| { | |
| "best_global_step": 1400, | |
| "best_metric": 0.5561918616294861, | |
| "best_model_checkpoint": ".../training_output/checkpoint-1400", | |
| "epoch": 10.0, | |
| "eval_steps": 50, | |
| "global_step": 1420, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07042253521126761, | |
| "grad_norm": 4.834911346435547, | |
| "learning_rate": 6.338028169014085e-07, | |
| "loss": 3.8016, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14084507042253522, | |
| "grad_norm": 3.7942707538604736, | |
| "learning_rate": 1.3380281690140844e-06, | |
| "loss": 3.7424, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2112676056338028, | |
| "grad_norm": 4.044854164123535, | |
| "learning_rate": 2.0422535211267608e-06, | |
| "loss": 3.6583, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28169014084507044, | |
| "grad_norm": 4.419040679931641, | |
| "learning_rate": 2.746478873239437e-06, | |
| "loss": 3.7144, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.352112676056338, | |
| "grad_norm": 4.994226455688477, | |
| "learning_rate": 3.4507042253521127e-06, | |
| "loss": 3.5497, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.352112676056338, | |
| "eval_loss": 1.7768433094024658, | |
| "eval_runtime": 10.0717, | |
| "eval_samples_per_second": 95.217, | |
| "eval_steps_per_second": 5.957, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4225352112676056, | |
| "grad_norm": 4.968951225280762, | |
| "learning_rate": 4.154929577464789e-06, | |
| "loss": 3.6716, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.49295774647887325, | |
| "grad_norm": 4.8160014152526855, | |
| "learning_rate": 4.859154929577465e-06, | |
| "loss": 3.5206, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5633802816901409, | |
| "grad_norm": 5.351637363433838, | |
| "learning_rate": 5.563380281690142e-06, | |
| "loss": 3.4553, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6338028169014085, | |
| "grad_norm": 5.762838363647461, | |
| "learning_rate": 6.267605633802818e-06, | |
| "loss": 3.4572, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.704225352112676, | |
| "grad_norm": 5.674283504486084, | |
| "learning_rate": 6.9718309859154935e-06, | |
| "loss": 3.4715, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.704225352112676, | |
| "eval_loss": 1.704325556755066, | |
| "eval_runtime": 9.3452, | |
| "eval_samples_per_second": 102.619, | |
| "eval_steps_per_second": 6.42, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7746478873239436, | |
| "grad_norm": 5.651138782501221, | |
| "learning_rate": 7.67605633802817e-06, | |
| "loss": 3.4693, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.8450704225352113, | |
| "grad_norm": 5.866046905517578, | |
| "learning_rate": 8.380281690140846e-06, | |
| "loss": 3.482, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9154929577464789, | |
| "grad_norm": 6.652071475982666, | |
| "learning_rate": 9.084507042253522e-06, | |
| "loss": 3.4296, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9859154929577465, | |
| "grad_norm": 8.077320098876953, | |
| "learning_rate": 9.788732394366198e-06, | |
| "loss": 3.3627, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.056338028169014, | |
| "grad_norm": 6.8231329917907715, | |
| "learning_rate": 9.945226917057903e-06, | |
| "loss": 3.3338, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.056338028169014, | |
| "eval_loss": 1.6177195310592651, | |
| "eval_runtime": 9.306, | |
| "eval_samples_per_second": 103.052, | |
| "eval_steps_per_second": 6.447, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1267605633802817, | |
| "grad_norm": 7.225569725036621, | |
| "learning_rate": 9.866979655712052e-06, | |
| "loss": 3.2974, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.1971830985915493, | |
| "grad_norm": 7.667593479156494, | |
| "learning_rate": 9.788732394366198e-06, | |
| "loss": 3.3756, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.267605633802817, | |
| "grad_norm": 7.092339992523193, | |
| "learning_rate": 9.710485133020345e-06, | |
| "loss": 3.3302, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.3380281690140845, | |
| "grad_norm": 7.544447898864746, | |
| "learning_rate": 9.632237871674493e-06, | |
| "loss": 3.2841, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.408450704225352, | |
| "grad_norm": 8.606858253479004, | |
| "learning_rate": 9.55399061032864e-06, | |
| "loss": 3.14, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.408450704225352, | |
| "eval_loss": 1.5555429458618164, | |
| "eval_runtime": 8.9418, | |
| "eval_samples_per_second": 107.249, | |
| "eval_steps_per_second": 6.71, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4788732394366197, | |
| "grad_norm": 9.349486351013184, | |
| "learning_rate": 9.475743348982786e-06, | |
| "loss": 3.2447, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.5492957746478875, | |
| "grad_norm": 10.120262145996094, | |
| "learning_rate": 9.397496087636934e-06, | |
| "loss": 3.301, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.619718309859155, | |
| "grad_norm": 8.683691024780273, | |
| "learning_rate": 9.31924882629108e-06, | |
| "loss": 3.2738, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.6901408450704225, | |
| "grad_norm": 8.391075134277344, | |
| "learning_rate": 9.241001564945227e-06, | |
| "loss": 3.1474, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.76056338028169, | |
| "grad_norm": 9.799640655517578, | |
| "learning_rate": 9.162754303599375e-06, | |
| "loss": 3.1417, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.76056338028169, | |
| "eval_loss": 1.5090895891189575, | |
| "eval_runtime": 9.059, | |
| "eval_samples_per_second": 105.862, | |
| "eval_steps_per_second": 6.623, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.8309859154929577, | |
| "grad_norm": 9.724505424499512, | |
| "learning_rate": 9.084507042253522e-06, | |
| "loss": 3.1931, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.9014084507042255, | |
| "grad_norm": 11.591618537902832, | |
| "learning_rate": 9.006259780907669e-06, | |
| "loss": 3.1649, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.971830985915493, | |
| "grad_norm": 9.90600299835205, | |
| "learning_rate": 8.928012519561817e-06, | |
| "loss": 3.2231, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0422535211267605, | |
| "grad_norm": 10.291154861450195, | |
| "learning_rate": 8.849765258215963e-06, | |
| "loss": 3.1172, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.112676056338028, | |
| "grad_norm": 11.620966911315918, | |
| "learning_rate": 8.77151799687011e-06, | |
| "loss": 3.0193, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.112676056338028, | |
| "eval_loss": 1.437499761581421, | |
| "eval_runtime": 8.9265, | |
| "eval_samples_per_second": 107.433, | |
| "eval_steps_per_second": 6.722, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.183098591549296, | |
| "grad_norm": 12.65719223022461, | |
| "learning_rate": 8.693270735524258e-06, | |
| "loss": 2.9451, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.2535211267605635, | |
| "grad_norm": 12.792698860168457, | |
| "learning_rate": 8.615023474178405e-06, | |
| "loss": 2.9669, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.323943661971831, | |
| "grad_norm": 12.682389259338379, | |
| "learning_rate": 8.536776212832551e-06, | |
| "loss": 2.9739, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.3943661971830985, | |
| "grad_norm": 12.96263313293457, | |
| "learning_rate": 8.4585289514867e-06, | |
| "loss": 2.8988, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.464788732394366, | |
| "grad_norm": 14.160724639892578, | |
| "learning_rate": 8.380281690140846e-06, | |
| "loss": 2.9009, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.464788732394366, | |
| "eval_loss": 1.3772871494293213, | |
| "eval_runtime": 9.6484, | |
| "eval_samples_per_second": 99.394, | |
| "eval_steps_per_second": 6.219, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.535211267605634, | |
| "grad_norm": 14.379948616027832, | |
| "learning_rate": 8.302034428794992e-06, | |
| "loss": 2.8597, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.6056338028169015, | |
| "grad_norm": 15.024813652038574, | |
| "learning_rate": 8.22378716744914e-06, | |
| "loss": 2.8603, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.676056338028169, | |
| "grad_norm": 15.947714805603027, | |
| "learning_rate": 8.145539906103287e-06, | |
| "loss": 2.9262, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.7464788732394365, | |
| "grad_norm": 15.812085151672363, | |
| "learning_rate": 8.067292644757434e-06, | |
| "loss": 2.9301, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.816901408450704, | |
| "grad_norm": 15.922052383422852, | |
| "learning_rate": 7.989045383411582e-06, | |
| "loss": 2.9446, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.816901408450704, | |
| "eval_loss": 1.3113040924072266, | |
| "eval_runtime": 8.9464, | |
| "eval_samples_per_second": 107.194, | |
| "eval_steps_per_second": 6.707, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.887323943661972, | |
| "grad_norm": 15.517271995544434, | |
| "learning_rate": 7.910798122065728e-06, | |
| "loss": 2.9103, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.9577464788732395, | |
| "grad_norm": 16.0577449798584, | |
| "learning_rate": 7.832550860719875e-06, | |
| "loss": 2.88, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.028169014084507, | |
| "grad_norm": 15.434414863586426, | |
| "learning_rate": 7.754303599374023e-06, | |
| "loss": 2.7008, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.0985915492957745, | |
| "grad_norm": 18.000938415527344, | |
| "learning_rate": 7.67605633802817e-06, | |
| "loss": 2.669, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.169014084507042, | |
| "grad_norm": 17.124902725219727, | |
| "learning_rate": 7.597809076682316e-06, | |
| "loss": 2.6466, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.169014084507042, | |
| "eval_loss": 1.2235872745513916, | |
| "eval_runtime": 8.9632, | |
| "eval_samples_per_second": 106.993, | |
| "eval_steps_per_second": 6.694, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.23943661971831, | |
| "grad_norm": 18.728208541870117, | |
| "learning_rate": 7.5195618153364636e-06, | |
| "loss": 2.6583, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.3098591549295775, | |
| "grad_norm": 20.174840927124023, | |
| "learning_rate": 7.441314553990611e-06, | |
| "loss": 2.5912, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.380281690140845, | |
| "grad_norm": 19.468631744384766, | |
| "learning_rate": 7.3630672926447575e-06, | |
| "loss": 2.5551, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.4507042253521125, | |
| "grad_norm": 19.333127975463867, | |
| "learning_rate": 7.284820031298905e-06, | |
| "loss": 2.5759, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.52112676056338, | |
| "grad_norm": 20.050857543945312, | |
| "learning_rate": 7.206572769953052e-06, | |
| "loss": 2.6059, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.52112676056338, | |
| "eval_loss": 1.1633683443069458, | |
| "eval_runtime": 8.9644, | |
| "eval_samples_per_second": 106.979, | |
| "eval_steps_per_second": 6.693, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.591549295774648, | |
| "grad_norm": 21.03053855895996, | |
| "learning_rate": 7.128325508607199e-06, | |
| "loss": 2.5128, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.6619718309859155, | |
| "grad_norm": 20.328510284423828, | |
| "learning_rate": 7.050078247261346e-06, | |
| "loss": 2.5397, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.732394366197183, | |
| "grad_norm": 21.760059356689453, | |
| "learning_rate": 6.9718309859154935e-06, | |
| "loss": 2.5198, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.802816901408451, | |
| "grad_norm": 20.767467498779297, | |
| "learning_rate": 6.89358372456964e-06, | |
| "loss": 2.4605, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.873239436619718, | |
| "grad_norm": 24.69249725341797, | |
| "learning_rate": 6.815336463223787e-06, | |
| "loss": 2.4911, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.873239436619718, | |
| "eval_loss": 1.0876926183700562, | |
| "eval_runtime": 8.9503, | |
| "eval_samples_per_second": 107.147, | |
| "eval_steps_per_second": 6.704, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.943661971830986, | |
| "grad_norm": 21.451522827148438, | |
| "learning_rate": 6.737089201877935e-06, | |
| "loss": 2.445, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.014084507042254, | |
| "grad_norm": 21.80568504333496, | |
| "learning_rate": 6.658841940532081e-06, | |
| "loss": 2.4261, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.084507042253521, | |
| "grad_norm": 23.770599365234375, | |
| "learning_rate": 6.580594679186229e-06, | |
| "loss": 2.2965, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.154929577464789, | |
| "grad_norm": 22.17987060546875, | |
| "learning_rate": 6.502347417840375e-06, | |
| "loss": 2.1889, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.225352112676056, | |
| "grad_norm": 22.51678466796875, | |
| "learning_rate": 6.424100156494523e-06, | |
| "loss": 2.2193, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.225352112676056, | |
| "eval_loss": 1.049774408340454, | |
| "eval_runtime": 8.9455, | |
| "eval_samples_per_second": 107.205, | |
| "eval_steps_per_second": 6.707, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.295774647887324, | |
| "grad_norm": 23.58041000366211, | |
| "learning_rate": 6.345852895148671e-06, | |
| "loss": 2.2024, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.366197183098592, | |
| "grad_norm": 24.69291877746582, | |
| "learning_rate": 6.267605633802818e-06, | |
| "loss": 2.2096, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.436619718309859, | |
| "grad_norm": 24.199907302856445, | |
| "learning_rate": 6.189358372456965e-06, | |
| "loss": 2.2598, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.507042253521127, | |
| "grad_norm": 23.92287826538086, | |
| "learning_rate": 6.111111111111112e-06, | |
| "loss": 2.2322, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.577464788732394, | |
| "grad_norm": 22.647499084472656, | |
| "learning_rate": 6.0328638497652595e-06, | |
| "loss": 2.1822, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.577464788732394, | |
| "eval_loss": 0.9610461592674255, | |
| "eval_runtime": 8.9698, | |
| "eval_samples_per_second": 106.915, | |
| "eval_steps_per_second": 6.689, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.647887323943662, | |
| "grad_norm": 23.590543746948242, | |
| "learning_rate": 5.954616588419406e-06, | |
| "loss": 2.1621, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.71830985915493, | |
| "grad_norm": 23.479148864746094, | |
| "learning_rate": 5.876369327073553e-06, | |
| "loss": 2.1543, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.788732394366197, | |
| "grad_norm": 26.514415740966797, | |
| "learning_rate": 5.798122065727701e-06, | |
| "loss": 2.2376, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.859154929577465, | |
| "grad_norm": 25.98239517211914, | |
| "learning_rate": 5.719874804381847e-06, | |
| "loss": 2.1652, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.929577464788732, | |
| "grad_norm": 25.147550582885742, | |
| "learning_rate": 5.641627543035995e-06, | |
| "loss": 2.1194, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.929577464788732, | |
| "eval_loss": 0.8909062743186951, | |
| "eval_runtime": 8.9807, | |
| "eval_samples_per_second": 106.785, | |
| "eval_steps_per_second": 6.681, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 24.74693489074707, | |
| "learning_rate": 5.563380281690142e-06, | |
| "loss": 2.0919, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 5.070422535211268, | |
| "grad_norm": 27.079837799072266, | |
| "learning_rate": 5.4851330203442885e-06, | |
| "loss": 1.9779, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.140845070422535, | |
| "grad_norm": 25.947982788085938, | |
| "learning_rate": 5.406885758998436e-06, | |
| "loss": 1.8619, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.211267605633803, | |
| "grad_norm": 27.102828979492188, | |
| "learning_rate": 5.328638497652583e-06, | |
| "loss": 1.9172, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.28169014084507, | |
| "grad_norm": 27.407745361328125, | |
| "learning_rate": 5.25039123630673e-06, | |
| "loss": 1.9355, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.28169014084507, | |
| "eval_loss": 0.8482040166854858, | |
| "eval_runtime": 9.8318, | |
| "eval_samples_per_second": 97.54, | |
| "eval_steps_per_second": 6.103, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.352112676056338, | |
| "grad_norm": 24.58213233947754, | |
| "learning_rate": 5.172143974960877e-06, | |
| "loss": 1.9393, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.422535211267606, | |
| "grad_norm": 24.623889923095703, | |
| "learning_rate": 5.093896713615024e-06, | |
| "loss": 1.8727, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.492957746478873, | |
| "grad_norm": 26.396087646484375, | |
| "learning_rate": 5.015649452269171e-06, | |
| "loss": 1.9599, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.563380281690141, | |
| "grad_norm": 28.786396026611328, | |
| "learning_rate": 4.9374021909233185e-06, | |
| "loss": 1.8271, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.633802816901408, | |
| "grad_norm": 25.507366180419922, | |
| "learning_rate": 4.859154929577465e-06, | |
| "loss": 1.9065, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.633802816901408, | |
| "eval_loss": 0.8195229768753052, | |
| "eval_runtime": 8.968, | |
| "eval_samples_per_second": 106.935, | |
| "eval_steps_per_second": 6.69, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.704225352112676, | |
| "grad_norm": 26.951005935668945, | |
| "learning_rate": 4.780907668231612e-06, | |
| "loss": 1.8742, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.774647887323944, | |
| "grad_norm": 27.19331169128418, | |
| "learning_rate": 4.70266040688576e-06, | |
| "loss": 1.8754, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.845070422535211, | |
| "grad_norm": 26.17990493774414, | |
| "learning_rate": 4.624413145539906e-06, | |
| "loss": 1.8412, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.915492957746479, | |
| "grad_norm": 28.198486328125, | |
| "learning_rate": 4.546165884194054e-06, | |
| "loss": 1.845, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.985915492957746, | |
| "grad_norm": 24.81572914123535, | |
| "learning_rate": 4.467918622848201e-06, | |
| "loss": 1.8035, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.985915492957746, | |
| "eval_loss": 0.7713097929954529, | |
| "eval_runtime": 8.914, | |
| "eval_samples_per_second": 107.583, | |
| "eval_steps_per_second": 6.731, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 6.056338028169014, | |
| "grad_norm": 27.168418884277344, | |
| "learning_rate": 4.3896713615023476e-06, | |
| "loss": 1.7062, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.126760563380282, | |
| "grad_norm": 25.584774017333984, | |
| "learning_rate": 4.311424100156495e-06, | |
| "loss": 1.6908, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.197183098591549, | |
| "grad_norm": 27.289413452148438, | |
| "learning_rate": 4.233176838810642e-06, | |
| "loss": 1.6868, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.267605633802817, | |
| "grad_norm": 26.789995193481445, | |
| "learning_rate": 4.154929577464789e-06, | |
| "loss": 1.686, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.338028169014084, | |
| "grad_norm": 27.878765106201172, | |
| "learning_rate": 4.076682316118936e-06, | |
| "loss": 1.564, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.338028169014084, | |
| "eval_loss": 0.7355391383171082, | |
| "eval_runtime": 8.9357, | |
| "eval_samples_per_second": 107.322, | |
| "eval_steps_per_second": 6.715, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.408450704225352, | |
| "grad_norm": 27.236083984375, | |
| "learning_rate": 3.998435054773084e-06, | |
| "loss": 1.7135, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 6.47887323943662, | |
| "grad_norm": 26.007850646972656, | |
| "learning_rate": 3.92018779342723e-06, | |
| "loss": 1.6614, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 6.549295774647887, | |
| "grad_norm": 26.08616065979004, | |
| "learning_rate": 3.8419405320813775e-06, | |
| "loss": 1.6442, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.619718309859155, | |
| "grad_norm": 26.021728515625, | |
| "learning_rate": 3.7636932707355245e-06, | |
| "loss": 1.6117, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.690140845070422, | |
| "grad_norm": 26.019577026367188, | |
| "learning_rate": 3.6854460093896714e-06, | |
| "loss": 1.6317, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.690140845070422, | |
| "eval_loss": 0.700791597366333, | |
| "eval_runtime": 8.9228, | |
| "eval_samples_per_second": 107.477, | |
| "eval_steps_per_second": 6.724, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.76056338028169, | |
| "grad_norm": 25.988859176635742, | |
| "learning_rate": 3.6071987480438188e-06, | |
| "loss": 1.6589, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.830985915492958, | |
| "grad_norm": 26.973453521728516, | |
| "learning_rate": 3.5289514866979657e-06, | |
| "loss": 1.673, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.901408450704225, | |
| "grad_norm": 27.36475372314453, | |
| "learning_rate": 3.4507042253521127e-06, | |
| "loss": 1.6018, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.971830985915493, | |
| "grad_norm": 28.897504806518555, | |
| "learning_rate": 3.37245696400626e-06, | |
| "loss": 1.6291, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 7.042253521126761, | |
| "grad_norm": 26.3133544921875, | |
| "learning_rate": 3.294209702660407e-06, | |
| "loss": 1.5407, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.042253521126761, | |
| "eval_loss": 0.6679695248603821, | |
| "eval_runtime": 8.9256, | |
| "eval_samples_per_second": 107.444, | |
| "eval_steps_per_second": 6.722, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.112676056338028, | |
| "grad_norm": 29.09581184387207, | |
| "learning_rate": 3.215962441314554e-06, | |
| "loss": 1.4842, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.183098591549296, | |
| "grad_norm": 27.576128005981445, | |
| "learning_rate": 3.137715179968701e-06, | |
| "loss": 1.5063, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.253521126760563, | |
| "grad_norm": 29.655752182006836, | |
| "learning_rate": 3.0594679186228483e-06, | |
| "loss": 1.5052, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 7.323943661971831, | |
| "grad_norm": 25.104883193969727, | |
| "learning_rate": 2.9812206572769952e-06, | |
| "loss": 1.42, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 7.394366197183099, | |
| "grad_norm": 27.621112823486328, | |
| "learning_rate": 2.902973395931143e-06, | |
| "loss": 1.4671, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.394366197183099, | |
| "eval_loss": 0.6450517177581787, | |
| "eval_runtime": 8.9546, | |
| "eval_samples_per_second": 107.096, | |
| "eval_steps_per_second": 6.7, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.464788732394366, | |
| "grad_norm": 27.425331115722656, | |
| "learning_rate": 2.82472613458529e-06, | |
| "loss": 1.4879, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 7.535211267605634, | |
| "grad_norm": 27.76177978515625, | |
| "learning_rate": 2.746478873239437e-06, | |
| "loss": 1.5012, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 7.605633802816901, | |
| "grad_norm": 24.934165954589844, | |
| "learning_rate": 2.6682316118935843e-06, | |
| "loss": 1.4386, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 7.676056338028169, | |
| "grad_norm": 28.298328399658203, | |
| "learning_rate": 2.5899843505477313e-06, | |
| "loss": 1.4544, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.746478873239437, | |
| "grad_norm": 24.99042320251465, | |
| "learning_rate": 2.5117370892018782e-06, | |
| "loss": 1.4076, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.746478873239437, | |
| "eval_loss": 0.6194617748260498, | |
| "eval_runtime": 8.9294, | |
| "eval_samples_per_second": 107.398, | |
| "eval_steps_per_second": 6.719, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.816901408450704, | |
| "grad_norm": 27.29831314086914, | |
| "learning_rate": 2.433489827856025e-06, | |
| "loss": 1.4838, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.887323943661972, | |
| "grad_norm": 27.53099250793457, | |
| "learning_rate": 2.3552425665101726e-06, | |
| "loss": 1.4745, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.957746478873239, | |
| "grad_norm": 27.09596824645996, | |
| "learning_rate": 2.2769953051643195e-06, | |
| "loss": 1.3781, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 8.028169014084508, | |
| "grad_norm": 26.090456008911133, | |
| "learning_rate": 2.1987480438184665e-06, | |
| "loss": 1.422, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 8.098591549295774, | |
| "grad_norm": 26.628990173339844, | |
| "learning_rate": 2.120500782472614e-06, | |
| "loss": 1.345, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.098591549295774, | |
| "eval_loss": 0.603298008441925, | |
| "eval_runtime": 8.923, | |
| "eval_samples_per_second": 107.475, | |
| "eval_steps_per_second": 6.724, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.169014084507042, | |
| "grad_norm": 26.968416213989258, | |
| "learning_rate": 2.0422535211267608e-06, | |
| "loss": 1.4478, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 8.23943661971831, | |
| "grad_norm": 28.169918060302734, | |
| "learning_rate": 1.9640062597809077e-06, | |
| "loss": 1.3111, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 8.309859154929578, | |
| "grad_norm": 28.084226608276367, | |
| "learning_rate": 1.885758998435055e-06, | |
| "loss": 1.3754, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 8.380281690140846, | |
| "grad_norm": 28.451416015625, | |
| "learning_rate": 1.807511737089202e-06, | |
| "loss": 1.3996, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 8.450704225352112, | |
| "grad_norm": 27.99500846862793, | |
| "learning_rate": 1.729264475743349e-06, | |
| "loss": 1.3818, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.450704225352112, | |
| "eval_loss": 0.5862457752227783, | |
| "eval_runtime": 8.9231, | |
| "eval_samples_per_second": 107.473, | |
| "eval_steps_per_second": 6.724, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.52112676056338, | |
| "grad_norm": 26.25287628173828, | |
| "learning_rate": 1.6510172143974962e-06, | |
| "loss": 1.348, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 8.591549295774648, | |
| "grad_norm": 27.26688003540039, | |
| "learning_rate": 1.5727699530516433e-06, | |
| "loss": 1.3567, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 8.661971830985916, | |
| "grad_norm": 25.22686767578125, | |
| "learning_rate": 1.4945226917057903e-06, | |
| "loss": 1.331, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 8.732394366197184, | |
| "grad_norm": 27.34234619140625, | |
| "learning_rate": 1.4162754303599375e-06, | |
| "loss": 1.2977, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.80281690140845, | |
| "grad_norm": 23.48381233215332, | |
| "learning_rate": 1.3380281690140844e-06, | |
| "loss": 1.2578, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.80281690140845, | |
| "eval_loss": 0.5742074847221375, | |
| "eval_runtime": 8.9308, | |
| "eval_samples_per_second": 107.381, | |
| "eval_steps_per_second": 6.718, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.873239436619718, | |
| "grad_norm": 25.575672149658203, | |
| "learning_rate": 1.2597809076682316e-06, | |
| "loss": 1.2807, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.943661971830986, | |
| "grad_norm": 27.154817581176758, | |
| "learning_rate": 1.1815336463223787e-06, | |
| "loss": 1.3597, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 9.014084507042254, | |
| "grad_norm": 28.042072296142578, | |
| "learning_rate": 1.103286384976526e-06, | |
| "loss": 1.3292, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 9.084507042253522, | |
| "grad_norm": 27.011009216308594, | |
| "learning_rate": 1.0250391236306729e-06, | |
| "loss": 1.2585, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 9.154929577464788, | |
| "grad_norm": 27.294761657714844, | |
| "learning_rate": 9.467918622848201e-07, | |
| "loss": 1.2552, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.154929577464788, | |
| "eval_loss": 0.5647178292274475, | |
| "eval_runtime": 8.9528, | |
| "eval_samples_per_second": 107.117, | |
| "eval_steps_per_second": 6.702, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.225352112676056, | |
| "grad_norm": 25.794416427612305, | |
| "learning_rate": 8.685446009389673e-07, | |
| "loss": 1.3339, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 9.295774647887324, | |
| "grad_norm": 26.401464462280273, | |
| "learning_rate": 7.902973395931143e-07, | |
| "loss": 1.2872, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 9.366197183098592, | |
| "grad_norm": 28.1634464263916, | |
| "learning_rate": 7.120500782472614e-07, | |
| "loss": 1.276, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 9.43661971830986, | |
| "grad_norm": 26.156156539916992, | |
| "learning_rate": 6.338028169014085e-07, | |
| "loss": 1.2846, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 9.507042253521126, | |
| "grad_norm": 26.216571807861328, | |
| "learning_rate": 5.555555555555555e-07, | |
| "loss": 1.2782, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.507042253521126, | |
| "eval_loss": 0.5621166229248047, | |
| "eval_runtime": 8.9665, | |
| "eval_samples_per_second": 106.953, | |
| "eval_steps_per_second": 6.692, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.577464788732394, | |
| "grad_norm": 26.63737678527832, | |
| "learning_rate": 4.773082942097027e-07, | |
| "loss": 1.2883, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 9.647887323943662, | |
| "grad_norm": 27.532459259033203, | |
| "learning_rate": 3.990610328638498e-07, | |
| "loss": 1.2515, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 9.71830985915493, | |
| "grad_norm": 23.93533706665039, | |
| "learning_rate": 3.208137715179969e-07, | |
| "loss": 1.3023, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 9.788732394366198, | |
| "grad_norm": 26.48802375793457, | |
| "learning_rate": 2.42566510172144e-07, | |
| "loss": 1.3053, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 9.859154929577464, | |
| "grad_norm": 25.6018123626709, | |
| "learning_rate": 1.643192488262911e-07, | |
| "loss": 1.2556, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.859154929577464, | |
| "eval_loss": 0.5561918616294861, | |
| "eval_runtime": 9.3282, | |
| "eval_samples_per_second": 102.806, | |
| "eval_steps_per_second": 6.432, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.929577464788732, | |
| "grad_norm": 26.17299461364746, | |
| "learning_rate": 8.607198748043818e-08, | |
| "loss": 1.2926, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 28.140972137451172, | |
| "learning_rate": 7.824726134585291e-09, | |
| "loss": 1.2199, | |
| "step": 1420 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1420, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 350, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |