| { | |
| "best_metric": 0.727655291557312, | |
| "best_model_checkpoint": "deberta_v3_finetuned_predicting_effective_arguments/checkpoint-10000", | |
| "epoch": 4.835287009063444, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2e-07, | |
| "loss": 1.0361, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 3.98e-07, | |
| "loss": 0.9751, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 5.979999999999999e-07, | |
| "loss": 0.9807, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 7.94e-07, | |
| "loss": 0.9513, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 9.94e-07, | |
| "loss": 0.9237, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 0.9464540481567383, | |
| "eval_runtime": 37.5769, | |
| "eval_samples_per_second": 97.587, | |
| "eval_steps_per_second": 24.403, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 9.955915919781764e-07, | |
| "loss": 0.9367, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 9.81900265076038e-07, | |
| "loss": 0.8993, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 9.594490241150311e-07, | |
| "loss": 0.8586, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 9.282057505552949e-07, | |
| "loss": 0.854, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 8.889381125453379e-07, | |
| "loss": 0.8642, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.8270628452301025, | |
| "eval_runtime": 28.2214, | |
| "eval_samples_per_second": 129.937, | |
| "eval_steps_per_second": 32.493, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 8.423819662432867e-07, | |
| "loss": 0.8314, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 7.894097508558568e-07, | |
| "loss": 0.8317, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 7.310141395581585e-07, | |
| "loss": 0.7939, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 6.682894372882701e-07, | |
| "loss": 0.803, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 6.024110740127264e-07, | |
| "loss": 0.8029, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 0.7635419368743896, | |
| "eval_runtime": 28.2809, | |
| "eval_samples_per_second": 129.663, | |
| "eval_steps_per_second": 32.425, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 5.346135777490083e-07, | |
| "loss": 0.7959, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.6616744011972247e-07, | |
| "loss": 0.7708, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 3.9835530796656867e-07, | |
| "loss": 0.7636, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 3.3244794718149894e-07, | |
| "loss": 0.7864, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2.696804291810131e-07, | |
| "loss": 0.8005, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 0.7537589073181152, | |
| "eval_runtime": 28.4034, | |
| "eval_samples_per_second": 129.104, | |
| "eval_steps_per_second": 32.285, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 2.1122898627730768e-07, | |
| "loss": 0.7924, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.5818896966520534e-07, | |
| "loss": 0.8072, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 1.1155432308130763e-07, | |
| "loss": 0.7997, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 7.219895678887305e-08, | |
| "loss": 0.7931, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.0860370930748286e-08, | |
| "loss": 0.7734, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "eval_loss": 0.7621562480926514, | |
| "eval_runtime": 28.9226, | |
| "eval_samples_per_second": 126.786, | |
| "eval_steps_per_second": 31.705, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.812583514064109e-08, | |
| "loss": 0.7487, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 4.421383399995704e-09, | |
| "loss": 0.7911, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 9.999961696285757e-07, | |
| "loss": 0.7794, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 9.950440411813353e-07, | |
| "loss": 0.7922, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 9.810025090182484e-07, | |
| "loss": 0.7842, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 0.748653769493103, | |
| "eval_runtime": 27.9531, | |
| "eval_samples_per_second": 131.184, | |
| "eval_steps_per_second": 32.805, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 9.581263257645987e-07, | |
| "loss": 0.7731, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 9.264785067160785e-07, | |
| "loss": 0.7849, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 8.868386909145071e-07, | |
| "loss": 0.7745, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 8.399497089461603e-07, | |
| "loss": 0.7611, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 7.872511787610445e-07, | |
| "loss": 0.7543, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "eval_loss": 0.7955728769302368, | |
| "eval_runtime": 28.0818, | |
| "eval_samples_per_second": 130.583, | |
| "eval_steps_per_second": 32.655, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 7.286677103241362e-07, | |
| "loss": 0.7793, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 6.657991218409446e-07, | |
| "loss": 0.7359, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 5.998235396346351e-07, | |
| "loss": 0.7602, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 5.319773135574745e-07, | |
| "loss": 0.7664, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 4.635318484047261e-07, | |
| "loss": 0.7883, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_loss": 0.7467952370643616, | |
| "eval_runtime": 28.6183, | |
| "eval_samples_per_second": 128.135, | |
| "eval_steps_per_second": 32.042, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 3.9576977841621134e-07, | |
| "loss": 0.7493, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 3.299609313436215e-07, | |
| "loss": 0.7312, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.6733853250510007e-07, | |
| "loss": 0.7387, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 2.0907609475136373e-07, | |
| "loss": 0.7492, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1.562654274139964e-07, | |
| "loss": 0.7515, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_loss": 0.7437878847122192, | |
| "eval_runtime": 27.6555, | |
| "eval_samples_per_second": 132.595, | |
| "eval_steps_per_second": 33.158, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.0989617633737314e-07, | |
| "loss": 0.7218, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 7.083727840391568e-08, | |
| "loss": 0.7554, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.00889953873248e-08, | |
| "loss": 0.768, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 1.7607339075303706e-08, | |
| "loss": 0.725, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 4.165483173136253e-09, | |
| "loss": 0.716, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_loss": 0.7440274953842163, | |
| "eval_runtime": 28.4825, | |
| "eval_samples_per_second": 128.746, | |
| "eval_steps_per_second": 32.195, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 9.999846785729901e-07, | |
| "loss": 0.7534, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 9.947653741800156e-07, | |
| "loss": 0.7644, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 9.80274410836591e-07, | |
| "loss": 0.7631, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 9.567833420446291e-07, | |
| "loss": 0.7624, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 9.24732378840687e-07, | |
| "loss": 0.743, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "eval_loss": 0.7312911152839661, | |
| "eval_runtime": 28.3334, | |
| "eval_samples_per_second": 129.423, | |
| "eval_steps_per_second": 32.365, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 8.847221404583831e-07, | |
| "loss": 0.7522, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 8.380074821716343e-07, | |
| "loss": 0.7367, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 7.845215744177952e-07, | |
| "loss": 0.7557, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 7.257038727678722e-07, | |
| "loss": 0.7329, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 6.626565919248464e-07, | |
| "loss": 0.726, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 0.7325854301452637, | |
| "eval_runtime": 27.7597, | |
| "eval_samples_per_second": 132.098, | |
| "eval_steps_per_second": 33.034, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 5.965612068185233e-07, | |
| "loss": 0.7804, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 5.286563123483964e-07, | |
| "loss": 0.7319, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.602144127264582e-07, | |
| "loss": 0.732, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 3.925180753762438e-07, | |
| "loss": 0.7665, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 3.268358962522407e-07, | |
| "loss": 0.7499, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_loss": 0.7382772564888, | |
| "eval_runtime": 27.6004, | |
| "eval_samples_per_second": 132.86, | |
| "eval_steps_per_second": 33.224, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 2.64398726977746e-07, | |
| "loss": 0.739, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 2.063766092928542e-07, | |
| "loss": 0.7201, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 1.5385684904965596e-07, | |
| "loss": 0.7013, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 1.0782364063714005e-07, | |
| "loss": 0.7234, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 6.913962366403719e-08, | |
| "loss": 0.7377, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "eval_loss": 0.7334151864051819, | |
| "eval_runtime": 27.6353, | |
| "eval_samples_per_second": 132.693, | |
| "eval_steps_per_second": 33.182, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 3.852971751825379e-08, | |
| "loss": 0.7272, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 1.6567536735284103e-08, | |
| "loss": 0.7323, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 3.664641744888397e-09, | |
| "loss": 0.6981, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 9.99937173568261e-07, | |
| "loss": 0.6917, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 9.94170412962655e-07, | |
| "loss": 0.7459, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "eval_loss": 0.7427003979682922, | |
| "eval_runtime": 27.4459, | |
| "eval_samples_per_second": 133.608, | |
| "eval_steps_per_second": 33.411, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 9.79143142686205e-07, | |
| "loss": 0.7406, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 9.551369663680862e-07, | |
| "loss": 0.7304, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 9.226017479056962e-07, | |
| "loss": 0.7278, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 8.821471812372215e-07, | |
| "loss": 0.7273, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 8.345313649781101e-07, | |
| "loss": 0.7186, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "eval_loss": 0.7306970953941345, | |
| "eval_runtime": 27.6504, | |
| "eval_samples_per_second": 132.62, | |
| "eval_steps_per_second": 33.164, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 7.806465960271379e-07, | |
| "loss": 0.6956, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 7.215026483628598e-07, | |
| "loss": 0.7794, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 6.582078503775666e-07, | |
| "loss": 0.7304, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 5.919483153501647e-07, | |
| "loss": 0.7477, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 5.239657142686591e-07, | |
| "loss": 0.723, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "eval_loss": 0.7324436902999878, | |
| "eval_runtime": 27.6456, | |
| "eval_samples_per_second": 132.643, | |
| "eval_steps_per_second": 33.17, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 4.555340075285952e-07, | |
| "loss": 0.7081, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 3.8793557154393083e-07, | |
| "loss": 0.717, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 3.2243716764587667e-07, | |
| "loss": 0.7229, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 2.602662036006858e-07, | |
| "loss": 0.713, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 2.025877325938261e-07, | |
| "loss": 0.7275, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_loss": 0.7325617074966431, | |
| "eval_runtime": 27.8844, | |
| "eval_samples_per_second": 131.507, | |
| "eval_steps_per_second": 32.886, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 1.504826207082261e-07, | |
| "loss": 0.693, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 1.049272920272809e-07, | |
| "loss": 0.689, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 6.677543092937843e-08, | |
| "loss": 0.7001, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 3.674198446393789e-08, | |
| "loss": 0.7398, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 1.538976459652064e-08, | |
| "loss": 0.7371, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "eval_loss": 0.7348757982254028, | |
| "eval_runtime": 27.6591, | |
| "eval_samples_per_second": 132.579, | |
| "eval_steps_per_second": 33.154, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 3.1189013903395455e-09, | |
| "loss": 0.7164, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 9.998406552336684e-07, | |
| "loss": 0.7045, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 9.934334446448806e-07, | |
| "loss": 0.7095, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 9.777795348078374e-07, | |
| "loss": 0.7265, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 9.531722722678393e-07, | |
| "loss": 0.7267, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "eval_loss": 0.727655291557312, | |
| "eval_runtime": 27.3264, | |
| "eval_samples_per_second": 134.193, | |
| "eval_steps_per_second": 33.557, | |
| "step": 10000 | |
| } | |
| ], | |
| "max_steps": 16544, | |
| "num_train_epochs": 8, | |
| "total_flos": 9147892997166300.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |