| { | |
| "best_metric": 0.6309406161308289, | |
| "best_model_checkpoint": "Action_agent/checkpoint-2600", | |
| "epoch": 10.0, | |
| "eval_steps": 100, | |
| "global_step": 3140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 6.1902642250061035, | |
| "learning_rate": 9.952229299363057e-06, | |
| "loss": 2.3093, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 8.609545707702637, | |
| "learning_rate": 9.904458598726116e-06, | |
| "loss": 2.2798, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 8.198623657226562, | |
| "learning_rate": 9.856687898089172e-06, | |
| "loss": 2.2163, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 7.1882829666137695, | |
| "learning_rate": 9.80891719745223e-06, | |
| "loss": 2.1529, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 8.259012222290039, | |
| "learning_rate": 9.761146496815288e-06, | |
| "loss": 2.114, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 9.213942527770996, | |
| "learning_rate": 9.713375796178345e-06, | |
| "loss": 2.039, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_accuracy": 0.4847619047619048, | |
| "eval_loss": 1.7706103324890137, | |
| "eval_runtime": 17.9171, | |
| "eval_samples_per_second": 58.603, | |
| "eval_steps_per_second": 7.367, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 8.229316711425781, | |
| "learning_rate": 9.665605095541401e-06, | |
| "loss": 1.9431, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 9.092002868652344, | |
| "learning_rate": 9.617834394904459e-06, | |
| "loss": 1.7759, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 12.155420303344727, | |
| "learning_rate": 9.570063694267517e-06, | |
| "loss": 1.7233, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 15.371952056884766, | |
| "learning_rate": 9.522292993630574e-06, | |
| "loss": 1.6441, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 15.928441047668457, | |
| "learning_rate": 9.47452229299363e-06, | |
| "loss": 1.5518, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 14.13963794708252, | |
| "learning_rate": 9.426751592356688e-06, | |
| "loss": 1.4362, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 22.86189842224121, | |
| "learning_rate": 9.378980891719746e-06, | |
| "loss": 1.3695, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_accuracy": 0.6457142857142857, | |
| "eval_loss": 1.0885976552963257, | |
| "eval_runtime": 13.391, | |
| "eval_samples_per_second": 78.411, | |
| "eval_steps_per_second": 9.857, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 18.267860412597656, | |
| "learning_rate": 9.331210191082803e-06, | |
| "loss": 1.2723, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 17.083765029907227, | |
| "learning_rate": 9.283439490445861e-06, | |
| "loss": 1.2886, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 13.116181373596191, | |
| "learning_rate": 9.235668789808919e-06, | |
| "loss": 1.1825, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 15.386600494384766, | |
| "learning_rate": 9.187898089171975e-06, | |
| "loss": 1.2512, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 36.78362274169922, | |
| "learning_rate": 9.140127388535032e-06, | |
| "loss": 1.1629, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 16.107894897460938, | |
| "learning_rate": 9.09235668789809e-06, | |
| "loss": 1.0824, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 15.064929962158203, | |
| "learning_rate": 9.044585987261148e-06, | |
| "loss": 1.099, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.68, | |
| "eval_loss": 0.9092212915420532, | |
| "eval_runtime": 14.0444, | |
| "eval_samples_per_second": 74.763, | |
| "eval_steps_per_second": 9.399, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 18.469070434570312, | |
| "learning_rate": 8.996815286624204e-06, | |
| "loss": 1.0649, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 17.13619613647461, | |
| "learning_rate": 8.949044585987261e-06, | |
| "loss": 1.1011, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 18.0528621673584, | |
| "learning_rate": 8.901273885350319e-06, | |
| "loss": 1.0787, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 14.162128448486328, | |
| "learning_rate": 8.853503184713377e-06, | |
| "loss": 1.0927, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 21.933330535888672, | |
| "learning_rate": 8.805732484076433e-06, | |
| "loss": 1.087, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 16.677528381347656, | |
| "learning_rate": 8.757961783439492e-06, | |
| "loss": 1.0011, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_accuracy": 0.7171428571428572, | |
| "eval_loss": 0.8182899951934814, | |
| "eval_runtime": 13.5557, | |
| "eval_samples_per_second": 77.458, | |
| "eval_steps_per_second": 9.738, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 13.870976448059082, | |
| "learning_rate": 8.710191082802548e-06, | |
| "loss": 0.9318, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 14.700624465942383, | |
| "learning_rate": 8.662420382165606e-06, | |
| "loss": 0.8846, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 17.18898582458496, | |
| "learning_rate": 8.614649681528664e-06, | |
| "loss": 1.0089, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 18.89067840576172, | |
| "learning_rate": 8.566878980891721e-06, | |
| "loss": 0.9356, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 15.696223258972168, | |
| "learning_rate": 8.519108280254777e-06, | |
| "loss": 0.8215, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 36.806602478027344, | |
| "learning_rate": 8.471337579617835e-06, | |
| "loss": 0.9668, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 13.624329566955566, | |
| "learning_rate": 8.423566878980893e-06, | |
| "loss": 0.8437, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_accuracy": 0.719047619047619, | |
| "eval_loss": 0.7674332857131958, | |
| "eval_runtime": 13.5992, | |
| "eval_samples_per_second": 77.211, | |
| "eval_steps_per_second": 9.706, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 15.652780532836914, | |
| "learning_rate": 8.37579617834395e-06, | |
| "loss": 0.8169, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 13.216598510742188, | |
| "learning_rate": 8.328025477707006e-06, | |
| "loss": 0.9283, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 19.80782127380371, | |
| "learning_rate": 8.280254777070064e-06, | |
| "loss": 0.8614, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 18.636619567871094, | |
| "learning_rate": 8.232484076433122e-06, | |
| "loss": 0.8656, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 17.433523178100586, | |
| "learning_rate": 8.18471337579618e-06, | |
| "loss": 0.8313, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 14.271307945251465, | |
| "learning_rate": 8.136942675159237e-06, | |
| "loss": 0.8857, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 16.59528923034668, | |
| "learning_rate": 8.089171974522295e-06, | |
| "loss": 0.8613, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_accuracy": 0.7409523809523809, | |
| "eval_loss": 0.7168479561805725, | |
| "eval_runtime": 13.4058, | |
| "eval_samples_per_second": 78.324, | |
| "eval_steps_per_second": 9.846, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 13.598006248474121, | |
| "learning_rate": 8.04140127388535e-06, | |
| "loss": 0.8694, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 13.958739280700684, | |
| "learning_rate": 7.993630573248408e-06, | |
| "loss": 0.9166, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 13.80545425415039, | |
| "learning_rate": 7.945859872611466e-06, | |
| "loss": 0.8781, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 14.67716121673584, | |
| "learning_rate": 7.898089171974524e-06, | |
| "loss": 0.7684, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 18.161645889282227, | |
| "learning_rate": 7.85031847133758e-06, | |
| "loss": 0.7923, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 11.349693298339844, | |
| "learning_rate": 7.802547770700637e-06, | |
| "loss": 0.7427, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_accuracy": 0.7352380952380952, | |
| "eval_loss": 0.7270055413246155, | |
| "eval_runtime": 13.6582, | |
| "eval_samples_per_second": 76.877, | |
| "eval_steps_per_second": 9.665, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 16.357065200805664, | |
| "learning_rate": 7.754777070063695e-06, | |
| "loss": 0.7645, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 15.481508255004883, | |
| "learning_rate": 7.707006369426753e-06, | |
| "loss": 0.7973, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 16.786632537841797, | |
| "learning_rate": 7.659235668789809e-06, | |
| "loss": 0.8256, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 11.738802909851074, | |
| "learning_rate": 7.611464968152867e-06, | |
| "loss": 0.8307, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 13.264825820922852, | |
| "learning_rate": 7.563694267515924e-06, | |
| "loss": 0.7431, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 15.430547714233398, | |
| "learning_rate": 7.515923566878982e-06, | |
| "loss": 0.7867, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 15.555388450622559, | |
| "learning_rate": 7.468152866242039e-06, | |
| "loss": 0.693, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_accuracy": 0.7676190476190476, | |
| "eval_loss": 0.6801217198371887, | |
| "eval_runtime": 13.4462, | |
| "eval_samples_per_second": 78.089, | |
| "eval_steps_per_second": 9.817, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 16.23971176147461, | |
| "learning_rate": 7.4203821656050955e-06, | |
| "loss": 0.7791, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 22.803543090820312, | |
| "learning_rate": 7.372611464968153e-06, | |
| "loss": 0.7564, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 15.14857292175293, | |
| "learning_rate": 7.32484076433121e-06, | |
| "loss": 0.6895, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 18.52122688293457, | |
| "learning_rate": 7.2770700636942685e-06, | |
| "loss": 0.7016, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "grad_norm": 11.38332748413086, | |
| "learning_rate": 7.2292993630573245e-06, | |
| "loss": 0.8174, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 22.539424896240234, | |
| "learning_rate": 7.181528662420383e-06, | |
| "loss": 0.7147, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 15.064950942993164, | |
| "learning_rate": 7.13375796178344e-06, | |
| "loss": 0.7789, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "eval_accuracy": 0.7590476190476191, | |
| "eval_loss": 0.6831705570220947, | |
| "eval_runtime": 13.5379, | |
| "eval_samples_per_second": 77.56, | |
| "eval_steps_per_second": 9.75, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 17.286598205566406, | |
| "learning_rate": 7.085987261146498e-06, | |
| "loss": 0.6743, | |
| "step": 915 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 19.45290756225586, | |
| "learning_rate": 7.0382165605095544e-06, | |
| "loss": 0.6263, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 26.688581466674805, | |
| "learning_rate": 6.990445859872612e-06, | |
| "loss": 0.7736, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 22.443763732910156, | |
| "learning_rate": 6.942675159235669e-06, | |
| "loss": 0.6392, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "grad_norm": 18.81976318359375, | |
| "learning_rate": 6.894904458598727e-06, | |
| "loss": 0.7653, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 12.312933921813965, | |
| "learning_rate": 6.8471337579617835e-06, | |
| "loss": 0.6863, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "eval_accuracy": 0.7752380952380953, | |
| "eval_loss": 0.665543794631958, | |
| "eval_runtime": 13.4956, | |
| "eval_samples_per_second": 77.803, | |
| "eval_steps_per_second": 9.781, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 15.371318817138672, | |
| "learning_rate": 6.799363057324841e-06, | |
| "loss": 0.7106, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 18.258623123168945, | |
| "learning_rate": 6.751592356687898e-06, | |
| "loss": 0.7305, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 16.52337074279785, | |
| "learning_rate": 6.7038216560509565e-06, | |
| "loss": 0.6947, | |
| "step": 1035 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "grad_norm": 18.67824363708496, | |
| "learning_rate": 6.6560509554140125e-06, | |
| "loss": 0.6669, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 16.26685905456543, | |
| "learning_rate": 6.608280254777071e-06, | |
| "loss": 0.6801, | |
| "step": 1065 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 13.744972229003906, | |
| "learning_rate": 6.560509554140128e-06, | |
| "loss": 0.6035, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "grad_norm": 12.479057312011719, | |
| "learning_rate": 6.5127388535031856e-06, | |
| "loss": 0.6437, | |
| "step": 1095 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_accuracy": 0.7771428571428571, | |
| "eval_loss": 0.6382023692131042, | |
| "eval_runtime": 13.3473, | |
| "eval_samples_per_second": 78.667, | |
| "eval_steps_per_second": 9.89, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "grad_norm": 14.826581954956055, | |
| "learning_rate": 6.464968152866242e-06, | |
| "loss": 0.7309, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "grad_norm": 12.955341339111328, | |
| "learning_rate": 6.4171974522293e-06, | |
| "loss": 0.6864, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "grad_norm": 14.903204917907715, | |
| "learning_rate": 6.369426751592357e-06, | |
| "loss": 0.6711, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 15.349693298339844, | |
| "learning_rate": 6.321656050955415e-06, | |
| "loss": 0.6362, | |
| "step": 1155 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "grad_norm": 25.346343994140625, | |
| "learning_rate": 6.2738853503184715e-06, | |
| "loss": 0.6359, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "grad_norm": 12.536116600036621, | |
| "learning_rate": 6.226114649681529e-06, | |
| "loss": 0.6991, | |
| "step": 1185 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "grad_norm": 19.788801193237305, | |
| "learning_rate": 6.178343949044586e-06, | |
| "loss": 0.6741, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "eval_accuracy": 0.7790476190476191, | |
| "eval_loss": 0.6445861458778381, | |
| "eval_runtime": 13.6114, | |
| "eval_samples_per_second": 77.141, | |
| "eval_steps_per_second": 9.698, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 16.279836654663086, | |
| "learning_rate": 6.1305732484076445e-06, | |
| "loss": 0.6977, | |
| "step": 1215 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 18.798139572143555, | |
| "learning_rate": 6.0828025477707005e-06, | |
| "loss": 0.653, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 17.142087936401367, | |
| "learning_rate": 6.035031847133759e-06, | |
| "loss": 0.6778, | |
| "step": 1245 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 17.632762908935547, | |
| "learning_rate": 5.987261146496816e-06, | |
| "loss": 0.5343, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 14.896882057189941, | |
| "learning_rate": 5.9394904458598736e-06, | |
| "loss": 0.5694, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "grad_norm": 19.62409019470215, | |
| "learning_rate": 5.89171974522293e-06, | |
| "loss": 0.5871, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_accuracy": 0.7838095238095238, | |
| "eval_loss": 0.6551438570022583, | |
| "eval_runtime": 13.8432, | |
| "eval_samples_per_second": 75.849, | |
| "eval_steps_per_second": 9.535, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 16.879796981811523, | |
| "learning_rate": 5.843949044585988e-06, | |
| "loss": 0.6344, | |
| "step": 1305 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 18.603700637817383, | |
| "learning_rate": 5.796178343949045e-06, | |
| "loss": 0.5768, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 16.433502197265625, | |
| "learning_rate": 5.748407643312103e-06, | |
| "loss": 0.5884, | |
| "step": 1335 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "grad_norm": 17.979280471801758, | |
| "learning_rate": 5.7006369426751594e-06, | |
| "loss": 0.6167, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "grad_norm": 20.778549194335938, | |
| "learning_rate": 5.652866242038217e-06, | |
| "loss": 0.6594, | |
| "step": 1365 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 14.834670066833496, | |
| "learning_rate": 5.605095541401274e-06, | |
| "loss": 0.6214, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "grad_norm": 19.214466094970703, | |
| "learning_rate": 5.5573248407643325e-06, | |
| "loss": 0.6051, | |
| "step": 1395 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "eval_accuracy": 0.7638095238095238, | |
| "eval_loss": 0.6970483660697937, | |
| "eval_runtime": 20.5305, | |
| "eval_samples_per_second": 51.143, | |
| "eval_steps_per_second": 6.429, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 16.332500457763672, | |
| "learning_rate": 5.5095541401273885e-06, | |
| "loss": 0.5996, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "grad_norm": 16.794343948364258, | |
| "learning_rate": 5.461783439490447e-06, | |
| "loss": 0.702, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "grad_norm": 21.159442901611328, | |
| "learning_rate": 5.414012738853504e-06, | |
| "loss": 0.5742, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "grad_norm": 26.400766372680664, | |
| "learning_rate": 5.3662420382165615e-06, | |
| "loss": 0.6288, | |
| "step": 1455 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "grad_norm": 19.17631721496582, | |
| "learning_rate": 5.318471337579618e-06, | |
| "loss": 0.5819, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "grad_norm": 18.10342025756836, | |
| "learning_rate": 5.270700636942676e-06, | |
| "loss": 0.5842, | |
| "step": 1485 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "grad_norm": 21.941911697387695, | |
| "learning_rate": 5.222929936305733e-06, | |
| "loss": 0.5175, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_accuracy": 0.7790476190476191, | |
| "eval_loss": 0.6552723050117493, | |
| "eval_runtime": 13.5024, | |
| "eval_samples_per_second": 77.764, | |
| "eval_steps_per_second": 9.776, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "grad_norm": 24.317623138427734, | |
| "learning_rate": 5.175159235668791e-06, | |
| "loss": 0.5984, | |
| "step": 1515 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "grad_norm": 14.877484321594238, | |
| "learning_rate": 5.1273885350318474e-06, | |
| "loss": 0.6142, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "grad_norm": 20.296701431274414, | |
| "learning_rate": 5.079617834394905e-06, | |
| "loss": 0.719, | |
| "step": 1545 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "grad_norm": 20.335296630859375, | |
| "learning_rate": 5.031847133757962e-06, | |
| "loss": 0.5651, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 17.09543228149414, | |
| "learning_rate": 4.98407643312102e-06, | |
| "loss": 0.4632, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "grad_norm": 15.416642189025879, | |
| "learning_rate": 4.9363057324840765e-06, | |
| "loss": 0.5795, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.7771428571428571, | |
| "eval_loss": 0.6666560173034668, | |
| "eval_runtime": 14.1067, | |
| "eval_samples_per_second": 74.433, | |
| "eval_steps_per_second": 9.357, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "grad_norm": 12.152099609375, | |
| "learning_rate": 4.888535031847134e-06, | |
| "loss": 0.6119, | |
| "step": 1605 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "grad_norm": 11.709696769714355, | |
| "learning_rate": 4.840764331210192e-06, | |
| "loss": 0.5521, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "grad_norm": 12.4248685836792, | |
| "learning_rate": 4.792993630573249e-06, | |
| "loss": 0.586, | |
| "step": 1635 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "grad_norm": 22.69182777404785, | |
| "learning_rate": 4.745222929936306e-06, | |
| "loss": 0.5848, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "grad_norm": 15.92928409576416, | |
| "learning_rate": 4.697452229299363e-06, | |
| "loss": 0.5922, | |
| "step": 1665 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "grad_norm": 25.377580642700195, | |
| "learning_rate": 4.649681528662421e-06, | |
| "loss": 0.6579, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "grad_norm": 12.89096450805664, | |
| "learning_rate": 4.601910828025479e-06, | |
| "loss": 0.4919, | |
| "step": 1695 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "eval_accuracy": 0.7904761904761904, | |
| "eval_loss": 0.6316953897476196, | |
| "eval_runtime": 13.547, | |
| "eval_samples_per_second": 77.508, | |
| "eval_steps_per_second": 9.744, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "grad_norm": 13.04831314086914, | |
| "learning_rate": 4.554140127388535e-06, | |
| "loss": 0.5459, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "grad_norm": 14.792088508605957, | |
| "learning_rate": 4.506369426751593e-06, | |
| "loss": 0.4729, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "grad_norm": 20.434284210205078, | |
| "learning_rate": 4.45859872611465e-06, | |
| "loss": 0.5285, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "grad_norm": 16.0216064453125, | |
| "learning_rate": 4.410828025477708e-06, | |
| "loss": 0.5891, | |
| "step": 1755 | |
| }, | |
| { | |
| "epoch": 5.64, | |
| "grad_norm": 14.537184715270996, | |
| "learning_rate": 4.3630573248407645e-06, | |
| "loss": 0.6203, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 5.68, | |
| "grad_norm": 16.755977630615234, | |
| "learning_rate": 4.315286624203822e-06, | |
| "loss": 0.5832, | |
| "step": 1785 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "grad_norm": 18.05998992919922, | |
| "learning_rate": 4.26751592356688e-06, | |
| "loss": 0.4986, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "eval_accuracy": 0.780952380952381, | |
| "eval_loss": 0.6485886573791504, | |
| "eval_runtime": 13.712, | |
| "eval_samples_per_second": 76.575, | |
| "eval_steps_per_second": 9.627, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "grad_norm": 13.940254211425781, | |
| "learning_rate": 4.219745222929937e-06, | |
| "loss": 0.5582, | |
| "step": 1815 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "grad_norm": 13.54953670501709, | |
| "learning_rate": 4.171974522292994e-06, | |
| "loss": 0.5189, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "grad_norm": 19.552183151245117, | |
| "learning_rate": 4.124203821656051e-06, | |
| "loss": 0.6037, | |
| "step": 1845 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "grad_norm": 13.757224082946777, | |
| "learning_rate": 4.076433121019109e-06, | |
| "loss": 0.5537, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "grad_norm": 24.593406677246094, | |
| "learning_rate": 4.0286624203821666e-06, | |
| "loss": 0.5527, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 22.236400604248047, | |
| "learning_rate": 3.980891719745223e-06, | |
| "loss": 0.5104, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_accuracy": 0.7742857142857142, | |
| "eval_loss": 0.6699539422988892, | |
| "eval_runtime": 13.5651, | |
| "eval_samples_per_second": 77.405, | |
| "eval_steps_per_second": 9.731, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 15.87308120727539, | |
| "learning_rate": 3.933121019108281e-06, | |
| "loss": 0.5268, | |
| "step": 1905 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "grad_norm": 13.48481273651123, | |
| "learning_rate": 3.885350318471338e-06, | |
| "loss": 0.5421, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "grad_norm": 13.895825386047363, | |
| "learning_rate": 3.837579617834396e-06, | |
| "loss": 0.6139, | |
| "step": 1935 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "grad_norm": 14.655675888061523, | |
| "learning_rate": 3.789808917197453e-06, | |
| "loss": 0.495, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "grad_norm": 21.782032012939453, | |
| "learning_rate": 3.7420382165605097e-06, | |
| "loss": 0.513, | |
| "step": 1965 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "grad_norm": 16.350772857666016, | |
| "learning_rate": 3.694267515923567e-06, | |
| "loss": 0.5182, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "grad_norm": 12.87532901763916, | |
| "learning_rate": 3.6464968152866242e-06, | |
| "loss": 0.4919, | |
| "step": 1995 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "eval_accuracy": 0.7819047619047619, | |
| "eval_loss": 0.6527658700942993, | |
| "eval_runtime": 13.9166, | |
| "eval_samples_per_second": 75.449, | |
| "eval_steps_per_second": 9.485, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 12.642027854919434, | |
| "learning_rate": 3.5987261146496815e-06, | |
| "loss": 0.5212, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "grad_norm": 13.786490440368652, | |
| "learning_rate": 3.5509554140127388e-06, | |
| "loss": 0.5004, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 28.24700927734375, | |
| "learning_rate": 3.5031847133757964e-06, | |
| "loss": 0.539, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "grad_norm": 10.891915321350098, | |
| "learning_rate": 3.4554140127388537e-06, | |
| "loss": 0.5316, | |
| "step": 2055 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "grad_norm": 21.343164443969727, | |
| "learning_rate": 3.407643312101911e-06, | |
| "loss": 0.5497, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "grad_norm": 15.246662139892578, | |
| "learning_rate": 3.3598726114649682e-06, | |
| "loss": 0.5212, | |
| "step": 2085 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "grad_norm": 18.424856185913086, | |
| "learning_rate": 3.3121019108280255e-06, | |
| "loss": 0.5144, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "eval_accuracy": 0.7876190476190477, | |
| "eval_loss": 0.6354712843894958, | |
| "eval_runtime": 13.6508, | |
| "eval_samples_per_second": 76.919, | |
| "eval_steps_per_second": 9.67, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "grad_norm": 15.570305824279785, | |
| "learning_rate": 3.2643312101910827e-06, | |
| "loss": 0.5892, | |
| "step": 2115 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "grad_norm": 16.673995971679688, | |
| "learning_rate": 3.2165605095541404e-06, | |
| "loss": 0.5079, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "grad_norm": 17.703060150146484, | |
| "learning_rate": 3.1687898089171977e-06, | |
| "loss": 0.496, | |
| "step": 2145 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "grad_norm": 14.203299522399902, | |
| "learning_rate": 3.121019108280255e-06, | |
| "loss": 0.5223, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "grad_norm": 14.10352897644043, | |
| "learning_rate": 3.0732484076433122e-06, | |
| "loss": 0.521, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "grad_norm": 13.882482528686523, | |
| "learning_rate": 3.0254777070063695e-06, | |
| "loss": 0.5554, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "eval_accuracy": 0.7771428571428571, | |
| "eval_loss": 0.6552413702011108, | |
| "eval_runtime": 13.0208, | |
| "eval_samples_per_second": 80.64, | |
| "eval_steps_per_second": 10.138, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 12.480643272399902, | |
| "learning_rate": 2.9777070063694267e-06, | |
| "loss": 0.4216, | |
| "step": 2205 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "grad_norm": 14.39759349822998, | |
| "learning_rate": 2.9299363057324844e-06, | |
| "loss": 0.4888, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "grad_norm": 17.724123001098633, | |
| "learning_rate": 2.8821656050955417e-06, | |
| "loss": 0.4579, | |
| "step": 2235 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "grad_norm": 14.149361610412598, | |
| "learning_rate": 2.834394904458599e-06, | |
| "loss": 0.5295, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "grad_norm": 18.39142608642578, | |
| "learning_rate": 2.786624203821656e-06, | |
| "loss": 0.4918, | |
| "step": 2265 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "grad_norm": 21.38290023803711, | |
| "learning_rate": 2.7388535031847135e-06, | |
| "loss": 0.5542, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "grad_norm": 21.44352912902832, | |
| "learning_rate": 2.6910828025477707e-06, | |
| "loss": 0.5389, | |
| "step": 2295 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "eval_accuracy": 0.7876190476190477, | |
| "eval_loss": 0.6360692381858826, | |
| "eval_runtime": 13.6127, | |
| "eval_samples_per_second": 77.134, | |
| "eval_steps_per_second": 9.697, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "grad_norm": 14.326496124267578, | |
| "learning_rate": 2.6433121019108284e-06, | |
| "loss": 0.4785, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "grad_norm": 17.98026466369629, | |
| "learning_rate": 2.5955414012738857e-06, | |
| "loss": 0.5193, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "grad_norm": 11.728538513183594, | |
| "learning_rate": 2.547770700636943e-06, | |
| "loss": 0.4371, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 17.007251739501953, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.4969, | |
| "step": 2355 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "grad_norm": 15.3156099319458, | |
| "learning_rate": 2.4522292993630575e-06, | |
| "loss": 0.5297, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "grad_norm": 13.694135665893555, | |
| "learning_rate": 2.4044585987261147e-06, | |
| "loss": 0.4651, | |
| "step": 2385 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "grad_norm": 16.395017623901367, | |
| "learning_rate": 2.356687898089172e-06, | |
| "loss": 0.5751, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "eval_accuracy": 0.7904761904761904, | |
| "eval_loss": 0.6376513838768005, | |
| "eval_runtime": 13.653, | |
| "eval_samples_per_second": 76.906, | |
| "eval_steps_per_second": 9.668, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "grad_norm": 21.47723388671875, | |
| "learning_rate": 2.3089171974522297e-06, | |
| "loss": 0.4762, | |
| "step": 2415 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "grad_norm": 17.56719970703125, | |
| "learning_rate": 2.261146496815287e-06, | |
| "loss": 0.6217, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "grad_norm": 12.036867141723633, | |
| "learning_rate": 2.213375796178344e-06, | |
| "loss": 0.4727, | |
| "step": 2445 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "grad_norm": 18.971595764160156, | |
| "learning_rate": 2.1656050955414015e-06, | |
| "loss": 0.4601, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 7.88, | |
| "grad_norm": 18.308382034301758, | |
| "learning_rate": 2.1178343949044587e-06, | |
| "loss": 0.5086, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "grad_norm": 13.486546516418457, | |
| "learning_rate": 2.070063694267516e-06, | |
| "loss": 0.4743, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "eval_accuracy": 0.7866666666666666, | |
| "eval_loss": 0.6417487859725952, | |
| "eval_runtime": 13.707, | |
| "eval_samples_per_second": 76.603, | |
| "eval_steps_per_second": 9.63, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "grad_norm": 12.4083833694458, | |
| "learning_rate": 2.0222929936305737e-06, | |
| "loss": 0.453, | |
| "step": 2505 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "grad_norm": 10.967087745666504, | |
| "learning_rate": 1.974522292993631e-06, | |
| "loss": 0.4937, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 11.663314819335938, | |
| "learning_rate": 1.926751592356688e-06, | |
| "loss": 0.5115, | |
| "step": 2535 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "grad_norm": 10.820151329040527, | |
| "learning_rate": 1.8789808917197455e-06, | |
| "loss": 0.4281, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "grad_norm": 15.378673553466797, | |
| "learning_rate": 1.8312101910828025e-06, | |
| "loss": 0.5341, | |
| "step": 2565 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "grad_norm": 18.212982177734375, | |
| "learning_rate": 1.78343949044586e-06, | |
| "loss": 0.5331, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "grad_norm": 25.97978401184082, | |
| "learning_rate": 1.7356687898089172e-06, | |
| "loss": 0.4519, | |
| "step": 2595 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "eval_accuracy": 0.7895238095238095, | |
| "eval_loss": 0.6309406161308289, | |
| "eval_runtime": 13.8301, | |
| "eval_samples_per_second": 75.921, | |
| "eval_steps_per_second": 9.544, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "grad_norm": 16.035921096801758, | |
| "learning_rate": 1.6878980891719745e-06, | |
| "loss": 0.4434, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "grad_norm": 33.498626708984375, | |
| "learning_rate": 1.640127388535032e-06, | |
| "loss": 0.5621, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "grad_norm": 22.457271575927734, | |
| "learning_rate": 1.5923566878980892e-06, | |
| "loss": 0.4843, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "grad_norm": 20.040433883666992, | |
| "learning_rate": 1.5445859872611465e-06, | |
| "loss": 0.5306, | |
| "step": 2655 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 25.173227310180664, | |
| "learning_rate": 1.496815286624204e-06, | |
| "loss": 0.5078, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "grad_norm": 14.426128387451172, | |
| "learning_rate": 1.4490445859872612e-06, | |
| "loss": 0.6197, | |
| "step": 2685 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "grad_norm": 21.540132522583008, | |
| "learning_rate": 1.4012738853503185e-06, | |
| "loss": 0.5058, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "eval_accuracy": 0.7866666666666666, | |
| "eval_loss": 0.6453108787536621, | |
| "eval_runtime": 13.9306, | |
| "eval_samples_per_second": 75.373, | |
| "eval_steps_per_second": 9.476, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "grad_norm": 17.211627960205078, | |
| "learning_rate": 1.353503184713376e-06, | |
| "loss": 0.6111, | |
| "step": 2715 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "grad_norm": 12.991823196411133, | |
| "learning_rate": 1.3057324840764332e-06, | |
| "loss": 0.4392, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "grad_norm": 10.215910911560059, | |
| "learning_rate": 1.2579617834394905e-06, | |
| "loss": 0.5505, | |
| "step": 2745 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "grad_norm": 16.872520446777344, | |
| "learning_rate": 1.210191082802548e-06, | |
| "loss": 0.4304, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "grad_norm": 14.673178672790527, | |
| "learning_rate": 1.1624203821656052e-06, | |
| "loss": 0.4886, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "grad_norm": 17.11809730529785, | |
| "learning_rate": 1.1146496815286625e-06, | |
| "loss": 0.4754, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "eval_accuracy": 0.7904761904761904, | |
| "eval_loss": 0.6414105296134949, | |
| "eval_runtime": 13.6725, | |
| "eval_samples_per_second": 76.797, | |
| "eval_steps_per_second": 9.654, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "grad_norm": 10.194275856018066, | |
| "learning_rate": 1.06687898089172e-06, | |
| "loss": 0.5041, | |
| "step": 2805 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "grad_norm": 24.478836059570312, | |
| "learning_rate": 1.0191082802547772e-06, | |
| "loss": 0.5016, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "grad_norm": 16.150724411010742, | |
| "learning_rate": 9.713375796178345e-07, | |
| "loss": 0.486, | |
| "step": 2835 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "grad_norm": 15.390515327453613, | |
| "learning_rate": 9.235668789808917e-07, | |
| "loss": 0.5024, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "grad_norm": 20.307998657226562, | |
| "learning_rate": 8.757961783439491e-07, | |
| "loss": 0.4982, | |
| "step": 2865 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "grad_norm": 18.18573760986328, | |
| "learning_rate": 8.280254777070064e-07, | |
| "loss": 0.5054, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "grad_norm": 15.57632827758789, | |
| "learning_rate": 7.802547770700637e-07, | |
| "loss": 0.4637, | |
| "step": 2895 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "eval_accuracy": 0.7904761904761904, | |
| "eval_loss": 0.6329751014709473, | |
| "eval_runtime": 13.6535, | |
| "eval_samples_per_second": 76.904, | |
| "eval_steps_per_second": 9.668, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "grad_norm": 11.808470726013184, | |
| "learning_rate": 7.324840764331211e-07, | |
| "loss": 0.5804, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "grad_norm": 19.781538009643555, | |
| "learning_rate": 6.847133757961784e-07, | |
| "loss": 0.5309, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "grad_norm": 15.966341972351074, | |
| "learning_rate": 6.369426751592357e-07, | |
| "loss": 0.536, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 11.989510536193848, | |
| "learning_rate": 5.89171974522293e-07, | |
| "loss": 0.4474, | |
| "step": 2955 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "grad_norm": 13.803847312927246, | |
| "learning_rate": 5.414012738853504e-07, | |
| "loss": 0.4868, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 9.51, | |
| "grad_norm": 16.266407012939453, | |
| "learning_rate": 4.936305732484077e-07, | |
| "loss": 0.4453, | |
| "step": 2985 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "grad_norm": 17.96660804748535, | |
| "learning_rate": 4.45859872611465e-07, | |
| "loss": 0.5028, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "eval_accuracy": 0.7857142857142857, | |
| "eval_loss": 0.641762375831604, | |
| "eval_runtime": 13.8591, | |
| "eval_samples_per_second": 75.763, | |
| "eval_steps_per_second": 9.524, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 15.89122200012207, | |
| "learning_rate": 3.980891719745223e-07, | |
| "loss": 0.4376, | |
| "step": 3015 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "grad_norm": 18.61841583251953, | |
| "learning_rate": 3.503184713375796e-07, | |
| "loss": 0.3695, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "grad_norm": 19.915699005126953, | |
| "learning_rate": 3.02547770700637e-07, | |
| "loss": 0.4777, | |
| "step": 3045 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "grad_norm": 22.503381729125977, | |
| "learning_rate": 2.547770700636943e-07, | |
| "loss": 0.6298, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "grad_norm": 16.984233856201172, | |
| "learning_rate": 2.070063694267516e-07, | |
| "loss": 0.4051, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "grad_norm": 19.879077911376953, | |
| "learning_rate": 1.5923566878980893e-07, | |
| "loss": 0.4227, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 9.87, | |
| "eval_accuracy": 0.7914285714285715, | |
| "eval_loss": 0.6412155628204346, | |
| "eval_runtime": 13.7603, | |
| "eval_samples_per_second": 76.307, | |
| "eval_steps_per_second": 9.593, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "grad_norm": 18.370866775512695, | |
| "learning_rate": 1.1146496815286625e-07, | |
| "loss": 0.4079, | |
| "step": 3105 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "grad_norm": 13.279521942138672, | |
| "learning_rate": 6.369426751592358e-08, | |
| "loss": 0.4035, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "grad_norm": 16.93092155456543, | |
| "learning_rate": 1.5923566878980894e-08, | |
| "loss": 0.4605, | |
| "step": 3135 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 3140, | |
| "total_flos": 3.265548125287219e+18, | |
| "train_loss": 0.7155859537944672, | |
| "train_runtime": 2626.1064, | |
| "train_samples_per_second": 38.212, | |
| "train_steps_per_second": 1.196 | |
| } | |
| ], | |
| "logging_steps": 15, | |
| "max_steps": 3140, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 3.265548125287219e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |