|
{ |
|
"best_metric": 0.7350361445783132, |
|
"best_model_checkpoint": "omarmomen/sf_babylm_1/finetune/qqp/checkpoint-7200", |
|
"epoch": 2.4178712220762155, |
|
"global_step": 9200, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.05, |
|
"eval_accuracy": 0.6677079796791077, |
|
"eval_f1": 0.6216067420488713, |
|
"eval_loss": 0.6047940850257874, |
|
"eval_mcc": 0.32731379930588683, |
|
"eval_runtime": 46.8159, |
|
"eval_samples_per_second": 574.356, |
|
"eval_steps_per_second": 71.813, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"eval_accuracy": 0.680315375328064, |
|
"eval_f1": 0.4777008141937052, |
|
"eval_loss": 0.6022979617118835, |
|
"eval_mcc": 0.3375634754226189, |
|
"eval_runtime": 46.8937, |
|
"eval_samples_per_second": 573.403, |
|
"eval_steps_per_second": 71.694, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 4.934296977660973e-05, |
|
"loss": 0.6024, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"eval_accuracy": 0.709323525428772, |
|
"eval_f1": 0.609941111887414, |
|
"eval_loss": 0.5577818751335144, |
|
"eval_mcc": 0.39238388281604025, |
|
"eval_runtime": 46.9096, |
|
"eval_samples_per_second": 573.209, |
|
"eval_steps_per_second": 71.67, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"eval_accuracy": 0.7173193693161011, |
|
"eval_f1": 0.637650760356581, |
|
"eval_loss": 0.5484437942504883, |
|
"eval_mcc": 0.4115118809231198, |
|
"eval_runtime": 46.7002, |
|
"eval_samples_per_second": 575.779, |
|
"eval_steps_per_second": 71.991, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 4.868593955321945e-05, |
|
"loss": 0.5498, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"eval_accuracy": 0.7279185056686401, |
|
"eval_f1": 0.647319706903201, |
|
"eval_loss": 0.531727135181427, |
|
"eval_mcc": 0.43332964524624523, |
|
"eval_runtime": 46.9206, |
|
"eval_samples_per_second": 573.075, |
|
"eval_steps_per_second": 71.653, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"eval_accuracy": 0.7243110537528992, |
|
"eval_f1": 0.5996219281663516, |
|
"eval_loss": 0.5354304909706116, |
|
"eval_mcc": 0.4284072164308066, |
|
"eval_runtime": 46.9481, |
|
"eval_samples_per_second": 572.739, |
|
"eval_steps_per_second": 71.611, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"eval_accuracy": 0.7361746430397034, |
|
"eval_f1": 0.6530712050078247, |
|
"eval_loss": 0.5389292240142822, |
|
"eval_mcc": 0.4504997444085917, |
|
"eval_runtime": 46.8939, |
|
"eval_samples_per_second": 573.401, |
|
"eval_steps_per_second": 71.694, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 4.8028909329829176e-05, |
|
"loss": 0.5294, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"eval_accuracy": 0.7398192286491394, |
|
"eval_f1": 0.6818553888130968, |
|
"eval_loss": 0.5164288878440857, |
|
"eval_mcc": 0.46257828041044885, |
|
"eval_runtime": 47.0487, |
|
"eval_samples_per_second": 571.514, |
|
"eval_steps_per_second": 71.458, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"eval_accuracy": 0.7426456809043884, |
|
"eval_f1": 0.644362216055093, |
|
"eval_loss": 0.5257071256637573, |
|
"eval_mcc": 0.4656119414084067, |
|
"eval_runtime": 47.1774, |
|
"eval_samples_per_second": 569.955, |
|
"eval_steps_per_second": 71.263, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 4.73718791064389e-05, |
|
"loss": 0.5119, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"eval_accuracy": 0.742757260799408, |
|
"eval_f1": 0.6479360716648852, |
|
"eval_loss": 0.5224727988243103, |
|
"eval_mcc": 0.46534565085485186, |
|
"eval_runtime": 46.7811, |
|
"eval_samples_per_second": 574.783, |
|
"eval_steps_per_second": 71.867, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"eval_accuracy": 0.7552902698516846, |
|
"eval_f1": 0.6936970486919282, |
|
"eval_loss": 0.4959010183811188, |
|
"eval_mcc": 0.49274075419468716, |
|
"eval_runtime": 46.9606, |
|
"eval_samples_per_second": 572.586, |
|
"eval_steps_per_second": 71.592, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"eval_accuracy": 0.7555878162384033, |
|
"eval_f1": 0.703188510522988, |
|
"eval_loss": 0.48696744441986084, |
|
"eval_mcc": 0.49591603513395327, |
|
"eval_runtime": 46.9543, |
|
"eval_samples_per_second": 572.664, |
|
"eval_steps_per_second": 71.602, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 4.6714848883048626e-05, |
|
"loss": 0.4987, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"eval_accuracy": 0.7640671133995056, |
|
"eval_f1": 0.7040216478492115, |
|
"eval_loss": 0.4819240868091583, |
|
"eval_mcc": 0.5109504518876297, |
|
"eval_runtime": 47.0804, |
|
"eval_samples_per_second": 571.129, |
|
"eval_steps_per_second": 71.41, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"eval_accuracy": 0.7513853311538696, |
|
"eval_f1": 0.6537166537166537, |
|
"eval_loss": 0.5181935429573059, |
|
"eval_mcc": 0.48548720145396584, |
|
"eval_runtime": 47.1423, |
|
"eval_samples_per_second": 570.379, |
|
"eval_steps_per_second": 71.316, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 4.605781865965835e-05, |
|
"loss": 0.4838, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"eval_accuracy": 0.7533935904502869, |
|
"eval_f1": 0.6517514836405651, |
|
"eval_loss": 0.5359770655632019, |
|
"eval_mcc": 0.49127620489734397, |
|
"eval_runtime": 46.8881, |
|
"eval_samples_per_second": 573.472, |
|
"eval_steps_per_second": 71.703, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"eval_accuracy": 0.7620216608047485, |
|
"eval_f1": 0.679328489100476, |
|
"eval_loss": 0.4843452274799347, |
|
"eval_mcc": 0.5065599262568212, |
|
"eval_runtime": 46.9451, |
|
"eval_samples_per_second": 572.776, |
|
"eval_steps_per_second": 71.616, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"eval_accuracy": 0.7511621713638306, |
|
"eval_f1": 0.6382656647023842, |
|
"eval_loss": 0.5113937854766846, |
|
"eval_mcc": 0.4899410983771862, |
|
"eval_runtime": 46.7915, |
|
"eval_samples_per_second": 574.656, |
|
"eval_steps_per_second": 71.851, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 4.540078843626807e-05, |
|
"loss": 0.4754, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"eval_accuracy": 0.7490051984786987, |
|
"eval_f1": 0.6292776709695139, |
|
"eval_loss": 0.5369313359260559, |
|
"eval_mcc": 0.48751389754786534, |
|
"eval_runtime": 46.9526, |
|
"eval_samples_per_second": 572.684, |
|
"eval_steps_per_second": 71.604, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.7707240581512451, |
|
"eval_f1": 0.7060786650774732, |
|
"eval_loss": 0.4762149453163147, |
|
"eval_mcc": 0.524223544413257, |
|
"eval_runtime": 46.9535, |
|
"eval_samples_per_second": 572.673, |
|
"eval_steps_per_second": 71.603, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 4.474375821287779e-05, |
|
"loss": 0.4432, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"eval_accuracy": 0.7651827931404114, |
|
"eval_f1": 0.6773633111905979, |
|
"eval_loss": 0.5329920649528503, |
|
"eval_mcc": 0.5147649506735034, |
|
"eval_runtime": 46.73, |
|
"eval_samples_per_second": 575.411, |
|
"eval_steps_per_second": 71.945, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"eval_accuracy": 0.7246829271316528, |
|
"eval_f1": 0.5548138793673703, |
|
"eval_loss": 0.6932922005653381, |
|
"eval_mcc": 0.4487472998665935, |
|
"eval_runtime": 46.7479, |
|
"eval_samples_per_second": 575.192, |
|
"eval_steps_per_second": 71.918, |
|
"step": 4200 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"eval_accuracy": 0.7754843831062317, |
|
"eval_f1": 0.7007089385751822, |
|
"eval_loss": 0.5217306017875671, |
|
"eval_mcc": 0.5351067454513799, |
|
"eval_runtime": 46.9072, |
|
"eval_samples_per_second": 573.238, |
|
"eval_steps_per_second": 71.673, |
|
"step": 4400 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"learning_rate": 4.408672798948752e-05, |
|
"loss": 0.4065, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"eval_accuracy": 0.7760422229766846, |
|
"eval_f1": 0.7029693203117293, |
|
"eval_loss": 0.5050324201583862, |
|
"eval_mcc": 0.5360631237667115, |
|
"eval_runtime": 46.9901, |
|
"eval_samples_per_second": 572.227, |
|
"eval_steps_per_second": 71.547, |
|
"step": 4600 |
|
}, |
|
{ |
|
"epoch": 1.26, |
|
"eval_accuracy": 0.7660753726959229, |
|
"eval_f1": 0.6718489148580968, |
|
"eval_loss": 0.5306209921836853, |
|
"eval_mcc": 0.51883128029507, |
|
"eval_runtime": 46.8476, |
|
"eval_samples_per_second": 573.967, |
|
"eval_steps_per_second": 71.765, |
|
"step": 4800 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"learning_rate": 4.342969776609724e-05, |
|
"loss": 0.4036, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"eval_accuracy": 0.7821785807609558, |
|
"eval_f1": 0.7167931918185775, |
|
"eval_loss": 0.4877411127090454, |
|
"eval_mcc": 0.5485301444175062, |
|
"eval_runtime": 46.8776, |
|
"eval_samples_per_second": 573.6, |
|
"eval_steps_per_second": 71.719, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"eval_accuracy": 0.7830339670181274, |
|
"eval_f1": 0.7211281070745698, |
|
"eval_loss": 0.479899138212204, |
|
"eval_mcc": 0.5502080054148385, |
|
"eval_runtime": 46.9924, |
|
"eval_samples_per_second": 572.199, |
|
"eval_steps_per_second": 71.543, |
|
"step": 5200 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"eval_accuracy": 0.7755959630012512, |
|
"eval_f1": 0.6952217395696535, |
|
"eval_loss": 0.5584419369697571, |
|
"eval_mcc": 0.5365814483252632, |
|
"eval_runtime": 46.7848, |
|
"eval_samples_per_second": 574.738, |
|
"eval_steps_per_second": 71.861, |
|
"step": 5400 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"learning_rate": 4.2772667542706967e-05, |
|
"loss": 0.4017, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"eval_accuracy": 0.7783108353614807, |
|
"eval_f1": 0.7025300663705774, |
|
"eval_loss": 0.499733030796051, |
|
"eval_mcc": 0.5415910362733949, |
|
"eval_runtime": 46.9262, |
|
"eval_samples_per_second": 573.006, |
|
"eval_steps_per_second": 71.644, |
|
"step": 5600 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"eval_accuracy": 0.780058741569519, |
|
"eval_f1": 0.7000405761817813, |
|
"eval_loss": 0.5046578645706177, |
|
"eval_mcc": 0.5466795199793227, |
|
"eval_runtime": 46.8838, |
|
"eval_samples_per_second": 573.524, |
|
"eval_steps_per_second": 71.709, |
|
"step": 5800 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"learning_rate": 4.211563731931669e-05, |
|
"loss": 0.3933, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"eval_accuracy": 0.7702777981758118, |
|
"eval_f1": 0.6732264719885732, |
|
"eval_loss": 0.5271868109703064, |
|
"eval_mcc": 0.5300802009673907, |
|
"eval_runtime": 46.8784, |
|
"eval_samples_per_second": 573.591, |
|
"eval_steps_per_second": 71.717, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"eval_accuracy": 0.776934802532196, |
|
"eval_f1": 0.6919363122752953, |
|
"eval_loss": 0.5182462930679321, |
|
"eval_mcc": 0.5411011440345556, |
|
"eval_runtime": 47.0585, |
|
"eval_samples_per_second": 571.395, |
|
"eval_steps_per_second": 71.443, |
|
"step": 6200 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"eval_accuracy": 0.7767488360404968, |
|
"eval_f1": 0.6880424050304007, |
|
"eval_loss": 0.5051272511482239, |
|
"eval_mcc": 0.5420832961943636, |
|
"eval_runtime": 46.9362, |
|
"eval_samples_per_second": 572.884, |
|
"eval_steps_per_second": 71.629, |
|
"step": 6400 |
|
}, |
|
{ |
|
"epoch": 1.71, |
|
"learning_rate": 4.1458607095926416e-05, |
|
"loss": 0.3949, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 1.73, |
|
"eval_accuracy": 0.7856744527816772, |
|
"eval_f1": 0.7125542421068383, |
|
"eval_loss": 0.49001672863960266, |
|
"eval_mcc": 0.5574799704127452, |
|
"eval_runtime": 46.7277, |
|
"eval_samples_per_second": 575.44, |
|
"eval_steps_per_second": 71.949, |
|
"step": 6600 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"eval_accuracy": 0.7814719676971436, |
|
"eval_f1": 0.699867197875166, |
|
"eval_loss": 0.4943256080150604, |
|
"eval_mcc": 0.5504778485474244, |
|
"eval_runtime": 46.8928, |
|
"eval_samples_per_second": 573.414, |
|
"eval_steps_per_second": 71.695, |
|
"step": 6800 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"learning_rate": 4.080157687253614e-05, |
|
"loss": 0.3868, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"eval_accuracy": 0.792517364025116, |
|
"eval_f1": 0.7269345602271059, |
|
"eval_loss": 0.4886585474014282, |
|
"eval_mcc": 0.5710093407604465, |
|
"eval_runtime": 46.9266, |
|
"eval_samples_per_second": 573.001, |
|
"eval_steps_per_second": 71.644, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"eval_accuracy": 0.7955297827720642, |
|
"eval_f1": 0.7350361445783132, |
|
"eval_loss": 0.47052621841430664, |
|
"eval_mcc": 0.5767997243339644, |
|
"eval_runtime": 47.1507, |
|
"eval_samples_per_second": 570.278, |
|
"eval_steps_per_second": 71.303, |
|
"step": 7200 |
|
}, |
|
{ |
|
"epoch": 1.94, |
|
"eval_accuracy": 0.7917736172676086, |
|
"eval_f1": 0.7264911337990327, |
|
"eval_loss": 0.45988404750823975, |
|
"eval_mcc": 0.5693223637509579, |
|
"eval_runtime": 47.1119, |
|
"eval_samples_per_second": 570.747, |
|
"eval_steps_per_second": 71.362, |
|
"step": 7400 |
|
}, |
|
{ |
|
"epoch": 1.97, |
|
"learning_rate": 4.0144546649145865e-05, |
|
"loss": 0.3873, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.7855628728866577, |
|
"eval_f1": 0.7087584604505506, |
|
"eval_loss": 0.5034728050231934, |
|
"eval_mcc": 0.5582978063516465, |
|
"eval_runtime": 46.4587, |
|
"eval_samples_per_second": 578.772, |
|
"eval_steps_per_second": 72.365, |
|
"step": 7600 |
|
}, |
|
{ |
|
"epoch": 2.05, |
|
"eval_accuracy": 0.7831083536148071, |
|
"eval_f1": 0.6980741354317664, |
|
"eval_loss": 0.591555118560791, |
|
"eval_mcc": 0.5556929016513195, |
|
"eval_runtime": 46.634, |
|
"eval_samples_per_second": 576.596, |
|
"eval_steps_per_second": 72.093, |
|
"step": 7800 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"learning_rate": 3.948751642575559e-05, |
|
"loss": 0.3159, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.1, |
|
"eval_accuracy": 0.7854140996932983, |
|
"eval_f1": 0.7026079785589114, |
|
"eval_loss": 0.5935361385345459, |
|
"eval_mcc": 0.5602146447885732, |
|
"eval_runtime": 46.8572, |
|
"eval_samples_per_second": 573.85, |
|
"eval_steps_per_second": 71.75, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 2.16, |
|
"eval_accuracy": 0.793744683265686, |
|
"eval_f1": 0.7241619417089427, |
|
"eval_loss": 0.5583649277687073, |
|
"eval_mcc": 0.5746964821159373, |
|
"eval_runtime": 46.4742, |
|
"eval_samples_per_second": 578.58, |
|
"eval_steps_per_second": 72.341, |
|
"step": 8200 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"eval_accuracy": 0.789393424987793, |
|
"eval_f1": 0.7119678551447027, |
|
"eval_loss": 0.5677328109741211, |
|
"eval_mcc": 0.5673760082808692, |
|
"eval_runtime": 47.0291, |
|
"eval_samples_per_second": 571.752, |
|
"eval_steps_per_second": 71.488, |
|
"step": 8400 |
|
}, |
|
{ |
|
"epoch": 2.23, |
|
"learning_rate": 3.8830486202365314e-05, |
|
"loss": 0.305, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"eval_accuracy": 0.7924429774284363, |
|
"eval_f1": 0.7181740140382771, |
|
"eval_loss": 0.5449676513671875, |
|
"eval_mcc": 0.5732624266819509, |
|
"eval_runtime": 46.808, |
|
"eval_samples_per_second": 574.453, |
|
"eval_steps_per_second": 71.825, |
|
"step": 8600 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"eval_accuracy": 0.7847074866294861, |
|
"eval_f1": 0.7036904335363668, |
|
"eval_loss": 0.5745317935943604, |
|
"eval_mcc": 0.5578160063550441, |
|
"eval_runtime": 46.518, |
|
"eval_samples_per_second": 578.035, |
|
"eval_steps_per_second": 72.273, |
|
"step": 8800 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"learning_rate": 3.817345597897504e-05, |
|
"loss": 0.3064, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"eval_accuracy": 0.7969430088996887, |
|
"eval_f1": 0.7335285505124451, |
|
"eval_loss": 0.5378233194351196, |
|
"eval_mcc": 0.5803196134149392, |
|
"eval_runtime": 46.5493, |
|
"eval_samples_per_second": 577.645, |
|
"eval_steps_per_second": 72.224, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"eval_accuracy": 0.7921082973480225, |
|
"eval_f1": 0.719546457957054, |
|
"eval_loss": 0.5811643600463867, |
|
"eval_mcc": 0.5719007005215488, |
|
"eval_runtime": 46.4778, |
|
"eval_samples_per_second": 578.534, |
|
"eval_steps_per_second": 72.336, |
|
"step": 9200 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"step": 9200, |
|
"total_flos": 1.5087744537126912e+16, |
|
"train_loss": 0.4304196527729864, |
|
"train_runtime": 4139.7163, |
|
"train_samples_per_second": 588.2, |
|
"train_steps_per_second": 9.191 |
|
} |
|
], |
|
"max_steps": 38050, |
|
"num_train_epochs": 10, |
|
"total_flos": 1.5087744537126912e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|