| { | |
| "best_metric": 0.5798903703689575, | |
| "best_model_checkpoint": "models/checkpoints/checkpoint-16000", | |
| "epoch": 14.47952951820855, | |
| "global_step": 16000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.9879336349924587e-05, | |
| "loss": 1.3189, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.9758672699849173e-05, | |
| "loss": 1.1007, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.9638009049773755e-05, | |
| "loss": 1.0933, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.951734539969834e-05, | |
| "loss": 1.0187, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1.9396681749622927e-05, | |
| "loss": 1.018, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.9501454830169678, | |
| "eval_runtime": 52.6762, | |
| "eval_samples_per_second": 149.194, | |
| "eval_steps_per_second": 9.34, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.9276018099547512e-05, | |
| "loss": 1.0293, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.9155354449472098e-05, | |
| "loss": 0.9875, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.9034690799396684e-05, | |
| "loss": 0.9752, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.891402714932127e-05, | |
| "loss": 0.979, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.8793363499245855e-05, | |
| "loss": 0.9513, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.8483895063400269, | |
| "eval_runtime": 32.7078, | |
| "eval_samples_per_second": 240.279, | |
| "eval_steps_per_second": 15.042, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.867269984917044e-05, | |
| "loss": 0.9691, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.8552036199095026e-05, | |
| "loss": 0.8995, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.843137254901961e-05, | |
| "loss": 0.9392, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.8310708898944194e-05, | |
| "loss": 0.9156, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 1.819004524886878e-05, | |
| "loss": 0.9142, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 0.8270823359489441, | |
| "eval_runtime": 30.0805, | |
| "eval_samples_per_second": 261.266, | |
| "eval_steps_per_second": 16.356, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.8069381598793365e-05, | |
| "loss": 0.8579, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.794871794871795e-05, | |
| "loss": 0.8649, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.7828054298642537e-05, | |
| "loss": 0.8935, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 1.770739064856712e-05, | |
| "loss": 0.8534, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.7586726998491705e-05, | |
| "loss": 0.8427, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "eval_loss": 0.8167164921760559, | |
| "eval_runtime": 29.67, | |
| "eval_samples_per_second": 264.88, | |
| "eval_steps_per_second": 16.582, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.746606334841629e-05, | |
| "loss": 0.8634, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.7345399698340876e-05, | |
| "loss": 0.8735, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.7224736048265462e-05, | |
| "loss": 0.8401, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.7104072398190047e-05, | |
| "loss": 0.8403, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 1.6983408748114633e-05, | |
| "loss": 0.8212, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 0.8272273540496826, | |
| "eval_runtime": 30.3829, | |
| "eval_samples_per_second": 258.665, | |
| "eval_steps_per_second": 16.193, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.686274509803922e-05, | |
| "loss": 0.8301, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 1.6742081447963804e-05, | |
| "loss": 0.8396, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 1.662141779788839e-05, | |
| "loss": 0.8067, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 1.6500754147812972e-05, | |
| "loss": 0.8231, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 1.6380090497737558e-05, | |
| "loss": 0.8125, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "eval_loss": 0.7901880741119385, | |
| "eval_runtime": 30.2297, | |
| "eval_samples_per_second": 259.976, | |
| "eval_steps_per_second": 16.275, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 1.6259426847662144e-05, | |
| "loss": 0.808, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.613876319758673e-05, | |
| "loss": 0.8359, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 1.6018099547511315e-05, | |
| "loss": 0.8017, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 1.5897435897435897e-05, | |
| "loss": 0.8226, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 1.5776772247360483e-05, | |
| "loss": 0.8121, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "eval_loss": 0.7621824741363525, | |
| "eval_runtime": 30.3432, | |
| "eval_samples_per_second": 259.004, | |
| "eval_steps_per_second": 16.215, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 1.565610859728507e-05, | |
| "loss": 0.7861, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 1.5535444947209654e-05, | |
| "loss": 0.7634, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 1.541478129713424e-05, | |
| "loss": 0.7941, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 1.5294117647058822e-05, | |
| "loss": 0.7628, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 1.517345399698341e-05, | |
| "loss": 0.7572, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "eval_loss": 0.7411866188049316, | |
| "eval_runtime": 29.8313, | |
| "eval_samples_per_second": 263.448, | |
| "eval_steps_per_second": 16.493, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 1.5052790346907995e-05, | |
| "loss": 0.7975, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.4932126696832581e-05, | |
| "loss": 0.7626, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.4811463046757167e-05, | |
| "loss": 0.7467, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 1.4690799396681752e-05, | |
| "loss": 0.7689, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 1.4570135746606336e-05, | |
| "loss": 0.7392, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "eval_loss": 0.7343409061431885, | |
| "eval_runtime": 30.138, | |
| "eval_samples_per_second": 260.767, | |
| "eval_steps_per_second": 16.325, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 1.444947209653092e-05, | |
| "loss": 0.7636, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 1.4328808446455506e-05, | |
| "loss": 0.768, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 1.4208144796380091e-05, | |
| "loss": 0.7234, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 1.4087481146304677e-05, | |
| "loss": 0.7868, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 1.3966817496229261e-05, | |
| "loss": 0.7538, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "eval_loss": 0.7230245471000671, | |
| "eval_runtime": 29.8163, | |
| "eval_samples_per_second": 263.581, | |
| "eval_steps_per_second": 16.501, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 0.7261, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 1.3725490196078432e-05, | |
| "loss": 0.7462, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 1.3604826546003018e-05, | |
| "loss": 0.7169, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 1.3484162895927604e-05, | |
| "loss": 0.7542, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 1.3363499245852188e-05, | |
| "loss": 0.71, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_loss": 0.7268965840339661, | |
| "eval_runtime": 29.6638, | |
| "eval_samples_per_second": 264.936, | |
| "eval_steps_per_second": 16.586, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 1.3242835595776773e-05, | |
| "loss": 0.7217, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 1.3122171945701359e-05, | |
| "loss": 0.7226, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 1.3001508295625945e-05, | |
| "loss": 0.7177, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 1.288084464555053e-05, | |
| "loss": 0.7038, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 1.2760180995475113e-05, | |
| "loss": 0.6966, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "eval_loss": 0.7119535803794861, | |
| "eval_runtime": 29.7811, | |
| "eval_samples_per_second": 263.892, | |
| "eval_steps_per_second": 16.521, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 1.2639517345399698e-05, | |
| "loss": 0.7167, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 1.2518853695324284e-05, | |
| "loss": 0.6907, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 1.239819004524887e-05, | |
| "loss": 0.6992, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 1.2277526395173455e-05, | |
| "loss": 0.7038, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 1.215686274509804e-05, | |
| "loss": 0.6995, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_loss": 0.7289978861808777, | |
| "eval_runtime": 30.2132, | |
| "eval_samples_per_second": 260.118, | |
| "eval_steps_per_second": 16.284, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 1.2036199095022625e-05, | |
| "loss": 0.6841, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 1.191553544494721e-05, | |
| "loss": 0.6827, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 1.1794871794871796e-05, | |
| "loss": 0.6664, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 1.1674208144796382e-05, | |
| "loss": 0.6911, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 1.1553544494720966e-05, | |
| "loss": 0.6719, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "eval_loss": 0.7228043675422668, | |
| "eval_runtime": 31.0054, | |
| "eval_samples_per_second": 253.472, | |
| "eval_steps_per_second": 15.868, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 1.1432880844645552e-05, | |
| "loss": 0.6785, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 1.1312217194570137e-05, | |
| "loss": 0.6908, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 1.1191553544494723e-05, | |
| "loss": 0.6775, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "learning_rate": 1.1070889894419309e-05, | |
| "loss": 0.6566, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 1.0950226244343893e-05, | |
| "loss": 0.6862, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "eval_loss": 0.6712033152580261, | |
| "eval_runtime": 29.8128, | |
| "eval_samples_per_second": 263.612, | |
| "eval_steps_per_second": 16.503, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 1.0829562594268476e-05, | |
| "loss": 0.6965, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 1.0708898944193062e-05, | |
| "loss": 0.6836, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 1.0588235294117648e-05, | |
| "loss": 0.6684, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "learning_rate": 1.0467571644042233e-05, | |
| "loss": 0.6767, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 1.0346907993966819e-05, | |
| "loss": 0.6775, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "eval_loss": 0.6799022555351257, | |
| "eval_runtime": 30.3013, | |
| "eval_samples_per_second": 259.362, | |
| "eval_steps_per_second": 16.237, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 1.0226244343891403e-05, | |
| "loss": 0.6371, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 1.0105580693815989e-05, | |
| "loss": 0.6623, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 9.984917043740574e-06, | |
| "loss": 0.6499, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 9.86425339366516e-06, | |
| "loss": 0.6457, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 9.743589743589744e-06, | |
| "loss": 0.6283, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "eval_loss": 0.6853426098823547, | |
| "eval_runtime": 29.9239, | |
| "eval_samples_per_second": 262.633, | |
| "eval_steps_per_second": 16.442, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 9.62292609351433e-06, | |
| "loss": 0.6151, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 9.502262443438914e-06, | |
| "loss": 0.6431, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 9.3815987933635e-06, | |
| "loss": 0.6293, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 9.260935143288085e-06, | |
| "loss": 0.6266, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 9.14027149321267e-06, | |
| "loss": 0.6028, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "eval_loss": 0.6904362440109253, | |
| "eval_runtime": 31.3612, | |
| "eval_samples_per_second": 250.596, | |
| "eval_steps_per_second": 15.688, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 9.019607843137256e-06, | |
| "loss": 0.6494, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 8.89894419306184e-06, | |
| "loss": 0.6014, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 8.778280542986426e-06, | |
| "loss": 0.6208, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 8.657616892911012e-06, | |
| "loss": 0.6461, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 8.536953242835596e-06, | |
| "loss": 0.6291, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "eval_loss": 0.6408899426460266, | |
| "eval_runtime": 31.2717, | |
| "eval_samples_per_second": 251.314, | |
| "eval_steps_per_second": 15.733, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 8.416289592760181e-06, | |
| "loss": 0.6056, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 8.295625942684767e-06, | |
| "loss": 0.6052, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 8.174962292609353e-06, | |
| "loss": 0.6128, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 8.054298642533938e-06, | |
| "loss": 0.5895, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 7.933634992458522e-06, | |
| "loss": 0.6304, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "eval_loss": 0.6174328327178955, | |
| "eval_runtime": 30.7682, | |
| "eval_samples_per_second": 255.426, | |
| "eval_steps_per_second": 15.991, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "learning_rate": 7.812971342383108e-06, | |
| "loss": 0.5882, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.6304, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 7.5716440422322776e-06, | |
| "loss": 0.6103, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 7.450980392156863e-06, | |
| "loss": 0.5952, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 7.330316742081448e-06, | |
| "loss": 0.6039, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "eval_loss": 0.6438130140304565, | |
| "eval_runtime": 31.9863, | |
| "eval_samples_per_second": 245.699, | |
| "eval_steps_per_second": 15.382, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 7.209653092006034e-06, | |
| "loss": 0.6079, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 7.088989441930619e-06, | |
| "loss": 0.6093, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 6.968325791855204e-06, | |
| "loss": 0.5787, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 6.84766214177979e-06, | |
| "loss": 0.6138, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 6.7269984917043755e-06, | |
| "loss": 0.6011, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "eval_loss": 0.6626730561256409, | |
| "eval_runtime": 30.1966, | |
| "eval_samples_per_second": 260.261, | |
| "eval_steps_per_second": 16.293, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 6.6063348416289595e-06, | |
| "loss": 0.6087, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 6.485671191553545e-06, | |
| "loss": 0.5929, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 10.23, | |
| "learning_rate": 6.36500754147813e-06, | |
| "loss": 0.5849, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 6.244343891402716e-06, | |
| "loss": 0.6042, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 6.123680241327301e-06, | |
| "loss": 0.637, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "eval_loss": 0.6357372403144836, | |
| "eval_runtime": 29.8866, | |
| "eval_samples_per_second": 262.961, | |
| "eval_steps_per_second": 16.462, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 6.003016591251885e-06, | |
| "loss": 0.5783, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.5739, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "learning_rate": 5.761689291101056e-06, | |
| "loss": 0.5617, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 5.641025641025641e-06, | |
| "loss": 0.5724, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "learning_rate": 5.520361990950227e-06, | |
| "loss": 0.5635, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 10.86, | |
| "eval_loss": 0.6293562054634094, | |
| "eval_runtime": 42.9677, | |
| "eval_samples_per_second": 182.905, | |
| "eval_steps_per_second": 11.45, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 10.95, | |
| "learning_rate": 5.399698340874812e-06, | |
| "loss": 0.6074, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 5.2790346907993975e-06, | |
| "loss": 0.5755, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 5.158371040723983e-06, | |
| "loss": 0.5653, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 5.037707390648567e-06, | |
| "loss": 0.5685, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 4.917043740573153e-06, | |
| "loss": 0.596, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "eval_loss": 0.619467556476593, | |
| "eval_runtime": 30.8886, | |
| "eval_samples_per_second": 254.431, | |
| "eval_steps_per_second": 15.928, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 4.7963800904977385e-06, | |
| "loss": 0.5765, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 4.675716440422323e-06, | |
| "loss": 0.563, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "learning_rate": 4.555052790346908e-06, | |
| "loss": 0.5871, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 4.434389140271493e-06, | |
| "loss": 0.5345, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 4.313725490196079e-06, | |
| "loss": 0.5647, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "eval_loss": 0.6173378825187683, | |
| "eval_runtime": 30.5561, | |
| "eval_samples_per_second": 257.199, | |
| "eval_steps_per_second": 16.102, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 4.193061840120664e-06, | |
| "loss": 0.5798, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 11.95, | |
| "learning_rate": 4.072398190045249e-06, | |
| "loss": 0.5553, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 3.951734539969834e-06, | |
| "loss": 0.5372, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 12.13, | |
| "learning_rate": 3.83107088989442e-06, | |
| "loss": 0.5432, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 3.710407239819005e-06, | |
| "loss": 0.5304, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "eval_loss": 0.6232503056526184, | |
| "eval_runtime": 30.2832, | |
| "eval_samples_per_second": 259.517, | |
| "eval_steps_per_second": 16.247, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 3.58974358974359e-06, | |
| "loss": 0.5423, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "learning_rate": 3.4690799396681753e-06, | |
| "loss": 0.546, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 3.34841628959276e-06, | |
| "loss": 0.5932, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 12.58, | |
| "learning_rate": 3.2277526395173458e-06, | |
| "loss": 0.5721, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 3.107088989441931e-06, | |
| "loss": 0.5815, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "eval_loss": 0.6265811324119568, | |
| "eval_runtime": 31.274, | |
| "eval_samples_per_second": 251.295, | |
| "eval_steps_per_second": 15.732, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 2.986425339366516e-06, | |
| "loss": 0.5646, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 2.865761689291101e-06, | |
| "loss": 0.5706, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "learning_rate": 2.7450980392156867e-06, | |
| "loss": 0.5561, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 2.624434389140272e-06, | |
| "loss": 0.5668, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 2.503770739064857e-06, | |
| "loss": 0.5555, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "eval_loss": 0.6143834590911865, | |
| "eval_runtime": 30.6444, | |
| "eval_samples_per_second": 256.458, | |
| "eval_steps_per_second": 16.055, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "learning_rate": 2.3831070889894425e-06, | |
| "loss": 0.5327, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 13.3, | |
| "learning_rate": 2.2624434389140273e-06, | |
| "loss": 0.5491, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "learning_rate": 2.1417797888386125e-06, | |
| "loss": 0.5633, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 13.48, | |
| "learning_rate": 2.0211161387631978e-06, | |
| "loss": 0.5612, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 13.57, | |
| "learning_rate": 1.9004524886877828e-06, | |
| "loss": 0.5725, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 13.57, | |
| "eval_loss": 0.6175289750099182, | |
| "eval_runtime": 30.1245, | |
| "eval_samples_per_second": 260.884, | |
| "eval_steps_per_second": 16.332, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 1.7797888386123682e-06, | |
| "loss": 0.5502, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 1.6591251885369533e-06, | |
| "loss": 0.5529, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 13.85, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.5351, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 1.4177978883861237e-06, | |
| "loss": 0.5563, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "learning_rate": 1.2971342383107092e-06, | |
| "loss": 0.5616, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 14.03, | |
| "eval_loss": 0.6149749755859375, | |
| "eval_runtime": 31.0977, | |
| "eval_samples_per_second": 252.719, | |
| "eval_steps_per_second": 15.821, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 0.5404, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 1.0558069381598795e-06, | |
| "loss": 0.5461, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 14.3, | |
| "learning_rate": 9.351432880844646e-07, | |
| "loss": 0.5449, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "learning_rate": 8.144796380090498e-07, | |
| "loss": 0.5296, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 6.938159879336351e-07, | |
| "loss": 0.5471, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "eval_loss": 0.5798903703689575, | |
| "eval_runtime": 29.2689, | |
| "eval_samples_per_second": 268.51, | |
| "eval_steps_per_second": 16.81, | |
| "step": 16000 | |
| } | |
| ], | |
| "max_steps": 16575, | |
| "num_train_epochs": 15, | |
| "total_flos": 5.59327922919324e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |