| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.23143277929728592, | |
| "eval_steps": 10, | |
| "global_step": 550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004207868714496108, | |
| "grad_norm": 10.497783660888672, | |
| "learning_rate": 0.0004909090909090909, | |
| "loss": 0.7938, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004207868714496108, | |
| "eval_accuracy": 0.5304018259048462, | |
| "eval_loss": 0.7331877946853638, | |
| "eval_runtime": 568.2697, | |
| "eval_samples_per_second": 8.364, | |
| "eval_steps_per_second": 2.092, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "grad_norm": 2.557694435119629, | |
| "learning_rate": 0.00048181818181818184, | |
| "loss": 0.7504, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008415737428992216, | |
| "eval_accuracy": 0.5634336471557617, | |
| "eval_loss": 0.7359612584114075, | |
| "eval_runtime": 559.3032, | |
| "eval_samples_per_second": 8.498, | |
| "eval_steps_per_second": 2.126, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "grad_norm": 3.9564919471740723, | |
| "learning_rate": 0.0004727272727272727, | |
| "loss": 0.7866, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012623606143488323, | |
| "eval_accuracy": 0.4085840582847595, | |
| "eval_loss": 0.7127184271812439, | |
| "eval_runtime": 567.0848, | |
| "eval_samples_per_second": 8.381, | |
| "eval_steps_per_second": 2.097, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "grad_norm": 29.89333724975586, | |
| "learning_rate": 0.00046363636363636366, | |
| "loss": 0.678, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.016831474857984433, | |
| "eval_accuracy": 0.6875657439231873, | |
| "eval_loss": 0.6557931303977966, | |
| "eval_runtime": 563.9541, | |
| "eval_samples_per_second": 8.428, | |
| "eval_steps_per_second": 2.108, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "grad_norm": 1.9482946395874023, | |
| "learning_rate": 0.00045454545454545455, | |
| "loss": 0.8634, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.021039343572480537, | |
| "eval_accuracy": 0.6572691202163696, | |
| "eval_loss": 0.613047182559967, | |
| "eval_runtime": 565.9198, | |
| "eval_samples_per_second": 8.399, | |
| "eval_steps_per_second": 2.101, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "grad_norm": 1.0520250797271729, | |
| "learning_rate": 0.00044545454545454543, | |
| "loss": 0.5117, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.025247212286976645, | |
| "eval_accuracy": 0.7416368722915649, | |
| "eval_loss": 0.5154983997344971, | |
| "eval_runtime": 566.7191, | |
| "eval_samples_per_second": 8.387, | |
| "eval_steps_per_second": 2.098, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "grad_norm": 3.958573341369629, | |
| "learning_rate": 0.00043636363636363637, | |
| "loss": 0.6066, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.029455081001472753, | |
| "eval_accuracy": 0.7858194708824158, | |
| "eval_loss": 0.5255141854286194, | |
| "eval_runtime": 561.6163, | |
| "eval_samples_per_second": 8.463, | |
| "eval_steps_per_second": 2.117, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "grad_norm": 7.255526542663574, | |
| "learning_rate": 0.00042727272727272726, | |
| "loss": 0.5282, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.033662949715968865, | |
| "eval_accuracy": 0.8045445084571838, | |
| "eval_loss": 0.6335883140563965, | |
| "eval_runtime": 565.1826, | |
| "eval_samples_per_second": 8.41, | |
| "eval_steps_per_second": 2.104, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "grad_norm": 4.439815521240234, | |
| "learning_rate": 0.00041818181818181814, | |
| "loss": 0.3694, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03787081843046497, | |
| "eval_accuracy": 0.8161161541938782, | |
| "eval_loss": 0.464495450258255, | |
| "eval_runtime": 562.0768, | |
| "eval_samples_per_second": 8.456, | |
| "eval_steps_per_second": 2.115, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "grad_norm": 1.75038743019104, | |
| "learning_rate": 0.00040909090909090913, | |
| "loss": 0.4786, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.042078687144961074, | |
| "eval_accuracy": 0.8079107999801636, | |
| "eval_loss": 0.4171961545944214, | |
| "eval_runtime": 555.7896, | |
| "eval_samples_per_second": 8.552, | |
| "eval_steps_per_second": 2.139, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "grad_norm": 2.398599624633789, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3394, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.046286555859457186, | |
| "eval_accuracy": 0.8739743232727051, | |
| "eval_loss": 0.353726327419281, | |
| "eval_runtime": 558.2722, | |
| "eval_samples_per_second": 8.514, | |
| "eval_steps_per_second": 2.13, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "grad_norm": 2.4667985439300537, | |
| "learning_rate": 0.00039090909090909096, | |
| "loss": 0.4313, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.05049442457395329, | |
| "eval_accuracy": 0.8889122605323792, | |
| "eval_loss": 0.2820029556751251, | |
| "eval_runtime": 563.3341, | |
| "eval_samples_per_second": 8.437, | |
| "eval_steps_per_second": 2.111, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "grad_norm": 1.1293903589248657, | |
| "learning_rate": 0.00038181818181818184, | |
| "loss": 0.3352, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.0547022932884494, | |
| "eval_accuracy": 0.8889122605323792, | |
| "eval_loss": 0.28811973333358765, | |
| "eval_runtime": 563.2561, | |
| "eval_samples_per_second": 8.438, | |
| "eval_steps_per_second": 2.111, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "grad_norm": 17.52765655517578, | |
| "learning_rate": 0.00037272727272727273, | |
| "loss": 0.3772, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05891016200294551, | |
| "eval_accuracy": 0.8116979002952576, | |
| "eval_loss": 0.4701159596443176, | |
| "eval_runtime": 565.5511, | |
| "eval_samples_per_second": 8.404, | |
| "eval_steps_per_second": 2.102, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "grad_norm": 2.1096954345703125, | |
| "learning_rate": 0.00036363636363636367, | |
| "loss": 0.3699, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06311803071744161, | |
| "eval_accuracy": 0.8693456649780273, | |
| "eval_loss": 0.32162848114967346, | |
| "eval_runtime": 561.7642, | |
| "eval_samples_per_second": 8.461, | |
| "eval_steps_per_second": 2.117, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "grad_norm": 0.7017167210578918, | |
| "learning_rate": 0.00035454545454545455, | |
| "loss": 0.4895, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.06732589943193773, | |
| "eval_accuracy": 0.8817588686943054, | |
| "eval_loss": 0.36469992995262146, | |
| "eval_runtime": 563.6361, | |
| "eval_samples_per_second": 8.433, | |
| "eval_steps_per_second": 2.11, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "grad_norm": 0.48568668961524963, | |
| "learning_rate": 0.00034545454545454544, | |
| "loss": 0.46, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07153376814643383, | |
| "eval_accuracy": 0.8449400663375854, | |
| "eval_loss": 0.35531896352767944, | |
| "eval_runtime": 566.3467, | |
| "eval_samples_per_second": 8.392, | |
| "eval_steps_per_second": 2.099, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "grad_norm": 0.6846579909324646, | |
| "learning_rate": 0.0003363636363636364, | |
| "loss": 0.3931, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07574163686092994, | |
| "eval_accuracy": 0.8916473984718323, | |
| "eval_loss": 0.2812640070915222, | |
| "eval_runtime": 565.8446, | |
| "eval_samples_per_second": 8.4, | |
| "eval_steps_per_second": 2.101, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "grad_norm": 2.391395092010498, | |
| "learning_rate": 0.00032727272727272726, | |
| "loss": 0.3666, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.07994950557542604, | |
| "eval_accuracy": 0.8321060538291931, | |
| "eval_loss": 0.4278256595134735, | |
| "eval_runtime": 562.6059, | |
| "eval_samples_per_second": 8.448, | |
| "eval_steps_per_second": 2.113, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "grad_norm": 1.708675503730774, | |
| "learning_rate": 0.0003181818181818182, | |
| "loss": 0.3471, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08415737428992215, | |
| "eval_accuracy": 0.9156322479248047, | |
| "eval_loss": 0.23789770901203156, | |
| "eval_runtime": 565.4642, | |
| "eval_samples_per_second": 8.405, | |
| "eval_steps_per_second": 2.103, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "grad_norm": 0.4107959270477295, | |
| "learning_rate": 0.0003090909090909091, | |
| "loss": 0.4351, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.08836524300441827, | |
| "eval_accuracy": 0.8889122605323792, | |
| "eval_loss": 0.29446446895599365, | |
| "eval_runtime": 562.2329, | |
| "eval_samples_per_second": 8.454, | |
| "eval_steps_per_second": 2.115, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "grad_norm": 10.296160697937012, | |
| "learning_rate": 0.0003, | |
| "loss": 0.262, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09257311171891437, | |
| "eval_accuracy": 0.9017462730407715, | |
| "eval_loss": 0.307280033826828, | |
| "eval_runtime": 561.3167, | |
| "eval_samples_per_second": 8.468, | |
| "eval_steps_per_second": 2.118, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "grad_norm": 0.6356618404388428, | |
| "learning_rate": 0.0002909090909090909, | |
| "loss": 0.3291, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.09678098043341048, | |
| "eval_accuracy": 0.8752366900444031, | |
| "eval_loss": 0.3042367994785309, | |
| "eval_runtime": 567.4141, | |
| "eval_samples_per_second": 8.377, | |
| "eval_steps_per_second": 2.095, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "grad_norm": 1.1508342027664185, | |
| "learning_rate": 0.0002818181818181818, | |
| "loss": 0.239, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10098884914790658, | |
| "eval_accuracy": 0.9150010347366333, | |
| "eval_loss": 0.20924389362335205, | |
| "eval_runtime": 564.144, | |
| "eval_samples_per_second": 8.425, | |
| "eval_steps_per_second": 2.108, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "grad_norm": 1.0988367795944214, | |
| "learning_rate": 0.00027272727272727274, | |
| "loss": 0.4306, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1051967178624027, | |
| "eval_accuracy": 0.905954122543335, | |
| "eval_loss": 0.24196282029151917, | |
| "eval_runtime": 565.5038, | |
| "eval_samples_per_second": 8.405, | |
| "eval_steps_per_second": 2.103, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "grad_norm": 1.5871909856796265, | |
| "learning_rate": 0.0002636363636363636, | |
| "loss": 0.2757, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.1094045865768988, | |
| "eval_accuracy": 0.8415737152099609, | |
| "eval_loss": 0.3900565505027771, | |
| "eval_runtime": 566.4852, | |
| "eval_samples_per_second": 8.39, | |
| "eval_steps_per_second": 2.099, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "grad_norm": 0.33946001529693604, | |
| "learning_rate": 0.0002545454545454545, | |
| "loss": 0.1851, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11361245529139491, | |
| "eval_accuracy": 0.9040606021881104, | |
| "eval_loss": 0.2729853093624115, | |
| "eval_runtime": 563.3502, | |
| "eval_samples_per_second": 8.437, | |
| "eval_steps_per_second": 2.111, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "grad_norm": 0.7806110978126526, | |
| "learning_rate": 0.00024545454545454545, | |
| "loss": 0.2655, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.11782032400589101, | |
| "eval_accuracy": 0.8645066022872925, | |
| "eval_loss": 0.3478306829929352, | |
| "eval_runtime": 565.4573, | |
| "eval_samples_per_second": 8.406, | |
| "eval_steps_per_second": 2.103, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "grad_norm": 1.8319846391677856, | |
| "learning_rate": 0.00023636363636363636, | |
| "loss": 0.2119, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12202819272038712, | |
| "eval_accuracy": 0.9112139940261841, | |
| "eval_loss": 0.23500923812389374, | |
| "eval_runtime": 563.7148, | |
| "eval_samples_per_second": 8.432, | |
| "eval_steps_per_second": 2.109, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "grad_norm": 1.5416840314865112, | |
| "learning_rate": 0.00022727272727272727, | |
| "loss": 0.1477, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.12623606143488322, | |
| "eval_accuracy": 0.9219440221786499, | |
| "eval_loss": 0.3015914261341095, | |
| "eval_runtime": 561.7407, | |
| "eval_samples_per_second": 8.461, | |
| "eval_steps_per_second": 2.117, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "grad_norm": 3.576409101486206, | |
| "learning_rate": 0.00021818181818181818, | |
| "loss": 0.2137, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13044393014937933, | |
| "eval_accuracy": 0.9282558560371399, | |
| "eval_loss": 0.30129292607307434, | |
| "eval_runtime": 563.5899, | |
| "eval_samples_per_second": 8.433, | |
| "eval_steps_per_second": 2.11, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "grad_norm": 9.926487922668457, | |
| "learning_rate": 0.00020909090909090907, | |
| "loss": 0.1136, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13465179886387546, | |
| "eval_accuracy": 0.9307805299758911, | |
| "eval_loss": 0.2483457773923874, | |
| "eval_runtime": 564.5525, | |
| "eval_samples_per_second": 8.419, | |
| "eval_steps_per_second": 2.106, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "grad_norm": 1.2627246379852295, | |
| "learning_rate": 0.0002, | |
| "loss": 0.197, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.13885966757837157, | |
| "eval_accuracy": 0.913528323173523, | |
| "eval_loss": 0.3319007456302643, | |
| "eval_runtime": 557.94, | |
| "eval_samples_per_second": 8.519, | |
| "eval_steps_per_second": 2.131, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "grad_norm": 14.595890998840332, | |
| "learning_rate": 0.00019090909090909092, | |
| "loss": 0.1492, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14306753629286767, | |
| "eval_accuracy": 0.8773406147956848, | |
| "eval_loss": 0.4927380383014679, | |
| "eval_runtime": 565.4025, | |
| "eval_samples_per_second": 8.406, | |
| "eval_steps_per_second": 2.103, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "grad_norm": 0.08015768229961395, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.0394, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.14727540500736377, | |
| "eval_accuracy": 0.8889122605323792, | |
| "eval_loss": 0.5241063833236694, | |
| "eval_runtime": 567.4727, | |
| "eval_samples_per_second": 8.376, | |
| "eval_steps_per_second": 2.095, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "grad_norm": 1.9829438924789429, | |
| "learning_rate": 0.00017272727272727272, | |
| "loss": 0.284, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15148327372185988, | |
| "eval_accuracy": 0.924679160118103, | |
| "eval_loss": 0.3283034563064575, | |
| "eval_runtime": 565.6368, | |
| "eval_samples_per_second": 8.403, | |
| "eval_steps_per_second": 2.102, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "grad_norm": 0.26033639907836914, | |
| "learning_rate": 0.00016363636363636363, | |
| "loss": 0.1096, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.15569114243635598, | |
| "eval_accuracy": 0.9179465770721436, | |
| "eval_loss": 0.2719731330871582, | |
| "eval_runtime": 568.9149, | |
| "eval_samples_per_second": 8.355, | |
| "eval_steps_per_second": 2.09, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "grad_norm": 0.4041542112827301, | |
| "learning_rate": 0.00015454545454545454, | |
| "loss": 0.3147, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1598990111508521, | |
| "eval_accuracy": 0.9389858841896057, | |
| "eval_loss": 0.20388783514499664, | |
| "eval_runtime": 576.8924, | |
| "eval_samples_per_second": 8.239, | |
| "eval_steps_per_second": 2.061, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "grad_norm": 0.7307087182998657, | |
| "learning_rate": 0.00014545454545454546, | |
| "loss": 0.1, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1641068798653482, | |
| "eval_accuracy": 0.9320428967475891, | |
| "eval_loss": 0.2630753219127655, | |
| "eval_runtime": 572.9749, | |
| "eval_samples_per_second": 8.295, | |
| "eval_steps_per_second": 2.075, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "grad_norm": 0.9732351303100586, | |
| "learning_rate": 0.00013636363636363637, | |
| "loss": 0.1954, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1683147485798443, | |
| "eval_accuracy": 0.93309485912323, | |
| "eval_loss": 0.265785813331604, | |
| "eval_runtime": 571.5753, | |
| "eval_samples_per_second": 8.316, | |
| "eval_steps_per_second": 2.08, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "grad_norm": 3.890988826751709, | |
| "learning_rate": 0.00012727272727272725, | |
| "loss": 0.1332, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17252261729434043, | |
| "eval_accuracy": 0.9469808340072632, | |
| "eval_loss": 0.2041083574295044, | |
| "eval_runtime": 572.7877, | |
| "eval_samples_per_second": 8.298, | |
| "eval_steps_per_second": 2.076, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "grad_norm": 0.22084006667137146, | |
| "learning_rate": 0.00011818181818181818, | |
| "loss": 0.116, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.17673048600883653, | |
| "eval_accuracy": 0.9162634015083313, | |
| "eval_loss": 0.3249155879020691, | |
| "eval_runtime": 572.8138, | |
| "eval_samples_per_second": 8.298, | |
| "eval_steps_per_second": 2.076, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "grad_norm": 0.09070427715778351, | |
| "learning_rate": 0.00010909090909090909, | |
| "loss": 0.3625, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18093835472333264, | |
| "eval_accuracy": 0.9541342258453369, | |
| "eval_loss": 0.16712911427021027, | |
| "eval_runtime": 570.8705, | |
| "eval_samples_per_second": 8.326, | |
| "eval_steps_per_second": 2.083, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "grad_norm": 3.719890594482422, | |
| "learning_rate": 0.0001, | |
| "loss": 0.203, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18514622343782874, | |
| "eval_accuracy": 0.9591836929321289, | |
| "eval_loss": 0.11601703613996506, | |
| "eval_runtime": 559.6205, | |
| "eval_samples_per_second": 8.493, | |
| "eval_steps_per_second": 2.125, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "grad_norm": 1.8163635730743408, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.1065, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.18935409215232485, | |
| "eval_accuracy": 0.9663370251655579, | |
| "eval_loss": 0.11517694592475891, | |
| "eval_runtime": 564.4747, | |
| "eval_samples_per_second": 8.42, | |
| "eval_steps_per_second": 2.106, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "grad_norm": 0.489663302898407, | |
| "learning_rate": 8.181818181818182e-05, | |
| "loss": 0.0797, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19356196086682095, | |
| "eval_accuracy": 0.9524511098861694, | |
| "eval_loss": 0.17924398183822632, | |
| "eval_runtime": 566.5969, | |
| "eval_samples_per_second": 8.389, | |
| "eval_steps_per_second": 2.098, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "grad_norm": 0.12014146894216537, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.158, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.19776982958131706, | |
| "eval_accuracy": 0.9545550346374512, | |
| "eval_loss": 0.17613738775253296, | |
| "eval_runtime": 565.847, | |
| "eval_samples_per_second": 8.4, | |
| "eval_steps_per_second": 2.101, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "grad_norm": 0.30841195583343506, | |
| "learning_rate": 6.363636363636363e-05, | |
| "loss": 0.1226, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20197769829581316, | |
| "eval_accuracy": 0.954344630241394, | |
| "eval_loss": 0.1746922880411148, | |
| "eval_runtime": 566.983, | |
| "eval_samples_per_second": 8.383, | |
| "eval_steps_per_second": 2.097, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 0.17474113404750824, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.0502, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "eval_accuracy": 0.9596044421195984, | |
| "eval_loss": 0.16830527782440186, | |
| "eval_runtime": 566.0043, | |
| "eval_samples_per_second": 8.397, | |
| "eval_steps_per_second": 2.101, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "grad_norm": 0.19428321719169617, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.0346, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2103934357248054, | |
| "eval_accuracy": 0.9585524797439575, | |
| "eval_loss": 0.17505739629268646, | |
| "eval_runtime": 564.2226, | |
| "eval_samples_per_second": 8.424, | |
| "eval_steps_per_second": 2.107, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "grad_norm": 0.13528193533420563, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0168, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2146013044393015, | |
| "eval_accuracy": 0.958131730556488, | |
| "eval_loss": 0.18199646472930908, | |
| "eval_runtime": 563.8378, | |
| "eval_samples_per_second": 8.43, | |
| "eval_steps_per_second": 2.109, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "grad_norm": 0.03537715971469879, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.1428, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.2188091731537976, | |
| "eval_accuracy": 0.9562381505966187, | |
| "eval_loss": 0.1906786561012268, | |
| "eval_runtime": 567.1515, | |
| "eval_samples_per_second": 8.38, | |
| "eval_steps_per_second": 2.096, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "grad_norm": 0.5100229978561401, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.0336, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22301704186829371, | |
| "eval_accuracy": 0.9503471255302429, | |
| "eval_loss": 0.2207876741886139, | |
| "eval_runtime": 564.0513, | |
| "eval_samples_per_second": 8.427, | |
| "eval_steps_per_second": 2.108, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "grad_norm": 5.639271259307861, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.1161, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.22722491058278982, | |
| "eval_accuracy": 0.9488744139671326, | |
| "eval_loss": 0.2279575914144516, | |
| "eval_runtime": 564.6939, | |
| "eval_samples_per_second": 8.417, | |
| "eval_steps_per_second": 2.106, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "grad_norm": 0.09282880276441574, | |
| "learning_rate": 0.0, | |
| "loss": 0.0377, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.23143277929728592, | |
| "eval_accuracy": 0.9495055675506592, | |
| "eval_loss": 0.22491338849067688, | |
| "eval_runtime": 556.6035, | |
| "eval_samples_per_second": 8.539, | |
| "eval_steps_per_second": 2.136, | |
| "step": 550 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.208452205931052e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |