "learning_rate": 1.153846153846154e-06, | |
| "loss": 0.8068, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 18.923076923076923, | |
| "grad_norm": 2.4404191970825195, | |
| "learning_rate": 1.076923076923077e-06, | |
| "loss": 0.788, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.8639049530029297, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.8081, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_loss": 0.6890266537666321, | |
| "eval_runtime": 0.7797, | |
| "eval_samples_per_second": 170.576, | |
| "eval_steps_per_second": 21.803, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 19.076923076923077, | |
| "grad_norm": 1.7245137691497803, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.7929, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 19.153846153846153, | |
| "grad_norm": 3.7959182262420654, | |
| "learning_rate": 8.461538461538463e-07, | |
| "loss": 0.7397, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 19.23076923076923, | |
| "grad_norm": 2.798788070678711, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 0.7928, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 19.307692307692307, | |
| "grad_norm": 2.1275336742401123, | |
| "learning_rate": 6.923076923076924e-07, | |
| "loss": 0.7672, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 19.384615384615383, | |
| "grad_norm": 2.9216866493225098, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.7918, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 19.46153846153846, | |
| "grad_norm": 2.3012797832489014, | |
| "learning_rate": 5.384615384615386e-07, | |
| "loss": 0.7418, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 19.53846153846154, | |
| "grad_norm": 2.5353312492370605, | |
| "learning_rate": 4.615384615384616e-07, | |
| "loss": 0.8115, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 19.615384615384617, | |
| "grad_norm": 3.469372510910034, | |
| "learning_rate": 3.846153846153847e-07, | |
| "loss": 0.7698, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 19.692307692307693, | |
| "grad_norm": 2.3621013164520264, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.6997, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 19.76923076923077, | |
| "grad_norm": 1.7231149673461914, | |
| "learning_rate": 2.307692307692308e-07, | |
| "loss": 0.7207, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 19.846153846153847, | |
| "grad_norm": 5.3792924880981445, | |
| "learning_rate": 1.5384615384615387e-07, | |
| "loss": 0.7656, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 19.923076923076923, | |
| "grad_norm": 1.9618691205978394, | |
| "learning_rate": 7.692307692307694e-08, | |
| "loss": 0.6919, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 4.051193714141846, | |
| "learning_rate": 0.0, | |
| "loss": 0.7916, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_loss": 0.6874601244926453, | |
| "eval_runtime": 0.8096, | |
| "eval_samples_per_second": 164.271, | |
| "eval_steps_per_second": 20.997, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2600, | |
| "total_flos": 2.0877820672794624e+17, | |
| "train_loss": 0.19350949709232038, | |
| "train_runtime": 49.806, | |
| "train_samples_per_second": 415.211, | |
| "train_steps_per_second": 52.203 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "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": 2.0877820672794624e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |