| { | |
| "best_metric": 9.23099136352539, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-600", | |
| "epoch": 0.706090026478376, | |
| "eval_steps": 50, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011768167107972932, | |
| "eval_loss": 10.379207611083984, | |
| "eval_runtime": 3.1746, | |
| "eval_samples_per_second": 451.087, | |
| "eval_steps_per_second": 112.772, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011768167107972934, | |
| "grad_norm": 0.5737844705581665, | |
| "learning_rate": 0.0002, | |
| "loss": 10.3536, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.023536334215945868, | |
| "grad_norm": 0.5810375809669495, | |
| "learning_rate": 0.0001998582695676762, | |
| "loss": 10.3108, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0353045013239188, | |
| "grad_norm": 0.5553200840950012, | |
| "learning_rate": 0.00019943348002101371, | |
| "loss": 10.2622, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.047072668431891736, | |
| "grad_norm": 0.559855043888092, | |
| "learning_rate": 0.00019872683547213446, | |
| "loss": 10.2096, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05884083553986467, | |
| "grad_norm": 0.5475898385047913, | |
| "learning_rate": 0.00019774033898178667, | |
| "loss": 10.1577, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05884083553986467, | |
| "eval_loss": 10.128434181213379, | |
| "eval_runtime": 3.1442, | |
| "eval_samples_per_second": 455.443, | |
| "eval_steps_per_second": 113.861, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0706090026478376, | |
| "grad_norm": 3.1386730670928955, | |
| "learning_rate": 0.0001964767868814516, | |
| "loss": 10.1068, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08237716975581054, | |
| "grad_norm": 0.5707929134368896, | |
| "learning_rate": 0.00019493976084683813, | |
| "loss": 10.0549, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09414533686378347, | |
| "grad_norm": 0.5704991221427917, | |
| "learning_rate": 0.00019313361774523385, | |
| "loss": 10.006, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1059135039717564, | |
| "grad_norm": 0.5538395643234253, | |
| "learning_rate": 0.00019106347728549135, | |
| "loss": 9.9593, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.11768167107972934, | |
| "grad_norm": 0.6376510858535767, | |
| "learning_rate": 0.00018873520750565718, | |
| "loss": 9.912, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11768167107972934, | |
| "eval_loss": 9.88634204864502, | |
| "eval_runtime": 3.1338, | |
| "eval_samples_per_second": 456.959, | |
| "eval_steps_per_second": 114.24, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12944983818770225, | |
| "grad_norm": 1.2404887676239014, | |
| "learning_rate": 0.0001861554081393806, | |
| "loss": 9.8677, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.1412180052956752, | |
| "grad_norm": 0.5645771622657776, | |
| "learning_rate": 0.0001833313919082515, | |
| "loss": 9.8202, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.15298617240364812, | |
| "grad_norm": 0.6336705088615417, | |
| "learning_rate": 0.00018027116379309638, | |
| "loss": 9.7735, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.16475433951162108, | |
| "grad_norm": 22.544200897216797, | |
| "learning_rate": 0.00017698339834299061, | |
| "loss": 9.7257, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.176522506619594, | |
| "grad_norm": 1.2222896814346313, | |
| "learning_rate": 0.00017347741508630672, | |
| "loss": 9.6954, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.176522506619594, | |
| "eval_loss": 9.663827896118164, | |
| "eval_runtime": 3.1098, | |
| "eval_samples_per_second": 460.483, | |
| "eval_steps_per_second": 115.121, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.18829067372756694, | |
| "grad_norm": 0.5674024224281311, | |
| "learning_rate": 0.0001697631521134985, | |
| "loss": 9.6526, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.20005884083553988, | |
| "grad_norm": 0.5667808651924133, | |
| "learning_rate": 0.00016585113790650388, | |
| "loss": 9.613, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2118270079435128, | |
| "grad_norm": 0.5394534468650818, | |
| "learning_rate": 0.0001617524614946192, | |
| "loss": 9.5741, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.22359517505148574, | |
| "grad_norm": 0.5556952357292175, | |
| "learning_rate": 0.0001574787410214407, | |
| "loss": 9.5348, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.23536334215945867, | |
| "grad_norm": 0.5671817064285278, | |
| "learning_rate": 0.00015304209081197425, | |
| "loss": 9.4987, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23536334215945867, | |
| "eval_loss": 9.482871055603027, | |
| "eval_runtime": 3.1933, | |
| "eval_samples_per_second": 448.444, | |
| "eval_steps_per_second": 112.111, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2471315092674316, | |
| "grad_norm": 0.7693024277687073, | |
| "learning_rate": 0.00014845508703326504, | |
| "loss": 9.4781, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.2588996763754045, | |
| "grad_norm": 3.6308693885803223, | |
| "learning_rate": 0.00014373073204588556, | |
| "loss": 9.4655, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.27066784348337747, | |
| "grad_norm": 0.5690402388572693, | |
| "learning_rate": 0.00013888241754733208, | |
| "loss": 9.4397, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.2824360105913504, | |
| "grad_norm": 0.5628238916397095, | |
| "learning_rate": 0.00013392388661180303, | |
| "loss": 9.4196, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.29420417769932333, | |
| "grad_norm": 0.5627617835998535, | |
| "learning_rate": 0.0001288691947339621, | |
| "loss": 9.3916, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.29420417769932333, | |
| "eval_loss": 9.378963470458984, | |
| "eval_runtime": 3.1537, | |
| "eval_samples_per_second": 454.067, | |
| "eval_steps_per_second": 113.517, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.30597234480729624, | |
| "grad_norm": 0.5669119954109192, | |
| "learning_rate": 0.0001237326699871115, | |
| "loss": 9.3743, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.3177405119152692, | |
| "grad_norm": 0.5632140040397644, | |
| "learning_rate": 0.00011852887240871145, | |
| "loss": 9.3444, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.32950867902324216, | |
| "grad_norm": 0.5589627623558044, | |
| "learning_rate": 0.00011327255272837221, | |
| "loss": 9.3176, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.34127684613121506, | |
| "grad_norm": 0.5277190208435059, | |
| "learning_rate": 0.00010797861055530831, | |
| "loss": 9.2899, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.353045013239188, | |
| "grad_norm": 0.5792800784111023, | |
| "learning_rate": 0.00010266205214377748, | |
| "loss": 9.2735, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.353045013239188, | |
| "eval_loss": 9.273290634155273, | |
| "eval_runtime": 3.1419, | |
| "eval_samples_per_second": 455.778, | |
| "eval_steps_per_second": 113.945, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3648131803471609, | |
| "grad_norm": 0.5586740374565125, | |
| "learning_rate": 9.733794785622253e-05, | |
| "loss": 9.2749, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.3765813474551339, | |
| "grad_norm": 0.5625619292259216, | |
| "learning_rate": 9.202138944469168e-05, | |
| "loss": 9.2613, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.3883495145631068, | |
| "grad_norm": 0.5586420893669128, | |
| "learning_rate": 8.672744727162781e-05, | |
| "loss": 9.2536, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.40011768167107975, | |
| "grad_norm": 0.5391138195991516, | |
| "learning_rate": 8.147112759128859e-05, | |
| "loss": 9.2467, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.41188584877905265, | |
| "grad_norm": 0.5625888705253601, | |
| "learning_rate": 7.626733001288851e-05, | |
| "loss": 9.2363, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.41188584877905265, | |
| "eval_loss": 9.241024017333984, | |
| "eval_runtime": 3.1378, | |
| "eval_samples_per_second": 456.375, | |
| "eval_steps_per_second": 114.094, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4236540158870256, | |
| "grad_norm": 0.5664399862289429, | |
| "learning_rate": 7.113080526603792e-05, | |
| "loss": 9.2488, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.4354221829949985, | |
| "grad_norm": 0.5664339661598206, | |
| "learning_rate": 6.607611338819697e-05, | |
| "loss": 9.2408, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.4471903501029715, | |
| "grad_norm": 42.019798278808594, | |
| "learning_rate": 6.111758245266794e-05, | |
| "loss": 9.2368, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4589585172109444, | |
| "grad_norm": 0.5508390069007874, | |
| "learning_rate": 5.626926795411447e-05, | |
| "loss": 9.2308, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.47072668431891734, | |
| "grad_norm": 0.5313293933868408, | |
| "learning_rate": 5.1544912966734994e-05, | |
| "loss": 9.2265, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.47072668431891734, | |
| "eval_loss": 9.23282241821289, | |
| "eval_runtime": 3.1777, | |
| "eval_samples_per_second": 450.641, | |
| "eval_steps_per_second": 112.66, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.48249485142689025, | |
| "grad_norm": 0.5782054662704468, | |
| "learning_rate": 4.695790918802576e-05, | |
| "loss": 9.2417, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.4942630185348632, | |
| "grad_norm": 0.5586299896240234, | |
| "learning_rate": 4.252125897855932e-05, | |
| "loss": 9.2365, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.5060311856428361, | |
| "grad_norm": 2.5999763011932373, | |
| "learning_rate": 3.824753850538082e-05, | |
| "loss": 9.2348, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.517799352750809, | |
| "grad_norm": 0.5547754764556885, | |
| "learning_rate": 3.414886209349615e-05, | |
| "loss": 9.2269, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.529567519858782, | |
| "grad_norm": 0.5392950773239136, | |
| "learning_rate": 3.0236847886501542e-05, | |
| "loss": 9.2234, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.529567519858782, | |
| "eval_loss": 9.231368064880371, | |
| "eval_runtime": 3.1373, | |
| "eval_samples_per_second": 456.443, | |
| "eval_steps_per_second": 114.111, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5413356869667549, | |
| "grad_norm": 0.5859640836715698, | |
| "learning_rate": 2.6522584913693294e-05, | |
| "loss": 9.2368, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.5531038540747278, | |
| "grad_norm": 0.5664204955101013, | |
| "learning_rate": 2.301660165700936e-05, | |
| "loss": 9.2346, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.5648720211827007, | |
| "grad_norm": 0.5508027672767639, | |
| "learning_rate": 1.9728836206903656e-05, | |
| "loss": 9.2328, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5766401882906738, | |
| "grad_norm": 0.5547236204147339, | |
| "learning_rate": 1.6668608091748495e-05, | |
| "loss": 9.2273, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.5884083553986467, | |
| "grad_norm": 0.5351949334144592, | |
| "learning_rate": 1.3844591860619383e-05, | |
| "loss": 9.2272, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5884083553986467, | |
| "eval_loss": 9.230931282043457, | |
| "eval_runtime": 3.1481, | |
| "eval_samples_per_second": 454.871, | |
| "eval_steps_per_second": 113.718, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6001765225066196, | |
| "grad_norm": 0.5742395520210266, | |
| "learning_rate": 1.1264792494342857e-05, | |
| "loss": 9.2388, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.6119446896145925, | |
| "grad_norm": 0.5664268732070923, | |
| "learning_rate": 8.936522714508678e-06, | |
| "loss": 9.234, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.6237128567225655, | |
| "grad_norm": 0.5586317777633667, | |
| "learning_rate": 6.866382254766157e-06, | |
| "loss": 9.231, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.6354810238305384, | |
| "grad_norm": 0.5682714581489563, | |
| "learning_rate": 5.060239153161872e-06, | |
| "loss": 9.227, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.6472491909385113, | |
| "grad_norm": 0.5156871676445007, | |
| "learning_rate": 3.5232131185484076e-06, | |
| "loss": 9.2246, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6472491909385113, | |
| "eval_loss": 9.23094654083252, | |
| "eval_runtime": 3.1842, | |
| "eval_samples_per_second": 449.718, | |
| "eval_steps_per_second": 112.429, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6590173580464843, | |
| "grad_norm": 0.5625162124633789, | |
| "learning_rate": 2.259661018213333e-06, | |
| "loss": 9.2384, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.6707855251544572, | |
| "grad_norm": 0.5742562413215637, | |
| "learning_rate": 1.2731645278655445e-06, | |
| "loss": 9.2365, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.6825536922624301, | |
| "grad_norm": 0.5625633001327515, | |
| "learning_rate": 5.665199789862907e-07, | |
| "loss": 9.2344, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.694321859370403, | |
| "grad_norm": 0.5586168766021729, | |
| "learning_rate": 1.4173043232380557e-07, | |
| "loss": 9.2276, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.706090026478376, | |
| "grad_norm": 0.5523204207420349, | |
| "learning_rate": 0.0, | |
| "loss": 9.2266, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.706090026478376, | |
| "eval_loss": 9.23099136352539, | |
| "eval_runtime": 3.1361, | |
| "eval_samples_per_second": 456.613, | |
| "eval_steps_per_second": 114.153, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 128442497236992.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |