| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.893337455557273, | |
| "eval_steps": 200, | |
| "global_step": 16000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0002981331673926571, | |
| "loss": 2.3038, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 1.589147925376892, | |
| "eval_runtime": 2.1684, | |
| "eval_samples_per_second": 461.163, | |
| "eval_steps_per_second": 57.645, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00029439950217797134, | |
| "loss": 1.5398, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.4633430242538452, | |
| "eval_runtime": 2.1071, | |
| "eval_samples_per_second": 474.58, | |
| "eval_steps_per_second": 59.323, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00029066583696328563, | |
| "loss": 1.4615, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 1.4296711683273315, | |
| "eval_runtime": 2.4812, | |
| "eval_samples_per_second": 403.032, | |
| "eval_steps_per_second": 50.379, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002869321717485998, | |
| "loss": 1.4244, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 1.3793567419052124, | |
| "eval_runtime": 2.9698, | |
| "eval_samples_per_second": 336.72, | |
| "eval_steps_per_second": 42.09, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0002831985065339141, | |
| "loss": 1.3921, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_loss": 1.3315461874008179, | |
| "eval_runtime": 2.7793, | |
| "eval_samples_per_second": 359.806, | |
| "eval_steps_per_second": 44.976, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00027946484131922836, | |
| "loss": 1.0958, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_loss": 0.7548955082893372, | |
| "eval_runtime": 2.7656, | |
| "eval_samples_per_second": 361.592, | |
| "eval_steps_per_second": 45.199, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002757311761045426, | |
| "loss": 0.6312, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 0.3990221917629242, | |
| "eval_runtime": 2.1502, | |
| "eval_samples_per_second": 465.071, | |
| "eval_steps_per_second": 58.134, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00027199751088985685, | |
| "loss": 0.4093, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 0.26113563776016235, | |
| "eval_runtime": 3.5103, | |
| "eval_samples_per_second": 284.875, | |
| "eval_steps_per_second": 35.609, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00026826384567517114, | |
| "loss": 0.2959, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.2783801257610321, | |
| "eval_runtime": 2.8441, | |
| "eval_samples_per_second": 351.603, | |
| "eval_steps_per_second": 43.95, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00026453018046048533, | |
| "loss": 0.2589, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 0.20705343782901764, | |
| "eval_runtime": 2.7524, | |
| "eval_samples_per_second": 363.322, | |
| "eval_steps_per_second": 45.415, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.0002607965152457996, | |
| "loss": 0.2246, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 0.15551678836345673, | |
| "eval_runtime": 2.15, | |
| "eval_samples_per_second": 465.117, | |
| "eval_steps_per_second": 58.14, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.00025706285003111387, | |
| "loss": 0.1991, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_loss": 0.15825262665748596, | |
| "eval_runtime": 2.1344, | |
| "eval_samples_per_second": 468.515, | |
| "eval_steps_per_second": 58.564, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.0002533291848164281, | |
| "loss": 0.1784, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.12008943408727646, | |
| "eval_runtime": 2.1414, | |
| "eval_samples_per_second": 466.985, | |
| "eval_steps_per_second": 58.373, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.00024959551960174235, | |
| "loss": 0.1598, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 0.12511701881885529, | |
| "eval_runtime": 2.55, | |
| "eval_samples_per_second": 392.155, | |
| "eval_steps_per_second": 49.019, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.0002458618543870566, | |
| "loss": 0.164, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_loss": 0.11049681156873703, | |
| "eval_runtime": 2.9765, | |
| "eval_samples_per_second": 335.97, | |
| "eval_steps_per_second": 41.996, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.00024212818917237084, | |
| "loss": 0.1475, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 0.0954003781080246, | |
| "eval_runtime": 2.8437, | |
| "eval_samples_per_second": 351.659, | |
| "eval_steps_per_second": 43.957, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0002383945239576851, | |
| "loss": 0.1388, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_loss": 0.10116879642009735, | |
| "eval_runtime": 2.7628, | |
| "eval_samples_per_second": 361.954, | |
| "eval_steps_per_second": 45.244, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.00023466085874299935, | |
| "loss": 0.1346, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_loss": 0.10693109035491943, | |
| "eval_runtime": 3.1697, | |
| "eval_samples_per_second": 315.491, | |
| "eval_steps_per_second": 39.436, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.00023092719352831362, | |
| "loss": 0.1232, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_loss": 0.09901304543018341, | |
| "eval_runtime": 2.1178, | |
| "eval_samples_per_second": 472.183, | |
| "eval_steps_per_second": 59.023, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.00022719352831362786, | |
| "loss": 0.1187, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_loss": 0.11418598890304565, | |
| "eval_runtime": 2.1348, | |
| "eval_samples_per_second": 468.423, | |
| "eval_steps_per_second": 58.553, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.0002234598630989421, | |
| "loss": 0.1133, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_loss": 0.0984039306640625, | |
| "eval_runtime": 2.1382, | |
| "eval_samples_per_second": 467.676, | |
| "eval_steps_per_second": 58.459, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 0.00021972619788425635, | |
| "loss": 0.1088, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "eval_loss": 0.07466612011194229, | |
| "eval_runtime": 2.8862, | |
| "eval_samples_per_second": 346.477, | |
| "eval_steps_per_second": 43.31, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.00021599253266957062, | |
| "loss": 0.1025, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 0.1227998435497284, | |
| "eval_runtime": 2.8738, | |
| "eval_samples_per_second": 347.966, | |
| "eval_steps_per_second": 43.496, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 0.00021225886745488486, | |
| "loss": 0.0971, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_loss": 0.07324225455522537, | |
| "eval_runtime": 2.2831, | |
| "eval_samples_per_second": 437.994, | |
| "eval_steps_per_second": 54.749, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 0.00020852520224019913, | |
| "loss": 0.0853, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "eval_loss": 0.07788190990686417, | |
| "eval_runtime": 2.1358, | |
| "eval_samples_per_second": 468.199, | |
| "eval_steps_per_second": 58.525, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 0.00020479153702551337, | |
| "loss": 0.0865, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_loss": 0.06575259566307068, | |
| "eval_runtime": 2.1474, | |
| "eval_samples_per_second": 465.679, | |
| "eval_steps_per_second": 58.21, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.0002010578718108276, | |
| "loss": 0.0768, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "eval_loss": 0.08183684200048447, | |
| "eval_runtime": 2.1211, | |
| "eval_samples_per_second": 471.453, | |
| "eval_steps_per_second": 58.932, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 0.00019732420659614186, | |
| "loss": 0.0738, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "eval_loss": 0.04662672430276871, | |
| "eval_runtime": 2.7913, | |
| "eval_samples_per_second": 358.253, | |
| "eval_steps_per_second": 44.782, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 0.0001935905413814561, | |
| "loss": 0.0622, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "eval_loss": 0.0433196946978569, | |
| "eval_runtime": 3.1597, | |
| "eval_samples_per_second": 316.49, | |
| "eval_steps_per_second": 39.561, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.00018985687616677037, | |
| "loss": 0.0671, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "eval_loss": 0.038382936269044876, | |
| "eval_runtime": 2.1009, | |
| "eval_samples_per_second": 475.976, | |
| "eval_steps_per_second": 59.497, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 0.0001861232109520846, | |
| "loss": 0.0545, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "eval_loss": 0.04082392156124115, | |
| "eval_runtime": 2.1346, | |
| "eval_samples_per_second": 468.481, | |
| "eval_steps_per_second": 58.56, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 0.00018238954573739888, | |
| "loss": 0.0564, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "eval_loss": 0.043197453022003174, | |
| "eval_runtime": 2.1169, | |
| "eval_samples_per_second": 472.389, | |
| "eval_steps_per_second": 59.049, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 0.0001786558805227131, | |
| "loss": 0.0523, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "eval_loss": 0.03342806547880173, | |
| "eval_runtime": 2.4926, | |
| "eval_samples_per_second": 401.182, | |
| "eval_steps_per_second": 50.148, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.00017492221530802736, | |
| "loss": 0.0456, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "eval_loss": 0.02744474820792675, | |
| "eval_runtime": 3.0, | |
| "eval_samples_per_second": 333.335, | |
| "eval_steps_per_second": 41.667, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 0.0001711885500933416, | |
| "loss": 0.0442, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "eval_loss": 0.024560416117310524, | |
| "eval_runtime": 2.6752, | |
| "eval_samples_per_second": 373.806, | |
| "eval_steps_per_second": 46.726, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 0.00016745488487865588, | |
| "loss": 0.0383, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "eval_loss": 0.018605533987283707, | |
| "eval_runtime": 2.1117, | |
| "eval_samples_per_second": 473.559, | |
| "eval_steps_per_second": 59.195, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.00016372121966397012, | |
| "loss": 0.0348, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "eval_loss": 0.01473915483802557, | |
| "eval_runtime": 2.1223, | |
| "eval_samples_per_second": 471.193, | |
| "eval_steps_per_second": 58.899, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 0.0001599875544492844, | |
| "loss": 0.0299, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "eval_loss": 0.025838036090135574, | |
| "eval_runtime": 2.1138, | |
| "eval_samples_per_second": 473.088, | |
| "eval_steps_per_second": 59.136, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 0.0001562538892345986, | |
| "loss": 0.0268, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "eval_loss": 0.01688736118376255, | |
| "eval_runtime": 2.1658, | |
| "eval_samples_per_second": 461.718, | |
| "eval_steps_per_second": 57.715, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.00015252022401991287, | |
| "loss": 0.0272, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "eval_loss": 0.020514091476798058, | |
| "eval_runtime": 2.1415, | |
| "eval_samples_per_second": 466.966, | |
| "eval_steps_per_second": 58.371, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 0.00014878655880522712, | |
| "loss": 0.0277, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "eval_loss": 0.018993763253092766, | |
| "eval_runtime": 2.3074, | |
| "eval_samples_per_second": 433.383, | |
| "eval_steps_per_second": 54.173, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 0.00014505289359054139, | |
| "loss": 0.0253, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "eval_loss": 0.0132982786744833, | |
| "eval_runtime": 2.7723, | |
| "eval_samples_per_second": 360.706, | |
| "eval_steps_per_second": 45.088, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.00014131922837585563, | |
| "loss": 0.0208, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "eval_loss": 0.011603164486587048, | |
| "eval_runtime": 2.2147, | |
| "eval_samples_per_second": 451.518, | |
| "eval_steps_per_second": 56.44, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 0.00013758556316116987, | |
| "loss": 0.019, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "eval_loss": 0.007933158427476883, | |
| "eval_runtime": 2.565, | |
| "eval_samples_per_second": 389.858, | |
| "eval_steps_per_second": 48.732, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 0.00013385189794648414, | |
| "loss": 0.0179, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "eval_loss": 0.00808796752244234, | |
| "eval_runtime": 2.157, | |
| "eval_samples_per_second": 463.605, | |
| "eval_steps_per_second": 57.951, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 0.00013011823273179835, | |
| "loss": 0.0136, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "eval_loss": 0.02137412503361702, | |
| "eval_runtime": 2.1642, | |
| "eval_samples_per_second": 462.06, | |
| "eval_steps_per_second": 57.758, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.00012638456751711262, | |
| "loss": 0.0196, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "eval_loss": 0.009271830320358276, | |
| "eval_runtime": 2.7483, | |
| "eval_samples_per_second": 363.865, | |
| "eval_steps_per_second": 45.483, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 0.00012265090230242687, | |
| "loss": 0.015, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_loss": 0.011388062499463558, | |
| "eval_runtime": 3.1063, | |
| "eval_samples_per_second": 321.931, | |
| "eval_steps_per_second": 40.241, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 0.00011891723708774112, | |
| "loss": 0.0196, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "eval_loss": 0.009324445389211178, | |
| "eval_runtime": 2.9695, | |
| "eval_samples_per_second": 336.759, | |
| "eval_steps_per_second": 42.095, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 0.00011518357187305538, | |
| "loss": 0.0192, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "eval_loss": 0.008494062349200249, | |
| "eval_runtime": 2.1785, | |
| "eval_samples_per_second": 459.035, | |
| "eval_steps_per_second": 57.379, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 0.00011144990665836963, | |
| "loss": 0.0155, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "eval_loss": 0.005131287965923548, | |
| "eval_runtime": 2.2151, | |
| "eval_samples_per_second": 451.441, | |
| "eval_steps_per_second": 56.43, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 0.00010771624144368388, | |
| "loss": 0.0182, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "eval_loss": 0.01033452432602644, | |
| "eval_runtime": 2.204, | |
| "eval_samples_per_second": 453.729, | |
| "eval_steps_per_second": 56.716, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 0.00010398257622899813, | |
| "loss": 0.0149, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "eval_loss": 0.006081216037273407, | |
| "eval_runtime": 2.6138, | |
| "eval_samples_per_second": 382.582, | |
| "eval_steps_per_second": 47.823, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 0.00010024891101431236, | |
| "loss": 0.0155, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "eval_loss": 0.008235114626586437, | |
| "eval_runtime": 2.9799, | |
| "eval_samples_per_second": 335.587, | |
| "eval_steps_per_second": 41.948, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 9.651524579962662e-05, | |
| "loss": 0.0125, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_loss": 0.0061024767346680164, | |
| "eval_runtime": 3.1763, | |
| "eval_samples_per_second": 314.832, | |
| "eval_steps_per_second": 39.354, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 9.278158058494087e-05, | |
| "loss": 0.0126, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "eval_loss": 0.0077368393540382385, | |
| "eval_runtime": 2.1677, | |
| "eval_samples_per_second": 461.31, | |
| "eval_steps_per_second": 57.664, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 8.904791537025512e-05, | |
| "loss": 0.016, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "eval_loss": 0.01462015975266695, | |
| "eval_runtime": 2.163, | |
| "eval_samples_per_second": 462.313, | |
| "eval_steps_per_second": 57.789, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 8.531425015556937e-05, | |
| "loss": 0.0168, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "eval_loss": 0.013114248402416706, | |
| "eval_runtime": 2.177, | |
| "eval_samples_per_second": 459.355, | |
| "eval_steps_per_second": 57.419, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "learning_rate": 8.158058494088363e-05, | |
| "loss": 0.0115, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "eval_loss": 0.0058467877097427845, | |
| "eval_runtime": 2.8432, | |
| "eval_samples_per_second": 351.72, | |
| "eval_steps_per_second": 43.965, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 7.784691972619787e-05, | |
| "loss": 0.0109, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "eval_loss": 0.007328983396291733, | |
| "eval_runtime": 2.9781, | |
| "eval_samples_per_second": 335.785, | |
| "eval_steps_per_second": 41.973, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 7.411325451151213e-05, | |
| "loss": 0.01, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "eval_loss": 0.00543447770178318, | |
| "eval_runtime": 2.13, | |
| "eval_samples_per_second": 469.492, | |
| "eval_steps_per_second": 58.686, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 7.037958929682637e-05, | |
| "loss": 0.0085, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "eval_loss": 0.005294375587254763, | |
| "eval_runtime": 2.1484, | |
| "eval_samples_per_second": 465.459, | |
| "eval_steps_per_second": 58.182, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 6.664592408214062e-05, | |
| "loss": 0.0105, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "eval_loss": 0.0051603252068161964, | |
| "eval_runtime": 2.1621, | |
| "eval_samples_per_second": 462.523, | |
| "eval_steps_per_second": 57.815, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 6.291225886745488e-05, | |
| "loss": 0.01, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "eval_loss": 0.005722519941627979, | |
| "eval_runtime": 2.7684, | |
| "eval_samples_per_second": 361.216, | |
| "eval_steps_per_second": 45.152, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 5.917859365276913e-05, | |
| "loss": 0.0071, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "eval_loss": 0.004564732778817415, | |
| "eval_runtime": 2.7551, | |
| "eval_samples_per_second": 362.961, | |
| "eval_steps_per_second": 45.37, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 5.5444928438083385e-05, | |
| "loss": 0.0065, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "eval_loss": 0.004461783915758133, | |
| "eval_runtime": 3.1705, | |
| "eval_samples_per_second": 315.412, | |
| "eval_steps_per_second": 39.426, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 5.171126322339763e-05, | |
| "loss": 0.0075, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "eval_loss": 0.004132562782615423, | |
| "eval_runtime": 3.5027, | |
| "eval_samples_per_second": 285.498, | |
| "eval_steps_per_second": 35.687, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 4.797759800871188e-05, | |
| "loss": 0.0072, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "eval_loss": 0.004298557061702013, | |
| "eval_runtime": 2.1516, | |
| "eval_samples_per_second": 464.775, | |
| "eval_steps_per_second": 58.097, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "learning_rate": 4.424393279402613e-05, | |
| "loss": 0.0077, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "eval_loss": 0.005747557617723942, | |
| "eval_runtime": 2.1174, | |
| "eval_samples_per_second": 472.272, | |
| "eval_steps_per_second": 59.034, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 4.051026757934038e-05, | |
| "loss": 0.009, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "eval_loss": 0.005076244939118624, | |
| "eval_runtime": 2.1715, | |
| "eval_samples_per_second": 460.514, | |
| "eval_steps_per_second": 57.564, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 3.677660236465463e-05, | |
| "loss": 0.0066, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "eval_loss": 0.004328867886215448, | |
| "eval_runtime": 2.1457, | |
| "eval_samples_per_second": 466.038, | |
| "eval_steps_per_second": 58.255, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 3.304293714996888e-05, | |
| "loss": 0.0065, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_loss": 0.004579309374094009, | |
| "eval_runtime": 2.5023, | |
| "eval_samples_per_second": 399.626, | |
| "eval_steps_per_second": 49.953, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 2.9309271935283136e-05, | |
| "loss": 0.0047, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_loss": 0.00406376738101244, | |
| "eval_runtime": 3.0193, | |
| "eval_samples_per_second": 331.204, | |
| "eval_steps_per_second": 41.401, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 2.5575606720597382e-05, | |
| "loss": 0.0049, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "eval_loss": 0.0037133253645151854, | |
| "eval_runtime": 2.5419, | |
| "eval_samples_per_second": 393.406, | |
| "eval_steps_per_second": 49.176, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 2.1841941505911635e-05, | |
| "loss": 0.0048, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "eval_loss": 0.0035180081613361835, | |
| "eval_runtime": 2.1535, | |
| "eval_samples_per_second": 464.362, | |
| "eval_steps_per_second": 58.045, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 1.8108276291225884e-05, | |
| "loss": 0.0045, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "eval_loss": 0.0041992985643446445, | |
| "eval_runtime": 2.1652, | |
| "eval_samples_per_second": 461.858, | |
| "eval_steps_per_second": 57.732, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 1.4374611076540135e-05, | |
| "loss": 0.0041, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "eval_loss": 0.003915323410183191, | |
| "eval_runtime": 2.7057, | |
| "eval_samples_per_second": 369.59, | |
| "eval_steps_per_second": 46.199, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 1.0640945861854385e-05, | |
| "loss": 0.0042, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "eval_loss": 0.0032798268366605043, | |
| "eval_runtime": 3.0263, | |
| "eval_samples_per_second": 330.438, | |
| "eval_steps_per_second": 41.305, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 6.907280647168636e-06, | |
| "loss": 0.0041, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "eval_loss": 0.003197046695277095, | |
| "eval_runtime": 2.2279, | |
| "eval_samples_per_second": 448.855, | |
| "eval_steps_per_second": 56.107, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 3.173615432482887e-06, | |
| "loss": 0.0039, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "eval_loss": 0.003054018598049879, | |
| "eval_runtime": 2.2116, | |
| "eval_samples_per_second": 452.155, | |
| "eval_steps_per_second": 56.519, | |
| "step": 16000 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 16170, | |
| "num_train_epochs": 10, | |
| "save_steps": 200, | |
| "total_flos": 6146864391499776.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |