| { | |
| "best_metric": 11.72977352142334, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-500", | |
| "epoch": 0.5830903790087464, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011661807580174927, | |
| "eval_loss": 11.762678146362305, | |
| "eval_runtime": 1.6672, | |
| "eval_samples_per_second": 216.536, | |
| "eval_steps_per_second": 54.584, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011661807580174927, | |
| "grad_norm": 0.1031990498304367, | |
| "learning_rate": 4.0600000000000004e-05, | |
| "loss": 11.7629, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.023323615160349854, | |
| "grad_norm": 0.12429989129304886, | |
| "learning_rate": 8.120000000000001e-05, | |
| "loss": 11.7609, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03498542274052478, | |
| "grad_norm": 0.188057541847229, | |
| "learning_rate": 0.00012179999999999999, | |
| "loss": 11.7631, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04664723032069971, | |
| "grad_norm": 0.19203342497348785, | |
| "learning_rate": 0.00016240000000000002, | |
| "loss": 11.7586, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05830903790087463, | |
| "grad_norm": 0.3170674443244934, | |
| "learning_rate": 0.000203, | |
| "loss": 11.7389, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05830903790087463, | |
| "eval_loss": 11.746705055236816, | |
| "eval_runtime": 1.6333, | |
| "eval_samples_per_second": 221.02, | |
| "eval_steps_per_second": 55.714, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06997084548104957, | |
| "grad_norm": 0.2140980362892151, | |
| "learning_rate": 0.00020275275110137215, | |
| "loss": 11.7573, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08163265306122448, | |
| "grad_norm": 0.1588020920753479, | |
| "learning_rate": 0.00020201220897726938, | |
| "loss": 11.7488, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09329446064139942, | |
| "grad_norm": 0.20971915125846863, | |
| "learning_rate": 0.00020078198147448128, | |
| "loss": 11.7447, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.10495626822157435, | |
| "grad_norm": 0.19178971648216248, | |
| "learning_rate": 0.00019906806213773937, | |
| "loss": 11.7421, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.11661807580174927, | |
| "grad_norm": 0.22939780354499817, | |
| "learning_rate": 0.0001968788010097697, | |
| "loss": 11.7342, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11661807580174927, | |
| "eval_loss": 11.742330551147461, | |
| "eval_runtime": 1.6574, | |
| "eval_samples_per_second": 217.817, | |
| "eval_steps_per_second": 54.907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1282798833819242, | |
| "grad_norm": 0.19177477061748505, | |
| "learning_rate": 0.00019422486395072398, | |
| "loss": 11.7501, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13994169096209913, | |
| "grad_norm": 0.15727351605892181, | |
| "learning_rate": 0.0001911191806751811, | |
| "loss": 11.7447, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.15160349854227406, | |
| "grad_norm": 0.1680230051279068, | |
| "learning_rate": 0.00018757688175987723, | |
| "loss": 11.735, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.16326530612244897, | |
| "grad_norm": 0.1868034452199936, | |
| "learning_rate": 0.00018361522492905716, | |
| "loss": 11.735, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.1749271137026239, | |
| "grad_norm": 0.3084157109260559, | |
| "learning_rate": 0.00017925351097657625, | |
| "loss": 11.7273, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.1749271137026239, | |
| "eval_loss": 11.739017486572266, | |
| "eval_runtime": 1.6264, | |
| "eval_samples_per_second": 221.96, | |
| "eval_steps_per_second": 55.951, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.18658892128279883, | |
| "grad_norm": 0.24353830516338348, | |
| "learning_rate": 0.00017451298973437308, | |
| "loss": 11.7474, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.19825072886297376, | |
| "grad_norm": 0.2174033671617508, | |
| "learning_rate": 0.0001694167565454241, | |
| "loss": 11.742, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2099125364431487, | |
| "grad_norm": 0.1835821270942688, | |
| "learning_rate": 0.0001639896397455543, | |
| "loss": 11.739, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.22157434402332363, | |
| "grad_norm": 0.18794667720794678, | |
| "learning_rate": 0.0001582580797022808, | |
| "loss": 11.7329, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.23323615160349853, | |
| "grad_norm": 0.3341284394264221, | |
| "learning_rate": 0.00015225, | |
| "loss": 11.7255, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23323615160349853, | |
| "eval_loss": 11.735800743103027, | |
| "eval_runtime": 1.6348, | |
| "eval_samples_per_second": 220.82, | |
| "eval_steps_per_second": 55.664, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.24489795918367346, | |
| "grad_norm": 0.19055917859077454, | |
| "learning_rate": 0.00014599467139909136, | |
| "loss": 11.7397, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.2565597667638484, | |
| "grad_norm": 0.19719476997852325, | |
| "learning_rate": 0.0001395225692317151, | |
| "loss": 11.7356, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.26822157434402333, | |
| "grad_norm": 0.17899419367313385, | |
| "learning_rate": 0.00013286522492905717, | |
| "loss": 11.7271, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.27988338192419826, | |
| "grad_norm": 0.2608484923839569, | |
| "learning_rate": 0.00012605507240336626, | |
| "loss": 11.729, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.2915451895043732, | |
| "grad_norm": 0.29318031668663025, | |
| "learning_rate": 0.00011912529003319345, | |
| "loss": 11.7252, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2915451895043732, | |
| "eval_loss": 11.734283447265625, | |
| "eval_runtime": 1.6363, | |
| "eval_samples_per_second": 220.625, | |
| "eval_steps_per_second": 55.615, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.3032069970845481, | |
| "grad_norm": 0.1815534234046936, | |
| "learning_rate": 0.00011210963902166683, | |
| "loss": 11.7412, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.31486880466472306, | |
| "grad_norm": 0.2676528990268707, | |
| "learning_rate": 0.00010504229891530386, | |
| "loss": 11.7368, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.32653061224489793, | |
| "grad_norm": 0.2543658912181854, | |
| "learning_rate": 9.795770108469618e-05, | |
| "loss": 11.7335, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.33819241982507287, | |
| "grad_norm": 0.25447171926498413, | |
| "learning_rate": 9.08903609783332e-05, | |
| "loss": 11.7292, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3498542274052478, | |
| "grad_norm": 0.3657534718513489, | |
| "learning_rate": 8.387470996680658e-05, | |
| "loss": 11.7254, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3498542274052478, | |
| "eval_loss": 11.732454299926758, | |
| "eval_runtime": 1.6359, | |
| "eval_samples_per_second": 220.675, | |
| "eval_steps_per_second": 55.627, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.36151603498542273, | |
| "grad_norm": 0.19558283686637878, | |
| "learning_rate": 7.694492759663374e-05, | |
| "loss": 11.7372, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.37317784256559766, | |
| "grad_norm": 0.29391545057296753, | |
| "learning_rate": 7.013477507094284e-05, | |
| "loss": 11.7345, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.3848396501457726, | |
| "grad_norm": 0.25074857473373413, | |
| "learning_rate": 6.347743076828492e-05, | |
| "loss": 11.7324, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.3965014577259475, | |
| "grad_norm": 0.1810549795627594, | |
| "learning_rate": 5.700532860090863e-05, | |
| "loss": 11.7274, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.40816326530612246, | |
| "grad_norm": 0.3190612196922302, | |
| "learning_rate": 5.075000000000002e-05, | |
| "loss": 11.7232, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.40816326530612246, | |
| "eval_loss": 11.731415748596191, | |
| "eval_runtime": 1.6273, | |
| "eval_samples_per_second": 221.846, | |
| "eval_steps_per_second": 55.922, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4198250728862974, | |
| "grad_norm": 0.2629506587982178, | |
| "learning_rate": 4.4741920297719214e-05, | |
| "loss": 11.7342, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.4314868804664723, | |
| "grad_norm": 0.24675559997558594, | |
| "learning_rate": 3.901036025444568e-05, | |
| "loss": 11.7363, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.44314868804664725, | |
| "grad_norm": 0.23566007614135742, | |
| "learning_rate": 3.358324345457592e-05, | |
| "loss": 11.73, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.45481049562682213, | |
| "grad_norm": 0.24955187737941742, | |
| "learning_rate": 2.8487010265626928e-05, | |
| "loss": 11.7289, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.46647230320699706, | |
| "grad_norm": 0.33412033319473267, | |
| "learning_rate": 2.3746489023423744e-05, | |
| "loss": 11.7211, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.46647230320699706, | |
| "eval_loss": 11.730313301086426, | |
| "eval_runtime": 1.6308, | |
| "eval_samples_per_second": 221.361, | |
| "eval_steps_per_second": 55.8, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.478134110787172, | |
| "grad_norm": 0.21241024136543274, | |
| "learning_rate": 1.9384775070942844e-05, | |
| "loss": 11.7359, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.4897959183673469, | |
| "grad_norm": 0.28444772958755493, | |
| "learning_rate": 1.5423118240122765e-05, | |
| "loss": 11.7339, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.5014577259475219, | |
| "grad_norm": 0.2678728401660919, | |
| "learning_rate": 1.188081932481891e-05, | |
| "loss": 11.7255, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.5131195335276968, | |
| "grad_norm": 0.3324980139732361, | |
| "learning_rate": 8.775136049276001e-06, | |
| "loss": 11.7318, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.5247813411078717, | |
| "grad_norm": 0.48563843965530396, | |
| "learning_rate": 6.121198990230306e-06, | |
| "loss": 11.7233, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5247813411078717, | |
| "eval_loss": 11.73011302947998, | |
| "eval_runtime": 1.6604, | |
| "eval_samples_per_second": 217.421, | |
| "eval_steps_per_second": 54.807, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5364431486880467, | |
| "grad_norm": 0.25317201018333435, | |
| "learning_rate": 3.931937862260632e-06, | |
| "loss": 11.7334, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.5481049562682215, | |
| "grad_norm": 0.28131625056266785, | |
| "learning_rate": 2.2180185255187225e-06, | |
| "loss": 11.7322, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.5597667638483965, | |
| "grad_norm": 0.21770986914634705, | |
| "learning_rate": 9.877910227306082e-07, | |
| "loss": 11.7306, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.31401193141937256, | |
| "learning_rate": 2.472488986278439e-07, | |
| "loss": 11.7282, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.5830903790087464, | |
| "grad_norm": 0.3276008665561676, | |
| "learning_rate": 0.0, | |
| "loss": 11.7215, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5830903790087464, | |
| "eval_loss": 11.72977352142334, | |
| "eval_runtime": 1.6296, | |
| "eval_samples_per_second": 221.53, | |
| "eval_steps_per_second": 55.843, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "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": 25896615936000.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |