| { | |
| "best_metric": 0.08956250548362732, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-600", | |
| "epoch": 2.608695652173913, | |
| "eval_steps": 50, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004347826086956522, | |
| "eval_loss": 2.616544723510742, | |
| "eval_runtime": 27.0944, | |
| "eval_samples_per_second": 14.32, | |
| "eval_steps_per_second": 3.58, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.043478260869565216, | |
| "grad_norm": 0.5260958671569824, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6161, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08695652173913043, | |
| "grad_norm": 0.49680450558662415, | |
| "learning_rate": 0.0001998582695676762, | |
| "loss": 0.3508, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13043478260869565, | |
| "grad_norm": 1.1319019794464111, | |
| "learning_rate": 0.00019943348002101371, | |
| "loss": 0.2446, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 2.221609115600586, | |
| "learning_rate": 0.00019872683547213446, | |
| "loss": 0.3257, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "grad_norm": 0.9951373338699341, | |
| "learning_rate": 0.00019774033898178667, | |
| "loss": 0.2999, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "eval_loss": 0.3432919681072235, | |
| "eval_runtime": 27.4695, | |
| "eval_samples_per_second": 14.125, | |
| "eval_steps_per_second": 3.531, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2608695652173913, | |
| "grad_norm": 0.349606454372406, | |
| "learning_rate": 0.0001964767868814516, | |
| "loss": 0.2521, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.30434782608695654, | |
| "grad_norm": 0.4386211335659027, | |
| "learning_rate": 0.00019493976084683813, | |
| "loss": 0.1782, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.5885153412818909, | |
| "learning_rate": 0.00019313361774523385, | |
| "loss": 0.1268, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.391304347826087, | |
| "grad_norm": 1.8868091106414795, | |
| "learning_rate": 0.00019106347728549135, | |
| "loss": 0.1951, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 1.164074420928955, | |
| "learning_rate": 0.00018873520750565718, | |
| "loss": 0.2354, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "eval_loss": 0.24758285284042358, | |
| "eval_runtime": 27.4327, | |
| "eval_samples_per_second": 14.144, | |
| "eval_steps_per_second": 3.536, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4782608695652174, | |
| "grad_norm": 0.29064542055130005, | |
| "learning_rate": 0.0001861554081393806, | |
| "loss": 0.1775, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 0.4147568643093109, | |
| "learning_rate": 0.0001833313919082515, | |
| "loss": 0.147, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5652173913043478, | |
| "grad_norm": 0.7357779741287231, | |
| "learning_rate": 0.00018027116379309638, | |
| "loss": 0.1242, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6086956521739131, | |
| "grad_norm": 1.05941903591156, | |
| "learning_rate": 0.00017698339834299061, | |
| "loss": 0.1637, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 4.279825687408447, | |
| "learning_rate": 0.00017347741508630672, | |
| "loss": 0.184, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "eval_loss": 0.303436279296875, | |
| "eval_runtime": 27.4825, | |
| "eval_samples_per_second": 14.118, | |
| "eval_steps_per_second": 3.53, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.33306750655174255, | |
| "learning_rate": 0.0001697631521134985, | |
| "loss": 0.189, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7391304347826086, | |
| "grad_norm": 0.3919294476509094, | |
| "learning_rate": 0.00016585113790650388, | |
| "loss": 0.1279, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "grad_norm": 0.617232620716095, | |
| "learning_rate": 0.0001617524614946192, | |
| "loss": 0.1031, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8260869565217391, | |
| "grad_norm": 1.3665099143981934, | |
| "learning_rate": 0.0001574787410214407, | |
| "loss": 0.1055, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 1.6653144359588623, | |
| "learning_rate": 0.00015304209081197425, | |
| "loss": 0.19, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 0.2233988642692566, | |
| "eval_runtime": 27.4566, | |
| "eval_samples_per_second": 14.131, | |
| "eval_steps_per_second": 3.533, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9130434782608695, | |
| "grad_norm": 0.357671320438385, | |
| "learning_rate": 0.00014845508703326504, | |
| "loss": 0.1606, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9565217391304348, | |
| "grad_norm": 0.7531294822692871, | |
| "learning_rate": 0.00014373073204588556, | |
| "loss": 0.0999, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.0645687580108643, | |
| "learning_rate": 0.00013888241754733208, | |
| "loss": 0.1406, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.27364978194236755, | |
| "learning_rate": 0.00013392388661180303, | |
| "loss": 0.1368, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "grad_norm": 0.3493202328681946, | |
| "learning_rate": 0.0001288691947339621, | |
| "loss": 0.0944, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.0869565217391304, | |
| "eval_loss": 0.13593319058418274, | |
| "eval_runtime": 27.5029, | |
| "eval_samples_per_second": 14.108, | |
| "eval_steps_per_second": 3.527, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1304347826086956, | |
| "grad_norm": 0.2853105962276459, | |
| "learning_rate": 0.0001237326699871115, | |
| "loss": 0.0725, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.1739130434782608, | |
| "grad_norm": 0.7938422560691833, | |
| "learning_rate": 0.00011852887240871145, | |
| "loss": 0.0838, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2173913043478262, | |
| "grad_norm": 1.0990744829177856, | |
| "learning_rate": 0.00011327255272837221, | |
| "loss": 0.1366, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.2608695652173914, | |
| "grad_norm": 0.2768363654613495, | |
| "learning_rate": 0.00010797861055530831, | |
| "loss": 0.1213, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.3043478260869565, | |
| "grad_norm": 0.20569927990436554, | |
| "learning_rate": 0.00010266205214377748, | |
| "loss": 0.0853, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3043478260869565, | |
| "eval_loss": 0.15111666917800903, | |
| "eval_runtime": 27.4713, | |
| "eval_samples_per_second": 14.124, | |
| "eval_steps_per_second": 3.531, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3478260869565217, | |
| "grad_norm": 0.8807733058929443, | |
| "learning_rate": 9.733794785622253e-05, | |
| "loss": 0.073, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 1.0780969858169556, | |
| "learning_rate": 9.202138944469168e-05, | |
| "loss": 0.0833, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.434782608695652, | |
| "grad_norm": 1.176543116569519, | |
| "learning_rate": 8.672744727162781e-05, | |
| "loss": 0.0913, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.4782608695652173, | |
| "grad_norm": 0.30861204862594604, | |
| "learning_rate": 8.147112759128859e-05, | |
| "loss": 0.1176, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.5217391304347827, | |
| "grad_norm": 0.2529921531677246, | |
| "learning_rate": 7.626733001288851e-05, | |
| "loss": 0.0809, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.5217391304347827, | |
| "eval_loss": 0.12669587135314941, | |
| "eval_runtime": 27.439, | |
| "eval_samples_per_second": 14.14, | |
| "eval_steps_per_second": 3.535, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.5652173913043477, | |
| "grad_norm": 0.3250602185726166, | |
| "learning_rate": 7.113080526603792e-05, | |
| "loss": 0.061, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.608695652173913, | |
| "grad_norm": 0.8286572098731995, | |
| "learning_rate": 6.607611338819697e-05, | |
| "loss": 0.0688, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.6521739130434783, | |
| "grad_norm": 0.8094994425773621, | |
| "learning_rate": 6.111758245266794e-05, | |
| "loss": 0.0633, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.6956521739130435, | |
| "grad_norm": 0.29753217101097107, | |
| "learning_rate": 5.626926795411447e-05, | |
| "loss": 0.0931, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.2855427861213684, | |
| "learning_rate": 5.1544912966734994e-05, | |
| "loss": 0.0691, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 0.10997624695301056, | |
| "eval_runtime": 27.483, | |
| "eval_samples_per_second": 14.118, | |
| "eval_steps_per_second": 3.529, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.7826086956521738, | |
| "grad_norm": 0.3453027606010437, | |
| "learning_rate": 4.695790918802576e-05, | |
| "loss": 0.0447, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.8260869565217392, | |
| "grad_norm": 0.8909457325935364, | |
| "learning_rate": 4.252125897855932e-05, | |
| "loss": 0.0641, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.8695652173913042, | |
| "grad_norm": 0.9604889154434204, | |
| "learning_rate": 3.824753850538082e-05, | |
| "loss": 0.0642, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "grad_norm": 0.3222946524620056, | |
| "learning_rate": 3.414886209349615e-05, | |
| "loss": 0.0763, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.9565217391304348, | |
| "grad_norm": 0.27602681517601013, | |
| "learning_rate": 3.0236847886501542e-05, | |
| "loss": 0.0506, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.9565217391304348, | |
| "eval_loss": 0.09027178585529327, | |
| "eval_runtime": 27.4281, | |
| "eval_samples_per_second": 14.146, | |
| "eval_steps_per_second": 3.537, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.6299391388893127, | |
| "learning_rate": 2.6522584913693294e-05, | |
| "loss": 0.0765, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.0434782608695654, | |
| "grad_norm": 0.17378808557987213, | |
| "learning_rate": 2.301660165700936e-05, | |
| "loss": 0.061, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.14308716356754303, | |
| "learning_rate": 1.9728836206903656e-05, | |
| "loss": 0.0433, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.130434782608696, | |
| "grad_norm": 0.17840631306171417, | |
| "learning_rate": 1.6668608091748495e-05, | |
| "loss": 0.0328, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.1739130434782608, | |
| "grad_norm": 0.7729475498199463, | |
| "learning_rate": 1.3844591860619383e-05, | |
| "loss": 0.0335, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.1739130434782608, | |
| "eval_loss": 0.08860904723405838, | |
| "eval_runtime": 27.5092, | |
| "eval_samples_per_second": 14.104, | |
| "eval_steps_per_second": 3.526, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.217391304347826, | |
| "grad_norm": 0.24761967360973358, | |
| "learning_rate": 1.1264792494342857e-05, | |
| "loss": 0.0361, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.260869565217391, | |
| "grad_norm": 0.2562055289745331, | |
| "learning_rate": 8.936522714508678e-06, | |
| "loss": 0.0598, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.3043478260869565, | |
| "grad_norm": 0.18212448060512543, | |
| "learning_rate": 6.866382254766157e-06, | |
| "loss": 0.0442, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.3478260869565215, | |
| "grad_norm": 0.2458052933216095, | |
| "learning_rate": 5.060239153161872e-06, | |
| "loss": 0.03, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.391304347826087, | |
| "grad_norm": 0.35940125584602356, | |
| "learning_rate": 3.5232131185484076e-06, | |
| "loss": 0.0232, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.391304347826087, | |
| "eval_loss": 0.08165618777275085, | |
| "eval_runtime": 27.4669, | |
| "eval_samples_per_second": 14.126, | |
| "eval_steps_per_second": 3.532, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.32912999391555786, | |
| "learning_rate": 2.259661018213333e-06, | |
| "loss": 0.0287, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.4782608695652173, | |
| "grad_norm": 0.18391677737236023, | |
| "learning_rate": 1.2731645278655445e-06, | |
| "loss": 0.0516, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.5217391304347827, | |
| "grad_norm": 0.17898060381412506, | |
| "learning_rate": 5.665199789862907e-07, | |
| "loss": 0.0425, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.5652173913043477, | |
| "grad_norm": 0.2860226333141327, | |
| "learning_rate": 1.4173043232380557e-07, | |
| "loss": 0.0326, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "grad_norm": 0.39208436012268066, | |
| "learning_rate": 0.0, | |
| "loss": 0.0258, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.608695652173913, | |
| "eval_loss": 0.08956250548362732, | |
| "eval_runtime": 27.4737, | |
| "eval_samples_per_second": 14.123, | |
| "eval_steps_per_second": 3.531, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 150, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.429935603923026e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |