| { | |
| "best_metric": 0.0007766146445646882, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-450", | |
| "epoch": 1.0, | |
| "eval_steps": 50, | |
| "global_step": 463, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0021598272138228943, | |
| "eval_loss": 0.11870575696229935, | |
| "eval_runtime": 28.391, | |
| "eval_samples_per_second": 6.868, | |
| "eval_steps_per_second": 1.726, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02159827213822894, | |
| "grad_norm": 0.6378040313720703, | |
| "learning_rate": 4.08e-05, | |
| "loss": 0.1198, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04319654427645788, | |
| "grad_norm": 0.3366822600364685, | |
| "learning_rate": 8.16e-05, | |
| "loss": 0.0585, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06479481641468683, | |
| "grad_norm": 1.0418509244918823, | |
| "learning_rate": 0.0001224, | |
| "loss": 0.0195, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08639308855291576, | |
| "grad_norm": 0.7536942362785339, | |
| "learning_rate": 0.0001632, | |
| "loss": 0.0173, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1079913606911447, | |
| "grad_norm": 2.1272497177124023, | |
| "learning_rate": 0.000204, | |
| "loss": 0.0183, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1079913606911447, | |
| "eval_loss": 0.011936341412365437, | |
| "eval_runtime": 28.1854, | |
| "eval_samples_per_second": 6.918, | |
| "eval_steps_per_second": 1.738, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12958963282937366, | |
| "grad_norm": 0.5203682780265808, | |
| "learning_rate": 0.00020370504185350093, | |
| "loss": 0.0316, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1511879049676026, | |
| "grad_norm": 0.8338865041732788, | |
| "learning_rate": 0.00020282187330239947, | |
| "loss": 0.0168, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17278617710583152, | |
| "grad_norm": 0.63815838098526, | |
| "learning_rate": 0.0002013556021458894, | |
| "loss": 0.0222, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19438444924406048, | |
| "grad_norm": 0.11844746768474579, | |
| "learning_rate": 0.00019931470855304105, | |
| "loss": 0.0154, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2159827213822894, | |
| "grad_norm": 4.035724639892578, | |
| "learning_rate": 0.00019671099601780242, | |
| "loss": 0.016, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2159827213822894, | |
| "eval_loss": 0.02815873548388481, | |
| "eval_runtime": 28.4968, | |
| "eval_samples_per_second": 6.843, | |
| "eval_steps_per_second": 1.719, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23758099352051837, | |
| "grad_norm": 0.14545699954032898, | |
| "learning_rate": 0.00019355952309357377, | |
| "loss": 0.0326, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2591792656587473, | |
| "grad_norm": 1.333457350730896, | |
| "learning_rate": 0.00018987851630216821, | |
| "loss": 0.0127, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.28077753779697623, | |
| "grad_norm": 0.19191227853298187, | |
| "learning_rate": 0.0001856892647208494, | |
| "loss": 0.0164, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.3023758099352052, | |
| "grad_norm": 0.6895551085472107, | |
| "learning_rate": 0.00018101599685710056, | |
| "loss": 0.0143, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.32397408207343414, | |
| "grad_norm": 0.06516972184181213, | |
| "learning_rate": 0.00017588574052321935, | |
| "loss": 0.0108, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.32397408207343414, | |
| "eval_loss": 0.007597790565341711, | |
| "eval_runtime": 28.5719, | |
| "eval_samples_per_second": 6.825, | |
| "eval_steps_per_second": 1.715, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.34557235421166305, | |
| "grad_norm": 0.01516179833561182, | |
| "learning_rate": 0.0001703281665211531, | |
| "loss": 0.0107, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.367170626349892, | |
| "grad_norm": 3.366034507751465, | |
| "learning_rate": 0.00016437541704162093, | |
| "loss": 0.0119, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.38876889848812096, | |
| "grad_norm": 0.2688906192779541, | |
| "learning_rate": 0.00015806191976997693, | |
| "loss": 0.0033, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4103671706263499, | |
| "grad_norm": 0.06040094047784805, | |
| "learning_rate": 0.00015142418877393032, | |
| "loss": 0.0028, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4319654427645788, | |
| "grad_norm": 0.10505006462335587, | |
| "learning_rate": 0.00014450061332468804, | |
| "loss": 0.0051, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4319654427645788, | |
| "eval_loss": 0.016910606995224953, | |
| "eval_runtime": 28.5258, | |
| "eval_samples_per_second": 6.836, | |
| "eval_steps_per_second": 1.718, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4535637149028078, | |
| "grad_norm": 0.40670064091682434, | |
| "learning_rate": 0.0001373312358728708, | |
| "loss": 0.0122, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.47516198704103674, | |
| "grad_norm": 1.2365318536758423, | |
| "learning_rate": 0.00012995752046327736, | |
| "loss": 0.0043, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.49676025917926564, | |
| "grad_norm": 0.05308518186211586, | |
| "learning_rate": 0.0001224221129278691, | |
| "loss": 0.0041, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5183585313174947, | |
| "grad_norm": 0.05913418531417847, | |
| "learning_rate": 0.00011476859424389611, | |
| "loss": 0.0024, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5399568034557235, | |
| "grad_norm": 0.029558856040239334, | |
| "learning_rate": 0.00010704122848361768, | |
| "loss": 0.0031, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5399568034557235, | |
| "eval_loss": 0.0030939881689846516, | |
| "eval_runtime": 28.2074, | |
| "eval_samples_per_second": 6.913, | |
| "eval_steps_per_second": 1.737, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5615550755939525, | |
| "grad_norm": 0.13922685384750366, | |
| "learning_rate": 9.928470681334698e-05, | |
| "loss": 0.0058, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5831533477321814, | |
| "grad_norm": 0.06830060482025146, | |
| "learning_rate": 9.154388902240052e-05, | |
| "loss": 0.0033, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6047516198704104, | |
| "grad_norm": 0.024560509249567986, | |
| "learning_rate": 8.386354407681761e-05, | |
| "loss": 0.0019, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6263498920086393, | |
| "grad_norm": 0.0010884521761909127, | |
| "learning_rate": 7.62880911983556e-05, | |
| "loss": 0.0004, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6479481641468683, | |
| "grad_norm": 0.04456557705998421, | |
| "learning_rate": 6.886134296622857e-05, | |
| "loss": 0.0003, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6479481641468683, | |
| "eval_loss": 0.001991268480196595, | |
| "eval_runtime": 28.47, | |
| "eval_samples_per_second": 6.849, | |
| "eval_steps_per_second": 1.721, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6695464362850972, | |
| "grad_norm": 0.012475444003939629, | |
| "learning_rate": 6.16262519273594e-05, | |
| "loss": 0.0013, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6911447084233261, | |
| "grad_norm": 0.19239909946918488, | |
| "learning_rate": 5.4624662180622114e-05, | |
| "loss": 0.0039, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.712742980561555, | |
| "grad_norm": 0.025792984291911125, | |
| "learning_rate": 4.7897067371786276e-05, | |
| "loss": 0.0023, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.734341252699784, | |
| "grad_norm": 0.013253681361675262, | |
| "learning_rate": 4.148237649879712e-05, | |
| "loss": 0.0021, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.755939524838013, | |
| "grad_norm": 0.0363602451980114, | |
| "learning_rate": 3.541768888185587e-05, | |
| "loss": 0.0029, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.755939524838013, | |
| "eval_loss": 0.0015392228960990906, | |
| "eval_runtime": 28.3248, | |
| "eval_samples_per_second": 6.884, | |
| "eval_steps_per_second": 1.73, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7775377969762419, | |
| "grad_norm": 0.004180555697530508, | |
| "learning_rate": 2.973807959975846e-05, | |
| "loss": 0.004, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7991360691144709, | |
| "grad_norm": 0.009106243029236794, | |
| "learning_rate": 2.447639663342025e-05, | |
| "loss": 0.0021, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8207343412526998, | |
| "grad_norm": 0.011521597392857075, | |
| "learning_rate": 1.9663070889806247e-05, | |
| "loss": 0.0054, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8423326133909287, | |
| "grad_norm": 0.0030731502920389175, | |
| "learning_rate": 1.5325940204991634e-05, | |
| "loss": 0.0011, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8639308855291576, | |
| "grad_norm": 0.04495791345834732, | |
| "learning_rate": 1.1490088344229713e-05, | |
| "loss": 0.0023, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8639308855291576, | |
| "eval_loss": 0.0008633602410554886, | |
| "eval_runtime": 28.3009, | |
| "eval_samples_per_second": 6.89, | |
| "eval_steps_per_second": 1.731, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8855291576673866, | |
| "grad_norm": 0.04856608808040619, | |
| "learning_rate": 8.177699930169379e-06, | |
| "loss": 0.0023, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.9071274298056156, | |
| "grad_norm": 0.004636272322386503, | |
| "learning_rate": 5.4079321382439e-06, | |
| "loss": 0.0015, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9287257019438445, | |
| "grad_norm": 0.12631860375404358, | |
| "learning_rate": 3.1968039012795417e-06, | |
| "loss": 0.0005, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9503239740820735, | |
| "grad_norm": 0.0010080524953082204, | |
| "learning_rate": 1.5571032641086151e-06, | |
| "loss": 0.0015, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.9719222462203023, | |
| "grad_norm": 0.02956795133650303, | |
| "learning_rate": 4.983134240007501e-07, | |
| "loss": 0.0003, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9719222462203023, | |
| "eval_loss": 0.0007766146445646882, | |
| "eval_runtime": 28.3411, | |
| "eval_samples_per_second": 6.88, | |
| "eval_steps_per_second": 1.729, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9935205183585313, | |
| "grad_norm": 0.060769353061914444, | |
| "learning_rate": 2.6557884657352427e-08, | |
| "loss": 0.0049, | |
| "step": 460 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 463, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "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": 2.3904763809103872e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |