| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9968387776606953, | |
| "eval_steps": 100, | |
| "global_step": 1422, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.8102548576192377, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.681, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.6409873968033384, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3735, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.4241968146548056, | |
| "learning_rate": 4.9993723817567996e-05, | |
| "loss": 0.3481, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.5192149828819788, | |
| "learning_rate": 4.997489842150924e-05, | |
| "loss": 0.3733, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.3829364414625014, | |
| "learning_rate": 4.994353326395334e-05, | |
| "loss": 0.3561, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 1.6320786389820312, | |
| "learning_rate": 4.989964409317637e-05, | |
| "loss": 0.355, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 1.4596195084848897, | |
| "learning_rate": 4.984325294569372e-05, | |
| "loss": 0.3444, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 1.4431914548343205, | |
| "learning_rate": 4.977438813519574e-05, | |
| "loss": 0.3373, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 2.498892153458844, | |
| "learning_rate": 4.969308423833152e-05, | |
| "loss": 0.3602, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 1.6601234439104857, | |
| "learning_rate": 4.9599382077348205e-05, | |
| "loss": 0.3436, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.34954115748405457, | |
| "eval_runtime": 46.7928, | |
| "eval_samples_per_second": 288.314, | |
| "eval_steps_per_second": 9.018, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 1.0930800520190758, | |
| "learning_rate": 4.949332869959432e-05, | |
| "loss": 0.352, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.5471349397538547, | |
| "learning_rate": 4.9374977353897566e-05, | |
| "loss": 0.3426, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 1.2518152452620943, | |
| "learning_rate": 4.9244387463828876e-05, | |
| "loss": 0.3422, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.0687037118070535, | |
| "learning_rate": 4.910162459786617e-05, | |
| "loss": 0.3378, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.2965517147404728, | |
| "learning_rate": 4.894676043647274e-05, | |
| "loss": 0.3381, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 1.077453296704333, | |
| "learning_rate": 4.8779872736106916e-05, | |
| "loss": 0.3214, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 1.1426693406130142, | |
| "learning_rate": 4.8601045290180946e-05, | |
| "loss": 0.3372, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 1.0178221568525938, | |
| "learning_rate": 4.84103678869888e-05, | |
| "loss": 0.3378, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.228896746635447, | |
| "learning_rate": 4.820793626462391e-05, | |
| "loss": 0.3218, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 1.7376473974923927, | |
| "learning_rate": 4.799385206290965e-05, | |
| "loss": 0.3306, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 0.33515387773513794, | |
| "eval_runtime": 47.1074, | |
| "eval_samples_per_second": 286.388, | |
| "eval_steps_per_second": 8.958, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.1779458621544412, | |
| "learning_rate": 4.7768222772366466e-05, | |
| "loss": 0.3288, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 1.0719044164695737, | |
| "learning_rate": 4.753116168024153e-05, | |
| "loss": 0.3052, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.3092426623376585, | |
| "learning_rate": 4.728278781362777e-05, | |
| "loss": 0.3329, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 1.3613973232931833, | |
| "learning_rate": 4.702322587970104e-05, | |
| "loss": 0.3305, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 1.379957402090315, | |
| "learning_rate": 4.6752606203105314e-05, | |
| "loss": 0.3293, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 1.2064354059145495, | |
| "learning_rate": 4.647106466051741e-05, | |
| "loss": 0.3276, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.9407218615903679, | |
| "learning_rate": 4.617874261242399e-05, | |
| "loss": 0.3148, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 1.102895278969822, | |
| "learning_rate": 4.5875786832145287e-05, | |
| "loss": 0.3131, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.1283794552904691, | |
| "learning_rate": 4.556234943214095e-05, | |
| "loss": 0.3222, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 1.3966644012679954, | |
| "learning_rate": 4.523858778763528e-05, | |
| "loss": 0.3417, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.32445600628852844, | |
| "eval_runtime": 47.1805, | |
| "eval_samples_per_second": 285.945, | |
| "eval_steps_per_second": 8.944, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 1.0630493028269177, | |
| "learning_rate": 4.490466445759988e-05, | |
| "loss": 0.3152, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 1.0652057869610336, | |
| "learning_rate": 4.456074710313378e-05, | |
| "loss": 0.3038, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.2341586090990317, | |
| "learning_rate": 4.420700840328162e-05, | |
| "loss": 0.3082, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.9252822939526227, | |
| "learning_rate": 4.38436259683325e-05, | |
| "loss": 0.3174, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.1914450398759333, | |
| "learning_rate": 4.347078225064276e-05, | |
| "loss": 0.3238, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 1.0380458098458347, | |
| "learning_rate": 4.308866445302766e-05, | |
| "loss": 0.3036, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.1100397252938874, | |
| "learning_rate": 4.269746443476787e-05, | |
| "loss": 0.3053, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.0584951830825808, | |
| "learning_rate": 4.2297378615277935e-05, | |
| "loss": 0.3082, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 1.1192502727228961, | |
| "learning_rate": 4.188860787548522e-05, | |
| "loss": 0.3126, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 1.0566732730775144, | |
| "learning_rate": 4.1471357456968556e-05, | |
| "loss": 0.3154, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 0.31601378321647644, | |
| "eval_runtime": 47.231, | |
| "eval_samples_per_second": 285.639, | |
| "eval_steps_per_second": 8.935, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 1.1380881980287134, | |
| "learning_rate": 4.1045836858907676e-05, | |
| "loss": 0.3144, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.9239785068723881, | |
| "learning_rate": 4.061225973289473e-05, | |
| "loss": 0.3054, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 1.1226362379499772, | |
| "learning_rate": 4.0170843775661025e-05, | |
| "loss": 0.3132, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1.055418877421314, | |
| "learning_rate": 3.9721810619772636e-05, | |
| "loss": 0.3096, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 1.260485940528199, | |
| "learning_rate": 3.926538572234991e-05, | |
| "loss": 0.3112, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.8290208161083492, | |
| "learning_rate": 3.880179825186667e-05, | |
| "loss": 0.32, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.9894500964017104, | |
| "learning_rate": 3.833128097308594e-05, | |
| "loss": 0.2963, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 0.8202492183589537, | |
| "learning_rate": 3.785407013019006e-05, | |
| "loss": 0.2216, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 0.8958874625557091, | |
| "learning_rate": 3.73704053281637e-05, | |
| "loss": 0.1707, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.7913123522777149, | |
| "learning_rate": 3.688052941248956e-05, | |
| "loss": 0.1577, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 0.3239268660545349, | |
| "eval_runtime": 47.1856, | |
| "eval_samples_per_second": 285.913, | |
| "eval_steps_per_second": 8.943, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 0.6951457322398226, | |
| "learning_rate": 3.6384688347216875e-05, | |
| "loss": 0.1549, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.3614607662338507, | |
| "learning_rate": 3.58831310914643e-05, | |
| "loss": 0.1597, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.8448116369901294, | |
| "learning_rate": 3.53761094744188e-05, | |
| "loss": 0.1492, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 0.9188827386043135, | |
| "learning_rate": 3.4863878068893625e-05, | |
| "loss": 0.1576, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.8781597082892243, | |
| "learning_rate": 3.434669406350866e-05, | |
| "loss": 0.152, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.6881353088868859, | |
| "learning_rate": 3.382481713355738e-05, | |
| "loss": 0.1581, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.2684676365868262, | |
| "learning_rate": 3.3298509310625363e-05, | |
| "loss": 0.1574, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 0.6337141659864904, | |
| "learning_rate": 3.276803485102557e-05, | |
| "loss": 0.1578, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.704714314928959, | |
| "learning_rate": 3.223366010311671e-05, | |
| "loss": 0.1514, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 0.6460825296146832, | |
| "learning_rate": 3.1695653373571196e-05, | |
| "loss": 0.159, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 0.32300886511802673, | |
| "eval_runtime": 47.067, | |
| "eval_samples_per_second": 286.634, | |
| "eval_steps_per_second": 8.966, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 0.8717091665319785, | |
| "learning_rate": 3.115428479265975e-05, | |
| "loss": 0.1527, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 0.8221409439379637, | |
| "learning_rate": 3.060982617862053e-05, | |
| "loss": 0.1542, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 0.7676631827886251, | |
| "learning_rate": 3.006255090118059e-05, | |
| "loss": 0.1574, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.8176603071255903, | |
| "learning_rate": 2.9512733744298482e-05, | |
| "loss": 0.1599, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 0.8478701648849646, | |
| "learning_rate": 2.8960650768196672e-05, | |
| "loss": 0.1527, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 0.8105241966593794, | |
| "learning_rate": 2.8406579170753205e-05, | |
| "loss": 0.1555, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 0.7341788063465178, | |
| "learning_rate": 2.785079714832216e-05, | |
| "loss": 0.1503, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 1.1238252352545457, | |
| "learning_rate": 2.7293583756052755e-05, | |
| "loss": 0.1615, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 0.7793278106904632, | |
| "learning_rate": 2.673521876777727e-05, | |
| "loss": 0.1527, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.7257177280067018, | |
| "learning_rate": 2.6175982535538098e-05, | |
| "loss": 0.1493, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_loss": 0.3175669312477112, | |
| "eval_runtime": 47.0992, | |
| "eval_samples_per_second": 286.438, | |
| "eval_steps_per_second": 8.96, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.801964046142728, | |
| "learning_rate": 2.561615584882453e-05, | |
| "loss": 0.1607, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.7155830027665171, | |
| "learning_rate": 2.5056019793589858e-05, | |
| "loss": 0.1492, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 0.7264384440518477, | |
| "learning_rate": 2.449585561111965e-05, | |
| "loss": 0.1324, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.8168302855335549, | |
| "learning_rate": 2.3935944556821966e-05, | |
| "loss": 0.1555, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 0.7446449421109613, | |
| "learning_rate": 2.3376567759010614e-05, | |
| "loss": 0.1528, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.6027586148022397, | |
| "learning_rate": 2.281800607775211e-05, | |
| "loss": 0.152, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 0.8554567150163206, | |
| "learning_rate": 2.2260539963847317e-05, | |
| "loss": 0.1534, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.7383522990234253, | |
| "learning_rate": 2.1704449318018692e-05, | |
| "loss": 0.1592, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 0.7958490328944942, | |
| "learning_rate": 2.1150013350373594e-05, | |
| "loss": 0.1426, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 0.9825338160992358, | |
| "learning_rate": 2.059751044021441e-05, | |
| "loss": 0.1498, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "eval_loss": 0.31580981612205505, | |
| "eval_runtime": 49.0093, | |
| "eval_samples_per_second": 275.274, | |
| "eval_steps_per_second": 8.611, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 0.7301707840582428, | |
| "learning_rate": 2.00472179962658e-05, | |
| "loss": 0.1437, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 0.9685123461176025, | |
| "learning_rate": 1.9499412317389305e-05, | |
| "loss": 0.156, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 2.10442864010648, | |
| "learning_rate": 1.895436845385516e-05, | |
| "loss": 0.1496, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 0.9146649049632373, | |
| "learning_rate": 1.8412360069241034e-05, | |
| "loss": 0.1454, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 1.4265670270210795, | |
| "learning_rate": 1.7873659303027052e-05, | |
| "loss": 0.1492, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 0.7559717203281935, | |
| "learning_rate": 1.733853663395602e-05, | |
| "loss": 0.1535, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 1.3311913559020063, | |
| "learning_rate": 1.6807260744227513e-05, | |
| "loss": 0.1479, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 0.7863355556565459, | |
| "learning_rate": 1.6280098384593966e-05, | |
| "loss": 0.1452, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.6405600189189222, | |
| "learning_rate": 1.5757314240426613e-05, | |
| "loss": 0.1417, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.9800841015815575, | |
| "learning_rate": 1.5239170798818381e-05, | |
| "loss": 0.147, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_loss": 0.3103533089160919, | |
| "eval_runtime": 47.2326, | |
| "eval_samples_per_second": 285.629, | |
| "eval_steps_per_second": 8.935, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.8072145902479386, | |
| "learning_rate": 1.472592821679048e-05, | |
| "loss": 0.1599, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.7396163055843903, | |
| "learning_rate": 1.4217844190669058e-05, | |
| "loss": 0.1413, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.7609487598644061, | |
| "learning_rate": 1.3715173826697209e-05, | |
| "loss": 0.1575, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 0.899248106175493, | |
| "learning_rate": 1.3218169512947542e-05, | |
| "loss": 0.1368, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.5492609472889338, | |
| "learning_rate": 1.2727080792599455e-05, | |
| "loss": 0.1356, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 0.5155984048617979, | |
| "learning_rate": 1.2242154238644879e-05, | |
| "loss": 0.0618, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.6456926556116795, | |
| "learning_rate": 1.1763633330085325e-05, | |
| "loss": 0.0638, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 0.6205542344499085, | |
| "learning_rate": 1.1291758329682358e-05, | |
| "loss": 0.0611, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 0.4847483179925426, | |
| "learning_rate": 1.0826766163322982e-05, | |
| "loss": 0.0575, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 0.5413371468713811, | |
| "learning_rate": 1.0368890301060457e-05, | |
| "loss": 0.0583, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 0.34516656398773193, | |
| "eval_runtime": 47.23, | |
| "eval_samples_per_second": 285.645, | |
| "eval_steps_per_second": 8.935, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 0.5404171416731017, | |
| "learning_rate": 9.918360639890187e-06, | |
| "loss": 0.0601, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.5620072184000532, | |
| "learning_rate": 9.475403388319752e-06, | |
| "loss": 0.0562, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 0.5612527587767797, | |
| "learning_rate": 9.040240952790765e-06, | |
| "loss": 0.06, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 0.556109347253415, | |
| "learning_rate": 8.613091826009884e-06, | |
| "loss": 0.0586, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 0.526323377202855, | |
| "learning_rate": 8.194170477244729e-06, | |
| "loss": 0.0555, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 0.4839147527811038, | |
| "learning_rate": 7.783687244640048e-06, | |
| "loss": 0.0586, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 0.5384902556315007, | |
| "learning_rate": 7.38184822960811e-06, | |
| "loss": 0.0563, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 0.5487199740187843, | |
| "learning_rate": 6.988855193346236e-06, | |
| "loss": 0.0591, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.558830265761421, | |
| "learning_rate": 6.604905455533625e-06, | |
| "loss": 0.0581, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.5905937756559718, | |
| "learning_rate": 6.230191795258228e-06, | |
| "loss": 0.0547, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_loss": 0.3394644260406494, | |
| "eval_runtime": 47.3785, | |
| "eval_samples_per_second": 284.75, | |
| "eval_steps_per_second": 8.907, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 0.5552812370662581, | |
| "learning_rate": 5.864902354223384e-06, | |
| "loss": 0.0647, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 0.5583176961830797, | |
| "learning_rate": 5.509220542282864e-06, | |
| "loss": 0.0579, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 0.527577115010199, | |
| "learning_rate": 5.163324945351841e-06, | |
| "loss": 0.0547, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.5723382741785656, | |
| "learning_rate": 4.82738923573983e-06, | |
| "loss": 0.0522, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 0.6612580191515155, | |
| "learning_rate": 4.501582084950715e-06, | |
| "loss": 0.0546, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 0.5398335429013659, | |
| "learning_rate": 4.186067078993794e-06, | |
| "loss": 0.0549, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 0.5344699336645098, | |
| "learning_rate": 3.8810026362480684e-06, | |
| "loss": 0.0587, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 0.5444597939849882, | |
| "learning_rate": 3.586541927921222e-06, | |
| "loss": 0.0581, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 0.5463988615566789, | |
| "learning_rate": 3.3028328011432157e-06, | |
| "loss": 0.0575, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 1.0764210232146447, | |
| "learning_rate": 3.030017704733043e-06, | |
| "loss": 0.0589, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 0.34168508648872375, | |
| "eval_runtime": 47.2752, | |
| "eval_samples_per_second": 285.372, | |
| "eval_steps_per_second": 8.926, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 0.4953517170839006, | |
| "learning_rate": 2.7682336176759195e-06, | |
| "loss": 0.0581, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 0.5541503548086418, | |
| "learning_rate": 2.51761198034689e-06, | |
| "loss": 0.0546, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 0.49468890807169186, | |
| "learning_rate": 2.278278628515354e-06, | |
| "loss": 0.0546, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 0.5409541969627151, | |
| "learning_rate": 2.050353730163554e-06, | |
| "loss": 0.0587, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 0.5627631577280182, | |
| "learning_rate": 1.8339517251509146e-06, | |
| "loss": 0.0591, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 0.6303369184474236, | |
| "learning_rate": 1.6291812677544121e-06, | |
| "loss": 0.0565, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 0.5950168840485455, | |
| "learning_rate": 1.436145172113834e-06, | |
| "loss": 0.0621, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.6169520124840763, | |
| "learning_rate": 1.2549403606093525e-06, | |
| "loss": 0.0585, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 0.6215760879408516, | |
| "learning_rate": 1.0856578151973246e-06, | |
| "loss": 0.0551, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 0.5481992179891889, | |
| "learning_rate": 9.283825317287692e-07, | |
| "loss": 0.0593, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_loss": 0.34135544300079346, | |
| "eval_runtime": 47.2086, | |
| "eval_samples_per_second": 285.774, | |
| "eval_steps_per_second": 8.939, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 0.5600873628791802, | |
| "learning_rate": 7.83193477273364e-07, | |
| "loss": 0.063, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 0.47881555590847813, | |
| "learning_rate": 6.501635504705422e-07, | |
| "loss": 0.0556, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.5702498469737556, | |
| "learning_rate": 5.293595449274685e-07, | |
| "loss": 0.0578, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 0.6344603201355465, | |
| "learning_rate": 4.208421156823239e-07, | |
| "loss": 0.0621, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.4557995262356338, | |
| "learning_rate": 3.2466574874975565e-07, | |
| "loss": 0.0615, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 0.60731847293474, | |
| "learning_rate": 2.408787337637164e-07, | |
| "loss": 0.0563, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 0.5748533793418215, | |
| "learning_rate": 1.6952313973152834e-07, | |
| "loss": 0.0557, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 0.5805376198014929, | |
| "learning_rate": 1.1063479391124898e-07, | |
| "loss": 0.0519, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "grad_norm": 0.5744481873667274, | |
| "learning_rate": 6.424326382299394e-08, | |
| "loss": 0.0569, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 0.6568824061735796, | |
| "learning_rate": 3.037184240325397e-08, | |
| "loss": 0.0624, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "eval_loss": 0.34196367859840393, | |
| "eval_runtime": 47.1898, | |
| "eval_samples_per_second": 285.888, | |
| "eval_steps_per_second": 8.943, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 0.5396314216567086, | |
| "learning_rate": 9.037536309636218e-09, | |
| "loss": 0.0549, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 0.6363488595941399, | |
| "learning_rate": 2.5105738190867033e-10, | |
| "loss": 0.0564, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 1422, | |
| "total_flos": 3.6498441825094205e+18, | |
| "train_loss": 0.18191866144936128, | |
| "train_runtime": 7053.867, | |
| "train_samples_per_second": 51.639, | |
| "train_steps_per_second": 0.202 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1422, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 3.6498441825094205e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |