| { | |
| "best_metric": 0.864070475101471, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-450", | |
| "epoch": 0.04722327162825841, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.444654325651681e-05, | |
| "eval_loss": 1.392892837524414, | |
| "eval_runtime": 256.0791, | |
| "eval_samples_per_second": 17.409, | |
| "eval_steps_per_second": 4.354, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0009444654325651682, | |
| "grad_norm": 0.7860835790634155, | |
| "learning_rate": 4.36e-05, | |
| "loss": 1.2044, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0018889308651303363, | |
| "grad_norm": 1.0544121265411377, | |
| "learning_rate": 8.72e-05, | |
| "loss": 1.1564, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0028333962976955043, | |
| "grad_norm": 1.270194411277771, | |
| "learning_rate": 0.0001308, | |
| "loss": 1.0678, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0037778617302606727, | |
| "grad_norm": 1.6109638214111328, | |
| "learning_rate": 0.0001744, | |
| "loss": 0.9977, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.004722327162825841, | |
| "grad_norm": 2.586062431335449, | |
| "learning_rate": 0.000218, | |
| "loss": 0.9803, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.004722327162825841, | |
| "eval_loss": 1.0977070331573486, | |
| "eval_runtime": 256.1997, | |
| "eval_samples_per_second": 17.4, | |
| "eval_steps_per_second": 4.352, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0056667925953910086, | |
| "grad_norm": 0.8779242634773254, | |
| "learning_rate": 0.00021773448147832086, | |
| "loss": 1.0668, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0066112580279561765, | |
| "grad_norm": 1.3890700340270996, | |
| "learning_rate": 0.0002169392194928312, | |
| "loss": 1.0106, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.007555723460521345, | |
| "grad_norm": 0.963046669960022, | |
| "learning_rate": 0.00021561808847998484, | |
| "loss": 1.0095, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.008500188893086513, | |
| "grad_norm": 1.1754647493362427, | |
| "learning_rate": 0.00021377752485727676, | |
| "loss": 0.9492, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.009444654325651681, | |
| "grad_norm": 1.9377000331878662, | |
| "learning_rate": 0.00021142649566566402, | |
| "loss": 0.7452, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.009444654325651681, | |
| "eval_loss": 1.0563247203826904, | |
| "eval_runtime": 256.3063, | |
| "eval_samples_per_second": 17.393, | |
| "eval_steps_per_second": 4.35, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.01038911975821685, | |
| "grad_norm": 0.8509911894798279, | |
| "learning_rate": 0.0002085764548830435, | |
| "loss": 1.0203, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.011333585190782017, | |
| "grad_norm": 0.8470553755760193, | |
| "learning_rate": 0.00020524128762162305, | |
| "loss": 0.9954, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.012278050623347185, | |
| "grad_norm": 0.9998864531517029, | |
| "learning_rate": 0.00020143724248105043, | |
| "loss": 0.9616, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.013222516055912353, | |
| "grad_norm": 1.3236018419265747, | |
| "learning_rate": 0.0001971828523868693, | |
| "loss": 1.0109, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.014166981488477521, | |
| "grad_norm": 1.5146818161010742, | |
| "learning_rate": 0.0001924988442999686, | |
| "loss": 0.8894, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.014166981488477521, | |
| "eval_loss": 1.04743492603302, | |
| "eval_runtime": 256.4077, | |
| "eval_samples_per_second": 17.386, | |
| "eval_steps_per_second": 4.349, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.01511144692104269, | |
| "grad_norm": 0.7381725907325745, | |
| "learning_rate": 0.00018740803823691298, | |
| "loss": 1.0541, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.016055912353607857, | |
| "grad_norm": 0.8298458456993103, | |
| "learning_rate": 0.00018193523609311556, | |
| "loss": 0.9556, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.017000377786173027, | |
| "grad_norm": 1.0353724956512451, | |
| "learning_rate": 0.00017610710081049675, | |
| "loss": 0.877, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.017944843218738193, | |
| "grad_norm": 1.3693569898605347, | |
| "learning_rate": 0.00016995202647831142, | |
| "loss": 0.9031, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.018889308651303362, | |
| "grad_norm": 1.9539626836776733, | |
| "learning_rate": 0.00016350000000000002, | |
| "loss": 0.878, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.018889308651303362, | |
| "eval_loss": 1.0043987035751343, | |
| "eval_runtime": 256.9248, | |
| "eval_samples_per_second": 17.351, | |
| "eval_steps_per_second": 4.34, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.019833774083868532, | |
| "grad_norm": 0.7500446438789368, | |
| "learning_rate": 0.00015678245500000943, | |
| "loss": 1.0322, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.0207782395164337, | |
| "grad_norm": 0.8743635416030884, | |
| "learning_rate": 0.00014983211868233444, | |
| "loss": 0.9676, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.021722704948998868, | |
| "grad_norm": 0.9597961902618408, | |
| "learning_rate": 0.00014268285238686927, | |
| "loss": 1.0018, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.022667170381564034, | |
| "grad_norm": 1.6248764991760254, | |
| "learning_rate": 0.00013536948662036378, | |
| "loss": 0.8842, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.023611635814129204, | |
| "grad_norm": 2.130323886871338, | |
| "learning_rate": 0.00012792765136569544, | |
| "loss": 0.8014, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.023611635814129204, | |
| "eval_loss": 0.9553887248039246, | |
| "eval_runtime": 256.8747, | |
| "eval_samples_per_second": 17.355, | |
| "eval_steps_per_second": 4.341, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.02455610124669437, | |
| "grad_norm": 0.6607593894004822, | |
| "learning_rate": 0.00012039360249617425, | |
| "loss": 0.9588, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.02550056667925954, | |
| "grad_norm": 1.0234694480895996, | |
| "learning_rate": 0.00011280404514057264, | |
| "loss": 0.9644, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.026445032111824706, | |
| "grad_norm": 1.1011518239974976, | |
| "learning_rate": 0.00010519595485942743, | |
| "loss": 0.8414, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.027389497544389876, | |
| "grad_norm": 1.0725549459457397, | |
| "learning_rate": 9.76063975038258e-05, | |
| "loss": 0.9655, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.028333962976955042, | |
| "grad_norm": 2.3623476028442383, | |
| "learning_rate": 9.00723486343046e-05, | |
| "loss": 0.7658, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.028333962976955042, | |
| "eval_loss": 0.9112545847892761, | |
| "eval_runtime": 256.7605, | |
| "eval_samples_per_second": 17.362, | |
| "eval_steps_per_second": 4.343, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.02927842840952021, | |
| "grad_norm": 0.5989169478416443, | |
| "learning_rate": 8.263051337963623e-05, | |
| "loss": 1.007, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.03022289384208538, | |
| "grad_norm": 0.7678573131561279, | |
| "learning_rate": 7.531714761313074e-05, | |
| "loss": 0.9324, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.031167359274650547, | |
| "grad_norm": 0.8592800498008728, | |
| "learning_rate": 6.816788131766559e-05, | |
| "loss": 0.8615, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.032111824707215714, | |
| "grad_norm": 1.0715807676315308, | |
| "learning_rate": 6.121754499999055e-05, | |
| "loss": 0.8406, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.03305629013978088, | |
| "grad_norm": 2.368269205093384, | |
| "learning_rate": 5.450000000000003e-05, | |
| "loss": 0.6401, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.03305629013978088, | |
| "eval_loss": 0.8866726160049438, | |
| "eval_runtime": 257.0405, | |
| "eval_samples_per_second": 17.344, | |
| "eval_steps_per_second": 4.338, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.03400075557234605, | |
| "grad_norm": 0.9228352308273315, | |
| "learning_rate": 4.804797352168861e-05, | |
| "loss": 0.9052, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.03494522100491122, | |
| "grad_norm": 0.7789422273635864, | |
| "learning_rate": 4.189289918950325e-05, | |
| "loss": 0.9935, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.035889686437476385, | |
| "grad_norm": 1.0351366996765137, | |
| "learning_rate": 3.606476390688449e-05, | |
| "loss": 0.9121, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.036834151870041555, | |
| "grad_norm": 0.9583853483200073, | |
| "learning_rate": 3.0591961763087043e-05, | |
| "loss": 0.8531, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.037778617302606725, | |
| "grad_norm": 1.4442740678787231, | |
| "learning_rate": 2.550115570003141e-05, | |
| "loss": 0.7264, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.037778617302606725, | |
| "eval_loss": 0.8729479908943176, | |
| "eval_runtime": 256.8229, | |
| "eval_samples_per_second": 17.358, | |
| "eval_steps_per_second": 4.342, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.038723082735171895, | |
| "grad_norm": 0.7448956966400146, | |
| "learning_rate": 2.081714761313074e-05, | |
| "loss": 0.917, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.039667548167737064, | |
| "grad_norm": 0.9566893577575684, | |
| "learning_rate": 1.656275751894957e-05, | |
| "loss": 1.004, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.04061201360030223, | |
| "grad_norm": 0.772722065448761, | |
| "learning_rate": 1.275871237837696e-05, | |
| "loss": 0.8142, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.0415564790328674, | |
| "grad_norm": 1.1294012069702148, | |
| "learning_rate": 9.423545116956494e-06, | |
| "loss": 0.7214, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.042500944465432566, | |
| "grad_norm": 3.888120174407959, | |
| "learning_rate": 6.573504334335994e-06, | |
| "loss": 0.7121, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.042500944465432566, | |
| "eval_loss": 0.864070475101471, | |
| "eval_runtime": 257.296, | |
| "eval_samples_per_second": 17.326, | |
| "eval_steps_per_second": 4.334, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.043445409897997736, | |
| "grad_norm": 0.5827313661575317, | |
| "learning_rate": 4.22247514272324e-06, | |
| "loss": 0.9635, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.0443898753305629, | |
| "grad_norm": 0.8031895756721497, | |
| "learning_rate": 2.38191152001518e-06, | |
| "loss": 0.9467, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.04533434076312807, | |
| "grad_norm": 0.6544033288955688, | |
| "learning_rate": 1.0607805071688306e-06, | |
| "loss": 0.8662, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.04627880619569324, | |
| "grad_norm": 1.268267273902893, | |
| "learning_rate": 2.655185216791625e-07, | |
| "loss": 0.8408, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.04722327162825841, | |
| "grad_norm": 2.5217607021331787, | |
| "learning_rate": 0.0, | |
| "loss": 0.7067, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04722327162825841, | |
| "eval_loss": 0.8645666837692261, | |
| "eval_runtime": 256.669, | |
| "eval_samples_per_second": 17.369, | |
| "eval_steps_per_second": 4.344, | |
| "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": 1 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.4194585739264e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |