| { | |
| "best_metric": 0.7672083377838135, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-450", | |
| "epoch": 0.17096939647803044, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00034193879295606086, | |
| "eval_loss": 4.036013603210449, | |
| "eval_runtime": 182.0782, | |
| "eval_samples_per_second": 6.766, | |
| "eval_steps_per_second": 1.692, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0034193879295606085, | |
| "grad_norm": 0.4218708574771881, | |
| "learning_rate": 4.0400000000000006e-05, | |
| "loss": 1.1371, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006838775859121217, | |
| "grad_norm": 0.48933151364326477, | |
| "learning_rate": 8.080000000000001e-05, | |
| "loss": 1.0934, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.010258163788681825, | |
| "grad_norm": 11.574728012084961, | |
| "learning_rate": 0.00012119999999999999, | |
| "loss": 6.2024, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.013677551718242434, | |
| "grad_norm": 5.524878025054932, | |
| "learning_rate": 0.00016160000000000002, | |
| "loss": 4.9511, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.017096939647803042, | |
| "grad_norm": 1.5277706384658813, | |
| "learning_rate": 0.000202, | |
| "loss": 1.4149, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.017096939647803042, | |
| "eval_loss": 2.1701488494873047, | |
| "eval_runtime": 181.9971, | |
| "eval_samples_per_second": 6.769, | |
| "eval_steps_per_second": 1.692, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02051632757736365, | |
| "grad_norm": 0.28429192304611206, | |
| "learning_rate": 0.00020175396907624226, | |
| "loss": 0.8977, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02393571550692426, | |
| "grad_norm": 0.2915061414241791, | |
| "learning_rate": 0.0002010170749428986, | |
| "loss": 0.8808, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.027355103436484868, | |
| "grad_norm": 9.972613334655762, | |
| "learning_rate": 0.00019979290767411438, | |
| "loss": 3.6864, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.030774491366045476, | |
| "grad_norm": 1.236708164215088, | |
| "learning_rate": 0.0001980874312897702, | |
| "loss": 2.9426, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.034193879295606085, | |
| "grad_norm": 1.045754313468933, | |
| "learning_rate": 0.00019590895469937675, | |
| "loss": 1.0005, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.034193879295606085, | |
| "eval_loss": 1.6803466081619263, | |
| "eval_runtime": 182.4878, | |
| "eval_samples_per_second": 6.751, | |
| "eval_steps_per_second": 1.688, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0376132672251667, | |
| "grad_norm": 0.32626786828041077, | |
| "learning_rate": 0.0001932680912219027, | |
| "loss": 0.8573, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.0410326551547273, | |
| "grad_norm": 0.27949294447898865, | |
| "learning_rate": 0.00019017770687875164, | |
| "loss": 0.8193, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.044452043084287914, | |
| "grad_norm": 5.386131763458252, | |
| "learning_rate": 0.000186652857711799, | |
| "loss": 2.9468, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.04787143101384852, | |
| "grad_norm": 0.46363380551338196, | |
| "learning_rate": 0.00018271071643186968, | |
| "loss": 1.296, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05129081894340913, | |
| "grad_norm": 1.1684648990631104, | |
| "learning_rate": 0.00017837048875501678, | |
| "loss": 0.8626, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.05129081894340913, | |
| "eval_loss": 1.3778539896011353, | |
| "eval_runtime": 181.9692, | |
| "eval_samples_per_second": 6.77, | |
| "eval_steps_per_second": 1.693, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.054710206872969736, | |
| "grad_norm": 0.3281489610671997, | |
| "learning_rate": 0.00017365331983420376, | |
| "loss": 0.8043, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.05812959480253035, | |
| "grad_norm": 0.2923721969127655, | |
| "learning_rate": 0.0001685821912422447, | |
| "loss": 0.7794, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.06154898273209095, | |
| "grad_norm": 6.542027950286865, | |
| "learning_rate": 0.00016318180900789148, | |
| "loss": 2.2373, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.06496837066165156, | |
| "grad_norm": 0.3875706195831299, | |
| "learning_rate": 0.00015747848325054544, | |
| "loss": 1.3301, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.06838775859121217, | |
| "grad_norm": 0.8955979943275452, | |
| "learning_rate": 0.0001515, | |
| "loss": 0.8364, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06838775859121217, | |
| "eval_loss": 1.3041934967041016, | |
| "eval_runtime": 182.0023, | |
| "eval_samples_per_second": 6.769, | |
| "eval_steps_per_second": 1.692, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.07180714652077279, | |
| "grad_norm": 0.32174670696258545, | |
| "learning_rate": 0.00014527548582569683, | |
| "loss": 0.8064, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.0752265344503334, | |
| "grad_norm": 0.29309213161468506, | |
| "learning_rate": 0.00013883526593500714, | |
| "loss": 0.7734, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.078645922379894, | |
| "grad_norm": 6.007425785064697, | |
| "learning_rate": 0.0001322107164318697, | |
| "loss": 2.042, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.0820653103094546, | |
| "grad_norm": 0.8852247595787048, | |
| "learning_rate": 0.00012543411145556643, | |
| "loss": 1.3562, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.08548469823901522, | |
| "grad_norm": 1.0789457559585571, | |
| "learning_rate": 0.00011853846594435998, | |
| "loss": 0.7506, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.08548469823901522, | |
| "eval_loss": 0.965094804763794, | |
| "eval_runtime": 182.2492, | |
| "eval_samples_per_second": 6.76, | |
| "eval_steps_per_second": 1.69, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.08890408616857583, | |
| "grad_norm": 0.33002084493637085, | |
| "learning_rate": 0.00011155737479003301, | |
| "loss": 0.8221, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.09232347409813643, | |
| "grad_norm": 0.34726133942604065, | |
| "learning_rate": 0.00010452484916695262, | |
| "loss": 0.7565, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.09574286202769704, | |
| "grad_norm": 2.3043720722198486, | |
| "learning_rate": 9.747515083304742e-05, | |
| "loss": 1.6552, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.09916224995725766, | |
| "grad_norm": 3.1896228790283203, | |
| "learning_rate": 9.044262520996702e-05, | |
| "loss": 0.7903, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.10258163788681826, | |
| "grad_norm": 1.102461814880371, | |
| "learning_rate": 8.346153405564004e-05, | |
| "loss": 0.7895, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.10258163788681826, | |
| "eval_loss": 0.8672501444816589, | |
| "eval_runtime": 181.9625, | |
| "eval_samples_per_second": 6.771, | |
| "eval_steps_per_second": 1.693, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.10600102581637887, | |
| "grad_norm": 0.28580135107040405, | |
| "learning_rate": 7.656588854443357e-05, | |
| "loss": 0.828, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.10942041374593947, | |
| "grad_norm": 0.3089773654937744, | |
| "learning_rate": 6.978928356813031e-05, | |
| "loss": 0.7365, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.11283980167550009, | |
| "grad_norm": 4.380869388580322, | |
| "learning_rate": 6.316473406499288e-05, | |
| "loss": 1.1384, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.1162591896050607, | |
| "grad_norm": 0.316327840089798, | |
| "learning_rate": 5.672451417430317e-05, | |
| "loss": 0.9138, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.1196785775346213, | |
| "grad_norm": 1.1205822229385376, | |
| "learning_rate": 5.050000000000002e-05, | |
| "loss": 0.6975, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.1196785775346213, | |
| "eval_loss": 0.8012982606887817, | |
| "eval_runtime": 182.4497, | |
| "eval_samples_per_second": 6.753, | |
| "eval_steps_per_second": 1.688, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.1230979654641819, | |
| "grad_norm": 0.3492680490016937, | |
| "learning_rate": 4.452151674945458e-05, | |
| "loss": 0.7909, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.1265173533937425, | |
| "grad_norm": 0.31265050172805786, | |
| "learning_rate": 3.8818190992108515e-05, | |
| "loss": 0.6951, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.12993674132330313, | |
| "grad_norm": 4.880677700042725, | |
| "learning_rate": 3.3417808757755355e-05, | |
| "loss": 0.9059, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.13335612925286375, | |
| "grad_norm": 0.3902217745780945, | |
| "learning_rate": 2.8346680165796253e-05, | |
| "loss": 0.7231, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.13677551718242434, | |
| "grad_norm": 0.9335039258003235, | |
| "learning_rate": 2.362951124498323e-05, | |
| "loss": 0.7406, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.13677551718242434, | |
| "eval_loss": 0.7848264575004578, | |
| "eval_runtime": 182.302, | |
| "eval_samples_per_second": 6.758, | |
| "eval_steps_per_second": 1.69, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14019490511198496, | |
| "grad_norm": 0.31434395909309387, | |
| "learning_rate": 1.928928356813032e-05, | |
| "loss": 0.7779, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.14361429304154558, | |
| "grad_norm": 0.28142374753952026, | |
| "learning_rate": 1.5347142288200977e-05, | |
| "loss": 0.7075, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.14703368097110617, | |
| "grad_norm": 2.3318331241607666, | |
| "learning_rate": 1.1822293121248375e-05, | |
| "loss": 0.7626, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.1504530689006668, | |
| "grad_norm": 0.3560838997364044, | |
| "learning_rate": 8.731908778097302e-06, | |
| "loss": 0.8547, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.15387245683022738, | |
| "grad_norm": 2.5311930179595947, | |
| "learning_rate": 6.09104530062326e-06, | |
| "loss": 0.8116, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.15387245683022738, | |
| "eval_loss": 0.7672083377838135, | |
| "eval_runtime": 182.3331, | |
| "eval_samples_per_second": 6.757, | |
| "eval_steps_per_second": 1.689, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.157291844759788, | |
| "grad_norm": 0.31098073720932007, | |
| "learning_rate": 3.912568710229791e-06, | |
| "loss": 0.7879, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.16071123268934862, | |
| "grad_norm": 6.655428886413574, | |
| "learning_rate": 2.2070923258856255e-06, | |
| "loss": 0.9999, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.1641306206189092, | |
| "grad_norm": 4.029115200042725, | |
| "learning_rate": 9.829250571013935e-07, | |
| "loss": 0.9647, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.16755000854846983, | |
| "grad_norm": 0.3827471435070038, | |
| "learning_rate": 2.4603092375775605e-07, | |
| "loss": 0.6842, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.17096939647803044, | |
| "grad_norm": 0.9424235224723816, | |
| "learning_rate": 0.0, | |
| "loss": 0.7972, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17096939647803044, | |
| "eval_loss": 0.779399037361145, | |
| "eval_runtime": 182.875, | |
| "eval_samples_per_second": 6.737, | |
| "eval_steps_per_second": 1.684, | |
| "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": 2.674007469981696e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |