| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 5.6, |
| "eval_steps": 100, |
| "global_step": 7000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 2.5604241602122784, |
| "epoch": 0.008, |
| "grad_norm": 360.0, |
| "learning_rate": 1.8e-07, |
| "loss": 6.7873, |
| "mean_token_accuracy": 0.21116912770085036, |
| "num_tokens": 175126.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 2.595553621649742, |
| "epoch": 0.016, |
| "grad_norm": 372.0, |
| "learning_rate": 3.8e-07, |
| "loss": 6.7228, |
| "mean_token_accuracy": 0.20832345820963383, |
| "num_tokens": 340696.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 2.6845590390264986, |
| "epoch": 0.024, |
| "grad_norm": 270.0, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 6.3033, |
| "mean_token_accuracy": 0.21763876075856387, |
| "num_tokens": 509066.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 2.8930466674268245, |
| "epoch": 0.032, |
| "grad_norm": 302.0, |
| "learning_rate": 7.8e-07, |
| "loss": 5.6306, |
| "mean_token_accuracy": 0.22912178803235292, |
| "num_tokens": 675930.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 3.081789229810238, |
| "epoch": 0.04, |
| "grad_norm": 123.0, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 4.9613, |
| "mean_token_accuracy": 0.24729052558541298, |
| "num_tokens": 841685.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 3.404916675388813, |
| "epoch": 0.048, |
| "grad_norm": 66.0, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 4.2301, |
| "mean_token_accuracy": 0.27200257908552883, |
| "num_tokens": 1007956.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 3.6782597199082376, |
| "epoch": 0.056, |
| "grad_norm": 38.0, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 3.7823, |
| "mean_token_accuracy": 0.3019493265077472, |
| "num_tokens": 1174735.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 3.4533004209399225, |
| "epoch": 0.064, |
| "grad_norm": 33.75, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 3.4895, |
| "mean_token_accuracy": 0.33877146458253266, |
| "num_tokens": 1340560.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 3.139299413561821, |
| "epoch": 0.072, |
| "grad_norm": 31.625, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 3.1861, |
| "mean_token_accuracy": 0.3694635771214962, |
| "num_tokens": 1510110.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 2.9895101204514503, |
| "epoch": 0.08, |
| "grad_norm": 21.25, |
| "learning_rate": 1.98e-06, |
| "loss": 2.9954, |
| "mean_token_accuracy": 0.39561883062124253, |
| "num_tokens": 1679432.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_coding_entropy": 2.0758819575309753, |
| "eval_coding_loss": 2.1979613304138184, |
| "eval_coding_mean_token_accuracy": 0.502328829407692, |
| "eval_coding_num_tokens": 1679432.0, |
| "eval_coding_runtime": 79.4698, |
| "eval_coding_samples_per_second": 6.292, |
| "eval_coding_steps_per_second": 3.146, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_physics_entropy": 2.9290144443511963, |
| "eval_physics_loss": 2.974918842315674, |
| "eval_physics_mean_token_accuracy": 0.40643512630462647, |
| "eval_physics_num_tokens": 1679432.0, |
| "eval_physics_runtime": 52.0573, |
| "eval_physics_samples_per_second": 9.605, |
| "eval_physics_steps_per_second": 4.802, |
| "step": 100 |
| }, |
| { |
| "entropy": 2.8267079159617423, |
| "epoch": 0.088, |
| "grad_norm": 21.125, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 2.8166, |
| "mean_token_accuracy": 0.4188435094431043, |
| "num_tokens": 1848521.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 2.731858339905739, |
| "epoch": 0.096, |
| "grad_norm": 19.625, |
| "learning_rate": 2.38e-06, |
| "loss": 2.7192, |
| "mean_token_accuracy": 0.43734086528420446, |
| "num_tokens": 2021570.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 2.742488440871239, |
| "epoch": 0.104, |
| "grad_norm": 20.5, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 2.7106, |
| "mean_token_accuracy": 0.4448789706453681, |
| "num_tokens": 2186993.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 2.546742835640907, |
| "epoch": 0.112, |
| "grad_norm": 19.875, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 2.5347, |
| "mean_token_accuracy": 0.46661383416503666, |
| "num_tokens": 2359297.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 2.5924400255084037, |
| "epoch": 0.12, |
| "grad_norm": 21.25, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 2.5665, |
| "mean_token_accuracy": 0.4689426450058818, |
| "num_tokens": 2522375.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 2.4927076123654843, |
| "epoch": 0.128, |
| "grad_norm": 21.125, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 2.4571, |
| "mean_token_accuracy": 0.4879589693620801, |
| "num_tokens": 2683168.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 2.4049933359026907, |
| "epoch": 0.136, |
| "grad_norm": 18.5, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 2.3829, |
| "mean_token_accuracy": 0.4920692486688495, |
| "num_tokens": 2850132.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 2.3835881277918816, |
| "epoch": 0.144, |
| "grad_norm": 20.5, |
| "learning_rate": 3.58e-06, |
| "loss": 2.3245, |
| "mean_token_accuracy": 0.5039734555408358, |
| "num_tokens": 3019426.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 2.3372473388910295, |
| "epoch": 0.152, |
| "grad_norm": 18.25, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 2.3194, |
| "mean_token_accuracy": 0.5051064729690552, |
| "num_tokens": 3182640.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 2.27973321005702, |
| "epoch": 0.16, |
| "grad_norm": 21.125, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 2.213, |
| "mean_token_accuracy": 0.5190408078953623, |
| "num_tokens": 3354390.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_coding_entropy": 1.9250439591407775, |
| "eval_coding_loss": 2.1208813190460205, |
| "eval_coding_mean_token_accuracy": 0.5191248273849487, |
| "eval_coding_num_tokens": 3354390.0, |
| "eval_coding_runtime": 79.1564, |
| "eval_coding_samples_per_second": 6.317, |
| "eval_coding_steps_per_second": 3.158, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_physics_entropy": 2.230250364303589, |
| "eval_physics_loss": 2.271095037460327, |
| "eval_physics_mean_token_accuracy": 0.5189732195138931, |
| "eval_physics_num_tokens": 3354390.0, |
| "eval_physics_runtime": 52.0613, |
| "eval_physics_samples_per_second": 9.604, |
| "eval_physics_steps_per_second": 4.802, |
| "step": 200 |
| }, |
| { |
| "entropy": 2.265408169478178, |
| "epoch": 0.168, |
| "grad_norm": 20.5, |
| "learning_rate": 4.18e-06, |
| "loss": 2.2294, |
| "mean_token_accuracy": 0.5198547914624214, |
| "num_tokens": 3521075.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 2.2567644871771337, |
| "epoch": 0.176, |
| "grad_norm": 26.375, |
| "learning_rate": 4.38e-06, |
| "loss": 2.1962, |
| "mean_token_accuracy": 0.5276498578488826, |
| "num_tokens": 3684374.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 2.2172055408358573, |
| "epoch": 0.184, |
| "grad_norm": 18.5, |
| "learning_rate": 4.58e-06, |
| "loss": 2.1651, |
| "mean_token_accuracy": 0.5290316773578525, |
| "num_tokens": 3848536.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 2.152971290051937, |
| "epoch": 0.192, |
| "grad_norm": 18.75, |
| "learning_rate": 4.78e-06, |
| "loss": 2.1125, |
| "mean_token_accuracy": 0.5370275607332587, |
| "num_tokens": 4023915.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 2.148153053969145, |
| "epoch": 0.2, |
| "grad_norm": 20.125, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 2.1104, |
| "mean_token_accuracy": 0.541122961603105, |
| "num_tokens": 4190485.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 2.0243859082460403, |
| "epoch": 0.208, |
| "grad_norm": 18.125, |
| "learning_rate": 5.18e-06, |
| "loss": 1.975, |
| "mean_token_accuracy": 0.558369823731482, |
| "num_tokens": 4352045.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 2.0731096126139166, |
| "epoch": 0.216, |
| "grad_norm": 19.625, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 2.0131, |
| "mean_token_accuracy": 0.5515947749838233, |
| "num_tokens": 4514042.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 2.0090589359402657, |
| "epoch": 0.224, |
| "grad_norm": 17.375, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 1.9505, |
| "mean_token_accuracy": 0.5629682159051299, |
| "num_tokens": 4681371.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 2.002196607738733, |
| "epoch": 0.232, |
| "grad_norm": 18.625, |
| "learning_rate": 5.78e-06, |
| "loss": 1.958, |
| "mean_token_accuracy": 0.5620508158579469, |
| "num_tokens": 4847650.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.9909410774707794, |
| "epoch": 0.24, |
| "grad_norm": 18.375, |
| "learning_rate": 5.98e-06, |
| "loss": 1.9375, |
| "mean_token_accuracy": 0.5657549701631069, |
| "num_tokens": 5021784.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_coding_entropy": 1.916715157032013, |
| "eval_coding_loss": 2.0919320583343506, |
| "eval_coding_mean_token_accuracy": 0.5256761227846145, |
| "eval_coding_num_tokens": 5021784.0, |
| "eval_coding_runtime": 79.1652, |
| "eval_coding_samples_per_second": 6.316, |
| "eval_coding_steps_per_second": 3.158, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_physics_entropy": 1.9569331226348876, |
| "eval_physics_loss": 1.9659218788146973, |
| "eval_physics_mean_token_accuracy": 0.5716171189546585, |
| "eval_physics_num_tokens": 5021784.0, |
| "eval_physics_runtime": 52.0616, |
| "eval_physics_samples_per_second": 9.604, |
| "eval_physics_steps_per_second": 4.802, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.9608435668051243, |
| "epoch": 0.248, |
| "grad_norm": 17.125, |
| "learning_rate": 6.18e-06, |
| "loss": 1.9066, |
| "mean_token_accuracy": 0.5718545202165842, |
| "num_tokens": 5201507.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.929166065901518, |
| "epoch": 0.256, |
| "grad_norm": 16.875, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 1.8819, |
| "mean_token_accuracy": 0.5770876344293356, |
| "num_tokens": 5367308.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.9197401903569697, |
| "epoch": 0.264, |
| "grad_norm": 18.5, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 1.8598, |
| "mean_token_accuracy": 0.5781059630215168, |
| "num_tokens": 5536442.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.9471155002713203, |
| "epoch": 0.272, |
| "grad_norm": 21.5, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 1.8791, |
| "mean_token_accuracy": 0.5742578504607081, |
| "num_tokens": 5699101.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.8642883323132993, |
| "epoch": 0.28, |
| "grad_norm": 17.875, |
| "learning_rate": 6.98e-06, |
| "loss": 1.8298, |
| "mean_token_accuracy": 0.587579975090921, |
| "num_tokens": 5860398.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.883735204488039, |
| "epoch": 0.288, |
| "grad_norm": 19.75, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 1.8306, |
| "mean_token_accuracy": 0.5885121200233698, |
| "num_tokens": 6022837.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.832478718459606, |
| "epoch": 0.296, |
| "grad_norm": 17.0, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 1.7789, |
| "mean_token_accuracy": 0.5987123249098658, |
| "num_tokens": 6183758.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.8392824359238147, |
| "epoch": 0.304, |
| "grad_norm": 15.1875, |
| "learning_rate": 7.58e-06, |
| "loss": 1.7763, |
| "mean_token_accuracy": 0.595327608473599, |
| "num_tokens": 6343826.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.8003668662160635, |
| "epoch": 0.312, |
| "grad_norm": 19.5, |
| "learning_rate": 7.78e-06, |
| "loss": 1.7664, |
| "mean_token_accuracy": 0.5983955936506391, |
| "num_tokens": 6502713.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.7905294828116893, |
| "epoch": 0.32, |
| "grad_norm": 15.3125, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 1.7435, |
| "mean_token_accuracy": 0.6020280076190829, |
| "num_tokens": 6670164.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_coding_entropy": 1.8930716361999511, |
| "eval_coding_loss": 2.0930795669555664, |
| "eval_coding_mean_token_accuracy": 0.5246047481298447, |
| "eval_coding_num_tokens": 6670164.0, |
| "eval_coding_runtime": 79.1697, |
| "eval_coding_samples_per_second": 6.316, |
| "eval_coding_steps_per_second": 3.158, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_physics_entropy": 1.7485348978042603, |
| "eval_physics_loss": 1.7591620683670044, |
| "eval_physics_mean_token_accuracy": 0.6077653738260269, |
| "eval_physics_num_tokens": 6670164.0, |
| "eval_physics_runtime": 52.1144, |
| "eval_physics_samples_per_second": 9.594, |
| "eval_physics_steps_per_second": 4.797, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.791950485110283, |
| "epoch": 0.328, |
| "grad_norm": 16.25, |
| "learning_rate": 8.18e-06, |
| "loss": 1.7309, |
| "mean_token_accuracy": 0.60343619287014, |
| "num_tokens": 6831503.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.6932228744029998, |
| "epoch": 0.336, |
| "grad_norm": 17.75, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 1.6384, |
| "mean_token_accuracy": 0.6203264124691487, |
| "num_tokens": 7003404.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.6966097623109817, |
| "epoch": 0.344, |
| "grad_norm": 19.75, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 1.6428, |
| "mean_token_accuracy": 0.6151626098901033, |
| "num_tokens": 7172909.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.688229451328516, |
| "epoch": 0.352, |
| "grad_norm": 18.625, |
| "learning_rate": 8.78e-06, |
| "loss": 1.6446, |
| "mean_token_accuracy": 0.619301825016737, |
| "num_tokens": 7342593.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.684376384317875, |
| "epoch": 0.36, |
| "grad_norm": 15.1875, |
| "learning_rate": 8.98e-06, |
| "loss": 1.6384, |
| "mean_token_accuracy": 0.6227533668279648, |
| "num_tokens": 7506313.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.6351792596280574, |
| "epoch": 0.368, |
| "grad_norm": 17.5, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 1.5883, |
| "mean_token_accuracy": 0.6302843233570457, |
| "num_tokens": 7675569.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.6966083988547325, |
| "epoch": 0.376, |
| "grad_norm": 18.625, |
| "learning_rate": 9.38e-06, |
| "loss": 1.6569, |
| "mean_token_accuracy": 0.6154216848313808, |
| "num_tokens": 7842654.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.6671229541301726, |
| "epoch": 0.384, |
| "grad_norm": 16.625, |
| "learning_rate": 9.58e-06, |
| "loss": 1.6255, |
| "mean_token_accuracy": 0.6238152993842959, |
| "num_tokens": 8012218.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.634845321625471, |
| "epoch": 0.392, |
| "grad_norm": 15.0, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 1.6026, |
| "mean_token_accuracy": 0.6318565379828215, |
| "num_tokens": 8181305.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.5764483150094748, |
| "epoch": 0.4, |
| "grad_norm": 16.125, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 1.5239, |
| "mean_token_accuracy": 0.6454750349745154, |
| "num_tokens": 8350120.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_coding_entropy": 1.9365736389160155, |
| "eval_coding_loss": 2.0959813594818115, |
| "eval_coding_mean_token_accuracy": 0.5250564186573029, |
| "eval_coding_num_tokens": 8350120.0, |
| "eval_coding_runtime": 79.1814, |
| "eval_coding_samples_per_second": 6.315, |
| "eval_coding_steps_per_second": 3.157, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_physics_entropy": 1.671481309890747, |
| "eval_physics_loss": 1.6045700311660767, |
| "eval_physics_mean_token_accuracy": 0.6350576342344284, |
| "eval_physics_num_tokens": 8350120.0, |
| "eval_physics_runtime": 51.8423, |
| "eval_physics_samples_per_second": 9.645, |
| "eval_physics_steps_per_second": 4.822, |
| "step": 500 |
| }, |
| { |
| "entropy": 1.5771338425576686, |
| "epoch": 0.408, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.018e-05, |
| "loss": 1.5352, |
| "mean_token_accuracy": 0.6389181109145283, |
| "num_tokens": 8520118.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 1.5821339197456836, |
| "epoch": 0.416, |
| "grad_norm": 16.125, |
| "learning_rate": 1.038e-05, |
| "loss": 1.5332, |
| "mean_token_accuracy": 0.6355980787426233, |
| "num_tokens": 8688773.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 1.5948798354715108, |
| "epoch": 0.424, |
| "grad_norm": 15.625, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 1.5238, |
| "mean_token_accuracy": 0.6424462385475636, |
| "num_tokens": 8855184.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 1.529571657255292, |
| "epoch": 0.432, |
| "grad_norm": 14.0625, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 1.4919, |
| "mean_token_accuracy": 0.6440655821934342, |
| "num_tokens": 9022094.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 1.5117546495050191, |
| "epoch": 0.44, |
| "grad_norm": 16.375, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 1.4678, |
| "mean_token_accuracy": 0.6497685741633177, |
| "num_tokens": 9190509.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 1.513111385703087, |
| "epoch": 0.448, |
| "grad_norm": 16.25, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 1.4693, |
| "mean_token_accuracy": 0.6491031926125288, |
| "num_tokens": 9360608.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 1.470874259248376, |
| "epoch": 0.456, |
| "grad_norm": 14.3125, |
| "learning_rate": 1.138e-05, |
| "loss": 1.415, |
| "mean_token_accuracy": 0.6601609066128731, |
| "num_tokens": 9522789.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 1.5079048488289117, |
| "epoch": 0.464, |
| "grad_norm": 15.875, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 1.4715, |
| "mean_token_accuracy": 0.6530725235119462, |
| "num_tokens": 9688896.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 1.4853499613702297, |
| "epoch": 0.472, |
| "grad_norm": 14.125, |
| "learning_rate": 1.178e-05, |
| "loss": 1.4262, |
| "mean_token_accuracy": 0.6561389841139317, |
| "num_tokens": 9858464.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 1.5173018813133239, |
| "epoch": 0.48, |
| "grad_norm": 21.375, |
| "learning_rate": 1.198e-05, |
| "loss": 1.4895, |
| "mean_token_accuracy": 0.6496954914182425, |
| "num_tokens": 10025864.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_coding_entropy": 1.953908012866974, |
| "eval_coding_loss": 2.1033880710601807, |
| "eval_coding_mean_token_accuracy": 0.5249351325035095, |
| "eval_coding_num_tokens": 10025864.0, |
| "eval_coding_runtime": 79.131, |
| "eval_coding_samples_per_second": 6.319, |
| "eval_coding_steps_per_second": 3.159, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_physics_entropy": 1.541918823480606, |
| "eval_physics_loss": 1.4983493089675903, |
| "eval_physics_mean_token_accuracy": 0.6534626665115356, |
| "eval_physics_num_tokens": 10025864.0, |
| "eval_physics_runtime": 52.0434, |
| "eval_physics_samples_per_second": 9.607, |
| "eval_physics_steps_per_second": 4.804, |
| "step": 600 |
| }, |
| { |
| "entropy": 1.4748347599059344, |
| "epoch": 0.488, |
| "grad_norm": 16.625, |
| "learning_rate": 1.218e-05, |
| "loss": 1.4246, |
| "mean_token_accuracy": 0.6571080841124057, |
| "num_tokens": 10191232.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 1.5049679655581714, |
| "epoch": 0.496, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 1.4596, |
| "mean_token_accuracy": 0.6528839386999608, |
| "num_tokens": 10362573.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 1.4375012002885341, |
| "epoch": 0.504, |
| "grad_norm": 14.4375, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 1.3841, |
| "mean_token_accuracy": 0.6608929991722107, |
| "num_tokens": 10528450.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 1.5176888063549996, |
| "epoch": 0.512, |
| "grad_norm": 14.25, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 1.4569, |
| "mean_token_accuracy": 0.6528515750542283, |
| "num_tokens": 10694440.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 1.4692929420620202, |
| "epoch": 0.52, |
| "grad_norm": 10.6875, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 1.4449, |
| "mean_token_accuracy": 0.655676519870758, |
| "num_tokens": 10857597.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 1.4483005907386541, |
| "epoch": 0.528, |
| "grad_norm": 12.125, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 1.4046, |
| "mean_token_accuracy": 0.665894140675664, |
| "num_tokens": 11022357.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 1.3869127415120601, |
| "epoch": 0.536, |
| "grad_norm": 12.8125, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 1.3467, |
| "mean_token_accuracy": 0.672945449501276, |
| "num_tokens": 11185045.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 1.4482205394655465, |
| "epoch": 0.544, |
| "grad_norm": 12.8125, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 1.3907, |
| "mean_token_accuracy": 0.6631886938586831, |
| "num_tokens": 11345911.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 1.4304094772785902, |
| "epoch": 0.552, |
| "grad_norm": 13.125, |
| "learning_rate": 1.378e-05, |
| "loss": 1.3881, |
| "mean_token_accuracy": 0.6646447636187076, |
| "num_tokens": 11518974.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 1.3900580175220967, |
| "epoch": 0.56, |
| "grad_norm": 10.875, |
| "learning_rate": 1.398e-05, |
| "loss": 1.34, |
| "mean_token_accuracy": 0.6744402695447207, |
| "num_tokens": 11687225.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_coding_entropy": 1.8969693536758423, |
| "eval_coding_loss": 2.1377148628234863, |
| "eval_coding_mean_token_accuracy": 0.5214611231088638, |
| "eval_coding_num_tokens": 11687225.0, |
| "eval_coding_runtime": 79.1442, |
| "eval_coding_samples_per_second": 6.318, |
| "eval_coding_steps_per_second": 3.159, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_physics_entropy": 1.3678591132164002, |
| "eval_physics_loss": 1.3962782621383667, |
| "eval_physics_mean_token_accuracy": 0.6698634608983993, |
| "eval_physics_num_tokens": 11687225.0, |
| "eval_physics_runtime": 52.0406, |
| "eval_physics_samples_per_second": 9.608, |
| "eval_physics_steps_per_second": 4.804, |
| "step": 700 |
| }, |
| { |
| "entropy": 1.3716834906488657, |
| "epoch": 0.568, |
| "grad_norm": 18.125, |
| "learning_rate": 1.418e-05, |
| "loss": 1.342, |
| "mean_token_accuracy": 0.67361662350595, |
| "num_tokens": 11848324.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 1.3629539489746094, |
| "epoch": 0.576, |
| "grad_norm": 12.9375, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 1.3293, |
| "mean_token_accuracy": 0.6735996751114726, |
| "num_tokens": 12015720.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 1.3565699491649865, |
| "epoch": 0.584, |
| "grad_norm": 13.0625, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 1.3375, |
| "mean_token_accuracy": 0.6773057371377945, |
| "num_tokens": 12178798.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 1.3613810572773217, |
| "epoch": 0.592, |
| "grad_norm": 12.125, |
| "learning_rate": 1.478e-05, |
| "loss": 1.319, |
| "mean_token_accuracy": 0.6774894919246435, |
| "num_tokens": 12343393.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 1.3767595831304789, |
| "epoch": 0.6, |
| "grad_norm": 11.6875, |
| "learning_rate": 1.498e-05, |
| "loss": 1.3265, |
| "mean_token_accuracy": 0.6750151777639986, |
| "num_tokens": 12509336.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 1.3187911182641983, |
| "epoch": 0.608, |
| "grad_norm": 12.5, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 1.2905, |
| "mean_token_accuracy": 0.6845489364117384, |
| "num_tokens": 12671474.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 1.3427253626286983, |
| "epoch": 0.616, |
| "grad_norm": 10.75, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 1.2963, |
| "mean_token_accuracy": 0.6806909158825875, |
| "num_tokens": 12835956.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 1.3635981816798448, |
| "epoch": 0.624, |
| "grad_norm": 11.0, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 1.3102, |
| "mean_token_accuracy": 0.6765737216919661, |
| "num_tokens": 12996077.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 1.3231649551540614, |
| "epoch": 0.632, |
| "grad_norm": 12.375, |
| "learning_rate": 1.578e-05, |
| "loss": 1.2993, |
| "mean_token_accuracy": 0.6843808237463236, |
| "num_tokens": 13162966.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 1.3815045531839132, |
| "epoch": 0.64, |
| "grad_norm": 11.0, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 1.3403, |
| "mean_token_accuracy": 0.676003416068852, |
| "num_tokens": 13325291.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_coding_entropy": 1.8999772543907165, |
| "eval_coding_loss": 2.137219190597534, |
| "eval_coding_mean_token_accuracy": 0.5216508195400238, |
| "eval_coding_num_tokens": 13325291.0, |
| "eval_coding_runtime": 79.1227, |
| "eval_coding_samples_per_second": 6.319, |
| "eval_coding_steps_per_second": 3.16, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_physics_entropy": 1.3213207037448884, |
| "eval_physics_loss": 1.3272991180419922, |
| "eval_physics_mean_token_accuracy": 0.6834507223367691, |
| "eval_physics_num_tokens": 13325291.0, |
| "eval_physics_runtime": 52.0498, |
| "eval_physics_samples_per_second": 9.606, |
| "eval_physics_steps_per_second": 4.803, |
| "step": 800 |
| }, |
| { |
| "entropy": 1.3278860542923212, |
| "epoch": 0.648, |
| "grad_norm": 12.0, |
| "learning_rate": 1.618e-05, |
| "loss": 1.2916, |
| "mean_token_accuracy": 0.6818249279633164, |
| "num_tokens": 13495304.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 1.2705524377524853, |
| "epoch": 0.656, |
| "grad_norm": 10.875, |
| "learning_rate": 1.638e-05, |
| "loss": 1.2362, |
| "mean_token_accuracy": 0.6954903973266482, |
| "num_tokens": 13671434.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 1.3271659668534994, |
| "epoch": 0.664, |
| "grad_norm": 13.375, |
| "learning_rate": 1.658e-05, |
| "loss": 1.2753, |
| "mean_token_accuracy": 0.6867586117237806, |
| "num_tokens": 13829692.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 1.2718054752796888, |
| "epoch": 0.672, |
| "grad_norm": 11.125, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 1.2239, |
| "mean_token_accuracy": 0.6929404513910413, |
| "num_tokens": 13993668.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 1.2376167189329863, |
| "epoch": 0.68, |
| "grad_norm": 9.75, |
| "learning_rate": 1.698e-05, |
| "loss": 1.2134, |
| "mean_token_accuracy": 0.6992333011701704, |
| "num_tokens": 14173446.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 1.3201266448944806, |
| "epoch": 0.688, |
| "grad_norm": 9.0625, |
| "learning_rate": 1.718e-05, |
| "loss": 1.2773, |
| "mean_token_accuracy": 0.6858949340879917, |
| "num_tokens": 14334907.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 1.2640008773654698, |
| "epoch": 0.696, |
| "grad_norm": 8.875, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 1.237, |
| "mean_token_accuracy": 0.6919090248644352, |
| "num_tokens": 14512555.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 1.2943935077637434, |
| "epoch": 0.704, |
| "grad_norm": 11.9375, |
| "learning_rate": 1.758e-05, |
| "loss": 1.2627, |
| "mean_token_accuracy": 0.6883153811097145, |
| "num_tokens": 14670639.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 1.2209337592124938, |
| "epoch": 0.712, |
| "grad_norm": 10.5, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 1.2024, |
| "mean_token_accuracy": 0.7000889934599399, |
| "num_tokens": 14832916.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 1.2753230314701796, |
| "epoch": 0.72, |
| "grad_norm": 9.0, |
| "learning_rate": 1.798e-05, |
| "loss": 1.2268, |
| "mean_token_accuracy": 0.6935238633304834, |
| "num_tokens": 14987995.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_coding_entropy": 1.9145963129997254, |
| "eval_coding_loss": 2.1864311695098877, |
| "eval_coding_mean_token_accuracy": 0.5160261226892471, |
| "eval_coding_num_tokens": 14987995.0, |
| "eval_coding_runtime": 79.3569, |
| "eval_coding_samples_per_second": 6.301, |
| "eval_coding_steps_per_second": 3.15, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_physics_entropy": 1.288757406234741, |
| "eval_physics_loss": 1.2604119777679443, |
| "eval_physics_mean_token_accuracy": 0.6937535845041275, |
| "eval_physics_num_tokens": 14987995.0, |
| "eval_physics_runtime": 52.2126, |
| "eval_physics_samples_per_second": 9.576, |
| "eval_physics_steps_per_second": 4.788, |
| "step": 900 |
| }, |
| { |
| "entropy": 1.208376468345523, |
| "epoch": 0.728, |
| "grad_norm": 10.3125, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 1.1886, |
| "mean_token_accuracy": 0.7017698954790831, |
| "num_tokens": 15158982.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 1.2472068972885608, |
| "epoch": 0.736, |
| "grad_norm": 9.6875, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 1.2129, |
| "mean_token_accuracy": 0.6974750630557537, |
| "num_tokens": 15324659.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 1.225362879037857, |
| "epoch": 0.744, |
| "grad_norm": 10.4375, |
| "learning_rate": 1.858e-05, |
| "loss": 1.183, |
| "mean_token_accuracy": 0.7002156199887395, |
| "num_tokens": 15493621.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 1.2113514818251132, |
| "epoch": 0.752, |
| "grad_norm": 11.3125, |
| "learning_rate": 1.878e-05, |
| "loss": 1.1911, |
| "mean_token_accuracy": 0.7010745037347078, |
| "num_tokens": 15657342.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 1.251608308404684, |
| "epoch": 0.76, |
| "grad_norm": 9.0, |
| "learning_rate": 1.898e-05, |
| "loss": 1.2108, |
| "mean_token_accuracy": 0.6982708286494017, |
| "num_tokens": 15815189.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 1.2250138234347105, |
| "epoch": 0.768, |
| "grad_norm": 9.375, |
| "learning_rate": 1.918e-05, |
| "loss": 1.2033, |
| "mean_token_accuracy": 0.6997085321694613, |
| "num_tokens": 15988993.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 1.2000590208917856, |
| "epoch": 0.776, |
| "grad_norm": 10.375, |
| "learning_rate": 1.938e-05, |
| "loss": 1.1522, |
| "mean_token_accuracy": 0.7049363922327757, |
| "num_tokens": 16163512.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 1.2060553971678019, |
| "epoch": 0.784, |
| "grad_norm": 7.875, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 1.1664, |
| "mean_token_accuracy": 0.7088450845330954, |
| "num_tokens": 16333857.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 1.1793909084051848, |
| "epoch": 0.792, |
| "grad_norm": 7.96875, |
| "learning_rate": 1.978e-05, |
| "loss": 1.1466, |
| "mean_token_accuracy": 0.709895571321249, |
| "num_tokens": 16494118.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 1.2104855172336102, |
| "epoch": 0.8, |
| "grad_norm": 10.0625, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 1.1781, |
| "mean_token_accuracy": 0.7040123168379069, |
| "num_tokens": 16664690.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_coding_entropy": 1.9454177751541137, |
| "eval_coding_loss": 2.2875986099243164, |
| "eval_coding_mean_token_accuracy": 0.49295639526844026, |
| "eval_coding_num_tokens": 16664690.0, |
| "eval_coding_runtime": 79.1544, |
| "eval_coding_samples_per_second": 6.317, |
| "eval_coding_steps_per_second": 3.158, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_physics_entropy": 1.249910979270935, |
| "eval_physics_loss": 1.215478539466858, |
| "eval_physics_mean_token_accuracy": 0.7021359713077545, |
| "eval_physics_num_tokens": 16664690.0, |
| "eval_physics_runtime": 51.8213, |
| "eval_physics_samples_per_second": 9.649, |
| "eval_physics_steps_per_second": 4.824, |
| "step": 1000 |
| }, |
| { |
| "entropy": 1.2079178389161824, |
| "epoch": 0.808, |
| "grad_norm": 10.6875, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 1.1621, |
| "mean_token_accuracy": 0.7027403622865677, |
| "num_tokens": 16829315.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 1.1786697298288344, |
| "epoch": 0.816, |
| "grad_norm": 8.0, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 1.1494, |
| "mean_token_accuracy": 0.7092785466462374, |
| "num_tokens": 16989307.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 1.1985579200088978, |
| "epoch": 0.824, |
| "grad_norm": 9.25, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 1.1467, |
| "mean_token_accuracy": 0.7067175708711148, |
| "num_tokens": 17155547.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 1.164666483551264, |
| "epoch": 0.832, |
| "grad_norm": 7.71875, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 1.1401, |
| "mean_token_accuracy": 0.7113697534427047, |
| "num_tokens": 17317524.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 1.1878154788166284, |
| "epoch": 0.84, |
| "grad_norm": 6.78125, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 1.1408, |
| "mean_token_accuracy": 0.7077680010348558, |
| "num_tokens": 17479138.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 1.1530823826789856, |
| "epoch": 0.848, |
| "grad_norm": 6.90625, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 1.1155, |
| "mean_token_accuracy": 0.71307920999825, |
| "num_tokens": 17650424.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 1.1825125798583032, |
| "epoch": 0.856, |
| "grad_norm": 7.8125, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 1.1398, |
| "mean_token_accuracy": 0.7109603315591813, |
| "num_tokens": 17814512.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 1.1390453603118658, |
| "epoch": 0.864, |
| "grad_norm": 6.53125, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 1.1154, |
| "mean_token_accuracy": 0.7171653687953949, |
| "num_tokens": 17985547.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 1.1706372637301683, |
| "epoch": 0.872, |
| "grad_norm": 8.0625, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 1.1272, |
| "mean_token_accuracy": 0.7104873843491077, |
| "num_tokens": 18147038.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 1.2016708802431821, |
| "epoch": 0.88, |
| "grad_norm": 7.9375, |
| "learning_rate": 1.978e-05, |
| "loss": 1.1414, |
| "mean_token_accuracy": 0.7089165017008782, |
| "num_tokens": 18315392.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_coding_entropy": 1.9011136808395386, |
| "eval_coding_loss": 2.2340805530548096, |
| "eval_coding_mean_token_accuracy": 0.5084248929023742, |
| "eval_coding_num_tokens": 18315392.0, |
| "eval_coding_runtime": 79.1322, |
| "eval_coding_samples_per_second": 6.319, |
| "eval_coding_steps_per_second": 3.159, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_physics_entropy": 1.1309143052101136, |
| "eval_physics_loss": 1.1532424688339233, |
| "eval_physics_mean_token_accuracy": 0.7144835109710693, |
| "eval_physics_num_tokens": 18315392.0, |
| "eval_physics_runtime": 52.1221, |
| "eval_physics_samples_per_second": 9.593, |
| "eval_physics_steps_per_second": 4.796, |
| "step": 1100 |
| }, |
| { |
| "entropy": 1.0508298195898533, |
| "epoch": 0.888, |
| "grad_norm": 8.25, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 1.0332, |
| "mean_token_accuracy": 0.7302042797207833, |
| "num_tokens": 18495457.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 1.1058499403297901, |
| "epoch": 0.896, |
| "grad_norm": 5.90625, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 1.0849, |
| "mean_token_accuracy": 0.7238155752420425, |
| "num_tokens": 18667124.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 1.1259642377495767, |
| "epoch": 0.904, |
| "grad_norm": 6.875, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 1.0818, |
| "mean_token_accuracy": 0.721034574508667, |
| "num_tokens": 18833366.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 1.145649480447173, |
| "epoch": 0.912, |
| "grad_norm": 7.65625, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 1.1107, |
| "mean_token_accuracy": 0.7126699529588223, |
| "num_tokens": 19002186.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 1.1212836857885122, |
| "epoch": 0.92, |
| "grad_norm": 6.5625, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 1.0845, |
| "mean_token_accuracy": 0.7232643336057663, |
| "num_tokens": 19167898.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 1.1022184763103724, |
| "epoch": 0.928, |
| "grad_norm": 7.78125, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 1.0742, |
| "mean_token_accuracy": 0.7239976346492767, |
| "num_tokens": 19335013.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 1.081941632181406, |
| "epoch": 0.936, |
| "grad_norm": 7.9375, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 1.0521, |
| "mean_token_accuracy": 0.7266536355018616, |
| "num_tokens": 19497952.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 1.1005669727921485, |
| "epoch": 0.944, |
| "grad_norm": 6.09375, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 1.0613, |
| "mean_token_accuracy": 0.7255159001797438, |
| "num_tokens": 19662733.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 1.0804655332118274, |
| "epoch": 0.952, |
| "grad_norm": 6.65625, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 1.0475, |
| "mean_token_accuracy": 0.728959995508194, |
| "num_tokens": 19829333.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 1.069432857632637, |
| "epoch": 0.96, |
| "grad_norm": 6.78125, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 1.0476, |
| "mean_token_accuracy": 0.7279221322387457, |
| "num_tokens": 19996789.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_coding_entropy": 1.8921072945594788, |
| "eval_coding_loss": 2.2613472938537598, |
| "eval_coding_mean_token_accuracy": 0.5047942440509796, |
| "eval_coding_num_tokens": 19996789.0, |
| "eval_coding_runtime": 79.1902, |
| "eval_coding_samples_per_second": 6.314, |
| "eval_coding_steps_per_second": 3.157, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_physics_entropy": 1.099768555045128, |
| "eval_physics_loss": 1.101595401763916, |
| "eval_physics_mean_token_accuracy": 0.7247749812602997, |
| "eval_physics_num_tokens": 19996789.0, |
| "eval_physics_runtime": 52.0796, |
| "eval_physics_samples_per_second": 9.601, |
| "eval_physics_steps_per_second": 4.8, |
| "step": 1200 |
| }, |
| { |
| "entropy": 1.0720167379826306, |
| "epoch": 0.968, |
| "grad_norm": 7.03125, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 1.0638, |
| "mean_token_accuracy": 0.7312380861490965, |
| "num_tokens": 20161977.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 1.088957086019218, |
| "epoch": 0.976, |
| "grad_norm": 7.34375, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 1.0538, |
| "mean_token_accuracy": 0.7287674982100725, |
| "num_tokens": 20332636.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 1.08813561424613, |
| "epoch": 0.984, |
| "grad_norm": 6.34375, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 1.0477, |
| "mean_token_accuracy": 0.729033438116312, |
| "num_tokens": 20495388.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 1.0714545506983995, |
| "epoch": 0.992, |
| "grad_norm": 6.53125, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 1.0405, |
| "mean_token_accuracy": 0.7327663522213698, |
| "num_tokens": 20662209.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 1.0615619242191314, |
| "epoch": 1.0, |
| "grad_norm": 6.78125, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 1.0355, |
| "mean_token_accuracy": 0.7302098982036114, |
| "num_tokens": 20832148.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.9550450762733818, |
| "epoch": 1.008, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 0.9201, |
| "mean_token_accuracy": 0.7552305307239294, |
| "num_tokens": 21005731.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 1.007374680787325, |
| "epoch": 1.016, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 0.972, |
| "mean_token_accuracy": 0.7435515383258462, |
| "num_tokens": 21167221.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.9439798355102539, |
| "epoch": 1.024, |
| "grad_norm": 6.0625, |
| "learning_rate": 1.938e-05, |
| "loss": 0.9237, |
| "mean_token_accuracy": 0.7548253701999783, |
| "num_tokens": 21335407.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.9643905458971858, |
| "epoch": 1.032, |
| "grad_norm": 5.8125, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 0.9309, |
| "mean_token_accuracy": 0.752015108242631, |
| "num_tokens": 21501759.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.9763634527102113, |
| "epoch": 1.04, |
| "grad_norm": 6.4375, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 0.9625, |
| "mean_token_accuracy": 0.7475563909858465, |
| "num_tokens": 21668490.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_coding_entropy": 1.7818276858329773, |
| "eval_coding_loss": 2.242872953414917, |
| "eval_coding_mean_token_accuracy": 0.5101647647619247, |
| "eval_coding_num_tokens": 21668490.0, |
| "eval_coding_runtime": 79.2066, |
| "eval_coding_samples_per_second": 6.313, |
| "eval_coding_steps_per_second": 3.156, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_physics_entropy": 1.046526582121849, |
| "eval_physics_loss": 1.0569711923599243, |
| "eval_physics_mean_token_accuracy": 0.7336113708019256, |
| "eval_physics_num_tokens": 21668490.0, |
| "eval_physics_runtime": 52.1165, |
| "eval_physics_samples_per_second": 9.594, |
| "eval_physics_steps_per_second": 4.797, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.9808619644492864, |
| "epoch": 1.048, |
| "grad_norm": 5.9375, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 0.95, |
| "mean_token_accuracy": 0.7478477224707604, |
| "num_tokens": 21835179.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.939110497944057, |
| "epoch": 1.056, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 0.9088, |
| "mean_token_accuracy": 0.755623085796833, |
| "num_tokens": 22006039.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.9877023957669735, |
| "epoch": 1.064, |
| "grad_norm": 5.25, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 0.9472, |
| "mean_token_accuracy": 0.7455626420676709, |
| "num_tokens": 22176185.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.9839252684265375, |
| "epoch": 1.072, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 0.9612, |
| "mean_token_accuracy": 0.7474370829761028, |
| "num_tokens": 22348206.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.9325820317491889, |
| "epoch": 1.08, |
| "grad_norm": 6.375, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 0.896, |
| "mean_token_accuracy": 0.7573953222483396, |
| "num_tokens": 22513925.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.9836602587252855, |
| "epoch": 1.088, |
| "grad_norm": 6.84375, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 0.9531, |
| "mean_token_accuracy": 0.7478942912071943, |
| "num_tokens": 22676163.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.9665768353268505, |
| "epoch": 1.096, |
| "grad_norm": 5.96875, |
| "learning_rate": 1.918e-05, |
| "loss": 0.9489, |
| "mean_token_accuracy": 0.7473849574103951, |
| "num_tokens": 22841009.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.9201742075383663, |
| "epoch": 1.104, |
| "grad_norm": 6.0, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 0.8934, |
| "mean_token_accuracy": 0.7597939491271972, |
| "num_tokens": 23009279.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.9489134527742863, |
| "epoch": 1.112, |
| "grad_norm": 5.71875, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 0.9142, |
| "mean_token_accuracy": 0.7531155597418546, |
| "num_tokens": 23177689.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.9486297775059939, |
| "epoch": 1.12, |
| "grad_norm": 6.1875, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 0.9228, |
| "mean_token_accuracy": 0.7521506313234567, |
| "num_tokens": 23346828.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_coding_entropy": 1.7531000714302063, |
| "eval_coding_loss": 2.281940460205078, |
| "eval_coding_mean_token_accuracy": 0.5089057025909424, |
| "eval_coding_num_tokens": 23346828.0, |
| "eval_coding_runtime": 79.1787, |
| "eval_coding_samples_per_second": 6.315, |
| "eval_coding_steps_per_second": 3.157, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_physics_entropy": 0.9558427100181579, |
| "eval_physics_loss": 1.0335971117019653, |
| "eval_physics_mean_token_accuracy": 0.7392593598365784, |
| "eval_physics_num_tokens": 23346828.0, |
| "eval_physics_runtime": 52.0901, |
| "eval_physics_samples_per_second": 9.599, |
| "eval_physics_steps_per_second": 4.799, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.9056999972090125, |
| "epoch": 1.1280000000000001, |
| "grad_norm": 6.5, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 0.8848, |
| "mean_token_accuracy": 0.7649097748100757, |
| "num_tokens": 23518156.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.9454022387042642, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 0.9168, |
| "mean_token_accuracy": 0.755837232619524, |
| "num_tokens": 23689161.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.9329152461141348, |
| "epoch": 1.144, |
| "grad_norm": 5.59375, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 0.9142, |
| "mean_token_accuracy": 0.758990440890193, |
| "num_tokens": 23851046.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.961996317282319, |
| "epoch": 1.152, |
| "grad_norm": 4.875, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 0.9288, |
| "mean_token_accuracy": 0.7505928639322519, |
| "num_tokens": 24016439.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.9266120318323374, |
| "epoch": 1.16, |
| "grad_norm": 6.09375, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 0.9021, |
| "mean_token_accuracy": 0.7583789460361003, |
| "num_tokens": 24180461.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.9504798075184226, |
| "epoch": 1.168, |
| "grad_norm": 6.1875, |
| "learning_rate": 1.898e-05, |
| "loss": 0.9328, |
| "mean_token_accuracy": 0.7518207874149084, |
| "num_tokens": 24344431.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.9474708326160908, |
| "epoch": 1.176, |
| "grad_norm": 5.8125, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 0.9112, |
| "mean_token_accuracy": 0.7553022742271424, |
| "num_tokens": 24507213.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.9036590620875359, |
| "epoch": 1.184, |
| "grad_norm": 6.21875, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 0.8795, |
| "mean_token_accuracy": 0.762787076830864, |
| "num_tokens": 24668087.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.9184596356004476, |
| "epoch": 1.192, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 0.8956, |
| "mean_token_accuracy": 0.7611413709819317, |
| "num_tokens": 24832180.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.921324947103858, |
| "epoch": 1.2, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 0.8789, |
| "mean_token_accuracy": 0.7592108882963657, |
| "num_tokens": 25000629.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_coding_entropy": 1.7903581528663635, |
| "eval_coding_loss": 2.2501142024993896, |
| "eval_coding_mean_token_accuracy": 0.513159777879715, |
| "eval_coding_num_tokens": 25000629.0, |
| "eval_coding_runtime": 79.2232, |
| "eval_coding_samples_per_second": 6.311, |
| "eval_coding_steps_per_second": 3.156, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_physics_entropy": 0.990718915104866, |
| "eval_physics_loss": 0.9999967813491821, |
| "eval_physics_mean_token_accuracy": 0.7454895977973938, |
| "eval_physics_num_tokens": 25000629.0, |
| "eval_physics_runtime": 51.8001, |
| "eval_physics_samples_per_second": 9.652, |
| "eval_physics_steps_per_second": 4.826, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.9167924275621772, |
| "epoch": 1.208, |
| "grad_norm": 5.9375, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 0.892, |
| "mean_token_accuracy": 0.7588412739336491, |
| "num_tokens": 25164342.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.9410123957321048, |
| "epoch": 1.216, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 0.9122, |
| "mean_token_accuracy": 0.7562039643526077, |
| "num_tokens": 25324324.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.9521240815520287, |
| "epoch": 1.224, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 0.8969, |
| "mean_token_accuracy": 0.7552298158407211, |
| "num_tokens": 25490292.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.8985677545890212, |
| "epoch": 1.232, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 0.8661, |
| "mean_token_accuracy": 0.76500195376575, |
| "num_tokens": 25658795.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.8914181362837553, |
| "epoch": 1.24, |
| "grad_norm": 5.90625, |
| "learning_rate": 1.878e-05, |
| "loss": 0.8667, |
| "mean_token_accuracy": 0.7672365583479405, |
| "num_tokens": 25823796.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.9169950911775231, |
| "epoch": 1.248, |
| "grad_norm": 7.0625, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 0.8949, |
| "mean_token_accuracy": 0.7595178321003914, |
| "num_tokens": 25993177.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.9169469196349382, |
| "epoch": 1.256, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 0.8825, |
| "mean_token_accuracy": 0.7615648206323385, |
| "num_tokens": 26162587.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.896455617621541, |
| "epoch": 1.264, |
| "grad_norm": 4.625, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 0.8742, |
| "mean_token_accuracy": 0.7666802130639553, |
| "num_tokens": 26329726.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.9278371959924698, |
| "epoch": 1.272, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 0.8965, |
| "mean_token_accuracy": 0.7571958255022764, |
| "num_tokens": 26500094.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.9765459138900041, |
| "epoch": 1.28, |
| "grad_norm": 6.15625, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 0.9344, |
| "mean_token_accuracy": 0.7511742908507586, |
| "num_tokens": 26664093.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_coding_entropy": 1.6453372964859008, |
| "eval_coding_loss": 2.296245574951172, |
| "eval_coding_mean_token_accuracy": 0.510434584736824, |
| "eval_coding_num_tokens": 26664093.0, |
| "eval_coding_runtime": 79.1166, |
| "eval_coding_samples_per_second": 6.32, |
| "eval_coding_steps_per_second": 3.16, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_physics_entropy": 0.9116508244276047, |
| "eval_physics_loss": 0.981630265712738, |
| "eval_physics_mean_token_accuracy": 0.7496122629642487, |
| "eval_physics_num_tokens": 26664093.0, |
| "eval_physics_runtime": 51.9369, |
| "eval_physics_samples_per_second": 9.627, |
| "eval_physics_steps_per_second": 4.814, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.862612402252853, |
| "epoch": 1.288, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 0.8454, |
| "mean_token_accuracy": 0.7705432698130608, |
| "num_tokens": 26834758.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.9043381884694099, |
| "epoch": 1.296, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 0.8718, |
| "mean_token_accuracy": 0.7643989242613316, |
| "num_tokens": 26994877.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.9238925974816083, |
| "epoch": 1.304, |
| "grad_norm": 4.375, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 0.8873, |
| "mean_token_accuracy": 0.7609186865389347, |
| "num_tokens": 27162719.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.9316502708941699, |
| "epoch": 1.312, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.858e-05, |
| "loss": 0.907, |
| "mean_token_accuracy": 0.7582122638821602, |
| "num_tokens": 27328198.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.9211711158975959, |
| "epoch": 1.32, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 0.8892, |
| "mean_token_accuracy": 0.7606807623058558, |
| "num_tokens": 27492744.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.892648309096694, |
| "epoch": 1.328, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 0.8718, |
| "mean_token_accuracy": 0.7630397897213698, |
| "num_tokens": 27658818.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.8834815826267004, |
| "epoch": 1.336, |
| "grad_norm": 5.96875, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 0.8589, |
| "mean_token_accuracy": 0.7691333398222924, |
| "num_tokens": 27821424.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.9194148594513536, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 5.25, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 0.8972, |
| "mean_token_accuracy": 0.7602852337062359, |
| "num_tokens": 27991108.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.9116265259683132, |
| "epoch": 1.3519999999999999, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 0.8777, |
| "mean_token_accuracy": 0.7609406780451536, |
| "num_tokens": 28154773.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.8890536414459348, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 0.8562, |
| "mean_token_accuracy": 0.7662287544459104, |
| "num_tokens": 28327775.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "eval_coding_entropy": 1.716024712085724, |
| "eval_coding_loss": 2.259434461593628, |
| "eval_coding_mean_token_accuracy": 0.5130739635229111, |
| "eval_coding_num_tokens": 28327775.0, |
| "eval_coding_runtime": 78.6033, |
| "eval_coding_samples_per_second": 6.361, |
| "eval_coding_steps_per_second": 3.181, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "eval_physics_entropy": 0.914379727602005, |
| "eval_physics_loss": 0.9594675302505493, |
| "eval_physics_mean_token_accuracy": 0.7536831333637237, |
| "eval_physics_num_tokens": 28327775.0, |
| "eval_physics_runtime": 51.7228, |
| "eval_physics_samples_per_second": 9.667, |
| "eval_physics_steps_per_second": 4.833, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.8726189162582159, |
| "epoch": 1.3679999999999999, |
| "grad_norm": 5.0, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 0.8495, |
| "mean_token_accuracy": 0.7711174737662077, |
| "num_tokens": 28498470.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.9009999068453908, |
| "epoch": 1.376, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 0.8623, |
| "mean_token_accuracy": 0.7660522252321244, |
| "num_tokens": 28666116.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.8770303575322032, |
| "epoch": 1.384, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 0.8659, |
| "mean_token_accuracy": 0.767541291192174, |
| "num_tokens": 28840893.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.8704624468460679, |
| "epoch": 1.392, |
| "grad_norm": 6.0, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 0.8293, |
| "mean_token_accuracy": 0.7710605084896087, |
| "num_tokens": 29004293.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.866730142198503, |
| "epoch": 1.4, |
| "grad_norm": 5.125, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 0.8459, |
| "mean_token_accuracy": 0.7679262559860944, |
| "num_tokens": 29167707.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.8981963096186518, |
| "epoch": 1.408, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 0.8666, |
| "mean_token_accuracy": 0.7673941906541586, |
| "num_tokens": 29330131.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.9249771688133478, |
| "epoch": 1.416, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 0.9063, |
| "mean_token_accuracy": 0.7583306394517422, |
| "num_tokens": 29491469.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.8935416838154197, |
| "epoch": 1.424, |
| "grad_norm": 4.375, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 0.8645, |
| "mean_token_accuracy": 0.7656872101128102, |
| "num_tokens": 29661850.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.8505860712379217, |
| "epoch": 1.432, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 0.8276, |
| "mean_token_accuracy": 0.7741888701915741, |
| "num_tokens": 29825641.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.862689713574946, |
| "epoch": 1.44, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 0.8296, |
| "mean_token_accuracy": 0.7719305366277694, |
| "num_tokens": 29999392.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_coding_entropy": 1.8027336740493773, |
| "eval_coding_loss": 2.246039628982544, |
| "eval_coding_mean_token_accuracy": 0.5113436381816864, |
| "eval_coding_num_tokens": 29999392.0, |
| "eval_coding_runtime": 79.2093, |
| "eval_coding_samples_per_second": 6.312, |
| "eval_coding_steps_per_second": 3.156, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_physics_entropy": 0.913623704791069, |
| "eval_physics_loss": 0.9421977996826172, |
| "eval_physics_mean_token_accuracy": 0.7563698604106903, |
| "eval_physics_num_tokens": 29999392.0, |
| "eval_physics_runtime": 51.9126, |
| "eval_physics_samples_per_second": 9.632, |
| "eval_physics_steps_per_second": 4.816, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.876362593844533, |
| "epoch": 1.448, |
| "grad_norm": 5.71875, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 0.8509, |
| "mean_token_accuracy": 0.7696236498653889, |
| "num_tokens": 30164371.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.847023755684495, |
| "epoch": 1.456, |
| "grad_norm": 4.625, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 0.8231, |
| "mean_token_accuracy": 0.7763396494090558, |
| "num_tokens": 30341004.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.871413410641253, |
| "epoch": 1.464, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 0.8386, |
| "mean_token_accuracy": 0.7727533802390099, |
| "num_tokens": 30502703.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.939422488771379, |
| "epoch": 1.472, |
| "grad_norm": 5.46875, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 0.9126, |
| "mean_token_accuracy": 0.7551350846886635, |
| "num_tokens": 30667689.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.8579436099156738, |
| "epoch": 1.48, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 0.851, |
| "mean_token_accuracy": 0.772059577703476, |
| "num_tokens": 30836874.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.8878307269886136, |
| "epoch": 1.488, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 0.8514, |
| "mean_token_accuracy": 0.7688593063503504, |
| "num_tokens": 30999697.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.8808000456541777, |
| "epoch": 1.496, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 0.8501, |
| "mean_token_accuracy": 0.770066749304533, |
| "num_tokens": 31166146.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.8454932847991585, |
| "epoch": 1.504, |
| "grad_norm": 4.875, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 0.8113, |
| "mean_token_accuracy": 0.7748839784413576, |
| "num_tokens": 31332775.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.8540815737098455, |
| "epoch": 1.512, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 0.8343, |
| "mean_token_accuracy": 0.7708807934075594, |
| "num_tokens": 31504675.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.9056571874767542, |
| "epoch": 1.52, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 0.8672, |
| "mean_token_accuracy": 0.7641353111714124, |
| "num_tokens": 31666521.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_coding_entropy": 1.6874914655685425, |
| "eval_coding_loss": 2.2765512466430664, |
| "eval_coding_mean_token_accuracy": 0.5112714154720306, |
| "eval_coding_num_tokens": 31666521.0, |
| "eval_coding_runtime": 79.0285, |
| "eval_coding_samples_per_second": 6.327, |
| "eval_coding_steps_per_second": 3.163, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_physics_entropy": 0.8699047968387604, |
| "eval_physics_loss": 0.9348219037055969, |
| "eval_physics_mean_token_accuracy": 0.7589149172306061, |
| "eval_physics_num_tokens": 31666521.0, |
| "eval_physics_runtime": 51.8968, |
| "eval_physics_samples_per_second": 9.635, |
| "eval_physics_steps_per_second": 4.817, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.8628190591931343, |
| "epoch": 1.528, |
| "grad_norm": 6.5, |
| "learning_rate": 1.798e-05, |
| "loss": 0.842, |
| "mean_token_accuracy": 0.7697089036926628, |
| "num_tokens": 31831967.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.9051342781633138, |
| "epoch": 1.536, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 0.8713, |
| "mean_token_accuracy": 0.7649960003793239, |
| "num_tokens": 31998232.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.8381768845021724, |
| "epoch": 1.544, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 0.8225, |
| "mean_token_accuracy": 0.7794440522789955, |
| "num_tokens": 32163122.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.8429474547505379, |
| "epoch": 1.552, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 0.8141, |
| "mean_token_accuracy": 0.77677460052073, |
| "num_tokens": 32328085.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.8625445317476987, |
| "epoch": 1.56, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 0.8221, |
| "mean_token_accuracy": 0.7729156997054816, |
| "num_tokens": 32498672.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.8660145524889231, |
| "epoch": 1.568, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 0.8336, |
| "mean_token_accuracy": 0.7706921048462391, |
| "num_tokens": 32656603.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.8720831394195556, |
| "epoch": 1.576, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 0.8522, |
| "mean_token_accuracy": 0.7667239267379046, |
| "num_tokens": 32815222.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.8575057543814182, |
| "epoch": 1.584, |
| "grad_norm": 4.875, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 0.8218, |
| "mean_token_accuracy": 0.7737087935209275, |
| "num_tokens": 32983769.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.8643774883821607, |
| "epoch": 1.592, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 0.8371, |
| "mean_token_accuracy": 0.7728578582406044, |
| "num_tokens": 33144736.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.8772571442648769, |
| "epoch": 1.6, |
| "grad_norm": 3.828125, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 0.8394, |
| "mean_token_accuracy": 0.7701976072043181, |
| "num_tokens": 33309282.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_coding_entropy": 1.7155875730514527, |
| "eval_coding_loss": 2.268036127090454, |
| "eval_coding_mean_token_accuracy": 0.5065519729852677, |
| "eval_coding_num_tokens": 33309282.0, |
| "eval_coding_runtime": 79.042, |
| "eval_coding_samples_per_second": 6.326, |
| "eval_coding_steps_per_second": 3.163, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_physics_entropy": 0.8701518812179565, |
| "eval_physics_loss": 0.9167826771736145, |
| "eval_physics_mean_token_accuracy": 0.7627079751491547, |
| "eval_physics_num_tokens": 33309282.0, |
| "eval_physics_runtime": 51.8141, |
| "eval_physics_samples_per_second": 9.65, |
| "eval_physics_steps_per_second": 4.825, |
| "step": 2000 |
| }, |
| { |
| "entropy": 0.8041736917570234, |
| "epoch": 1.608, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.7757777777777777e-05, |
| "loss": 0.7911, |
| "mean_token_accuracy": 0.7850766744464636, |
| "num_tokens": 33485091.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 0.8584910430014133, |
| "epoch": 1.616, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.7735555555555558e-05, |
| "loss": 0.822, |
| "mean_token_accuracy": 0.7741672981530427, |
| "num_tokens": 33647616.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 0.8141503371298313, |
| "epoch": 1.624, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.7713333333333335e-05, |
| "loss": 0.7946, |
| "mean_token_accuracy": 0.7806556668132544, |
| "num_tokens": 33814904.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 0.8944115854799748, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.7691111111111113e-05, |
| "loss": 0.8494, |
| "mean_token_accuracy": 0.7674026219174266, |
| "num_tokens": 33977586.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 0.8180335260927677, |
| "epoch": 1.6400000000000001, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.766888888888889e-05, |
| "loss": 0.7943, |
| "mean_token_accuracy": 0.7807611677795648, |
| "num_tokens": 34143742.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 0.8157250108197331, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 5.5, |
| "learning_rate": 1.7646666666666668e-05, |
| "loss": 0.786, |
| "mean_token_accuracy": 0.7834763184189797, |
| "num_tokens": 34312436.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 0.8433098591864109, |
| "epoch": 1.6560000000000001, |
| "grad_norm": 5.25, |
| "learning_rate": 1.7624444444444446e-05, |
| "loss": 0.8172, |
| "mean_token_accuracy": 0.777406994253397, |
| "num_tokens": 34478504.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 0.8418531702831388, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 6.6875, |
| "learning_rate": 1.7602222222222223e-05, |
| "loss": 0.8182, |
| "mean_token_accuracy": 0.7756667707115412, |
| "num_tokens": 34641549.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 0.825640257447958, |
| "epoch": 1.6720000000000002, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.758e-05, |
| "loss": 0.7972, |
| "mean_token_accuracy": 0.7794421594589949, |
| "num_tokens": 34808044.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 0.8450808398425579, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.755777777777778e-05, |
| "loss": 0.8058, |
| "mean_token_accuracy": 0.7755126718431711, |
| "num_tokens": 34977031.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "eval_coding_entropy": 1.6850582294464111, |
| "eval_coding_loss": 2.2593472003936768, |
| "eval_coding_mean_token_accuracy": 0.5131129772663117, |
| "eval_coding_num_tokens": 34977031.0, |
| "eval_coding_runtime": 79.2037, |
| "eval_coding_samples_per_second": 6.313, |
| "eval_coding_steps_per_second": 3.156, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "eval_physics_entropy": 0.8579872615337372, |
| "eval_physics_loss": 0.9012077450752258, |
| "eval_physics_mean_token_accuracy": 0.7655161898136139, |
| "eval_physics_num_tokens": 34977031.0, |
| "eval_physics_runtime": 52.1742, |
| "eval_physics_samples_per_second": 9.583, |
| "eval_physics_steps_per_second": 4.792, |
| "step": 2100 |
| }, |
| { |
| "entropy": 0.8239427028223872, |
| "epoch": 1.688, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.7535555555555556e-05, |
| "loss": 0.8111, |
| "mean_token_accuracy": 0.7777137096971274, |
| "num_tokens": 35141418.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 0.8669725690037012, |
| "epoch": 1.696, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.7513333333333334e-05, |
| "loss": 0.8166, |
| "mean_token_accuracy": 0.774109473079443, |
| "num_tokens": 35300962.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 0.8681604472920299, |
| "epoch": 1.704, |
| "grad_norm": 5.125, |
| "learning_rate": 1.749111111111111e-05, |
| "loss": 0.8338, |
| "mean_token_accuracy": 0.7683083459734916, |
| "num_tokens": 35465304.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 0.8354681877419352, |
| "epoch": 1.712, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.746888888888889e-05, |
| "loss": 0.7984, |
| "mean_token_accuracy": 0.7774088971316815, |
| "num_tokens": 35642662.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 0.7997790666297078, |
| "epoch": 1.72, |
| "grad_norm": 4.375, |
| "learning_rate": 1.7446666666666667e-05, |
| "loss": 0.7903, |
| "mean_token_accuracy": 0.783223495259881, |
| "num_tokens": 35810260.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 0.8489536764100194, |
| "epoch": 1.728, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.7424444444444444e-05, |
| "loss": 0.8092, |
| "mean_token_accuracy": 0.7768806152045726, |
| "num_tokens": 35980993.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 0.8214310843497514, |
| "epoch": 1.736, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.7402222222222222e-05, |
| "loss": 0.7963, |
| "mean_token_accuracy": 0.7785018008202315, |
| "num_tokens": 36158360.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 0.8393697302788496, |
| "epoch": 1.744, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 0.8124, |
| "mean_token_accuracy": 0.7777101572602987, |
| "num_tokens": 36326797.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 0.8256299896165729, |
| "epoch": 1.752, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.735777777777778e-05, |
| "loss": 0.8052, |
| "mean_token_accuracy": 0.7793278936296701, |
| "num_tokens": 36495982.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 0.8003837876021862, |
| "epoch": 1.76, |
| "grad_norm": 4.75, |
| "learning_rate": 1.7335555555555558e-05, |
| "loss": 0.7815, |
| "mean_token_accuracy": 0.7837960578501224, |
| "num_tokens": 36654797.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_coding_entropy": 1.7353993301391601, |
| "eval_coding_loss": 2.24050235748291, |
| "eval_coding_mean_token_accuracy": 0.5156019266843795, |
| "eval_coding_num_tokens": 36654797.0, |
| "eval_coding_runtime": 79.2517, |
| "eval_coding_samples_per_second": 6.309, |
| "eval_coding_steps_per_second": 3.155, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_physics_entropy": 0.8497708789110183, |
| "eval_physics_loss": 0.8915674090385437, |
| "eval_physics_mean_token_accuracy": 0.7682892513275147, |
| "eval_physics_num_tokens": 36654797.0, |
| "eval_physics_runtime": 52.1968, |
| "eval_physics_samples_per_second": 9.579, |
| "eval_physics_steps_per_second": 4.79, |
| "step": 2200 |
| }, |
| { |
| "entropy": 0.8612198643386364, |
| "epoch": 1.768, |
| "grad_norm": 5.84375, |
| "learning_rate": 1.7313333333333336e-05, |
| "loss": 0.8251, |
| "mean_token_accuracy": 0.7738977476954461, |
| "num_tokens": 36817463.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 0.8459191296249629, |
| "epoch": 1.776, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.7291111111111113e-05, |
| "loss": 0.8264, |
| "mean_token_accuracy": 0.7742797108367085, |
| "num_tokens": 36989310.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 0.8024844925850629, |
| "epoch": 1.784, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.726888888888889e-05, |
| "loss": 0.7582, |
| "mean_token_accuracy": 0.7858391430228948, |
| "num_tokens": 37157452.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 0.8630851672962307, |
| "epoch": 1.792, |
| "grad_norm": 5.59375, |
| "learning_rate": 1.724666666666667e-05, |
| "loss": 0.8317, |
| "mean_token_accuracy": 0.7712806731462478, |
| "num_tokens": 37318535.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 0.8015124432742595, |
| "epoch": 1.8, |
| "grad_norm": 3.734375, |
| "learning_rate": 1.7224444444444446e-05, |
| "loss": 0.7698, |
| "mean_token_accuracy": 0.7848561979830265, |
| "num_tokens": 37487831.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 0.8111644266173244, |
| "epoch": 1.808, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.7202222222222224e-05, |
| "loss": 0.7993, |
| "mean_token_accuracy": 0.7826206222176552, |
| "num_tokens": 37658889.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 0.8139818359166384, |
| "epoch": 1.8159999999999998, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.718e-05, |
| "loss": 0.7889, |
| "mean_token_accuracy": 0.7834100909531116, |
| "num_tokens": 37829943.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 0.8303235996514559, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.715777777777778e-05, |
| "loss": 0.795, |
| "mean_token_accuracy": 0.7808313373476267, |
| "num_tokens": 37995198.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 0.8542025130242109, |
| "epoch": 1.8319999999999999, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.7135555555555557e-05, |
| "loss": 0.8313, |
| "mean_token_accuracy": 0.7719877079129219, |
| "num_tokens": 38155297.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 0.8089347897097469, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 4.0, |
| "learning_rate": 1.7113333333333334e-05, |
| "loss": 0.7779, |
| "mean_token_accuracy": 0.7852071691304445, |
| "num_tokens": 38325542.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "eval_coding_entropy": 1.7236487927436828, |
| "eval_coding_loss": 2.2375903129577637, |
| "eval_coding_mean_token_accuracy": 0.5160671151876449, |
| "eval_coding_num_tokens": 38325542.0, |
| "eval_coding_runtime": 79.4113, |
| "eval_coding_samples_per_second": 6.296, |
| "eval_coding_steps_per_second": 3.148, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "eval_physics_entropy": 0.8456154173612594, |
| "eval_physics_loss": 0.8841537833213806, |
| "eval_physics_mean_token_accuracy": 0.7688510994911194, |
| "eval_physics_num_tokens": 38325542.0, |
| "eval_physics_runtime": 52.4055, |
| "eval_physics_samples_per_second": 9.541, |
| "eval_physics_steps_per_second": 4.77, |
| "step": 2300 |
| }, |
| { |
| "entropy": 0.805744357407093, |
| "epoch": 1.8479999999999999, |
| "grad_norm": 3.84375, |
| "learning_rate": 1.7091111111111112e-05, |
| "loss": 0.7829, |
| "mean_token_accuracy": 0.784181559085846, |
| "num_tokens": 38497246.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 0.8343328120186925, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.706888888888889e-05, |
| "loss": 0.7888, |
| "mean_token_accuracy": 0.7806125942617654, |
| "num_tokens": 38668148.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 0.8110112180933357, |
| "epoch": 1.8639999999999999, |
| "grad_norm": 4.5, |
| "learning_rate": 1.704666666666667e-05, |
| "loss": 0.7752, |
| "mean_token_accuracy": 0.7813364051282405, |
| "num_tokens": 38840232.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 0.8125607699155808, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.7024444444444445e-05, |
| "loss": 0.7895, |
| "mean_token_accuracy": 0.7821429204195738, |
| "num_tokens": 39001652.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 0.8465429963544011, |
| "epoch": 1.88, |
| "grad_norm": 5.78125, |
| "learning_rate": 1.7002222222222226e-05, |
| "loss": 0.8126, |
| "mean_token_accuracy": 0.7767612311989069, |
| "num_tokens": 39169312.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 0.8441216722130775, |
| "epoch": 1.888, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.698e-05, |
| "loss": 0.8063, |
| "mean_token_accuracy": 0.7776072409003973, |
| "num_tokens": 39333474.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 0.8453333882614971, |
| "epoch": 1.896, |
| "grad_norm": 3.71875, |
| "learning_rate": 1.695777777777778e-05, |
| "loss": 0.813, |
| "mean_token_accuracy": 0.7757103960961104, |
| "num_tokens": 39499971.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 0.8293590556830168, |
| "epoch": 1.904, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.6935555555555555e-05, |
| "loss": 0.8131, |
| "mean_token_accuracy": 0.7780664421617984, |
| "num_tokens": 39663055.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 0.8597976418212057, |
| "epoch": 1.912, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.6913333333333336e-05, |
| "loss": 0.8246, |
| "mean_token_accuracy": 0.7739259105175733, |
| "num_tokens": 39823268.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 0.8054354287683964, |
| "epoch": 1.92, |
| "grad_norm": 5.5, |
| "learning_rate": 1.689111111111111e-05, |
| "loss": 0.7896, |
| "mean_token_accuracy": 0.7838379312306643, |
| "num_tokens": 39989824.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_coding_entropy": 1.7244014859199523, |
| "eval_coding_loss": 2.2286953926086426, |
| "eval_coding_mean_token_accuracy": 0.5165949345827102, |
| "eval_coding_num_tokens": 39989824.0, |
| "eval_coding_runtime": 79.4303, |
| "eval_coding_samples_per_second": 6.295, |
| "eval_coding_steps_per_second": 3.147, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_physics_entropy": 0.8283440818786622, |
| "eval_physics_loss": 0.8724973201751709, |
| "eval_physics_mean_token_accuracy": 0.7722779376506805, |
| "eval_physics_num_tokens": 39989824.0, |
| "eval_physics_runtime": 52.2349, |
| "eval_physics_samples_per_second": 9.572, |
| "eval_physics_steps_per_second": 4.786, |
| "step": 2400 |
| }, |
| { |
| "entropy": 0.8490746941417455, |
| "epoch": 1.928, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.686888888888889e-05, |
| "loss": 0.8138, |
| "mean_token_accuracy": 0.7763742931187153, |
| "num_tokens": 40159086.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 0.8009230157360434, |
| "epoch": 1.936, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.684666666666667e-05, |
| "loss": 0.782, |
| "mean_token_accuracy": 0.7832826662808656, |
| "num_tokens": 40325070.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 0.8089035371318459, |
| "epoch": 1.944, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.6824444444444447e-05, |
| "loss": 0.7736, |
| "mean_token_accuracy": 0.7844303779304027, |
| "num_tokens": 40489400.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 0.7484890200197697, |
| "epoch": 1.952, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.6802222222222224e-05, |
| "loss": 0.736, |
| "mean_token_accuracy": 0.7953914526849986, |
| "num_tokens": 40659614.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 0.8137759083881975, |
| "epoch": 1.96, |
| "grad_norm": 4.875, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 0.7801, |
| "mean_token_accuracy": 0.7849993612617254, |
| "num_tokens": 40824844.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 0.8361523715779186, |
| "epoch": 1.968, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.675777777777778e-05, |
| "loss": 0.799, |
| "mean_token_accuracy": 0.7781310323625803, |
| "num_tokens": 40991508.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 0.8114897888153791, |
| "epoch": 1.976, |
| "grad_norm": 4.0, |
| "learning_rate": 1.6735555555555557e-05, |
| "loss": 0.7787, |
| "mean_token_accuracy": 0.7807551980018616, |
| "num_tokens": 41160674.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 0.8438869906589389, |
| "epoch": 1.984, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.6713333333333335e-05, |
| "loss": 0.8216, |
| "mean_token_accuracy": 0.7774552527815104, |
| "num_tokens": 41328779.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 0.8189408538863063, |
| "epoch": 1.992, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.6691111111111112e-05, |
| "loss": 0.7889, |
| "mean_token_accuracy": 0.782219960168004, |
| "num_tokens": 41495501.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 0.8064166329801082, |
| "epoch": 2.0, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.666888888888889e-05, |
| "loss": 0.7833, |
| "mean_token_accuracy": 0.7833917330950498, |
| "num_tokens": 41664296.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_coding_entropy": 1.717286424636841, |
| "eval_coding_loss": 2.2317516803741455, |
| "eval_coding_mean_token_accuracy": 0.5175079522132874, |
| "eval_coding_num_tokens": 41664296.0, |
| "eval_coding_runtime": 79.4337, |
| "eval_coding_samples_per_second": 6.295, |
| "eval_coding_steps_per_second": 3.147, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_physics_entropy": 0.8259405951499939, |
| "eval_physics_loss": 0.8628295660018921, |
| "eval_physics_mean_token_accuracy": 0.7733515868186951, |
| "eval_physics_num_tokens": 41664296.0, |
| "eval_physics_runtime": 51.845, |
| "eval_physics_samples_per_second": 9.644, |
| "eval_physics_steps_per_second": 4.822, |
| "step": 2500 |
| }, |
| { |
| "entropy": 0.6267373934388161, |
| "epoch": 2.008, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.6646666666666668e-05, |
| "loss": 0.6037, |
| "mean_token_accuracy": 0.8214042343199253, |
| "num_tokens": 41832842.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 0.613454102165997, |
| "epoch": 2.016, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.6624444444444445e-05, |
| "loss": 0.5774, |
| "mean_token_accuracy": 0.8276482291519642, |
| "num_tokens": 42001193.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 0.6082960139028728, |
| "epoch": 2.024, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.6602222222222223e-05, |
| "loss": 0.5701, |
| "mean_token_accuracy": 0.8264455772936344, |
| "num_tokens": 42168656.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 0.618254185281694, |
| "epoch": 2.032, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.658e-05, |
| "loss": 0.5945, |
| "mean_token_accuracy": 0.822199922427535, |
| "num_tokens": 42340757.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 0.6162297690287233, |
| "epoch": 2.04, |
| "grad_norm": 4.375, |
| "learning_rate": 1.6557777777777778e-05, |
| "loss": 0.5882, |
| "mean_token_accuracy": 0.8243944302201272, |
| "num_tokens": 42505053.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 0.6441289829090238, |
| "epoch": 2.048, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.6535555555555556e-05, |
| "loss": 0.6276, |
| "mean_token_accuracy": 0.8147114396095276, |
| "num_tokens": 42670324.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 0.6393495127558708, |
| "epoch": 2.056, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.6513333333333333e-05, |
| "loss": 0.6024, |
| "mean_token_accuracy": 0.8184194505214691, |
| "num_tokens": 42840053.0, |
| "step": 2570 |
| }, |
| { |
| "entropy": 0.589662010036409, |
| "epoch": 2.064, |
| "grad_norm": 3.890625, |
| "learning_rate": 1.6491111111111114e-05, |
| "loss": 0.5638, |
| "mean_token_accuracy": 0.8299374941736459, |
| "num_tokens": 43010231.0, |
| "step": 2580 |
| }, |
| { |
| "entropy": 0.6177625317126513, |
| "epoch": 2.072, |
| "grad_norm": 4.75, |
| "learning_rate": 1.646888888888889e-05, |
| "loss": 0.5889, |
| "mean_token_accuracy": 0.822924955189228, |
| "num_tokens": 43175953.0, |
| "step": 2590 |
| }, |
| { |
| "entropy": 0.632154917344451, |
| "epoch": 2.08, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.644666666666667e-05, |
| "loss": 0.6016, |
| "mean_token_accuracy": 0.8200040526688099, |
| "num_tokens": 43340227.0, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_coding_entropy": 1.37443935918808, |
| "eval_coding_loss": 2.450657367706299, |
| "eval_coding_mean_token_accuracy": 0.510794317483902, |
| "eval_coding_num_tokens": 43340227.0, |
| "eval_coding_runtime": 79.4427, |
| "eval_coding_samples_per_second": 6.294, |
| "eval_coding_steps_per_second": 3.147, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_physics_entropy": 0.6555066047906876, |
| "eval_physics_loss": 0.9201055765151978, |
| "eval_physics_mean_token_accuracy": 0.7700548541545867, |
| "eval_physics_num_tokens": 43340227.0, |
| "eval_physics_runtime": 52.1793, |
| "eval_physics_samples_per_second": 9.582, |
| "eval_physics_steps_per_second": 4.791, |
| "step": 2600 |
| }, |
| { |
| "entropy": 0.639796886779368, |
| "epoch": 2.088, |
| "grad_norm": 4.625, |
| "learning_rate": 1.6424444444444444e-05, |
| "loss": 0.6096, |
| "mean_token_accuracy": 0.8176310174167156, |
| "num_tokens": 43505627.0, |
| "step": 2610 |
| }, |
| { |
| "entropy": 0.5980126017704606, |
| "epoch": 2.096, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.6402222222222225e-05, |
| "loss": 0.5813, |
| "mean_token_accuracy": 0.8265918381512165, |
| "num_tokens": 43673099.0, |
| "step": 2620 |
| }, |
| { |
| "entropy": 0.6023792052641511, |
| "epoch": 2.104, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.638e-05, |
| "loss": 0.5792, |
| "mean_token_accuracy": 0.825976712629199, |
| "num_tokens": 43833934.0, |
| "step": 2630 |
| }, |
| { |
| "entropy": 0.6025502832606435, |
| "epoch": 2.112, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.635777777777778e-05, |
| "loss": 0.5783, |
| "mean_token_accuracy": 0.8258870415389538, |
| "num_tokens": 44005986.0, |
| "step": 2640 |
| }, |
| { |
| "entropy": 0.5969322631135583, |
| "epoch": 2.12, |
| "grad_norm": 3.765625, |
| "learning_rate": 1.6335555555555558e-05, |
| "loss": 0.5691, |
| "mean_token_accuracy": 0.826336657628417, |
| "num_tokens": 44166266.0, |
| "step": 2650 |
| }, |
| { |
| "entropy": 0.6238168029114604, |
| "epoch": 2.128, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.6313333333333335e-05, |
| "loss": 0.606, |
| "mean_token_accuracy": 0.8218153961002826, |
| "num_tokens": 44321465.0, |
| "step": 2660 |
| }, |
| { |
| "entropy": 0.6158078715205193, |
| "epoch": 2.136, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.6291111111111113e-05, |
| "loss": 0.5882, |
| "mean_token_accuracy": 0.8230926766991615, |
| "num_tokens": 44489375.0, |
| "step": 2670 |
| }, |
| { |
| "entropy": 0.6018604909069836, |
| "epoch": 2.144, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.626888888888889e-05, |
| "loss": 0.5823, |
| "mean_token_accuracy": 0.8254737179726362, |
| "num_tokens": 44661027.0, |
| "step": 2680 |
| }, |
| { |
| "entropy": 0.6252726474776864, |
| "epoch": 2.152, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.6246666666666668e-05, |
| "loss": 0.6081, |
| "mean_token_accuracy": 0.8179498245939613, |
| "num_tokens": 44827429.0, |
| "step": 2690 |
| }, |
| { |
| "entropy": 0.6539112368598581, |
| "epoch": 2.16, |
| "grad_norm": 4.375, |
| "learning_rate": 1.6224444444444446e-05, |
| "loss": 0.6191, |
| "mean_token_accuracy": 0.8172940738499165, |
| "num_tokens": 44994236.0, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.16, |
| "eval_coding_entropy": 1.3508017370700836, |
| "eval_coding_loss": 2.459696054458618, |
| "eval_coding_mean_token_accuracy": 0.5095216584205627, |
| "eval_coding_num_tokens": 44994236.0, |
| "eval_coding_runtime": 79.7189, |
| "eval_coding_samples_per_second": 6.272, |
| "eval_coding_steps_per_second": 3.136, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.16, |
| "eval_physics_entropy": 0.6559373445510864, |
| "eval_physics_loss": 0.913827657699585, |
| "eval_physics_mean_token_accuracy": 0.7716447129249573, |
| "eval_physics_num_tokens": 44994236.0, |
| "eval_physics_runtime": 52.1899, |
| "eval_physics_samples_per_second": 9.58, |
| "eval_physics_steps_per_second": 4.79, |
| "step": 2700 |
| }, |
| { |
| "entropy": 0.6039218049496412, |
| "epoch": 2.168, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.6202222222222223e-05, |
| "loss": 0.5657, |
| "mean_token_accuracy": 0.8246539358049632, |
| "num_tokens": 45161571.0, |
| "step": 2710 |
| }, |
| { |
| "entropy": 0.603247781470418, |
| "epoch": 2.176, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.618e-05, |
| "loss": 0.5867, |
| "mean_token_accuracy": 0.8266920678317546, |
| "num_tokens": 45334280.0, |
| "step": 2720 |
| }, |
| { |
| "entropy": 0.5918160032480955, |
| "epoch": 2.184, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.615777777777778e-05, |
| "loss": 0.5677, |
| "mean_token_accuracy": 0.8290746930986643, |
| "num_tokens": 45499402.0, |
| "step": 2730 |
| }, |
| { |
| "entropy": 0.6160703864879906, |
| "epoch": 2.192, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.6135555555555556e-05, |
| "loss": 0.5889, |
| "mean_token_accuracy": 0.824155556038022, |
| "num_tokens": 45666954.0, |
| "step": 2740 |
| }, |
| { |
| "entropy": 0.5953169440850615, |
| "epoch": 2.2, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.6113333333333334e-05, |
| "loss": 0.5826, |
| "mean_token_accuracy": 0.8253328789025545, |
| "num_tokens": 45834132.0, |
| "step": 2750 |
| }, |
| { |
| "entropy": 0.6015375588089228, |
| "epoch": 2.208, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.609111111111111e-05, |
| "loss": 0.5723, |
| "mean_token_accuracy": 0.8269878134131432, |
| "num_tokens": 45999546.0, |
| "step": 2760 |
| }, |
| { |
| "entropy": 0.6001254281029105, |
| "epoch": 2.216, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.606888888888889e-05, |
| "loss": 0.5876, |
| "mean_token_accuracy": 0.8253753263503313, |
| "num_tokens": 46167159.0, |
| "step": 2770 |
| }, |
| { |
| "entropy": 0.6099058744497597, |
| "epoch": 2.224, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.6046666666666667e-05, |
| "loss": 0.5841, |
| "mean_token_accuracy": 0.8269340958446264, |
| "num_tokens": 46333809.0, |
| "step": 2780 |
| }, |
| { |
| "entropy": 0.5924582777544856, |
| "epoch": 2.232, |
| "grad_norm": 4.625, |
| "learning_rate": 1.6024444444444444e-05, |
| "loss": 0.5688, |
| "mean_token_accuracy": 0.8282918892800808, |
| "num_tokens": 46501691.0, |
| "step": 2790 |
| }, |
| { |
| "entropy": 0.6166877252981067, |
| "epoch": 2.24, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.6002222222222222e-05, |
| "loss": 0.5935, |
| "mean_token_accuracy": 0.8221744630485773, |
| "num_tokens": 46673328.0, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_coding_entropy": 1.3449536728858948, |
| "eval_coding_loss": 2.4668049812316895, |
| "eval_coding_mean_token_accuracy": 0.5107591571807861, |
| "eval_coding_num_tokens": 46673328.0, |
| "eval_coding_runtime": 79.6587, |
| "eval_coding_samples_per_second": 6.277, |
| "eval_coding_steps_per_second": 3.138, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_physics_entropy": 0.6617927324771881, |
| "eval_physics_loss": 0.9057582020759583, |
| "eval_physics_mean_token_accuracy": 0.7725264024734497, |
| "eval_physics_num_tokens": 46673328.0, |
| "eval_physics_runtime": 52.2066, |
| "eval_physics_samples_per_second": 9.577, |
| "eval_physics_steps_per_second": 4.789, |
| "step": 2800 |
| }, |
| { |
| "entropy": 0.6337659023702145, |
| "epoch": 2.248, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 0.5989, |
| "mean_token_accuracy": 0.8209787521511316, |
| "num_tokens": 46841085.0, |
| "step": 2810 |
| }, |
| { |
| "entropy": 0.6291999282315374, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 4.375, |
| "learning_rate": 1.5957777777777777e-05, |
| "loss": 0.6039, |
| "mean_token_accuracy": 0.8203245922923088, |
| "num_tokens": 47001509.0, |
| "step": 2820 |
| }, |
| { |
| "entropy": 0.5945435255765915, |
| "epoch": 2.2640000000000002, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.5935555555555558e-05, |
| "loss": 0.5754, |
| "mean_token_accuracy": 0.8288602285087109, |
| "num_tokens": 47171016.0, |
| "step": 2830 |
| }, |
| { |
| "entropy": 0.6188051411882043, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.5913333333333332e-05, |
| "loss": 0.5876, |
| "mean_token_accuracy": 0.8232857622206211, |
| "num_tokens": 47340142.0, |
| "step": 2840 |
| }, |
| { |
| "entropy": 0.6111969035118818, |
| "epoch": 2.2800000000000002, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.5891111111111113e-05, |
| "loss": 0.5884, |
| "mean_token_accuracy": 0.8232033301144839, |
| "num_tokens": 47507916.0, |
| "step": 2850 |
| }, |
| { |
| "entropy": 0.6393554072827101, |
| "epoch": 2.288, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.5868888888888888e-05, |
| "loss": 0.614, |
| "mean_token_accuracy": 0.8167192228138447, |
| "num_tokens": 47677550.0, |
| "step": 2860 |
| }, |
| { |
| "entropy": 0.6181843418627977, |
| "epoch": 2.296, |
| "grad_norm": 3.78125, |
| "learning_rate": 1.584666666666667e-05, |
| "loss": 0.5814, |
| "mean_token_accuracy": 0.8251254186034203, |
| "num_tokens": 47841479.0, |
| "step": 2870 |
| }, |
| { |
| "entropy": 0.6250437296926975, |
| "epoch": 2.304, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.5824444444444446e-05, |
| "loss": 0.6038, |
| "mean_token_accuracy": 0.8207253012806177, |
| "num_tokens": 47993851.0, |
| "step": 2880 |
| }, |
| { |
| "entropy": 0.6118797030299902, |
| "epoch": 2.312, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.5802222222222224e-05, |
| "loss": 0.5885, |
| "mean_token_accuracy": 0.8238303918391466, |
| "num_tokens": 48164405.0, |
| "step": 2890 |
| }, |
| { |
| "entropy": 0.604522593691945, |
| "epoch": 2.32, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.578e-05, |
| "loss": 0.5861, |
| "mean_token_accuracy": 0.8257577165961265, |
| "num_tokens": 48332577.0, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.32, |
| "eval_coding_entropy": 1.3523815784454345, |
| "eval_coding_loss": 2.4751594066619873, |
| "eval_coding_mean_token_accuracy": 0.5096428718566894, |
| "eval_coding_num_tokens": 48332577.0, |
| "eval_coding_runtime": 79.4259, |
| "eval_coding_samples_per_second": 6.295, |
| "eval_coding_steps_per_second": 3.148, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.32, |
| "eval_physics_entropy": 0.65406782579422, |
| "eval_physics_loss": 0.9018837809562683, |
| "eval_physics_mean_token_accuracy": 0.7731419174671174, |
| "eval_physics_num_tokens": 48332577.0, |
| "eval_physics_runtime": 52.3012, |
| "eval_physics_samples_per_second": 9.56, |
| "eval_physics_steps_per_second": 4.78, |
| "step": 2900 |
| }, |
| { |
| "entropy": 0.5681330621242523, |
| "epoch": 2.328, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.575777777777778e-05, |
| "loss": 0.5495, |
| "mean_token_accuracy": 0.8327588770538569, |
| "num_tokens": 48502589.0, |
| "step": 2910 |
| }, |
| { |
| "entropy": 0.6044447740539909, |
| "epoch": 2.336, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.5735555555555557e-05, |
| "loss": 0.5778, |
| "mean_token_accuracy": 0.826114010065794, |
| "num_tokens": 48676623.0, |
| "step": 2920 |
| }, |
| { |
| "entropy": 0.5827158512547612, |
| "epoch": 2.344, |
| "grad_norm": 4.625, |
| "learning_rate": 1.5713333333333334e-05, |
| "loss": 0.5586, |
| "mean_token_accuracy": 0.8299524106085301, |
| "num_tokens": 48842782.0, |
| "step": 2930 |
| }, |
| { |
| "entropy": 0.5938589986413717, |
| "epoch": 2.352, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.5691111111111112e-05, |
| "loss": 0.5696, |
| "mean_token_accuracy": 0.8275059584528208, |
| "num_tokens": 49007770.0, |
| "step": 2940 |
| }, |
| { |
| "entropy": 0.5799241781234741, |
| "epoch": 2.36, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.5668888888888893e-05, |
| "loss": 0.5673, |
| "mean_token_accuracy": 0.8314082741737365, |
| "num_tokens": 49185671.0, |
| "step": 2950 |
| }, |
| { |
| "entropy": 0.6408661613240838, |
| "epoch": 2.368, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.5646666666666667e-05, |
| "loss": 0.6059, |
| "mean_token_accuracy": 0.8191751647740603, |
| "num_tokens": 49350016.0, |
| "step": 2960 |
| }, |
| { |
| "entropy": 0.6003497345373034, |
| "epoch": 2.376, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.5624444444444448e-05, |
| "loss": 0.5894, |
| "mean_token_accuracy": 0.8256981022655964, |
| "num_tokens": 49512200.0, |
| "step": 2970 |
| }, |
| { |
| "entropy": 0.6165332013741136, |
| "epoch": 2.384, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.5602222222222222e-05, |
| "loss": 0.5927, |
| "mean_token_accuracy": 0.823698042705655, |
| "num_tokens": 49679074.0, |
| "step": 2980 |
| }, |
| { |
| "entropy": 0.611846811696887, |
| "epoch": 2.392, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 0.6, |
| "mean_token_accuracy": 0.8226090833544731, |
| "num_tokens": 49848577.0, |
| "step": 2990 |
| }, |
| { |
| "entropy": 0.6002186056226492, |
| "epoch": 2.4, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.5557777777777778e-05, |
| "loss": 0.5749, |
| "mean_token_accuracy": 0.8277635291218758, |
| "num_tokens": 50020648.0, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_coding_entropy": 1.349739289045334, |
| "eval_coding_loss": 2.481516122817993, |
| "eval_coding_mean_token_accuracy": 0.5096702026128769, |
| "eval_coding_num_tokens": 50020648.0, |
| "eval_coding_runtime": 79.305, |
| "eval_coding_samples_per_second": 6.305, |
| "eval_coding_steps_per_second": 3.152, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_physics_entropy": 0.6552768393754959, |
| "eval_physics_loss": 0.8977621793746948, |
| "eval_physics_mean_token_accuracy": 0.7747766389846802, |
| "eval_physics_num_tokens": 50020648.0, |
| "eval_physics_runtime": 51.7134, |
| "eval_physics_samples_per_second": 9.669, |
| "eval_physics_steps_per_second": 4.834, |
| "step": 3000 |
| }, |
| { |
| "entropy": 0.6143279479816556, |
| "epoch": 2.408, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.553555555555556e-05, |
| "loss": 0.5952, |
| "mean_token_accuracy": 0.8243662029504776, |
| "num_tokens": 50196721.0, |
| "step": 3010 |
| }, |
| { |
| "entropy": 0.5889019390568138, |
| "epoch": 2.416, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.5513333333333333e-05, |
| "loss": 0.5781, |
| "mean_token_accuracy": 0.8261144172400237, |
| "num_tokens": 50370135.0, |
| "step": 3020 |
| }, |
| { |
| "entropy": 0.61474132174626, |
| "epoch": 2.424, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.5491111111111114e-05, |
| "loss": 0.5791, |
| "mean_token_accuracy": 0.8264197196811438, |
| "num_tokens": 50529922.0, |
| "step": 3030 |
| }, |
| { |
| "entropy": 0.621494947001338, |
| "epoch": 2.432, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.546888888888889e-05, |
| "loss": 0.6078, |
| "mean_token_accuracy": 0.8213183127343655, |
| "num_tokens": 50695363.0, |
| "step": 3040 |
| }, |
| { |
| "entropy": 0.6155845869332552, |
| "epoch": 2.44, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.544666666666667e-05, |
| "loss": 0.5859, |
| "mean_token_accuracy": 0.8250884145498276, |
| "num_tokens": 50861141.0, |
| "step": 3050 |
| }, |
| { |
| "entropy": 0.6297358063980937, |
| "epoch": 2.448, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.5424444444444447e-05, |
| "loss": 0.6097, |
| "mean_token_accuracy": 0.8201751086860896, |
| "num_tokens": 51023498.0, |
| "step": 3060 |
| }, |
| { |
| "entropy": 0.6233988754451275, |
| "epoch": 2.456, |
| "grad_norm": 4.125, |
| "learning_rate": 1.5402222222222224e-05, |
| "loss": 0.5963, |
| "mean_token_accuracy": 0.8221795730292797, |
| "num_tokens": 51187205.0, |
| "step": 3070 |
| }, |
| { |
| "entropy": 0.6175283070653677, |
| "epoch": 2.464, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 0.5789, |
| "mean_token_accuracy": 0.823941082879901, |
| "num_tokens": 51349816.0, |
| "step": 3080 |
| }, |
| { |
| "entropy": 0.606114044599235, |
| "epoch": 2.472, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.535777777777778e-05, |
| "loss": 0.5836, |
| "mean_token_accuracy": 0.8250056773424148, |
| "num_tokens": 51518966.0, |
| "step": 3090 |
| }, |
| { |
| "entropy": 0.6060268938541412, |
| "epoch": 2.48, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.5335555555555557e-05, |
| "loss": 0.5785, |
| "mean_token_accuracy": 0.8250746950507164, |
| "num_tokens": 51680507.0, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.48, |
| "eval_coding_entropy": 1.35354470205307, |
| "eval_coding_loss": 2.473125457763672, |
| "eval_coding_mean_token_accuracy": 0.510221528172493, |
| "eval_coding_num_tokens": 51680507.0, |
| "eval_coding_runtime": 79.0945, |
| "eval_coding_samples_per_second": 6.322, |
| "eval_coding_steps_per_second": 3.161, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.48, |
| "eval_physics_entropy": 0.6581197297573089, |
| "eval_physics_loss": 0.8899555206298828, |
| "eval_physics_mean_token_accuracy": 0.7764303793907166, |
| "eval_physics_num_tokens": 51680507.0, |
| "eval_physics_runtime": 52.0283, |
| "eval_physics_samples_per_second": 9.61, |
| "eval_physics_steps_per_second": 4.805, |
| "step": 3100 |
| }, |
| { |
| "entropy": 0.620896311942488, |
| "epoch": 2.488, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.5313333333333335e-05, |
| "loss": 0.6086, |
| "mean_token_accuracy": 0.8216223731637001, |
| "num_tokens": 51850021.0, |
| "step": 3110 |
| }, |
| { |
| "entropy": 0.6087856724858284, |
| "epoch": 2.496, |
| "grad_norm": 4.25, |
| "learning_rate": 1.5291111111111112e-05, |
| "loss": 0.5757, |
| "mean_token_accuracy": 0.8269111402332783, |
| "num_tokens": 52017752.0, |
| "step": 3120 |
| }, |
| { |
| "entropy": 0.5899860902689398, |
| "epoch": 2.504, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.526888888888889e-05, |
| "loss": 0.5661, |
| "mean_token_accuracy": 0.8284287117421627, |
| "num_tokens": 52182225.0, |
| "step": 3130 |
| }, |
| { |
| "entropy": 0.6004647793248296, |
| "epoch": 2.512, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.5246666666666668e-05, |
| "loss": 0.5699, |
| "mean_token_accuracy": 0.8293101724237204, |
| "num_tokens": 52345439.0, |
| "step": 3140 |
| }, |
| { |
| "entropy": 0.6012964773923158, |
| "epoch": 2.52, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.5224444444444447e-05, |
| "loss": 0.5882, |
| "mean_token_accuracy": 0.8248716313391924, |
| "num_tokens": 52512638.0, |
| "step": 3150 |
| }, |
| { |
| "entropy": 0.5881863052025438, |
| "epoch": 2.528, |
| "grad_norm": 3.953125, |
| "learning_rate": 1.5202222222222223e-05, |
| "loss": 0.5618, |
| "mean_token_accuracy": 0.8315121363848448, |
| "num_tokens": 52683886.0, |
| "step": 3160 |
| }, |
| { |
| "entropy": 0.5888467213138938, |
| "epoch": 2.536, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 0.565, |
| "mean_token_accuracy": 0.829793405905366, |
| "num_tokens": 52843943.0, |
| "step": 3170 |
| }, |
| { |
| "entropy": 0.6155722702853381, |
| "epoch": 2.544, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.5157777777777778e-05, |
| "loss": 0.5873, |
| "mean_token_accuracy": 0.8239083737134933, |
| "num_tokens": 53004529.0, |
| "step": 3180 |
| }, |
| { |
| "entropy": 0.6100942814722657, |
| "epoch": 2.552, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.5135555555555557e-05, |
| "loss": 0.5926, |
| "mean_token_accuracy": 0.8236753419041634, |
| "num_tokens": 53168822.0, |
| "step": 3190 |
| }, |
| { |
| "entropy": 0.5878218809142709, |
| "epoch": 2.56, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.5113333333333335e-05, |
| "loss": 0.5591, |
| "mean_token_accuracy": 0.830387581884861, |
| "num_tokens": 53336280.0, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_coding_entropy": 1.3497926185131073, |
| "eval_coding_loss": 2.468592882156372, |
| "eval_coding_mean_token_accuracy": 0.5101335799694061, |
| "eval_coding_num_tokens": 53336280.0, |
| "eval_coding_runtime": 79.5126, |
| "eval_coding_samples_per_second": 6.288, |
| "eval_coding_steps_per_second": 3.144, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_physics_entropy": 0.6507383289337159, |
| "eval_physics_loss": 0.8879098892211914, |
| "eval_physics_mean_token_accuracy": 0.7767288887500763, |
| "eval_physics_num_tokens": 53336280.0, |
| "eval_physics_runtime": 52.214, |
| "eval_physics_samples_per_second": 9.576, |
| "eval_physics_steps_per_second": 4.788, |
| "step": 3200 |
| }, |
| { |
| "entropy": 0.583949718810618, |
| "epoch": 2.568, |
| "grad_norm": 4.375, |
| "learning_rate": 1.5091111111111113e-05, |
| "loss": 0.5666, |
| "mean_token_accuracy": 0.8300550568848848, |
| "num_tokens": 53500372.0, |
| "step": 3210 |
| }, |
| { |
| "entropy": 0.6028478074818849, |
| "epoch": 2.576, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.506888888888889e-05, |
| "loss": 0.5727, |
| "mean_token_accuracy": 0.827161419019103, |
| "num_tokens": 53667647.0, |
| "step": 3220 |
| }, |
| { |
| "entropy": 0.6196060003712773, |
| "epoch": 2.584, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.5046666666666668e-05, |
| "loss": 0.5955, |
| "mean_token_accuracy": 0.8225338887423277, |
| "num_tokens": 53829770.0, |
| "step": 3230 |
| }, |
| { |
| "entropy": 0.5970783233642578, |
| "epoch": 2.592, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.5024444444444445e-05, |
| "loss": 0.5666, |
| "mean_token_accuracy": 0.8276160791516304, |
| "num_tokens": 53997614.0, |
| "step": 3240 |
| }, |
| { |
| "entropy": 0.6138347880914807, |
| "epoch": 2.6, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.5002222222222223e-05, |
| "loss": 0.5778, |
| "mean_token_accuracy": 0.8222551662474871, |
| "num_tokens": 54162416.0, |
| "step": 3250 |
| }, |
| { |
| "entropy": 0.5776338375173509, |
| "epoch": 2.608, |
| "grad_norm": 4.125, |
| "learning_rate": 1.498e-05, |
| "loss": 0.5567, |
| "mean_token_accuracy": 0.8313048735260964, |
| "num_tokens": 54330195.0, |
| "step": 3260 |
| }, |
| { |
| "entropy": 0.5890184611082077, |
| "epoch": 2.616, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.495777777777778e-05, |
| "loss": 0.5625, |
| "mean_token_accuracy": 0.829730398952961, |
| "num_tokens": 54496496.0, |
| "step": 3270 |
| }, |
| { |
| "entropy": 0.6196051860228181, |
| "epoch": 2.624, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.4935555555555556e-05, |
| "loss": 0.584, |
| "mean_token_accuracy": 0.8229093950241804, |
| "num_tokens": 54664342.0, |
| "step": 3280 |
| }, |
| { |
| "entropy": 0.6057103559374809, |
| "epoch": 2.632, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.4913333333333335e-05, |
| "loss": 0.5759, |
| "mean_token_accuracy": 0.8247067466378212, |
| "num_tokens": 54825560.0, |
| "step": 3290 |
| }, |
| { |
| "entropy": 0.6122049475088716, |
| "epoch": 2.64, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.4891111111111111e-05, |
| "loss": 0.5833, |
| "mean_token_accuracy": 0.8250450536608696, |
| "num_tokens": 54993626.0, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.64, |
| "eval_coding_entropy": 1.3738399562835693, |
| "eval_coding_loss": 2.4581685066223145, |
| "eval_coding_mean_token_accuracy": 0.5096644060611725, |
| "eval_coding_num_tokens": 54993626.0, |
| "eval_coding_runtime": 79.3766, |
| "eval_coding_samples_per_second": 6.299, |
| "eval_coding_steps_per_second": 3.15, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.64, |
| "eval_physics_entropy": 0.660741778254509, |
| "eval_physics_loss": 0.8812330365180969, |
| "eval_physics_mean_token_accuracy": 0.7774314482212067, |
| "eval_physics_num_tokens": 54993626.0, |
| "eval_physics_runtime": 52.1478, |
| "eval_physics_samples_per_second": 9.588, |
| "eval_physics_steps_per_second": 4.794, |
| "step": 3300 |
| }, |
| { |
| "entropy": 0.6118903381749987, |
| "epoch": 2.648, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.486888888888889e-05, |
| "loss": 0.5827, |
| "mean_token_accuracy": 0.8242136374115944, |
| "num_tokens": 55168382.0, |
| "step": 3310 |
| }, |
| { |
| "entropy": 0.6567736398428679, |
| "epoch": 2.656, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.4846666666666666e-05, |
| "loss": 0.63, |
| "mean_token_accuracy": 0.8162180844694376, |
| "num_tokens": 55339962.0, |
| "step": 3320 |
| }, |
| { |
| "entropy": 0.6124452227726579, |
| "epoch": 2.664, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.4824444444444446e-05, |
| "loss": 0.597, |
| "mean_token_accuracy": 0.8222682312130928, |
| "num_tokens": 55504628.0, |
| "step": 3330 |
| }, |
| { |
| "entropy": 0.6048196353949606, |
| "epoch": 2.672, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.4802222222222225e-05, |
| "loss": 0.5789, |
| "mean_token_accuracy": 0.8265079196542502, |
| "num_tokens": 55670567.0, |
| "step": 3340 |
| }, |
| { |
| "entropy": 0.5801108043640852, |
| "epoch": 2.68, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.478e-05, |
| "loss": 0.5571, |
| "mean_token_accuracy": 0.8314849089831113, |
| "num_tokens": 55837991.0, |
| "step": 3350 |
| }, |
| { |
| "entropy": 0.6023631945252419, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.475777777777778e-05, |
| "loss": 0.5846, |
| "mean_token_accuracy": 0.8274341113865376, |
| "num_tokens": 56003156.0, |
| "step": 3360 |
| }, |
| { |
| "entropy": 0.6052138676866889, |
| "epoch": 2.6959999999999997, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.4735555555555556e-05, |
| "loss": 0.5783, |
| "mean_token_accuracy": 0.8274508312344551, |
| "num_tokens": 56177679.0, |
| "step": 3370 |
| }, |
| { |
| "entropy": 0.6188904494047165, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.4713333333333335e-05, |
| "loss": 0.5907, |
| "mean_token_accuracy": 0.8227757509797812, |
| "num_tokens": 56342926.0, |
| "step": 3380 |
| }, |
| { |
| "entropy": 0.603462403267622, |
| "epoch": 2.7119999999999997, |
| "grad_norm": 4.5, |
| "learning_rate": 1.4691111111111111e-05, |
| "loss": 0.585, |
| "mean_token_accuracy": 0.8255855433642865, |
| "num_tokens": 56510821.0, |
| "step": 3390 |
| }, |
| { |
| "entropy": 0.5923863802105188, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.466888888888889e-05, |
| "loss": 0.57, |
| "mean_token_accuracy": 0.8282286584377289, |
| "num_tokens": 56677521.0, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_coding_entropy": 1.3335860939025879, |
| "eval_coding_loss": 2.4741172790527344, |
| "eval_coding_mean_token_accuracy": 0.510094462275505, |
| "eval_coding_num_tokens": 56677521.0, |
| "eval_coding_runtime": 79.2133, |
| "eval_coding_samples_per_second": 6.312, |
| "eval_coding_steps_per_second": 3.156, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_physics_entropy": 0.6450637406110764, |
| "eval_physics_loss": 0.8810954689979553, |
| "eval_physics_mean_token_accuracy": 0.7775574955940247, |
| "eval_physics_num_tokens": 56677521.0, |
| "eval_physics_runtime": 52.1872, |
| "eval_physics_samples_per_second": 9.581, |
| "eval_physics_steps_per_second": 4.79, |
| "step": 3400 |
| }, |
| { |
| "entropy": 0.611242413148284, |
| "epoch": 2.7279999999999998, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.4646666666666666e-05, |
| "loss": 0.5847, |
| "mean_token_accuracy": 0.8257013656198978, |
| "num_tokens": 56838328.0, |
| "step": 3410 |
| }, |
| { |
| "entropy": 0.619679344072938, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.4624444444444446e-05, |
| "loss": 0.5947, |
| "mean_token_accuracy": 0.8227478194981813, |
| "num_tokens": 57003161.0, |
| "step": 3420 |
| }, |
| { |
| "entropy": 0.6120741838589311, |
| "epoch": 2.7439999999999998, |
| "grad_norm": 4.875, |
| "learning_rate": 1.4602222222222225e-05, |
| "loss": 0.5832, |
| "mean_token_accuracy": 0.8231907237321139, |
| "num_tokens": 57163139.0, |
| "step": 3430 |
| }, |
| { |
| "entropy": 0.6255141733214259, |
| "epoch": 2.752, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 0.6017, |
| "mean_token_accuracy": 0.821520210057497, |
| "num_tokens": 57335850.0, |
| "step": 3440 |
| }, |
| { |
| "entropy": 0.5965338239446283, |
| "epoch": 2.76, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.455777777777778e-05, |
| "loss": 0.5663, |
| "mean_token_accuracy": 0.8295173797756433, |
| "num_tokens": 57505353.0, |
| "step": 3450 |
| }, |
| { |
| "entropy": 0.590160558000207, |
| "epoch": 2.768, |
| "grad_norm": 4.375, |
| "learning_rate": 1.4535555555555556e-05, |
| "loss": 0.5709, |
| "mean_token_accuracy": 0.8295429788529873, |
| "num_tokens": 57672649.0, |
| "step": 3460 |
| }, |
| { |
| "entropy": 0.6105528209358454, |
| "epoch": 2.776, |
| "grad_norm": 3.84375, |
| "learning_rate": 1.4513333333333336e-05, |
| "loss": 0.5868, |
| "mean_token_accuracy": 0.8259994424879551, |
| "num_tokens": 57837671.0, |
| "step": 3470 |
| }, |
| { |
| "entropy": 0.5995156250894069, |
| "epoch": 2.784, |
| "grad_norm": 4.375, |
| "learning_rate": 1.4491111111111111e-05, |
| "loss": 0.5781, |
| "mean_token_accuracy": 0.8267749030143022, |
| "num_tokens": 58007431.0, |
| "step": 3480 |
| }, |
| { |
| "entropy": 0.6301500285044312, |
| "epoch": 2.792, |
| "grad_norm": 4.25, |
| "learning_rate": 1.446888888888889e-05, |
| "loss": 0.6029, |
| "mean_token_accuracy": 0.8213560383766889, |
| "num_tokens": 58166452.0, |
| "step": 3490 |
| }, |
| { |
| "entropy": 0.6354689218103886, |
| "epoch": 2.8, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.4446666666666668e-05, |
| "loss": 0.6117, |
| "mean_token_accuracy": 0.8186302322894334, |
| "num_tokens": 58332730.0, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_coding_entropy": 1.3431692762374878, |
| "eval_coding_loss": 2.4674572944641113, |
| "eval_coding_mean_token_accuracy": 0.5104503653049469, |
| "eval_coding_num_tokens": 58332730.0, |
| "eval_coding_runtime": 79.3012, |
| "eval_coding_samples_per_second": 6.305, |
| "eval_coding_steps_per_second": 3.153, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_physics_entropy": 0.6469574912786483, |
| "eval_physics_loss": 0.8759974837303162, |
| "eval_physics_mean_token_accuracy": 0.7790948405265808, |
| "eval_physics_num_tokens": 58332730.0, |
| "eval_physics_runtime": 51.8278, |
| "eval_physics_samples_per_second": 9.647, |
| "eval_physics_steps_per_second": 4.824, |
| "step": 3500 |
| }, |
| { |
| "entropy": 0.581175766699016, |
| "epoch": 2.808, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.4424444444444446e-05, |
| "loss": 0.5625, |
| "mean_token_accuracy": 0.8309446167200804, |
| "num_tokens": 58498673.0, |
| "step": 3510 |
| }, |
| { |
| "entropy": 0.5683499926701188, |
| "epoch": 2.816, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.4402222222222224e-05, |
| "loss": 0.5488, |
| "mean_token_accuracy": 0.8349772330373526, |
| "num_tokens": 58672106.0, |
| "step": 3520 |
| }, |
| { |
| "entropy": 0.625259586237371, |
| "epoch": 2.824, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 0.5958, |
| "mean_token_accuracy": 0.8214506339281797, |
| "num_tokens": 58829550.0, |
| "step": 3530 |
| }, |
| { |
| "entropy": 0.5870909972116352, |
| "epoch": 2.832, |
| "grad_norm": 4.875, |
| "learning_rate": 1.4357777777777779e-05, |
| "loss": 0.5634, |
| "mean_token_accuracy": 0.8310813024640084, |
| "num_tokens": 58997592.0, |
| "step": 3540 |
| }, |
| { |
| "entropy": 0.5966650864109397, |
| "epoch": 2.84, |
| "grad_norm": 4.375, |
| "learning_rate": 1.4335555555555556e-05, |
| "loss": 0.5702, |
| "mean_token_accuracy": 0.8293530613183975, |
| "num_tokens": 59165798.0, |
| "step": 3550 |
| }, |
| { |
| "entropy": 0.593536170385778, |
| "epoch": 2.848, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.4313333333333334e-05, |
| "loss": 0.5852, |
| "mean_token_accuracy": 0.8260917115956545, |
| "num_tokens": 59326336.0, |
| "step": 3560 |
| }, |
| { |
| "entropy": 0.5985551407560706, |
| "epoch": 2.856, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.4291111111111113e-05, |
| "loss": 0.5806, |
| "mean_token_accuracy": 0.8293380744755268, |
| "num_tokens": 59497403.0, |
| "step": 3570 |
| }, |
| { |
| "entropy": 0.5909693138673902, |
| "epoch": 2.864, |
| "grad_norm": 4.375, |
| "learning_rate": 1.426888888888889e-05, |
| "loss": 0.5613, |
| "mean_token_accuracy": 0.8294834043830633, |
| "num_tokens": 59668586.0, |
| "step": 3580 |
| }, |
| { |
| "entropy": 0.5909178427420556, |
| "epoch": 2.872, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.4246666666666669e-05, |
| "loss": 0.5649, |
| "mean_token_accuracy": 0.8311422824859619, |
| "num_tokens": 59833072.0, |
| "step": 3590 |
| }, |
| { |
| "entropy": 0.6086006159894168, |
| "epoch": 2.88, |
| "grad_norm": 5.0, |
| "learning_rate": 1.4224444444444445e-05, |
| "loss": 0.584, |
| "mean_token_accuracy": 0.8246811926364899, |
| "num_tokens": 59993572.0, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_coding_entropy": 1.3511269943714141, |
| "eval_coding_loss": 2.4732460975646973, |
| "eval_coding_mean_token_accuracy": 0.5091971122026443, |
| "eval_coding_num_tokens": 59993572.0, |
| "eval_coding_runtime": 79.3468, |
| "eval_coding_samples_per_second": 6.301, |
| "eval_coding_steps_per_second": 3.151, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_physics_entropy": 0.662506309390068, |
| "eval_physics_loss": 0.8710473775863647, |
| "eval_physics_mean_token_accuracy": 0.7800338654518127, |
| "eval_physics_num_tokens": 59993572.0, |
| "eval_physics_runtime": 52.1568, |
| "eval_physics_samples_per_second": 9.586, |
| "eval_physics_steps_per_second": 4.793, |
| "step": 3600 |
| }, |
| { |
| "entropy": 0.5712122654542326, |
| "epoch": 2.888, |
| "grad_norm": 4.5, |
| "learning_rate": 1.4202222222222224e-05, |
| "loss": 0.5519, |
| "mean_token_accuracy": 0.8347968529909849, |
| "num_tokens": 60161738.0, |
| "step": 3610 |
| }, |
| { |
| "entropy": 0.606316895596683, |
| "epoch": 2.896, |
| "grad_norm": 4.125, |
| "learning_rate": 1.418e-05, |
| "loss": 0.5809, |
| "mean_token_accuracy": 0.8250825423747301, |
| "num_tokens": 60324128.0, |
| "step": 3620 |
| }, |
| { |
| "entropy": 0.5778268726542592, |
| "epoch": 2.904, |
| "grad_norm": 4.5, |
| "learning_rate": 1.4157777777777779e-05, |
| "loss": 0.5582, |
| "mean_token_accuracy": 0.8324020821601152, |
| "num_tokens": 60486604.0, |
| "step": 3630 |
| }, |
| { |
| "entropy": 0.5976108478382229, |
| "epoch": 2.912, |
| "grad_norm": 4.5, |
| "learning_rate": 1.4135555555555555e-05, |
| "loss": 0.5764, |
| "mean_token_accuracy": 0.8290127456188202, |
| "num_tokens": 60657399.0, |
| "step": 3640 |
| }, |
| { |
| "entropy": 0.5928749725222587, |
| "epoch": 2.92, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.4113333333333334e-05, |
| "loss": 0.5669, |
| "mean_token_accuracy": 0.8293946173042059, |
| "num_tokens": 60816620.0, |
| "step": 3650 |
| }, |
| { |
| "entropy": 0.5822400348260999, |
| "epoch": 2.928, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.4091111111111114e-05, |
| "loss": 0.5543, |
| "mean_token_accuracy": 0.8312558867037296, |
| "num_tokens": 60984711.0, |
| "step": 3660 |
| }, |
| { |
| "entropy": 0.599770469404757, |
| "epoch": 2.936, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.406888888888889e-05, |
| "loss": 0.5594, |
| "mean_token_accuracy": 0.8292138613760471, |
| "num_tokens": 61150576.0, |
| "step": 3670 |
| }, |
| { |
| "entropy": 0.5714768514968455, |
| "epoch": 2.944, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.4046666666666669e-05, |
| "loss": 0.5638, |
| "mean_token_accuracy": 0.832502580806613, |
| "num_tokens": 61321359.0, |
| "step": 3680 |
| }, |
| { |
| "entropy": 0.6112455248832702, |
| "epoch": 2.952, |
| "grad_norm": 3.671875, |
| "learning_rate": 1.4024444444444445e-05, |
| "loss": 0.5904, |
| "mean_token_accuracy": 0.8250285256654024, |
| "num_tokens": 61490234.0, |
| "step": 3690 |
| }, |
| { |
| "entropy": 0.5833005086518824, |
| "epoch": 2.96, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.4002222222222224e-05, |
| "loss": 0.5583, |
| "mean_token_accuracy": 0.8299073360860347, |
| "num_tokens": 61659042.0, |
| "step": 3700 |
| }, |
| { |
| "epoch": 2.96, |
| "eval_coding_entropy": 1.346010504245758, |
| "eval_coding_loss": 2.472853422164917, |
| "eval_coding_mean_token_accuracy": 0.5105656070709228, |
| "eval_coding_num_tokens": 61659042.0, |
| "eval_coding_runtime": 79.3058, |
| "eval_coding_samples_per_second": 6.305, |
| "eval_coding_steps_per_second": 3.152, |
| "step": 3700 |
| }, |
| { |
| "epoch": 2.96, |
| "eval_physics_entropy": 0.6391666326522827, |
| "eval_physics_loss": 0.8725790977478027, |
| "eval_physics_mean_token_accuracy": 0.779846668958664, |
| "eval_physics_num_tokens": 61659042.0, |
| "eval_physics_runtime": 52.1233, |
| "eval_physics_samples_per_second": 9.593, |
| "eval_physics_steps_per_second": 4.796, |
| "step": 3700 |
| }, |
| { |
| "entropy": 0.6079537514597178, |
| "epoch": 2.968, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.398e-05, |
| "loss": 0.5779, |
| "mean_token_accuracy": 0.8248198796063662, |
| "num_tokens": 61823323.0, |
| "step": 3710 |
| }, |
| { |
| "entropy": 0.600875335931778, |
| "epoch": 2.976, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.395777777777778e-05, |
| "loss": 0.5674, |
| "mean_token_accuracy": 0.8291897263377905, |
| "num_tokens": 61993595.0, |
| "step": 3720 |
| }, |
| { |
| "entropy": 0.577336086705327, |
| "epoch": 2.984, |
| "grad_norm": 4.75, |
| "learning_rate": 1.3935555555555557e-05, |
| "loss": 0.5596, |
| "mean_token_accuracy": 0.8334302712231875, |
| "num_tokens": 62163881.0, |
| "step": 3730 |
| }, |
| { |
| "entropy": 0.5730939632281661, |
| "epoch": 2.992, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.3913333333333335e-05, |
| "loss": 0.5481, |
| "mean_token_accuracy": 0.8316211476922035, |
| "num_tokens": 62334012.0, |
| "step": 3740 |
| }, |
| { |
| "entropy": 0.5661907571367919, |
| "epoch": 3.0, |
| "grad_norm": 5.125, |
| "learning_rate": 1.3891111111111114e-05, |
| "loss": 0.5451, |
| "mean_token_accuracy": 0.8358493153005838, |
| "num_tokens": 62496444.0, |
| "step": 3750 |
| }, |
| { |
| "entropy": 0.42234776820987463, |
| "epoch": 3.008, |
| "grad_norm": 5.46875, |
| "learning_rate": 1.386888888888889e-05, |
| "loss": 0.3698, |
| "mean_token_accuracy": 0.8843890953809023, |
| "num_tokens": 62660928.0, |
| "step": 3760 |
| }, |
| { |
| "entropy": 0.382783761434257, |
| "epoch": 3.016, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.3846666666666669e-05, |
| "loss": 0.3522, |
| "mean_token_accuracy": 0.8870693776756525, |
| "num_tokens": 62822736.0, |
| "step": 3770 |
| }, |
| { |
| "entropy": 0.38227535421028735, |
| "epoch": 3.024, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.3824444444444445e-05, |
| "loss": 0.3412, |
| "mean_token_accuracy": 0.8905686508864165, |
| "num_tokens": 62992670.0, |
| "step": 3780 |
| }, |
| { |
| "entropy": 0.3676112797111273, |
| "epoch": 3.032, |
| "grad_norm": 4.625, |
| "learning_rate": 1.3802222222222224e-05, |
| "loss": 0.3445, |
| "mean_token_accuracy": 0.8898772832006216, |
| "num_tokens": 63166479.0, |
| "step": 3790 |
| }, |
| { |
| "entropy": 0.37529820706695316, |
| "epoch": 3.04, |
| "grad_norm": 5.25, |
| "learning_rate": 1.378e-05, |
| "loss": 0.3428, |
| "mean_token_accuracy": 0.8897315964102746, |
| "num_tokens": 63335157.0, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_coding_entropy": 1.0409069275856018, |
| "eval_coding_loss": 2.957873582839966, |
| "eval_coding_mean_token_accuracy": 0.4940963124036789, |
| "eval_coding_num_tokens": 63335157.0, |
| "eval_coding_runtime": 79.3324, |
| "eval_coding_samples_per_second": 6.303, |
| "eval_coding_steps_per_second": 3.151, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_physics_entropy": 0.46495462536811827, |
| "eval_physics_loss": 1.0424754619598389, |
| "eval_physics_mean_token_accuracy": 0.7717642667293548, |
| "eval_physics_num_tokens": 63335157.0, |
| "eval_physics_runtime": 52.1506, |
| "eval_physics_samples_per_second": 9.588, |
| "eval_physics_steps_per_second": 4.794, |
| "step": 3800 |
| }, |
| { |
| "entropy": 0.38329875115305184, |
| "epoch": 3.048, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.375777777777778e-05, |
| "loss": 0.3463, |
| "mean_token_accuracy": 0.8881552763283252, |
| "num_tokens": 63495448.0, |
| "step": 3810 |
| }, |
| { |
| "entropy": 0.3797042270191014, |
| "epoch": 3.056, |
| "grad_norm": 5.25, |
| "learning_rate": 1.3735555555555557e-05, |
| "loss": 0.3494, |
| "mean_token_accuracy": 0.8880059469491244, |
| "num_tokens": 63662314.0, |
| "step": 3820 |
| }, |
| { |
| "entropy": 0.36236086413264273, |
| "epoch": 3.064, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.3713333333333335e-05, |
| "loss": 0.3344, |
| "mean_token_accuracy": 0.8910818416625261, |
| "num_tokens": 63827771.0, |
| "step": 3830 |
| }, |
| { |
| "entropy": 0.39001957830041645, |
| "epoch": 3.072, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.3691111111111112e-05, |
| "loss": 0.3613, |
| "mean_token_accuracy": 0.8849859081208706, |
| "num_tokens": 63997780.0, |
| "step": 3840 |
| }, |
| { |
| "entropy": 0.3801506935618818, |
| "epoch": 3.08, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.366888888888889e-05, |
| "loss": 0.3514, |
| "mean_token_accuracy": 0.8855122305452824, |
| "num_tokens": 64160135.0, |
| "step": 3850 |
| }, |
| { |
| "entropy": 0.37905457550659777, |
| "epoch": 3.088, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.3646666666666668e-05, |
| "loss": 0.3466, |
| "mean_token_accuracy": 0.8873339295387268, |
| "num_tokens": 64325274.0, |
| "step": 3860 |
| }, |
| { |
| "entropy": 0.3634565350599587, |
| "epoch": 3.096, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.3624444444444445e-05, |
| "loss": 0.3372, |
| "mean_token_accuracy": 0.8906644720584154, |
| "num_tokens": 64498675.0, |
| "step": 3870 |
| }, |
| { |
| "entropy": 0.37181077813729646, |
| "epoch": 3.104, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.3602222222222223e-05, |
| "loss": 0.3422, |
| "mean_token_accuracy": 0.8906843084841967, |
| "num_tokens": 64659683.0, |
| "step": 3880 |
| }, |
| { |
| "entropy": 0.3587542022578418, |
| "epoch": 3.112, |
| "grad_norm": 4.625, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 0.3373, |
| "mean_token_accuracy": 0.8922356773167849, |
| "num_tokens": 64826996.0, |
| "step": 3890 |
| }, |
| { |
| "entropy": 0.3602323802188039, |
| "epoch": 3.12, |
| "grad_norm": 5.125, |
| "learning_rate": 1.3557777777777778e-05, |
| "loss": 0.3346, |
| "mean_token_accuracy": 0.8933298390358686, |
| "num_tokens": 64998212.0, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.12, |
| "eval_coding_entropy": 1.0204580907821654, |
| "eval_coding_loss": 2.995056629180908, |
| "eval_coding_mean_token_accuracy": 0.49392418837547303, |
| "eval_coding_num_tokens": 64998212.0, |
| "eval_coding_runtime": 79.6061, |
| "eval_coding_samples_per_second": 6.281, |
| "eval_coding_steps_per_second": 3.14, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.12, |
| "eval_physics_entropy": 0.4634668634533882, |
| "eval_physics_loss": 1.0529135465621948, |
| "eval_physics_mean_token_accuracy": 0.7717892789840698, |
| "eval_physics_num_tokens": 64998212.0, |
| "eval_physics_runtime": 52.0804, |
| "eval_physics_samples_per_second": 9.601, |
| "eval_physics_steps_per_second": 4.8, |
| "step": 3900 |
| }, |
| { |
| "entropy": 0.39407063769176603, |
| "epoch": 3.128, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.3535555555555557e-05, |
| "loss": 0.3602, |
| "mean_token_accuracy": 0.8827631033957004, |
| "num_tokens": 65155653.0, |
| "step": 3910 |
| }, |
| { |
| "entropy": 0.37749654073268174, |
| "epoch": 3.136, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.3513333333333333e-05, |
| "loss": 0.35, |
| "mean_token_accuracy": 0.8872763376682997, |
| "num_tokens": 65328436.0, |
| "step": 3920 |
| }, |
| { |
| "entropy": 0.35364336604252455, |
| "epoch": 3.144, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.3491111111111113e-05, |
| "loss": 0.3243, |
| "mean_token_accuracy": 0.8945908039808274, |
| "num_tokens": 65495949.0, |
| "step": 3930 |
| }, |
| { |
| "entropy": 0.36987975845113397, |
| "epoch": 3.152, |
| "grad_norm": 5.125, |
| "learning_rate": 1.3468888888888888e-05, |
| "loss": 0.3438, |
| "mean_token_accuracy": 0.8887593563646078, |
| "num_tokens": 65659346.0, |
| "step": 3940 |
| }, |
| { |
| "entropy": 0.386718874424696, |
| "epoch": 3.16, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.3446666666666668e-05, |
| "loss": 0.3595, |
| "mean_token_accuracy": 0.8847455944865942, |
| "num_tokens": 65822312.0, |
| "step": 3950 |
| }, |
| { |
| "entropy": 0.37097716331481934, |
| "epoch": 3.168, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.3424444444444447e-05, |
| "loss": 0.3382, |
| "mean_token_accuracy": 0.8905388720333576, |
| "num_tokens": 65986382.0, |
| "step": 3960 |
| }, |
| { |
| "entropy": 0.3660844849422574, |
| "epoch": 3.176, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.3402222222222223e-05, |
| "loss": 0.336, |
| "mean_token_accuracy": 0.8902653686702251, |
| "num_tokens": 66160738.0, |
| "step": 3970 |
| }, |
| { |
| "entropy": 0.3804160048253834, |
| "epoch": 3.184, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 0.3503, |
| "mean_token_accuracy": 0.8860555302351714, |
| "num_tokens": 66324308.0, |
| "step": 3980 |
| }, |
| { |
| "entropy": 0.38136240411549804, |
| "epoch": 3.192, |
| "grad_norm": 5.71875, |
| "learning_rate": 1.3357777777777778e-05, |
| "loss": 0.3511, |
| "mean_token_accuracy": 0.8879099935293198, |
| "num_tokens": 66493224.0, |
| "step": 3990 |
| }, |
| { |
| "entropy": 0.37074621077626946, |
| "epoch": 3.2, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.3335555555555558e-05, |
| "loss": 0.344, |
| "mean_token_accuracy": 0.8887821324169636, |
| "num_tokens": 66658712.0, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_coding_entropy": 1.01984184217453, |
| "eval_coding_loss": 3.005119800567627, |
| "eval_coding_mean_token_accuracy": 0.4930288236141205, |
| "eval_coding_num_tokens": 66658712.0, |
| "eval_coding_runtime": 79.3471, |
| "eval_coding_samples_per_second": 6.301, |
| "eval_coding_steps_per_second": 3.151, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_physics_entropy": 0.4624620456695557, |
| "eval_physics_loss": 1.0565595626831055, |
| "eval_physics_mean_token_accuracy": 0.7716201021671295, |
| "eval_physics_num_tokens": 66658712.0, |
| "eval_physics_runtime": 52.0937, |
| "eval_physics_samples_per_second": 9.598, |
| "eval_physics_steps_per_second": 4.799, |
| "step": 4000 |
| }, |
| { |
| "entropy": 0.3687359469011426, |
| "epoch": 3.208, |
| "grad_norm": 4.875, |
| "learning_rate": 1.3313333333333333e-05, |
| "loss": 0.3382, |
| "mean_token_accuracy": 0.889621015265584, |
| "num_tokens": 66825006.0, |
| "step": 4010 |
| }, |
| { |
| "entropy": 0.37608798453584313, |
| "epoch": 3.216, |
| "grad_norm": 5.0, |
| "learning_rate": 1.3291111111111113e-05, |
| "loss": 0.346, |
| "mean_token_accuracy": 0.8872137345373631, |
| "num_tokens": 66990142.0, |
| "step": 4020 |
| }, |
| { |
| "entropy": 0.395880725979805, |
| "epoch": 3.224, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.3268888888888889e-05, |
| "loss": 0.3632, |
| "mean_token_accuracy": 0.8830111533403396, |
| "num_tokens": 67152359.0, |
| "step": 4030 |
| }, |
| { |
| "entropy": 0.3963136670179665, |
| "epoch": 3.232, |
| "grad_norm": 5.125, |
| "learning_rate": 1.3246666666666668e-05, |
| "loss": 0.3668, |
| "mean_token_accuracy": 0.8830410160124302, |
| "num_tokens": 67323231.0, |
| "step": 4040 |
| }, |
| { |
| "entropy": 0.3608850891701877, |
| "epoch": 3.24, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.3224444444444446e-05, |
| "loss": 0.3385, |
| "mean_token_accuracy": 0.8910890750586986, |
| "num_tokens": 67493986.0, |
| "step": 4050 |
| }, |
| { |
| "entropy": 0.36570190070196984, |
| "epoch": 3.248, |
| "grad_norm": 5.875, |
| "learning_rate": 1.3202222222222223e-05, |
| "loss": 0.3435, |
| "mean_token_accuracy": 0.8898316625505686, |
| "num_tokens": 67664253.0, |
| "step": 4060 |
| }, |
| { |
| "entropy": 0.37633528099395336, |
| "epoch": 3.2560000000000002, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 0.3427, |
| "mean_token_accuracy": 0.8867505926638841, |
| "num_tokens": 67827707.0, |
| "step": 4070 |
| }, |
| { |
| "entropy": 0.37384027782827617, |
| "epoch": 3.2640000000000002, |
| "grad_norm": 5.9375, |
| "learning_rate": 1.3157777777777778e-05, |
| "loss": 0.3474, |
| "mean_token_accuracy": 0.8890785362571478, |
| "num_tokens": 67991082.0, |
| "step": 4080 |
| }, |
| { |
| "entropy": 0.37050414048135283, |
| "epoch": 3.2720000000000002, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.3135555555555558e-05, |
| "loss": 0.3389, |
| "mean_token_accuracy": 0.8890818245708942, |
| "num_tokens": 68157996.0, |
| "step": 4090 |
| }, |
| { |
| "entropy": 0.3820228569675237, |
| "epoch": 3.2800000000000002, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.3113333333333334e-05, |
| "loss": 0.3541, |
| "mean_token_accuracy": 0.886331158131361, |
| "num_tokens": 68317737.0, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.2800000000000002, |
| "eval_coding_entropy": 1.0199904294013977, |
| "eval_coding_loss": 3.0209267139434814, |
| "eval_coding_mean_token_accuracy": 0.4920610988140106, |
| "eval_coding_num_tokens": 68317737.0, |
| "eval_coding_runtime": 79.261, |
| "eval_coding_samples_per_second": 6.308, |
| "eval_coding_steps_per_second": 3.154, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.2800000000000002, |
| "eval_physics_entropy": 0.4655421174764633, |
| "eval_physics_loss": 1.0537148714065552, |
| "eval_physics_mean_token_accuracy": 0.7707213397026063, |
| "eval_physics_num_tokens": 68317737.0, |
| "eval_physics_runtime": 52.2055, |
| "eval_physics_samples_per_second": 9.578, |
| "eval_physics_steps_per_second": 4.789, |
| "step": 4100 |
| }, |
| { |
| "entropy": 0.3780054041184485, |
| "epoch": 3.288, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.3091111111111113e-05, |
| "loss": 0.3586, |
| "mean_token_accuracy": 0.8851760346442461, |
| "num_tokens": 68487161.0, |
| "step": 4110 |
| }, |
| { |
| "entropy": 0.3680135501548648, |
| "epoch": 3.296, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.306888888888889e-05, |
| "loss": 0.3426, |
| "mean_token_accuracy": 0.8898434318602085, |
| "num_tokens": 68653995.0, |
| "step": 4120 |
| }, |
| { |
| "entropy": 0.38406856255605815, |
| "epoch": 3.304, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.3046666666666668e-05, |
| "loss": 0.3494, |
| "mean_token_accuracy": 0.8867152512073517, |
| "num_tokens": 68823169.0, |
| "step": 4130 |
| }, |
| { |
| "entropy": 0.3735280029475689, |
| "epoch": 3.312, |
| "grad_norm": 5.25, |
| "learning_rate": 1.3024444444444446e-05, |
| "loss": 0.346, |
| "mean_token_accuracy": 0.8880814719945193, |
| "num_tokens": 68994300.0, |
| "step": 4140 |
| }, |
| { |
| "entropy": 0.3876368122175336, |
| "epoch": 3.32, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.3002222222222223e-05, |
| "loss": 0.3647, |
| "mean_token_accuracy": 0.8845827709883451, |
| "num_tokens": 69162700.0, |
| "step": 4150 |
| }, |
| { |
| "entropy": 0.3814154784195125, |
| "epoch": 3.328, |
| "grad_norm": 5.625, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 0.3543, |
| "mean_token_accuracy": 0.8850255236029625, |
| "num_tokens": 69323456.0, |
| "step": 4160 |
| }, |
| { |
| "entropy": 0.38359872410073875, |
| "epoch": 3.336, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.2957777777777779e-05, |
| "loss": 0.3582, |
| "mean_token_accuracy": 0.8844977922737598, |
| "num_tokens": 69489946.0, |
| "step": 4170 |
| }, |
| { |
| "entropy": 0.3773769561201334, |
| "epoch": 3.344, |
| "grad_norm": 5.875, |
| "learning_rate": 1.2935555555555556e-05, |
| "loss": 0.3487, |
| "mean_token_accuracy": 0.887412928044796, |
| "num_tokens": 69659373.0, |
| "step": 4180 |
| }, |
| { |
| "entropy": 0.38140337420627474, |
| "epoch": 3.352, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.2913333333333336e-05, |
| "loss": 0.3554, |
| "mean_token_accuracy": 0.8859492279589176, |
| "num_tokens": 69830145.0, |
| "step": 4190 |
| }, |
| { |
| "entropy": 0.39355383040383457, |
| "epoch": 3.36, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.2891111111111112e-05, |
| "loss": 0.3643, |
| "mean_token_accuracy": 0.8822440423071385, |
| "num_tokens": 69998761.0, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_coding_entropy": 1.0311934192180634, |
| "eval_coding_loss": 2.995810031890869, |
| "eval_coding_mean_token_accuracy": 0.49246379554271696, |
| "eval_coding_num_tokens": 69998761.0, |
| "eval_coding_runtime": 79.2672, |
| "eval_coding_samples_per_second": 6.308, |
| "eval_coding_steps_per_second": 3.154, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_physics_entropy": 0.47156879740953445, |
| "eval_physics_loss": 1.0502831935882568, |
| "eval_physics_mean_token_accuracy": 0.7711660912036896, |
| "eval_physics_num_tokens": 69998761.0, |
| "eval_physics_runtime": 52.1213, |
| "eval_physics_samples_per_second": 9.593, |
| "eval_physics_steps_per_second": 4.797, |
| "step": 4200 |
| }, |
| { |
| "entropy": 0.38113170452415945, |
| "epoch": 3.368, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.2868888888888891e-05, |
| "loss": 0.3495, |
| "mean_token_accuracy": 0.887039466202259, |
| "num_tokens": 70173483.0, |
| "step": 4210 |
| }, |
| { |
| "entropy": 0.3615466058254242, |
| "epoch": 3.376, |
| "grad_norm": 5.125, |
| "learning_rate": 1.2846666666666667e-05, |
| "loss": 0.3385, |
| "mean_token_accuracy": 0.8919301256537437, |
| "num_tokens": 70344663.0, |
| "step": 4220 |
| }, |
| { |
| "entropy": 0.38931633485481143, |
| "epoch": 3.384, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.2824444444444446e-05, |
| "loss": 0.3549, |
| "mean_token_accuracy": 0.8840402740985155, |
| "num_tokens": 70510115.0, |
| "step": 4230 |
| }, |
| { |
| "entropy": 0.3735740644857287, |
| "epoch": 3.392, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.2802222222222222e-05, |
| "loss": 0.3486, |
| "mean_token_accuracy": 0.886800118163228, |
| "num_tokens": 70671470.0, |
| "step": 4240 |
| }, |
| { |
| "entropy": 0.3768125709146261, |
| "epoch": 3.4, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 0.3495, |
| "mean_token_accuracy": 0.8869495436549186, |
| "num_tokens": 70835486.0, |
| "step": 4250 |
| }, |
| { |
| "entropy": 0.3772420261986554, |
| "epoch": 3.408, |
| "grad_norm": 5.96875, |
| "learning_rate": 1.2757777777777777e-05, |
| "loss": 0.3519, |
| "mean_token_accuracy": 0.8866618055850267, |
| "num_tokens": 71003257.0, |
| "step": 4260 |
| }, |
| { |
| "entropy": 0.3700935058295727, |
| "epoch": 3.416, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.2735555555555557e-05, |
| "loss": 0.3487, |
| "mean_token_accuracy": 0.8873854592442513, |
| "num_tokens": 71165000.0, |
| "step": 4270 |
| }, |
| { |
| "entropy": 0.37494869697839023, |
| "epoch": 3.424, |
| "grad_norm": 5.875, |
| "learning_rate": 1.2713333333333336e-05, |
| "loss": 0.3449, |
| "mean_token_accuracy": 0.887766731157899, |
| "num_tokens": 71333695.0, |
| "step": 4280 |
| }, |
| { |
| "entropy": 0.3676827215589583, |
| "epoch": 3.432, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.2691111111111112e-05, |
| "loss": 0.3513, |
| "mean_token_accuracy": 0.8884382307529449, |
| "num_tokens": 71502153.0, |
| "step": 4290 |
| }, |
| { |
| "entropy": 0.4056678608059883, |
| "epoch": 3.44, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.2668888888888891e-05, |
| "loss": 0.3749, |
| "mean_token_accuracy": 0.8803832571953535, |
| "num_tokens": 71672438.0, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.44, |
| "eval_coding_entropy": 1.0257436592578888, |
| "eval_coding_loss": 3.0044476985931396, |
| "eval_coding_mean_token_accuracy": 0.4924482258558273, |
| "eval_coding_num_tokens": 71672438.0, |
| "eval_coding_runtime": 79.3364, |
| "eval_coding_samples_per_second": 6.302, |
| "eval_coding_steps_per_second": 3.151, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.44, |
| "eval_physics_entropy": 0.47017036139965057, |
| "eval_physics_loss": 1.0506194829940796, |
| "eval_physics_mean_token_accuracy": 0.7719645640850067, |
| "eval_physics_num_tokens": 71672438.0, |
| "eval_physics_runtime": 52.1443, |
| "eval_physics_samples_per_second": 9.589, |
| "eval_physics_steps_per_second": 4.794, |
| "step": 4300 |
| }, |
| { |
| "entropy": 0.3925869864411652, |
| "epoch": 3.448, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.2646666666666667e-05, |
| "loss": 0.3583, |
| "mean_token_accuracy": 0.8844016410410405, |
| "num_tokens": 71842590.0, |
| "step": 4310 |
| }, |
| { |
| "entropy": 0.37298523718491194, |
| "epoch": 3.456, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.2624444444444446e-05, |
| "loss": 0.354, |
| "mean_token_accuracy": 0.887511258944869, |
| "num_tokens": 72009302.0, |
| "step": 4320 |
| }, |
| { |
| "entropy": 0.38410011427477003, |
| "epoch": 3.464, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.2602222222222222e-05, |
| "loss": 0.3601, |
| "mean_token_accuracy": 0.8846664246171713, |
| "num_tokens": 72168371.0, |
| "step": 4330 |
| }, |
| { |
| "entropy": 0.380076789110899, |
| "epoch": 3.472, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 0.3507, |
| "mean_token_accuracy": 0.8869646117091179, |
| "num_tokens": 72334603.0, |
| "step": 4340 |
| }, |
| { |
| "entropy": 0.3751876854337752, |
| "epoch": 3.48, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.2557777777777779e-05, |
| "loss": 0.3464, |
| "mean_token_accuracy": 0.8875335864722729, |
| "num_tokens": 72501103.0, |
| "step": 4350 |
| }, |
| { |
| "entropy": 0.37113183857873083, |
| "epoch": 3.488, |
| "grad_norm": 5.78125, |
| "learning_rate": 1.2535555555555557e-05, |
| "loss": 0.3468, |
| "mean_token_accuracy": 0.8888764757663011, |
| "num_tokens": 72669209.0, |
| "step": 4360 |
| }, |
| { |
| "entropy": 0.3841777864843607, |
| "epoch": 3.496, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.2513333333333334e-05, |
| "loss": 0.3635, |
| "mean_token_accuracy": 0.8854869689792395, |
| "num_tokens": 72833408.0, |
| "step": 4370 |
| }, |
| { |
| "entropy": 0.3892757879570127, |
| "epoch": 3.504, |
| "grad_norm": 6.15625, |
| "learning_rate": 1.2491111111111112e-05, |
| "loss": 0.3544, |
| "mean_token_accuracy": 0.8844669658690691, |
| "num_tokens": 73001765.0, |
| "step": 4380 |
| }, |
| { |
| "entropy": 0.3706186875700951, |
| "epoch": 3.512, |
| "grad_norm": 5.0, |
| "learning_rate": 1.246888888888889e-05, |
| "loss": 0.3447, |
| "mean_token_accuracy": 0.8879940524697304, |
| "num_tokens": 73168103.0, |
| "step": 4390 |
| }, |
| { |
| "entropy": 0.40220920694991946, |
| "epoch": 3.52, |
| "grad_norm": 6.09375, |
| "learning_rate": 1.2446666666666667e-05, |
| "loss": 0.3719, |
| "mean_token_accuracy": 0.8829085052013397, |
| "num_tokens": 73326681.0, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_coding_entropy": 1.0151909618377686, |
| "eval_coding_loss": 3.0377590656280518, |
| "eval_coding_mean_token_accuracy": 0.49079420840740207, |
| "eval_coding_num_tokens": 73326681.0, |
| "eval_coding_runtime": 79.5834, |
| "eval_coding_samples_per_second": 6.283, |
| "eval_coding_steps_per_second": 3.141, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_physics_entropy": 0.4613168597817421, |
| "eval_physics_loss": 1.055752158164978, |
| "eval_physics_mean_token_accuracy": 0.7722527477741241, |
| "eval_physics_num_tokens": 73326681.0, |
| "eval_physics_runtime": 52.4454, |
| "eval_physics_samples_per_second": 9.534, |
| "eval_physics_steps_per_second": 4.767, |
| "step": 4400 |
| }, |
| { |
| "entropy": 0.3821963082998991, |
| "epoch": 3.528, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.2424444444444445e-05, |
| "loss": 0.3619, |
| "mean_token_accuracy": 0.8844816010445357, |
| "num_tokens": 73494034.0, |
| "step": 4410 |
| }, |
| { |
| "entropy": 0.38084353255108, |
| "epoch": 3.536, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.2402222222222222e-05, |
| "loss": 0.3493, |
| "mean_token_accuracy": 0.8873479556292295, |
| "num_tokens": 73658724.0, |
| "step": 4420 |
| }, |
| { |
| "entropy": 0.364745448064059, |
| "epoch": 3.544, |
| "grad_norm": 5.25, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 0.3409, |
| "mean_token_accuracy": 0.8900253400206566, |
| "num_tokens": 73824282.0, |
| "step": 4430 |
| }, |
| { |
| "entropy": 0.38108069701120256, |
| "epoch": 3.552, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.235777777777778e-05, |
| "loss": 0.3576, |
| "mean_token_accuracy": 0.8847320571541786, |
| "num_tokens": 73990446.0, |
| "step": 4440 |
| }, |
| { |
| "entropy": 0.37883115112781524, |
| "epoch": 3.56, |
| "grad_norm": 5.375, |
| "learning_rate": 1.2335555555555557e-05, |
| "loss": 0.351, |
| "mean_token_accuracy": 0.8880171198397875, |
| "num_tokens": 74158749.0, |
| "step": 4450 |
| }, |
| { |
| "entropy": 0.37149322861805556, |
| "epoch": 3.568, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.2313333333333335e-05, |
| "loss": 0.3399, |
| "mean_token_accuracy": 0.8887308649718761, |
| "num_tokens": 74332932.0, |
| "step": 4460 |
| }, |
| { |
| "entropy": 0.38722019726410506, |
| "epoch": 3.576, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.2291111111111112e-05, |
| "loss": 0.3629, |
| "mean_token_accuracy": 0.8837050903588534, |
| "num_tokens": 74495632.0, |
| "step": 4470 |
| }, |
| { |
| "entropy": 0.4009826176799834, |
| "epoch": 3.584, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.226888888888889e-05, |
| "loss": 0.3705, |
| "mean_token_accuracy": 0.8816609259694814, |
| "num_tokens": 74660938.0, |
| "step": 4480 |
| }, |
| { |
| "entropy": 0.3549527461640537, |
| "epoch": 3.592, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.2246666666666667e-05, |
| "loss": 0.3285, |
| "mean_token_accuracy": 0.8931238904595376, |
| "num_tokens": 74841302.0, |
| "step": 4490 |
| }, |
| { |
| "entropy": 0.39866647506132724, |
| "epoch": 3.6, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.2224444444444445e-05, |
| "loss": 0.3702, |
| "mean_token_accuracy": 0.880438070371747, |
| "num_tokens": 75004640.0, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.6, |
| "eval_coding_entropy": 1.0210343375205995, |
| "eval_coding_loss": 3.013296604156494, |
| "eval_coding_mean_token_accuracy": 0.49208451342582704, |
| "eval_coding_num_tokens": 75004640.0, |
| "eval_coding_runtime": 79.5242, |
| "eval_coding_samples_per_second": 6.287, |
| "eval_coding_steps_per_second": 3.144, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.6, |
| "eval_physics_entropy": 0.4691944249272347, |
| "eval_physics_loss": 1.05069899559021, |
| "eval_physics_mean_token_accuracy": 0.7724122807979584, |
| "eval_physics_num_tokens": 75004640.0, |
| "eval_physics_runtime": 51.9377, |
| "eval_physics_samples_per_second": 9.627, |
| "eval_physics_steps_per_second": 4.813, |
| "step": 4500 |
| }, |
| { |
| "entropy": 0.3892392609268427, |
| "epoch": 3.608, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.2202222222222224e-05, |
| "loss": 0.3622, |
| "mean_token_accuracy": 0.8845055736601353, |
| "num_tokens": 75172766.0, |
| "step": 4510 |
| }, |
| { |
| "entropy": 0.36131666516885164, |
| "epoch": 3.616, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.218e-05, |
| "loss": 0.3417, |
| "mean_token_accuracy": 0.8905558951199055, |
| "num_tokens": 75344044.0, |
| "step": 4520 |
| }, |
| { |
| "entropy": 0.3780773578211665, |
| "epoch": 3.624, |
| "grad_norm": 5.875, |
| "learning_rate": 1.215777777777778e-05, |
| "loss": 0.3561, |
| "mean_token_accuracy": 0.8866020537912845, |
| "num_tokens": 75513247.0, |
| "step": 4530 |
| }, |
| { |
| "entropy": 0.3854208854958415, |
| "epoch": 3.632, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.2135555555555556e-05, |
| "loss": 0.354, |
| "mean_token_accuracy": 0.8842881310731172, |
| "num_tokens": 75676495.0, |
| "step": 4540 |
| }, |
| { |
| "entropy": 0.3840908773243427, |
| "epoch": 3.64, |
| "grad_norm": 5.84375, |
| "learning_rate": 1.2113333333333335e-05, |
| "loss": 0.3553, |
| "mean_token_accuracy": 0.886628508195281, |
| "num_tokens": 75838858.0, |
| "step": 4550 |
| }, |
| { |
| "entropy": 0.3843832564540207, |
| "epoch": 3.648, |
| "grad_norm": 5.0, |
| "learning_rate": 1.209111111111111e-05, |
| "loss": 0.365, |
| "mean_token_accuracy": 0.8850967317819596, |
| "num_tokens": 76005637.0, |
| "step": 4560 |
| }, |
| { |
| "entropy": 0.37927645137533544, |
| "epoch": 3.656, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.206888888888889e-05, |
| "loss": 0.3516, |
| "mean_token_accuracy": 0.8872136309742927, |
| "num_tokens": 76171337.0, |
| "step": 4570 |
| }, |
| { |
| "entropy": 0.379787333868444, |
| "epoch": 3.664, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.204666666666667e-05, |
| "loss": 0.3458, |
| "mean_token_accuracy": 0.8858440466225147, |
| "num_tokens": 76344141.0, |
| "step": 4580 |
| }, |
| { |
| "entropy": 0.3895066253840923, |
| "epoch": 3.672, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.2024444444444445e-05, |
| "loss": 0.3596, |
| "mean_token_accuracy": 0.8838235948234796, |
| "num_tokens": 76504454.0, |
| "step": 4590 |
| }, |
| { |
| "entropy": 0.3706223049201071, |
| "epoch": 3.68, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.2002222222222225e-05, |
| "loss": 0.3427, |
| "mean_token_accuracy": 0.8891801279038191, |
| "num_tokens": 76666670.0, |
| "step": 4600 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_coding_entropy": 1.0267571895122527, |
| "eval_coding_loss": 3.0143237113952637, |
| "eval_coding_mean_token_accuracy": 0.49128677988052366, |
| "eval_coding_num_tokens": 76666670.0, |
| "eval_coding_runtime": 79.1622, |
| "eval_coding_samples_per_second": 6.316, |
| "eval_coding_steps_per_second": 3.158, |
| "step": 4600 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_physics_entropy": 0.4683635458946228, |
| "eval_physics_loss": 1.0507175922393799, |
| "eval_physics_mean_token_accuracy": 0.7722031376361846, |
| "eval_physics_num_tokens": 76666670.0, |
| "eval_physics_runtime": 52.0967, |
| "eval_physics_samples_per_second": 9.598, |
| "eval_physics_steps_per_second": 4.799, |
| "step": 4600 |
| }, |
| { |
| "entropy": 0.3896908285096288, |
| "epoch": 3.6879999999999997, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.198e-05, |
| "loss": 0.3619, |
| "mean_token_accuracy": 0.882770498469472, |
| "num_tokens": 76834877.0, |
| "step": 4610 |
| }, |
| { |
| "entropy": 0.3899038337171078, |
| "epoch": 3.6959999999999997, |
| "grad_norm": 5.9375, |
| "learning_rate": 1.195777777777778e-05, |
| "loss": 0.3616, |
| "mean_token_accuracy": 0.8839272748678922, |
| "num_tokens": 76997576.0, |
| "step": 4620 |
| }, |
| { |
| "entropy": 0.36462077740579846, |
| "epoch": 3.7039999999999997, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.1935555555555556e-05, |
| "loss": 0.3391, |
| "mean_token_accuracy": 0.8902773730456829, |
| "num_tokens": 77169776.0, |
| "step": 4630 |
| }, |
| { |
| "entropy": 0.38682004110887647, |
| "epoch": 3.7119999999999997, |
| "grad_norm": 5.59375, |
| "learning_rate": 1.1913333333333335e-05, |
| "loss": 0.3538, |
| "mean_token_accuracy": 0.884754529222846, |
| "num_tokens": 77333230.0, |
| "step": 4640 |
| }, |
| { |
| "entropy": 0.3708122307434678, |
| "epoch": 3.7199999999999998, |
| "grad_norm": 6.125, |
| "learning_rate": 1.1891111111111111e-05, |
| "loss": 0.3451, |
| "mean_token_accuracy": 0.889562564715743, |
| "num_tokens": 77497116.0, |
| "step": 4650 |
| }, |
| { |
| "entropy": 0.38634428698569534, |
| "epoch": 3.7279999999999998, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.186888888888889e-05, |
| "loss": 0.3684, |
| "mean_token_accuracy": 0.8828420627862215, |
| "num_tokens": 77662989.0, |
| "step": 4660 |
| }, |
| { |
| "entropy": 0.3869167296215892, |
| "epoch": 3.7359999999999998, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.1846666666666668e-05, |
| "loss": 0.3597, |
| "mean_token_accuracy": 0.8858137641102075, |
| "num_tokens": 77828135.0, |
| "step": 4670 |
| }, |
| { |
| "entropy": 0.37564525082707406, |
| "epoch": 3.7439999999999998, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.1824444444444445e-05, |
| "loss": 0.3575, |
| "mean_token_accuracy": 0.8865491844713688, |
| "num_tokens": 77990559.0, |
| "step": 4680 |
| }, |
| { |
| "entropy": 0.38707160549238323, |
| "epoch": 3.752, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.1802222222222223e-05, |
| "loss": 0.3561, |
| "mean_token_accuracy": 0.885225672647357, |
| "num_tokens": 78156913.0, |
| "step": 4690 |
| }, |
| { |
| "entropy": 0.37813973939046264, |
| "epoch": 3.76, |
| "grad_norm": 5.5, |
| "learning_rate": 1.178e-05, |
| "loss": 0.3537, |
| "mean_token_accuracy": 0.8867236789315939, |
| "num_tokens": 78328058.0, |
| "step": 4700 |
| }, |
| { |
| "epoch": 3.76, |
| "eval_coding_entropy": 1.0269946234226226, |
| "eval_coding_loss": 3.0190868377685547, |
| "eval_coding_mean_token_accuracy": 0.49062985932826997, |
| "eval_coding_num_tokens": 78328058.0, |
| "eval_coding_runtime": 79.1694, |
| "eval_coding_samples_per_second": 6.316, |
| "eval_coding_steps_per_second": 3.158, |
| "step": 4700 |
| }, |
| { |
| "epoch": 3.76, |
| "eval_physics_entropy": 0.46678994792699813, |
| "eval_physics_loss": 1.0511798858642578, |
| "eval_physics_mean_token_accuracy": 0.7726296453475953, |
| "eval_physics_num_tokens": 78328058.0, |
| "eval_physics_runtime": 52.0771, |
| "eval_physics_samples_per_second": 9.601, |
| "eval_physics_steps_per_second": 4.801, |
| "step": 4700 |
| }, |
| { |
| "entropy": 0.385709291882813, |
| "epoch": 3.768, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.1757777777777778e-05, |
| "loss": 0.3554, |
| "mean_token_accuracy": 0.8869376711547374, |
| "num_tokens": 78494373.0, |
| "step": 4710 |
| }, |
| { |
| "entropy": 0.3720444541424513, |
| "epoch": 3.776, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.1735555555555556e-05, |
| "loss": 0.3399, |
| "mean_token_accuracy": 0.8888941917568445, |
| "num_tokens": 78656796.0, |
| "step": 4720 |
| }, |
| { |
| "entropy": 0.3962340967729688, |
| "epoch": 3.784, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.1713333333333334e-05, |
| "loss": 0.3684, |
| "mean_token_accuracy": 0.8822142042219638, |
| "num_tokens": 78826840.0, |
| "step": 4730 |
| }, |
| { |
| "entropy": 0.3647663754411042, |
| "epoch": 3.792, |
| "grad_norm": 5.25, |
| "learning_rate": 1.1691111111111113e-05, |
| "loss": 0.3392, |
| "mean_token_accuracy": 0.889918739721179, |
| "num_tokens": 79000001.0, |
| "step": 4740 |
| }, |
| { |
| "entropy": 0.37643586173653604, |
| "epoch": 3.8, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.1668888888888889e-05, |
| "loss": 0.3552, |
| "mean_token_accuracy": 0.8864477183669806, |
| "num_tokens": 79164463.0, |
| "step": 4750 |
| }, |
| { |
| "entropy": 0.3730422070249915, |
| "epoch": 3.808, |
| "grad_norm": 6.0, |
| "learning_rate": 1.1646666666666668e-05, |
| "loss": 0.3428, |
| "mean_token_accuracy": 0.8881399553269148, |
| "num_tokens": 79332103.0, |
| "step": 4760 |
| }, |
| { |
| "entropy": 0.38533862996846435, |
| "epoch": 3.816, |
| "grad_norm": 6.59375, |
| "learning_rate": 1.1624444444444446e-05, |
| "loss": 0.3641, |
| "mean_token_accuracy": 0.8849582724273205, |
| "num_tokens": 79493414.0, |
| "step": 4770 |
| }, |
| { |
| "entropy": 0.3922125525772572, |
| "epoch": 3.824, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.1602222222222223e-05, |
| "loss": 0.3646, |
| "mean_token_accuracy": 0.8824360508471727, |
| "num_tokens": 79658616.0, |
| "step": 4780 |
| }, |
| { |
| "entropy": 0.38893508845940233, |
| "epoch": 3.832, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 0.3595, |
| "mean_token_accuracy": 0.8834522284567357, |
| "num_tokens": 79821970.0, |
| "step": 4790 |
| }, |
| { |
| "entropy": 0.37476310413330793, |
| "epoch": 3.84, |
| "grad_norm": 5.75, |
| "learning_rate": 1.1557777777777779e-05, |
| "loss": 0.353, |
| "mean_token_accuracy": 0.8865226674824953, |
| "num_tokens": 79994326.0, |
| "step": 4800 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_coding_entropy": 1.031722464799881, |
| "eval_coding_loss": 3.0076234340667725, |
| "eval_coding_mean_token_accuracy": 0.491949645280838, |
| "eval_coding_num_tokens": 79994326.0, |
| "eval_coding_runtime": 79.1836, |
| "eval_coding_samples_per_second": 6.314, |
| "eval_coding_steps_per_second": 3.157, |
| "step": 4800 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_physics_entropy": 0.4703272629380226, |
| "eval_physics_loss": 1.0469328165054321, |
| "eval_physics_mean_token_accuracy": 0.7728608014583588, |
| "eval_physics_num_tokens": 79994326.0, |
| "eval_physics_runtime": 52.1543, |
| "eval_physics_samples_per_second": 9.587, |
| "eval_physics_steps_per_second": 4.793, |
| "step": 4800 |
| }, |
| { |
| "entropy": 0.3820637227036059, |
| "epoch": 3.848, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.1535555555555556e-05, |
| "loss": 0.3493, |
| "mean_token_accuracy": 0.8855405732989311, |
| "num_tokens": 80167071.0, |
| "step": 4810 |
| }, |
| { |
| "entropy": 0.37587422747164967, |
| "epoch": 3.856, |
| "grad_norm": 5.25, |
| "learning_rate": 1.1513333333333334e-05, |
| "loss": 0.3484, |
| "mean_token_accuracy": 0.887658753991127, |
| "num_tokens": 80332596.0, |
| "step": 4820 |
| }, |
| { |
| "entropy": 0.3753030957654119, |
| "epoch": 3.864, |
| "grad_norm": 5.46875, |
| "learning_rate": 1.1491111111111113e-05, |
| "loss": 0.3441, |
| "mean_token_accuracy": 0.8888305995613337, |
| "num_tokens": 80499477.0, |
| "step": 4830 |
| }, |
| { |
| "entropy": 0.3800191371701658, |
| "epoch": 3.872, |
| "grad_norm": 5.90625, |
| "learning_rate": 1.1468888888888889e-05, |
| "loss": 0.3541, |
| "mean_token_accuracy": 0.8869428627192975, |
| "num_tokens": 80662674.0, |
| "step": 4840 |
| }, |
| { |
| "entropy": 0.3862288841046393, |
| "epoch": 3.88, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.1446666666666668e-05, |
| "loss": 0.3673, |
| "mean_token_accuracy": 0.8838630292564631, |
| "num_tokens": 80828611.0, |
| "step": 4850 |
| }, |
| { |
| "entropy": 0.3846357778646052, |
| "epoch": 3.888, |
| "grad_norm": 6.3125, |
| "learning_rate": 1.1424444444444444e-05, |
| "loss": 0.3575, |
| "mean_token_accuracy": 0.8853116244077682, |
| "num_tokens": 80995198.0, |
| "step": 4860 |
| }, |
| { |
| "entropy": 0.3945392989553511, |
| "epoch": 3.896, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.1402222222222224e-05, |
| "loss": 0.3673, |
| "mean_token_accuracy": 0.8817976925522089, |
| "num_tokens": 81164554.0, |
| "step": 4870 |
| }, |
| { |
| "entropy": 0.4017768774181604, |
| "epoch": 3.904, |
| "grad_norm": 6.21875, |
| "learning_rate": 1.138e-05, |
| "loss": 0.3705, |
| "mean_token_accuracy": 0.8811160765588284, |
| "num_tokens": 81331457.0, |
| "step": 4880 |
| }, |
| { |
| "entropy": 0.38470241650938986, |
| "epoch": 3.912, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.1357777777777779e-05, |
| "loss": 0.3648, |
| "mean_token_accuracy": 0.8828286323696375, |
| "num_tokens": 81498040.0, |
| "step": 4890 |
| }, |
| { |
| "entropy": 0.3750403059646487, |
| "epoch": 3.92, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.1335555555555558e-05, |
| "loss": 0.3523, |
| "mean_token_accuracy": 0.8868987247347832, |
| "num_tokens": 81668354.0, |
| "step": 4900 |
| }, |
| { |
| "epoch": 3.92, |
| "eval_coding_entropy": 1.037083692073822, |
| "eval_coding_loss": 3.0055859088897705, |
| "eval_coding_mean_token_accuracy": 0.4919183658361435, |
| "eval_coding_num_tokens": 81668354.0, |
| "eval_coding_runtime": 79.1767, |
| "eval_coding_samples_per_second": 6.315, |
| "eval_coding_steps_per_second": 3.157, |
| "step": 4900 |
| }, |
| { |
| "epoch": 3.92, |
| "eval_physics_entropy": 0.46753990530967715, |
| "eval_physics_loss": 1.050595998764038, |
| "eval_physics_mean_token_accuracy": 0.7724316239356994, |
| "eval_physics_num_tokens": 81668354.0, |
| "eval_physics_runtime": 52.3012, |
| "eval_physics_samples_per_second": 9.56, |
| "eval_physics_steps_per_second": 4.78, |
| "step": 4900 |
| }, |
| { |
| "entropy": 0.3862785813398659, |
| "epoch": 3.928, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.1313333333333334e-05, |
| "loss": 0.3585, |
| "mean_token_accuracy": 0.8859840672463178, |
| "num_tokens": 81828903.0, |
| "step": 4910 |
| }, |
| { |
| "entropy": 0.3703251021914184, |
| "epoch": 3.936, |
| "grad_norm": 5.78125, |
| "learning_rate": 1.1291111111111113e-05, |
| "loss": 0.3482, |
| "mean_token_accuracy": 0.8889971677213907, |
| "num_tokens": 81994993.0, |
| "step": 4920 |
| }, |
| { |
| "entropy": 0.38606200283393266, |
| "epoch": 3.944, |
| "grad_norm": 5.8125, |
| "learning_rate": 1.126888888888889e-05, |
| "loss": 0.3518, |
| "mean_token_accuracy": 0.8859570454806089, |
| "num_tokens": 82159265.0, |
| "step": 4930 |
| }, |
| { |
| "entropy": 0.371717198099941, |
| "epoch": 3.952, |
| "grad_norm": 6.90625, |
| "learning_rate": 1.1246666666666669e-05, |
| "loss": 0.3485, |
| "mean_token_accuracy": 0.8871124267578125, |
| "num_tokens": 82327299.0, |
| "step": 4940 |
| }, |
| { |
| "entropy": 0.36819626530632377, |
| "epoch": 3.96, |
| "grad_norm": 4.75, |
| "learning_rate": 1.1224444444444444e-05, |
| "loss": 0.3387, |
| "mean_token_accuracy": 0.8902491096407175, |
| "num_tokens": 82493501.0, |
| "step": 4950 |
| }, |
| { |
| "entropy": 0.37841445291414855, |
| "epoch": 3.968, |
| "grad_norm": 5.71875, |
| "learning_rate": 1.1202222222222224e-05, |
| "loss": 0.3534, |
| "mean_token_accuracy": 0.8856508497148752, |
| "num_tokens": 82653218.0, |
| "step": 4960 |
| }, |
| { |
| "entropy": 0.3760814812034369, |
| "epoch": 3.976, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 0.3518, |
| "mean_token_accuracy": 0.8869401030242443, |
| "num_tokens": 82825734.0, |
| "step": 4970 |
| }, |
| { |
| "entropy": 0.3701860437169671, |
| "epoch": 3.984, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.1157777777777779e-05, |
| "loss": 0.3408, |
| "mean_token_accuracy": 0.888851211220026, |
| "num_tokens": 82997429.0, |
| "step": 4980 |
| }, |
| { |
| "entropy": 0.3758525485172868, |
| "epoch": 3.992, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.1135555555555557e-05, |
| "loss": 0.3492, |
| "mean_token_accuracy": 0.8871539425104856, |
| "num_tokens": 83163004.0, |
| "step": 4990 |
| }, |
| { |
| "entropy": 0.36695010410621764, |
| "epoch": 4.0, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.1113333333333334e-05, |
| "loss": 0.3487, |
| "mean_token_accuracy": 0.8889952551573514, |
| "num_tokens": 83328592.0, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_coding_entropy": 1.0181054289340974, |
| "eval_coding_loss": 3.0115039348602295, |
| "eval_coding_mean_token_accuracy": 0.4926280071735382, |
| "eval_coding_num_tokens": 83328592.0, |
| "eval_coding_runtime": 79.2723, |
| "eval_coding_samples_per_second": 6.307, |
| "eval_coding_steps_per_second": 3.154, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_physics_entropy": 0.46515926730632784, |
| "eval_physics_loss": 1.0475009679794312, |
| "eval_physics_mean_token_accuracy": 0.7731966464519501, |
| "eval_physics_num_tokens": 83328592.0, |
| "eval_physics_runtime": 52.0572, |
| "eval_physics_samples_per_second": 9.605, |
| "eval_physics_steps_per_second": 4.802, |
| "step": 5000 |
| }, |
| { |
| "entropy": 0.28828687416389587, |
| "epoch": 4.008, |
| "grad_norm": 6.625, |
| "learning_rate": 1.1091111111111112e-05, |
| "loss": 0.2104, |
| "mean_token_accuracy": 0.935462799668312, |
| "num_tokens": 83490945.0, |
| "step": 5010 |
| }, |
| { |
| "entropy": 0.24608047427609564, |
| "epoch": 4.016, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.106888888888889e-05, |
| "loss": 0.1993, |
| "mean_token_accuracy": 0.9376339934766292, |
| "num_tokens": 83655133.0, |
| "step": 5020 |
| }, |
| { |
| "entropy": 0.2571288766339421, |
| "epoch": 4.024, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.1046666666666667e-05, |
| "loss": 0.2, |
| "mean_token_accuracy": 0.9383496355265379, |
| "num_tokens": 83829903.0, |
| "step": 5030 |
| }, |
| { |
| "entropy": 0.25705197467468677, |
| "epoch": 4.032, |
| "grad_norm": 4.875, |
| "learning_rate": 1.1024444444444445e-05, |
| "loss": 0.1965, |
| "mean_token_accuracy": 0.9380440566688776, |
| "num_tokens": 83995704.0, |
| "step": 5040 |
| }, |
| { |
| "entropy": 0.2525853905826807, |
| "epoch": 4.04, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.1002222222222222e-05, |
| "loss": 0.1952, |
| "mean_token_accuracy": 0.9391312584280967, |
| "num_tokens": 84155510.0, |
| "step": 5050 |
| }, |
| { |
| "entropy": 0.25542440814897416, |
| "epoch": 4.048, |
| "grad_norm": 4.5, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 0.2046, |
| "mean_token_accuracy": 0.9370964780449867, |
| "num_tokens": 84323657.0, |
| "step": 5060 |
| }, |
| { |
| "entropy": 0.24826187775470315, |
| "epoch": 4.056, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.0957777777777778e-05, |
| "loss": 0.1925, |
| "mean_token_accuracy": 0.9384117465466261, |
| "num_tokens": 84498853.0, |
| "step": 5070 |
| }, |
| { |
| "entropy": 0.250696533638984, |
| "epoch": 4.064, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.0935555555555557e-05, |
| "loss": 0.1956, |
| "mean_token_accuracy": 0.9400551084429025, |
| "num_tokens": 84662055.0, |
| "step": 5080 |
| }, |
| { |
| "entropy": 0.25483012036420405, |
| "epoch": 4.072, |
| "grad_norm": 4.875, |
| "learning_rate": 1.0913333333333333e-05, |
| "loss": 0.1971, |
| "mean_token_accuracy": 0.9384329654276371, |
| "num_tokens": 84826253.0, |
| "step": 5090 |
| }, |
| { |
| "entropy": 0.24928630511276423, |
| "epoch": 4.08, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.0891111111111112e-05, |
| "loss": 0.1932, |
| "mean_token_accuracy": 0.9397989839315415, |
| "num_tokens": 84990884.0, |
| "step": 5100 |
| }, |
| { |
| "epoch": 4.08, |
| "eval_coding_entropy": 0.9160269091129303, |
| "eval_coding_loss": 3.3617751598358154, |
| "eval_coding_mean_token_accuracy": 0.48005122113227844, |
| "eval_coding_num_tokens": 84990884.0, |
| "eval_coding_runtime": 79.2713, |
| "eval_coding_samples_per_second": 6.307, |
| "eval_coding_steps_per_second": 3.154, |
| "step": 5100 |
| }, |
| { |
| "epoch": 4.08, |
| "eval_physics_entropy": 0.3862661864757538, |
| "eval_physics_loss": 1.2060325145721436, |
| "eval_physics_mean_token_accuracy": 0.7663328912258148, |
| "eval_physics_num_tokens": 84990884.0, |
| "eval_physics_runtime": 52.1405, |
| "eval_physics_samples_per_second": 9.589, |
| "eval_physics_steps_per_second": 4.795, |
| "step": 5100 |
| }, |
| { |
| "entropy": 0.25156633332371714, |
| "epoch": 4.088, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.0868888888888888e-05, |
| "loss": 0.1983, |
| "mean_token_accuracy": 0.9387177176773548, |
| "num_tokens": 85165005.0, |
| "step": 5110 |
| }, |
| { |
| "entropy": 0.2385215105023235, |
| "epoch": 4.096, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.0846666666666667e-05, |
| "loss": 0.1899, |
| "mean_token_accuracy": 0.9417340341955424, |
| "num_tokens": 85334346.0, |
| "step": 5120 |
| }, |
| { |
| "entropy": 0.2598756125662476, |
| "epoch": 4.104, |
| "grad_norm": 5.0, |
| "learning_rate": 1.0824444444444447e-05, |
| "loss": 0.197, |
| "mean_token_accuracy": 0.9379688289016486, |
| "num_tokens": 85492249.0, |
| "step": 5130 |
| }, |
| { |
| "entropy": 0.24627253361977636, |
| "epoch": 4.112, |
| "grad_norm": 5.46875, |
| "learning_rate": 1.0802222222222223e-05, |
| "loss": 0.1941, |
| "mean_token_accuracy": 0.9398119896650314, |
| "num_tokens": 85654431.0, |
| "step": 5140 |
| }, |
| { |
| "entropy": 0.24774275501258672, |
| "epoch": 4.12, |
| "grad_norm": 4.5, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 0.1932, |
| "mean_token_accuracy": 0.9394119538366794, |
| "num_tokens": 85819390.0, |
| "step": 5150 |
| }, |
| { |
| "entropy": 0.2618040062021464, |
| "epoch": 4.128, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.0757777777777778e-05, |
| "loss": 0.2056, |
| "mean_token_accuracy": 0.9360566876828671, |
| "num_tokens": 85984868.0, |
| "step": 5160 |
| }, |
| { |
| "entropy": 0.2598070489708334, |
| "epoch": 4.136, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.0735555555555557e-05, |
| "loss": 0.2022, |
| "mean_token_accuracy": 0.9361321475356817, |
| "num_tokens": 86144185.0, |
| "step": 5170 |
| }, |
| { |
| "entropy": 0.24930871804244817, |
| "epoch": 4.144, |
| "grad_norm": 4.875, |
| "learning_rate": 1.0713333333333333e-05, |
| "loss": 0.1962, |
| "mean_token_accuracy": 0.9392761684954166, |
| "num_tokens": 86313610.0, |
| "step": 5180 |
| }, |
| { |
| "entropy": 0.25302718207240105, |
| "epoch": 4.152, |
| "grad_norm": 5.0, |
| "learning_rate": 1.0691111111111112e-05, |
| "loss": 0.1951, |
| "mean_token_accuracy": 0.9386980183422565, |
| "num_tokens": 86474213.0, |
| "step": 5190 |
| }, |
| { |
| "entropy": 0.24344518510624766, |
| "epoch": 4.16, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.0668888888888892e-05, |
| "loss": 0.1965, |
| "mean_token_accuracy": 0.9387242361903191, |
| "num_tokens": 86648308.0, |
| "step": 5200 |
| }, |
| { |
| "epoch": 4.16, |
| "eval_coding_entropy": 0.9100202202796936, |
| "eval_coding_loss": 3.384716272354126, |
| "eval_coding_mean_token_accuracy": 0.4790853888988495, |
| "eval_coding_num_tokens": 86648308.0, |
| "eval_coding_runtime": 79.1343, |
| "eval_coding_samples_per_second": 6.318, |
| "eval_coding_steps_per_second": 3.159, |
| "step": 5200 |
| }, |
| { |
| "epoch": 4.16, |
| "eval_physics_entropy": 0.38321507781744, |
| "eval_physics_loss": 1.2163034677505493, |
| "eval_physics_mean_token_accuracy": 0.7663352589607239, |
| "eval_physics_num_tokens": 86648308.0, |
| "eval_physics_runtime": 52.0727, |
| "eval_physics_samples_per_second": 9.602, |
| "eval_physics_steps_per_second": 4.801, |
| "step": 5200 |
| }, |
| { |
| "entropy": 0.2545318819116801, |
| "epoch": 4.168, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.0646666666666668e-05, |
| "loss": 0.1996, |
| "mean_token_accuracy": 0.9381030213087798, |
| "num_tokens": 86810965.0, |
| "step": 5210 |
| }, |
| { |
| "entropy": 0.2521157009061426, |
| "epoch": 4.176, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.0624444444444447e-05, |
| "loss": 0.1976, |
| "mean_token_accuracy": 0.9371171236038208, |
| "num_tokens": 86979720.0, |
| "step": 5220 |
| }, |
| { |
| "entropy": 0.24082404510118066, |
| "epoch": 4.184, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.0602222222222223e-05, |
| "loss": 0.1932, |
| "mean_token_accuracy": 0.9395224828273058, |
| "num_tokens": 87151536.0, |
| "step": 5230 |
| }, |
| { |
| "entropy": 0.24610360157676042, |
| "epoch": 4.192, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 0.195, |
| "mean_token_accuracy": 0.937963343039155, |
| "num_tokens": 87322940.0, |
| "step": 5240 |
| }, |
| { |
| "entropy": 0.24998870017006994, |
| "epoch": 4.2, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.0557777777777778e-05, |
| "loss": 0.199, |
| "mean_token_accuracy": 0.9386361390352249, |
| "num_tokens": 87491615.0, |
| "step": 5250 |
| }, |
| { |
| "entropy": 0.2593475771136582, |
| "epoch": 4.208, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.0535555555555557e-05, |
| "loss": 0.2019, |
| "mean_token_accuracy": 0.9366621308028698, |
| "num_tokens": 87654476.0, |
| "step": 5260 |
| }, |
| { |
| "entropy": 0.2611536079552025, |
| "epoch": 4.216, |
| "grad_norm": 4.875, |
| "learning_rate": 1.0513333333333333e-05, |
| "loss": 0.2111, |
| "mean_token_accuracy": 0.9348096769303084, |
| "num_tokens": 87821280.0, |
| "step": 5270 |
| }, |
| { |
| "entropy": 0.27446596780791876, |
| "epoch": 4.224, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.0491111111111112e-05, |
| "loss": 0.218, |
| "mean_token_accuracy": 0.9308515418320894, |
| "num_tokens": 87986180.0, |
| "step": 5280 |
| }, |
| { |
| "entropy": 0.2559265170712024, |
| "epoch": 4.232, |
| "grad_norm": 5.84375, |
| "learning_rate": 1.046888888888889e-05, |
| "loss": 0.1996, |
| "mean_token_accuracy": 0.9377362191677093, |
| "num_tokens": 88147013.0, |
| "step": 5290 |
| }, |
| { |
| "entropy": 0.2470927257090807, |
| "epoch": 4.24, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.0446666666666668e-05, |
| "loss": 0.2002, |
| "mean_token_accuracy": 0.938216557353735, |
| "num_tokens": 88322431.0, |
| "step": 5300 |
| }, |
| { |
| "epoch": 4.24, |
| "eval_coding_entropy": 0.9060706837177277, |
| "eval_coding_loss": 3.38850998878479, |
| "eval_coding_mean_token_accuracy": 0.47926331424713137, |
| "eval_coding_num_tokens": 88322431.0, |
| "eval_coding_runtime": 79.1582, |
| "eval_coding_samples_per_second": 6.316, |
| "eval_coding_steps_per_second": 3.158, |
| "step": 5300 |
| }, |
| { |
| "epoch": 4.24, |
| "eval_physics_entropy": 0.3824515398740768, |
| "eval_physics_loss": 1.2171977758407593, |
| "eval_physics_mean_token_accuracy": 0.7662743992805481, |
| "eval_physics_num_tokens": 88322431.0, |
| "eval_physics_runtime": 52.1497, |
| "eval_physics_samples_per_second": 9.588, |
| "eval_physics_steps_per_second": 4.794, |
| "step": 5300 |
| }, |
| { |
| "entropy": 0.25588256628252565, |
| "epoch": 4.248, |
| "grad_norm": 4.875, |
| "learning_rate": 1.0424444444444445e-05, |
| "loss": 0.2012, |
| "mean_token_accuracy": 0.9374486811459064, |
| "num_tokens": 88488277.0, |
| "step": 5310 |
| }, |
| { |
| "entropy": 0.24605262177065015, |
| "epoch": 4.256, |
| "grad_norm": 5.0, |
| "learning_rate": 1.0402222222222223e-05, |
| "loss": 0.1939, |
| "mean_token_accuracy": 0.9394217438995838, |
| "num_tokens": 88650585.0, |
| "step": 5320 |
| }, |
| { |
| "entropy": 0.25547544257715343, |
| "epoch": 4.264, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.038e-05, |
| "loss": 0.2048, |
| "mean_token_accuracy": 0.9368464518338442, |
| "num_tokens": 88822042.0, |
| "step": 5330 |
| }, |
| { |
| "entropy": 0.25734468596056104, |
| "epoch": 4.272, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.0357777777777778e-05, |
| "loss": 0.2076, |
| "mean_token_accuracy": 0.9357014950364828, |
| "num_tokens": 88983258.0, |
| "step": 5340 |
| }, |
| { |
| "entropy": 0.24844884043559431, |
| "epoch": 4.28, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.0335555555555556e-05, |
| "loss": 0.1969, |
| "mean_token_accuracy": 0.9376862313598394, |
| "num_tokens": 89151148.0, |
| "step": 5350 |
| }, |
| { |
| "entropy": 0.2563304465264082, |
| "epoch": 4.288, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.0313333333333335e-05, |
| "loss": 0.2017, |
| "mean_token_accuracy": 0.9364820554852485, |
| "num_tokens": 89306176.0, |
| "step": 5360 |
| }, |
| { |
| "entropy": 0.2508170139044523, |
| "epoch": 4.296, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.0291111111111111e-05, |
| "loss": 0.1986, |
| "mean_token_accuracy": 0.9370678186416626, |
| "num_tokens": 89471001.0, |
| "step": 5370 |
| }, |
| { |
| "entropy": 0.2509464251808822, |
| "epoch": 4.304, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.026888888888889e-05, |
| "loss": 0.2024, |
| "mean_token_accuracy": 0.9376652576029301, |
| "num_tokens": 89635604.0, |
| "step": 5380 |
| }, |
| { |
| "entropy": 0.25435569281689824, |
| "epoch": 4.312, |
| "grad_norm": 5.0, |
| "learning_rate": 1.0246666666666666e-05, |
| "loss": 0.2029, |
| "mean_token_accuracy": 0.9358005274087191, |
| "num_tokens": 89797573.0, |
| "step": 5390 |
| }, |
| { |
| "entropy": 0.2524940128903836, |
| "epoch": 4.32, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.0224444444444446e-05, |
| "loss": 0.1995, |
| "mean_token_accuracy": 0.937572841718793, |
| "num_tokens": 89961895.0, |
| "step": 5400 |
| }, |
| { |
| "epoch": 4.32, |
| "eval_coding_entropy": 0.9091968116760254, |
| "eval_coding_loss": 3.3865809440612793, |
| "eval_coding_mean_token_accuracy": 0.47897004640102386, |
| "eval_coding_num_tokens": 89961895.0, |
| "eval_coding_runtime": 80.0942, |
| "eval_coding_samples_per_second": 6.243, |
| "eval_coding_steps_per_second": 3.121, |
| "step": 5400 |
| }, |
| { |
| "epoch": 4.32, |
| "eval_physics_entropy": 0.38306553304195406, |
| "eval_physics_loss": 1.21638023853302, |
| "eval_physics_mean_token_accuracy": 0.7656607213020324, |
| "eval_physics_num_tokens": 89961895.0, |
| "eval_physics_runtime": 53.4819, |
| "eval_physics_samples_per_second": 9.349, |
| "eval_physics_steps_per_second": 4.674, |
| "step": 5400 |
| }, |
| { |
| "entropy": 0.24834470530040562, |
| "epoch": 4.328, |
| "grad_norm": 5.625, |
| "learning_rate": 1.0202222222222221e-05, |
| "loss": 0.1966, |
| "mean_token_accuracy": 0.9372350744903087, |
| "num_tokens": 90125773.0, |
| "step": 5410 |
| }, |
| { |
| "entropy": 0.2483433459419757, |
| "epoch": 4.336, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.018e-05, |
| "loss": 0.1955, |
| "mean_token_accuracy": 0.9382834773510694, |
| "num_tokens": 90293218.0, |
| "step": 5420 |
| }, |
| { |
| "entropy": 0.2522097827401012, |
| "epoch": 4.344, |
| "grad_norm": 5.125, |
| "learning_rate": 1.0157777777777777e-05, |
| "loss": 0.2048, |
| "mean_token_accuracy": 0.935153954476118, |
| "num_tokens": 90458384.0, |
| "step": 5430 |
| }, |
| { |
| "entropy": 0.23947314182296395, |
| "epoch": 4.352, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.0135555555555556e-05, |
| "loss": 0.1887, |
| "mean_token_accuracy": 0.9403942737728357, |
| "num_tokens": 90631006.0, |
| "step": 5440 |
| }, |
| { |
| "entropy": 0.25989792472682893, |
| "epoch": 4.36, |
| "grad_norm": 5.375, |
| "learning_rate": 1.0113333333333335e-05, |
| "loss": 0.2101, |
| "mean_token_accuracy": 0.9348955221474171, |
| "num_tokens": 90788240.0, |
| "step": 5450 |
| }, |
| { |
| "entropy": 0.2579982799477875, |
| "epoch": 4.368, |
| "grad_norm": 5.625, |
| "learning_rate": 1.0091111111111111e-05, |
| "loss": 0.2062, |
| "mean_token_accuracy": 0.9346442516893149, |
| "num_tokens": 90951742.0, |
| "step": 5460 |
| }, |
| { |
| "entropy": 0.2625875208992511, |
| "epoch": 4.376, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.006888888888889e-05, |
| "loss": 0.2084, |
| "mean_token_accuracy": 0.9342471640557051, |
| "num_tokens": 91118386.0, |
| "step": 5470 |
| }, |
| { |
| "entropy": 0.2513840536121279, |
| "epoch": 4.384, |
| "grad_norm": 5.375, |
| "learning_rate": 1.0046666666666666e-05, |
| "loss": 0.204, |
| "mean_token_accuracy": 0.9365727048367262, |
| "num_tokens": 91290391.0, |
| "step": 5480 |
| }, |
| { |
| "entropy": 0.25519236200489104, |
| "epoch": 4.392, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.0024444444444446e-05, |
| "loss": 0.1996, |
| "mean_token_accuracy": 0.9369384657591582, |
| "num_tokens": 91456191.0, |
| "step": 5490 |
| }, |
| { |
| "entropy": 0.25372437653131785, |
| "epoch": 4.4, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.0002222222222222e-05, |
| "loss": 0.2025, |
| "mean_token_accuracy": 0.9364530324935914, |
| "num_tokens": 91623212.0, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.4, |
| "eval_coding_entropy": 0.902980350971222, |
| "eval_coding_loss": 3.3784267902374268, |
| "eval_coding_mean_token_accuracy": 0.48033658194541934, |
| "eval_coding_num_tokens": 91623212.0, |
| "eval_coding_runtime": 80.7394, |
| "eval_coding_samples_per_second": 6.193, |
| "eval_coding_steps_per_second": 3.096, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.4, |
| "eval_physics_entropy": 0.38216441035270693, |
| "eval_physics_loss": 1.215621829032898, |
| "eval_physics_mean_token_accuracy": 0.7667231605052948, |
| "eval_physics_num_tokens": 91623212.0, |
| "eval_physics_runtime": 53.9567, |
| "eval_physics_samples_per_second": 9.267, |
| "eval_physics_steps_per_second": 4.633, |
| "step": 5500 |
| }, |
| { |
| "entropy": 0.25119374208152295, |
| "epoch": 4.408, |
| "grad_norm": 4.96875, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 0.2038, |
| "mean_token_accuracy": 0.9372086472809315, |
| "num_tokens": 91795278.0, |
| "step": 5510 |
| }, |
| { |
| "entropy": 0.2537687565200031, |
| "epoch": 4.416, |
| "grad_norm": 5.03125, |
| "learning_rate": 9.957777777777779e-06, |
| "loss": 0.2001, |
| "mean_token_accuracy": 0.9366831291466952, |
| "num_tokens": 91963726.0, |
| "step": 5520 |
| }, |
| { |
| "entropy": 0.24195127468556166, |
| "epoch": 4.424, |
| "grad_norm": 5.21875, |
| "learning_rate": 9.935555555555556e-06, |
| "loss": 0.1957, |
| "mean_token_accuracy": 0.9390779342502356, |
| "num_tokens": 92131643.0, |
| "step": 5530 |
| }, |
| { |
| "entropy": 0.2512932606507093, |
| "epoch": 4.432, |
| "grad_norm": 5.09375, |
| "learning_rate": 9.913333333333334e-06, |
| "loss": 0.2051, |
| "mean_token_accuracy": 0.9367362342774868, |
| "num_tokens": 92295130.0, |
| "step": 5540 |
| }, |
| { |
| "entropy": 0.26577539909631015, |
| "epoch": 4.44, |
| "grad_norm": 5.28125, |
| "learning_rate": 9.891111111111113e-06, |
| "loss": 0.2138, |
| "mean_token_accuracy": 0.9331799853593111, |
| "num_tokens": 92457017.0, |
| "step": 5550 |
| }, |
| { |
| "entropy": 0.2568955378606915, |
| "epoch": 4.448, |
| "grad_norm": 5.28125, |
| "learning_rate": 9.86888888888889e-06, |
| "loss": 0.2043, |
| "mean_token_accuracy": 0.9350587010383606, |
| "num_tokens": 92619914.0, |
| "step": 5560 |
| }, |
| { |
| "entropy": 0.24414967973716556, |
| "epoch": 4.456, |
| "grad_norm": 4.625, |
| "learning_rate": 9.846666666666668e-06, |
| "loss": 0.1955, |
| "mean_token_accuracy": 0.9384665235877037, |
| "num_tokens": 92786646.0, |
| "step": 5570 |
| }, |
| { |
| "entropy": 0.2534818566869944, |
| "epoch": 4.464, |
| "grad_norm": 4.875, |
| "learning_rate": 9.824444444444446e-06, |
| "loss": 0.2075, |
| "mean_token_accuracy": 0.9359779760241509, |
| "num_tokens": 92954334.0, |
| "step": 5580 |
| }, |
| { |
| "entropy": 0.24864718355238438, |
| "epoch": 4.4719999999999995, |
| "grad_norm": 5.28125, |
| "learning_rate": 9.802222222222224e-06, |
| "loss": 0.1979, |
| "mean_token_accuracy": 0.9378546915948391, |
| "num_tokens": 93123391.0, |
| "step": 5590 |
| }, |
| { |
| "entropy": 0.2416323603130877, |
| "epoch": 4.48, |
| "grad_norm": 4.9375, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 0.1949, |
| "mean_token_accuracy": 0.9384495489299297, |
| "num_tokens": 93288025.0, |
| "step": 5600 |
| }, |
| { |
| "epoch": 4.48, |
| "eval_coding_entropy": 0.9062890486717224, |
| "eval_coding_loss": 3.387852430343628, |
| "eval_coding_mean_token_accuracy": 0.47923202538490295, |
| "eval_coding_num_tokens": 93288025.0, |
| "eval_coding_runtime": 81.3446, |
| "eval_coding_samples_per_second": 6.147, |
| "eval_coding_steps_per_second": 3.073, |
| "step": 5600 |
| }, |
| { |
| "epoch": 4.48, |
| "eval_physics_entropy": 0.3837174729704857, |
| "eval_physics_loss": 1.218315601348877, |
| "eval_physics_mean_token_accuracy": 0.7657727310657502, |
| "eval_physics_num_tokens": 93288025.0, |
| "eval_physics_runtime": 54.0042, |
| "eval_physics_samples_per_second": 9.259, |
| "eval_physics_steps_per_second": 4.629, |
| "step": 5600 |
| }, |
| { |
| "entropy": 0.2603446855209768, |
| "epoch": 4.4879999999999995, |
| "grad_norm": 5.53125, |
| "learning_rate": 9.757777777777779e-06, |
| "loss": 0.2086, |
| "mean_token_accuracy": 0.9351207684725523, |
| "num_tokens": 93450303.0, |
| "step": 5610 |
| }, |
| { |
| "entropy": 0.2522587388753891, |
| "epoch": 4.496, |
| "grad_norm": 5.34375, |
| "learning_rate": 9.735555555555556e-06, |
| "loss": 0.2042, |
| "mean_token_accuracy": 0.9354668237268925, |
| "num_tokens": 93621454.0, |
| "step": 5620 |
| }, |
| { |
| "entropy": 0.25489661609753966, |
| "epoch": 4.504, |
| "grad_norm": 4.8125, |
| "learning_rate": 9.713333333333334e-06, |
| "loss": 0.2027, |
| "mean_token_accuracy": 0.935209759324789, |
| "num_tokens": 93786045.0, |
| "step": 5630 |
| }, |
| { |
| "entropy": 0.25900470023043454, |
| "epoch": 4.5120000000000005, |
| "grad_norm": 5.375, |
| "learning_rate": 9.691111111111112e-06, |
| "loss": 0.2076, |
| "mean_token_accuracy": 0.9344288654625416, |
| "num_tokens": 93957847.0, |
| "step": 5640 |
| }, |
| { |
| "entropy": 0.25965928924269976, |
| "epoch": 4.52, |
| "grad_norm": 6.09375, |
| "learning_rate": 9.66888888888889e-06, |
| "loss": 0.2126, |
| "mean_token_accuracy": 0.9333121538162231, |
| "num_tokens": 94121741.0, |
| "step": 5650 |
| }, |
| { |
| "entropy": 0.24905031314119697, |
| "epoch": 4.5280000000000005, |
| "grad_norm": 5.65625, |
| "learning_rate": 9.646666666666667e-06, |
| "loss": 0.1998, |
| "mean_token_accuracy": 0.9364268813282252, |
| "num_tokens": 94287674.0, |
| "step": 5660 |
| }, |
| { |
| "entropy": 0.24139762027189135, |
| "epoch": 4.536, |
| "grad_norm": 5.40625, |
| "learning_rate": 9.624444444444445e-06, |
| "loss": 0.1907, |
| "mean_token_accuracy": 0.9397859137505293, |
| "num_tokens": 94450369.0, |
| "step": 5670 |
| }, |
| { |
| "entropy": 0.25715916510671377, |
| "epoch": 4.5440000000000005, |
| "grad_norm": 4.96875, |
| "learning_rate": 9.602222222222222e-06, |
| "loss": 0.2091, |
| "mean_token_accuracy": 0.9357426490634679, |
| "num_tokens": 94613456.0, |
| "step": 5680 |
| }, |
| { |
| "entropy": 0.2616544866934419, |
| "epoch": 4.552, |
| "grad_norm": 5.75, |
| "learning_rate": 9.58e-06, |
| "loss": 0.2065, |
| "mean_token_accuracy": 0.9356221269816161, |
| "num_tokens": 94775395.0, |
| "step": 5690 |
| }, |
| { |
| "entropy": 0.24898776262998581, |
| "epoch": 4.5600000000000005, |
| "grad_norm": 5.34375, |
| "learning_rate": 9.557777777777777e-06, |
| "loss": 0.1995, |
| "mean_token_accuracy": 0.9373305406421423, |
| "num_tokens": 94941684.0, |
| "step": 5700 |
| }, |
| { |
| "epoch": 4.5600000000000005, |
| "eval_coding_entropy": 0.8988055062294006, |
| "eval_coding_loss": 3.395606756210327, |
| "eval_coding_mean_token_accuracy": 0.4791694487333298, |
| "eval_coding_num_tokens": 94941684.0, |
| "eval_coding_runtime": 80.3019, |
| "eval_coding_samples_per_second": 6.227, |
| "eval_coding_steps_per_second": 3.113, |
| "step": 5700 |
| }, |
| { |
| "epoch": 4.5600000000000005, |
| "eval_physics_entropy": 0.38017728835344317, |
| "eval_physics_loss": 1.2212059497833252, |
| "eval_physics_mean_token_accuracy": 0.7663404207229614, |
| "eval_physics_num_tokens": 94941684.0, |
| "eval_physics_runtime": 53.908, |
| "eval_physics_samples_per_second": 9.275, |
| "eval_physics_steps_per_second": 4.638, |
| "step": 5700 |
| }, |
| { |
| "entropy": 0.2539431510027498, |
| "epoch": 4.568, |
| "grad_norm": 5.09375, |
| "learning_rate": 9.535555555555557e-06, |
| "loss": 0.2082, |
| "mean_token_accuracy": 0.935172364860773, |
| "num_tokens": 95109854.0, |
| "step": 5710 |
| }, |
| { |
| "entropy": 0.2610381402075291, |
| "epoch": 4.576, |
| "grad_norm": 5.65625, |
| "learning_rate": 9.513333333333334e-06, |
| "loss": 0.2066, |
| "mean_token_accuracy": 0.9338071327656507, |
| "num_tokens": 95278037.0, |
| "step": 5720 |
| }, |
| { |
| "entropy": 0.2522063947748393, |
| "epoch": 4.584, |
| "grad_norm": 4.90625, |
| "learning_rate": 9.491111111111112e-06, |
| "loss": 0.1991, |
| "mean_token_accuracy": 0.9396138455718756, |
| "num_tokens": 95442690.0, |
| "step": 5730 |
| }, |
| { |
| "entropy": 0.25343138808384535, |
| "epoch": 4.592, |
| "grad_norm": 4.9375, |
| "learning_rate": 9.46888888888889e-06, |
| "loss": 0.2012, |
| "mean_token_accuracy": 0.9365191299468278, |
| "num_tokens": 95611182.0, |
| "step": 5740 |
| }, |
| { |
| "entropy": 0.24988892613910138, |
| "epoch": 4.6, |
| "grad_norm": 5.34375, |
| "learning_rate": 9.446666666666667e-06, |
| "loss": 0.2027, |
| "mean_token_accuracy": 0.9366266682744027, |
| "num_tokens": 95780146.0, |
| "step": 5750 |
| }, |
| { |
| "entropy": 0.2660195084288716, |
| "epoch": 4.608, |
| "grad_norm": 5.28125, |
| "learning_rate": 9.424444444444445e-06, |
| "loss": 0.2141, |
| "mean_token_accuracy": 0.933330875262618, |
| "num_tokens": 95952986.0, |
| "step": 5760 |
| }, |
| { |
| "entropy": 0.2439029183704406, |
| "epoch": 4.616, |
| "grad_norm": 4.78125, |
| "learning_rate": 9.402222222222222e-06, |
| "loss": 0.1963, |
| "mean_token_accuracy": 0.9382667493075132, |
| "num_tokens": 96126079.0, |
| "step": 5770 |
| }, |
| { |
| "entropy": 0.2458340112119913, |
| "epoch": 4.624, |
| "grad_norm": 4.8125, |
| "learning_rate": 9.38e-06, |
| "loss": 0.2006, |
| "mean_token_accuracy": 0.9368267845362425, |
| "num_tokens": 96297175.0, |
| "step": 5780 |
| }, |
| { |
| "entropy": 0.2636739745736122, |
| "epoch": 4.632, |
| "grad_norm": 5.15625, |
| "learning_rate": 9.35777777777778e-06, |
| "loss": 0.2115, |
| "mean_token_accuracy": 0.9343665629625321, |
| "num_tokens": 96464963.0, |
| "step": 5790 |
| }, |
| { |
| "entropy": 0.27045974591746924, |
| "epoch": 4.64, |
| "grad_norm": 5.34375, |
| "learning_rate": 9.335555555555557e-06, |
| "loss": 0.2197, |
| "mean_token_accuracy": 0.9319406561553478, |
| "num_tokens": 96630392.0, |
| "step": 5800 |
| }, |
| { |
| "epoch": 4.64, |
| "eval_coding_entropy": 0.903528913974762, |
| "eval_coding_loss": 3.396648406982422, |
| "eval_coding_mean_token_accuracy": 0.47869827890396116, |
| "eval_coding_num_tokens": 96630392.0, |
| "eval_coding_runtime": 80.7716, |
| "eval_coding_samples_per_second": 6.19, |
| "eval_coding_steps_per_second": 3.095, |
| "step": 5800 |
| }, |
| { |
| "epoch": 4.64, |
| "eval_physics_entropy": 0.3829757045507431, |
| "eval_physics_loss": 1.2199360132217407, |
| "eval_physics_mean_token_accuracy": 0.766262964963913, |
| "eval_physics_num_tokens": 96630392.0, |
| "eval_physics_runtime": 54.6679, |
| "eval_physics_samples_per_second": 9.146, |
| "eval_physics_steps_per_second": 4.573, |
| "step": 5800 |
| }, |
| { |
| "entropy": 0.2534672657493502, |
| "epoch": 4.648, |
| "grad_norm": 4.9375, |
| "learning_rate": 9.313333333333335e-06, |
| "loss": 0.1977, |
| "mean_token_accuracy": 0.9366308826953172, |
| "num_tokens": 96801676.0, |
| "step": 5810 |
| }, |
| { |
| "entropy": 0.2465435629710555, |
| "epoch": 4.656, |
| "grad_norm": 4.8125, |
| "learning_rate": 9.291111111111112e-06, |
| "loss": 0.2008, |
| "mean_token_accuracy": 0.9363819014281034, |
| "num_tokens": 96978935.0, |
| "step": 5820 |
| }, |
| { |
| "entropy": 0.2436069377232343, |
| "epoch": 4.664, |
| "grad_norm": 5.15625, |
| "learning_rate": 9.26888888888889e-06, |
| "loss": 0.1959, |
| "mean_token_accuracy": 0.937981928139925, |
| "num_tokens": 97148367.0, |
| "step": 5830 |
| }, |
| { |
| "entropy": 0.2596503809094429, |
| "epoch": 4.672, |
| "grad_norm": 4.90625, |
| "learning_rate": 9.246666666666667e-06, |
| "loss": 0.216, |
| "mean_token_accuracy": 0.9330236673355102, |
| "num_tokens": 97316431.0, |
| "step": 5840 |
| }, |
| { |
| "entropy": 0.25156848281621935, |
| "epoch": 4.68, |
| "grad_norm": 5.0625, |
| "learning_rate": 9.224444444444445e-06, |
| "loss": 0.2016, |
| "mean_token_accuracy": 0.9370012372732163, |
| "num_tokens": 97489452.0, |
| "step": 5850 |
| }, |
| { |
| "entropy": 0.25523227341473104, |
| "epoch": 4.688, |
| "grad_norm": 4.8125, |
| "learning_rate": 9.202222222222224e-06, |
| "loss": 0.207, |
| "mean_token_accuracy": 0.9347957037389278, |
| "num_tokens": 97665534.0, |
| "step": 5860 |
| }, |
| { |
| "entropy": 0.25535209095105527, |
| "epoch": 4.696, |
| "grad_norm": 5.5, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 0.2023, |
| "mean_token_accuracy": 0.9348052326589823, |
| "num_tokens": 97833817.0, |
| "step": 5870 |
| }, |
| { |
| "entropy": 0.2495630129240453, |
| "epoch": 4.704, |
| "grad_norm": 4.96875, |
| "learning_rate": 9.15777777777778e-06, |
| "loss": 0.2027, |
| "mean_token_accuracy": 0.9376333773136138, |
| "num_tokens": 98004052.0, |
| "step": 5880 |
| }, |
| { |
| "entropy": 0.2485922292340547, |
| "epoch": 4.712, |
| "grad_norm": 5.4375, |
| "learning_rate": 9.135555555555557e-06, |
| "loss": 0.1983, |
| "mean_token_accuracy": 0.9373996194452048, |
| "num_tokens": 98162610.0, |
| "step": 5890 |
| }, |
| { |
| "entropy": 0.24784396877512335, |
| "epoch": 4.72, |
| "grad_norm": 4.5625, |
| "learning_rate": 9.113333333333335e-06, |
| "loss": 0.1968, |
| "mean_token_accuracy": 0.9383329566568136, |
| "num_tokens": 98325785.0, |
| "step": 5900 |
| }, |
| { |
| "epoch": 4.72, |
| "eval_coding_entropy": 0.8986214983463288, |
| "eval_coding_loss": 3.412487268447876, |
| "eval_coding_mean_token_accuracy": 0.4778224585056305, |
| "eval_coding_num_tokens": 98325785.0, |
| "eval_coding_runtime": 80.9299, |
| "eval_coding_samples_per_second": 6.178, |
| "eval_coding_steps_per_second": 3.089, |
| "step": 5900 |
| }, |
| { |
| "epoch": 4.72, |
| "eval_physics_entropy": 0.38050036400556564, |
| "eval_physics_loss": 1.2213542461395264, |
| "eval_physics_mean_token_accuracy": 0.7665274612903595, |
| "eval_physics_num_tokens": 98325785.0, |
| "eval_physics_runtime": 52.7024, |
| "eval_physics_samples_per_second": 9.487, |
| "eval_physics_steps_per_second": 4.744, |
| "step": 5900 |
| }, |
| { |
| "entropy": 0.2475348873063922, |
| "epoch": 4.728, |
| "grad_norm": 4.78125, |
| "learning_rate": 9.091111111111112e-06, |
| "loss": 0.1973, |
| "mean_token_accuracy": 0.9381827648729086, |
| "num_tokens": 98489878.0, |
| "step": 5910 |
| }, |
| { |
| "entropy": 0.26135987960733476, |
| "epoch": 4.736, |
| "grad_norm": 5.28125, |
| "learning_rate": 9.06888888888889e-06, |
| "loss": 0.2079, |
| "mean_token_accuracy": 0.9345921602100133, |
| "num_tokens": 98649969.0, |
| "step": 5920 |
| }, |
| { |
| "entropy": 0.24196100910194218, |
| "epoch": 4.744, |
| "grad_norm": 5.34375, |
| "learning_rate": 9.046666666666668e-06, |
| "loss": 0.1969, |
| "mean_token_accuracy": 0.9369668692350388, |
| "num_tokens": 98820789.0, |
| "step": 5930 |
| }, |
| { |
| "entropy": 0.2546594746876508, |
| "epoch": 4.752, |
| "grad_norm": 4.4375, |
| "learning_rate": 9.024444444444445e-06, |
| "loss": 0.199, |
| "mean_token_accuracy": 0.9356756627559661, |
| "num_tokens": 98987145.0, |
| "step": 5940 |
| }, |
| { |
| "entropy": 0.2511145379859954, |
| "epoch": 4.76, |
| "grad_norm": 4.53125, |
| "learning_rate": 9.002222222222223e-06, |
| "loss": 0.2028, |
| "mean_token_accuracy": 0.9371317740529775, |
| "num_tokens": 99150591.0, |
| "step": 5950 |
| }, |
| { |
| "entropy": 0.2575173297896981, |
| "epoch": 4.768, |
| "grad_norm": 5.3125, |
| "learning_rate": 8.98e-06, |
| "loss": 0.2094, |
| "mean_token_accuracy": 0.934956020489335, |
| "num_tokens": 99309770.0, |
| "step": 5960 |
| }, |
| { |
| "entropy": 0.2507347702980042, |
| "epoch": 4.776, |
| "grad_norm": 4.625, |
| "learning_rate": 8.957777777777778e-06, |
| "loss": 0.1975, |
| "mean_token_accuracy": 0.9379363503307104, |
| "num_tokens": 99477209.0, |
| "step": 5970 |
| }, |
| { |
| "entropy": 0.25367600335739554, |
| "epoch": 4.784, |
| "grad_norm": 5.40625, |
| "learning_rate": 8.935555555555556e-06, |
| "loss": 0.201, |
| "mean_token_accuracy": 0.9361171681433916, |
| "num_tokens": 99643583.0, |
| "step": 5980 |
| }, |
| { |
| "entropy": 0.27054069209843873, |
| "epoch": 4.792, |
| "grad_norm": 5.375, |
| "learning_rate": 8.913333333333333e-06, |
| "loss": 0.2146, |
| "mean_token_accuracy": 0.9322415180504322, |
| "num_tokens": 99805112.0, |
| "step": 5990 |
| }, |
| { |
| "entropy": 0.255790955806151, |
| "epoch": 4.8, |
| "grad_norm": 5.40625, |
| "learning_rate": 8.891111111111111e-06, |
| "loss": 0.2079, |
| "mean_token_accuracy": 0.9351296197623015, |
| "num_tokens": 99972302.0, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.8, |
| "eval_coding_entropy": 0.9033038120269775, |
| "eval_coding_loss": 3.40761137008667, |
| "eval_coding_mean_token_accuracy": 0.4778556535243988, |
| "eval_coding_num_tokens": 99972302.0, |
| "eval_coding_runtime": 136.2941, |
| "eval_coding_samples_per_second": 3.669, |
| "eval_coding_steps_per_second": 1.834, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.8, |
| "eval_physics_entropy": 0.38075472819805145, |
| "eval_physics_loss": 1.2201876640319824, |
| "eval_physics_mean_token_accuracy": 0.7671071858406067, |
| "eval_physics_num_tokens": 99972302.0, |
| "eval_physics_runtime": 52.1151, |
| "eval_physics_samples_per_second": 9.594, |
| "eval_physics_steps_per_second": 4.797, |
| "step": 6000 |
| }, |
| { |
| "entropy": 0.2585463082417846, |
| "epoch": 4.808, |
| "grad_norm": 6.125, |
| "learning_rate": 8.868888888888888e-06, |
| "loss": 0.2085, |
| "mean_token_accuracy": 0.9343608073890209, |
| "num_tokens": 100132659.0, |
| "step": 6010 |
| }, |
| { |
| "entropy": 0.2561819693539292, |
| "epoch": 4.816, |
| "grad_norm": 5.0625, |
| "learning_rate": 8.846666666666668e-06, |
| "loss": 0.2052, |
| "mean_token_accuracy": 0.9355102214962244, |
| "num_tokens": 100290023.0, |
| "step": 6020 |
| }, |
| { |
| "entropy": 0.24513509073294698, |
| "epoch": 4.824, |
| "grad_norm": 4.84375, |
| "learning_rate": 8.824444444444445e-06, |
| "loss": 0.1954, |
| "mean_token_accuracy": 0.9386078182607889, |
| "num_tokens": 100469081.0, |
| "step": 6030 |
| }, |
| { |
| "entropy": 0.2420899815391749, |
| "epoch": 4.832, |
| "grad_norm": 5.0, |
| "learning_rate": 8.802222222222223e-06, |
| "loss": 0.2002, |
| "mean_token_accuracy": 0.9370839197188616, |
| "num_tokens": 100643312.0, |
| "step": 6040 |
| }, |
| { |
| "entropy": 0.24983442863449454, |
| "epoch": 4.84, |
| "grad_norm": 5.59375, |
| "learning_rate": 8.78e-06, |
| "loss": 0.2036, |
| "mean_token_accuracy": 0.9359896499663591, |
| "num_tokens": 100811205.0, |
| "step": 6050 |
| }, |
| { |
| "entropy": 0.26226755380630495, |
| "epoch": 4.848, |
| "grad_norm": 5.46875, |
| "learning_rate": 8.757777777777778e-06, |
| "loss": 0.2084, |
| "mean_token_accuracy": 0.934832938015461, |
| "num_tokens": 100977034.0, |
| "step": 6060 |
| }, |
| { |
| "entropy": 0.2544243303127587, |
| "epoch": 4.856, |
| "grad_norm": 5.0, |
| "learning_rate": 8.735555555555556e-06, |
| "loss": 0.2094, |
| "mean_token_accuracy": 0.9346483610570431, |
| "num_tokens": 101148036.0, |
| "step": 6070 |
| }, |
| { |
| "entropy": 0.24702603537589313, |
| "epoch": 4.864, |
| "grad_norm": 5.15625, |
| "learning_rate": 8.713333333333333e-06, |
| "loss": 0.2014, |
| "mean_token_accuracy": 0.9368080459535122, |
| "num_tokens": 101309531.0, |
| "step": 6080 |
| }, |
| { |
| "entropy": 0.2597447274252772, |
| "epoch": 4.872, |
| "grad_norm": 5.53125, |
| "learning_rate": 8.691111111111111e-06, |
| "loss": 0.2065, |
| "mean_token_accuracy": 0.9353206589818001, |
| "num_tokens": 101477622.0, |
| "step": 6090 |
| }, |
| { |
| "entropy": 0.2612296444363892, |
| "epoch": 4.88, |
| "grad_norm": 5.84375, |
| "learning_rate": 8.66888888888889e-06, |
| "loss": 0.2111, |
| "mean_token_accuracy": 0.9329081155359745, |
| "num_tokens": 101646955.0, |
| "step": 6100 |
| }, |
| { |
| "epoch": 4.88, |
| "eval_coding_entropy": 0.8974457256793976, |
| "eval_coding_loss": 3.405156135559082, |
| "eval_coding_mean_token_accuracy": 0.47816654479503634, |
| "eval_coding_num_tokens": 101646955.0, |
| "eval_coding_runtime": 79.7896, |
| "eval_coding_samples_per_second": 6.266, |
| "eval_coding_steps_per_second": 3.133, |
| "step": 6100 |
| }, |
| { |
| "epoch": 4.88, |
| "eval_physics_entropy": 0.38089080721139906, |
| "eval_physics_loss": 1.2193241119384766, |
| "eval_physics_mean_token_accuracy": 0.7663243329524994, |
| "eval_physics_num_tokens": 101646955.0, |
| "eval_physics_runtime": 52.7727, |
| "eval_physics_samples_per_second": 9.475, |
| "eval_physics_steps_per_second": 4.737, |
| "step": 6100 |
| }, |
| { |
| "entropy": 0.2502961977384984, |
| "epoch": 4.888, |
| "grad_norm": 6.0, |
| "learning_rate": 8.646666666666668e-06, |
| "loss": 0.1994, |
| "mean_token_accuracy": 0.9364591002464294, |
| "num_tokens": 101815839.0, |
| "step": 6110 |
| }, |
| { |
| "entropy": 0.2527038010302931, |
| "epoch": 4.896, |
| "grad_norm": 5.09375, |
| "learning_rate": 8.624444444444446e-06, |
| "loss": 0.1996, |
| "mean_token_accuracy": 0.937300831079483, |
| "num_tokens": 101981516.0, |
| "step": 6120 |
| }, |
| { |
| "entropy": 0.25195587510243056, |
| "epoch": 4.904, |
| "grad_norm": 5.3125, |
| "learning_rate": 8.602222222222223e-06, |
| "loss": 0.2075, |
| "mean_token_accuracy": 0.9359634295105934, |
| "num_tokens": 102145097.0, |
| "step": 6130 |
| }, |
| { |
| "entropy": 0.2508444357663393, |
| "epoch": 4.912, |
| "grad_norm": 5.0, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 0.2013, |
| "mean_token_accuracy": 0.9370537139475346, |
| "num_tokens": 102312681.0, |
| "step": 6140 |
| }, |
| { |
| "entropy": 0.26045903414487837, |
| "epoch": 4.92, |
| "grad_norm": 5.21875, |
| "learning_rate": 8.557777777777778e-06, |
| "loss": 0.2121, |
| "mean_token_accuracy": 0.933551586791873, |
| "num_tokens": 102479899.0, |
| "step": 6150 |
| }, |
| { |
| "entropy": 0.24932825425639749, |
| "epoch": 4.928, |
| "grad_norm": 5.0625, |
| "learning_rate": 8.535555555555556e-06, |
| "loss": 0.2015, |
| "mean_token_accuracy": 0.9369784589856863, |
| "num_tokens": 102647889.0, |
| "step": 6160 |
| }, |
| { |
| "entropy": 0.253802378103137, |
| "epoch": 4.936, |
| "grad_norm": 4.5625, |
| "learning_rate": 8.513333333333335e-06, |
| "loss": 0.2045, |
| "mean_token_accuracy": 0.9349048610776662, |
| "num_tokens": 102816503.0, |
| "step": 6170 |
| }, |
| { |
| "entropy": 0.2582039737608284, |
| "epoch": 4.944, |
| "grad_norm": 5.0625, |
| "learning_rate": 8.491111111111113e-06, |
| "loss": 0.2073, |
| "mean_token_accuracy": 0.9342209905385971, |
| "num_tokens": 102985809.0, |
| "step": 6180 |
| }, |
| { |
| "entropy": 0.2591179468203336, |
| "epoch": 4.952, |
| "grad_norm": 5.0625, |
| "learning_rate": 8.46888888888889e-06, |
| "loss": 0.2118, |
| "mean_token_accuracy": 0.9337667029350996, |
| "num_tokens": 103151372.0, |
| "step": 6190 |
| }, |
| { |
| "entropy": 0.24958483809605242, |
| "epoch": 4.96, |
| "grad_norm": 4.96875, |
| "learning_rate": 8.446666666666668e-06, |
| "loss": 0.2033, |
| "mean_token_accuracy": 0.9371551293879747, |
| "num_tokens": 103320682.0, |
| "step": 6200 |
| }, |
| { |
| "epoch": 4.96, |
| "eval_coding_entropy": 0.8931297225952148, |
| "eval_coding_loss": 3.408529281616211, |
| "eval_coding_mean_token_accuracy": 0.47882734644412994, |
| "eval_coding_num_tokens": 103320682.0, |
| "eval_coding_runtime": 79.5523, |
| "eval_coding_samples_per_second": 6.285, |
| "eval_coding_steps_per_second": 3.143, |
| "step": 6200 |
| }, |
| { |
| "epoch": 4.96, |
| "eval_physics_entropy": 0.3808671704530716, |
| "eval_physics_loss": 1.2207255363464355, |
| "eval_physics_mean_token_accuracy": 0.7665228016376495, |
| "eval_physics_num_tokens": 103320682.0, |
| "eval_physics_runtime": 52.4117, |
| "eval_physics_samples_per_second": 9.54, |
| "eval_physics_steps_per_second": 4.77, |
| "step": 6200 |
| }, |
| { |
| "entropy": 0.2500837286002934, |
| "epoch": 4.968, |
| "grad_norm": 5.75, |
| "learning_rate": 8.424444444444446e-06, |
| "loss": 0.2035, |
| "mean_token_accuracy": 0.9360714457929135, |
| "num_tokens": 103490848.0, |
| "step": 6210 |
| }, |
| { |
| "entropy": 0.25305729256942866, |
| "epoch": 4.976, |
| "grad_norm": 4.96875, |
| "learning_rate": 8.402222222222223e-06, |
| "loss": 0.2038, |
| "mean_token_accuracy": 0.9358239717781544, |
| "num_tokens": 103656297.0, |
| "step": 6220 |
| }, |
| { |
| "entropy": 0.25323430392891166, |
| "epoch": 4.984, |
| "grad_norm": 5.09375, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 0.2027, |
| "mean_token_accuracy": 0.9357287209481001, |
| "num_tokens": 103818187.0, |
| "step": 6230 |
| }, |
| { |
| "entropy": 0.2474314162507653, |
| "epoch": 4.992, |
| "grad_norm": 5.5625, |
| "learning_rate": 8.357777777777779e-06, |
| "loss": 0.1949, |
| "mean_token_accuracy": 0.9377578873187303, |
| "num_tokens": 103984906.0, |
| "step": 6240 |
| }, |
| { |
| "entropy": 0.24971986413002015, |
| "epoch": 5.0, |
| "grad_norm": 5.25, |
| "learning_rate": 8.335555555555556e-06, |
| "loss": 0.2037, |
| "mean_token_accuracy": 0.9367631152272224, |
| "num_tokens": 104160740.0, |
| "step": 6250 |
| }, |
| { |
| "entropy": 0.2031078478321433, |
| "epoch": 5.008, |
| "grad_norm": 4.28125, |
| "learning_rate": 8.313333333333334e-06, |
| "loss": 0.125, |
| "mean_token_accuracy": 0.9659961301833391, |
| "num_tokens": 104329988.0, |
| "step": 6260 |
| }, |
| { |
| "entropy": 0.2012895429506898, |
| "epoch": 5.016, |
| "grad_norm": 3.9375, |
| "learning_rate": 8.291111111111112e-06, |
| "loss": 0.1287, |
| "mean_token_accuracy": 0.9643992695957422, |
| "num_tokens": 104500448.0, |
| "step": 6270 |
| }, |
| { |
| "entropy": 0.2011468899436295, |
| "epoch": 5.024, |
| "grad_norm": 3.734375, |
| "learning_rate": 8.268888888888889e-06, |
| "loss": 0.1275, |
| "mean_token_accuracy": 0.9643775910139084, |
| "num_tokens": 104670082.0, |
| "step": 6280 |
| }, |
| { |
| "entropy": 0.187453023949638, |
| "epoch": 5.032, |
| "grad_norm": 3.6875, |
| "learning_rate": 8.246666666666667e-06, |
| "loss": 0.1166, |
| "mean_token_accuracy": 0.9668762777000666, |
| "num_tokens": 104839109.0, |
| "step": 6290 |
| }, |
| { |
| "entropy": 0.19810657487250866, |
| "epoch": 5.04, |
| "grad_norm": 3.8125, |
| "learning_rate": 8.224444444444444e-06, |
| "loss": 0.1315, |
| "mean_token_accuracy": 0.9638035885989666, |
| "num_tokens": 105012648.0, |
| "step": 6300 |
| }, |
| { |
| "epoch": 5.04, |
| "eval_coding_entropy": 0.8877947137355805, |
| "eval_coding_loss": 3.545646905899048, |
| "eval_coding_mean_token_accuracy": 0.4724773551225662, |
| "eval_coding_num_tokens": 105012648.0, |
| "eval_coding_runtime": 79.5298, |
| "eval_coding_samples_per_second": 6.287, |
| "eval_coding_steps_per_second": 3.143, |
| "step": 6300 |
| }, |
| { |
| "epoch": 5.04, |
| "eval_physics_entropy": 0.3605686429142952, |
| "eval_physics_loss": 1.289768934249878, |
| "eval_physics_mean_token_accuracy": 0.7637167615890503, |
| "eval_physics_num_tokens": 105012648.0, |
| "eval_physics_runtime": 52.4609, |
| "eval_physics_samples_per_second": 9.531, |
| "eval_physics_steps_per_second": 4.765, |
| "step": 6300 |
| }, |
| { |
| "entropy": 0.20116808991879226, |
| "epoch": 5.048, |
| "grad_norm": 4.0, |
| "learning_rate": 8.202222222222222e-06, |
| "loss": 0.1233, |
| "mean_token_accuracy": 0.9658465381711722, |
| "num_tokens": 105175353.0, |
| "step": 6310 |
| }, |
| { |
| "entropy": 0.19210361344739796, |
| "epoch": 5.056, |
| "grad_norm": 4.21875, |
| "learning_rate": 8.18e-06, |
| "loss": 0.1233, |
| "mean_token_accuracy": 0.965586156770587, |
| "num_tokens": 105340074.0, |
| "step": 6320 |
| }, |
| { |
| "entropy": 0.19798338054679335, |
| "epoch": 5.064, |
| "grad_norm": 3.84375, |
| "learning_rate": 8.157777777777779e-06, |
| "loss": 0.1265, |
| "mean_token_accuracy": 0.9641043458133935, |
| "num_tokens": 105509206.0, |
| "step": 6330 |
| }, |
| { |
| "entropy": 0.19844807693734764, |
| "epoch": 5.072, |
| "grad_norm": 3.75, |
| "learning_rate": 8.135555555555557e-06, |
| "loss": 0.1293, |
| "mean_token_accuracy": 0.9633849579840899, |
| "num_tokens": 105674220.0, |
| "step": 6340 |
| }, |
| { |
| "entropy": 0.20063547580502927, |
| "epoch": 5.08, |
| "grad_norm": 4.0625, |
| "learning_rate": 8.113333333333334e-06, |
| "loss": 0.1294, |
| "mean_token_accuracy": 0.9632270362228155, |
| "num_tokens": 105837080.0, |
| "step": 6350 |
| }, |
| { |
| "entropy": 0.19153314703144134, |
| "epoch": 5.088, |
| "grad_norm": 4.40625, |
| "learning_rate": 8.091111111111112e-06, |
| "loss": 0.1239, |
| "mean_token_accuracy": 0.9651885312050581, |
| "num_tokens": 106008744.0, |
| "step": 6360 |
| }, |
| { |
| "entropy": 0.19501116001047195, |
| "epoch": 5.096, |
| "grad_norm": 4.53125, |
| "learning_rate": 8.06888888888889e-06, |
| "loss": 0.1265, |
| "mean_token_accuracy": 0.9651710100471973, |
| "num_tokens": 106173891.0, |
| "step": 6370 |
| }, |
| { |
| "entropy": 0.19993340387009084, |
| "epoch": 5.104, |
| "grad_norm": 4.15625, |
| "learning_rate": 8.046666666666667e-06, |
| "loss": 0.1294, |
| "mean_token_accuracy": 0.9630896743386984, |
| "num_tokens": 106332594.0, |
| "step": 6380 |
| }, |
| { |
| "entropy": 0.18695482835173607, |
| "epoch": 5.112, |
| "grad_norm": 3.671875, |
| "learning_rate": 8.024444444444445e-06, |
| "loss": 0.1203, |
| "mean_token_accuracy": 0.9662993267178536, |
| "num_tokens": 106503940.0, |
| "step": 6390 |
| }, |
| { |
| "entropy": 0.19542481317184865, |
| "epoch": 5.12, |
| "grad_norm": 4.09375, |
| "learning_rate": 8.002222222222222e-06, |
| "loss": 0.1264, |
| "mean_token_accuracy": 0.9647392224520445, |
| "num_tokens": 106670394.0, |
| "step": 6400 |
| }, |
| { |
| "epoch": 5.12, |
| "eval_coding_entropy": 0.8728853042125702, |
| "eval_coding_loss": 3.5656251907348633, |
| "eval_coding_mean_token_accuracy": 0.47321837043762205, |
| "eval_coding_num_tokens": 106670394.0, |
| "eval_coding_runtime": 79.8087, |
| "eval_coding_samples_per_second": 6.265, |
| "eval_coding_steps_per_second": 3.132, |
| "step": 6400 |
| }, |
| { |
| "epoch": 5.12, |
| "eval_physics_entropy": 0.3538407863378525, |
| "eval_physics_loss": 1.3026418685913086, |
| "eval_physics_mean_token_accuracy": 0.7633731746673584, |
| "eval_physics_num_tokens": 106670394.0, |
| "eval_physics_runtime": 53.0398, |
| "eval_physics_samples_per_second": 9.427, |
| "eval_physics_steps_per_second": 4.713, |
| "step": 6400 |
| }, |
| { |
| "entropy": 0.19087084899656476, |
| "epoch": 5.128, |
| "grad_norm": 3.453125, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 0.1234, |
| "mean_token_accuracy": 0.9646917715668678, |
| "num_tokens": 106843359.0, |
| "step": 6410 |
| }, |
| { |
| "entropy": 0.19899238399229943, |
| "epoch": 5.136, |
| "grad_norm": 4.125, |
| "learning_rate": 7.957777777777779e-06, |
| "loss": 0.1308, |
| "mean_token_accuracy": 0.9636497907340527, |
| "num_tokens": 107010814.0, |
| "step": 6420 |
| }, |
| { |
| "entropy": 0.1939126175828278, |
| "epoch": 5.144, |
| "grad_norm": 3.984375, |
| "learning_rate": 7.935555555555557e-06, |
| "loss": 0.1278, |
| "mean_token_accuracy": 0.9645107124000788, |
| "num_tokens": 107184103.0, |
| "step": 6430 |
| }, |
| { |
| "entropy": 0.19105937178246676, |
| "epoch": 5.152, |
| "grad_norm": 3.671875, |
| "learning_rate": 7.913333333333334e-06, |
| "loss": 0.1233, |
| "mean_token_accuracy": 0.9650929987430572, |
| "num_tokens": 107353217.0, |
| "step": 6440 |
| }, |
| { |
| "entropy": 0.19350384399294854, |
| "epoch": 5.16, |
| "grad_norm": 4.4375, |
| "learning_rate": 7.891111111111112e-06, |
| "loss": 0.1262, |
| "mean_token_accuracy": 0.9641659058630466, |
| "num_tokens": 107529705.0, |
| "step": 6450 |
| }, |
| { |
| "entropy": 0.1894688137806952, |
| "epoch": 5.168, |
| "grad_norm": 3.90625, |
| "learning_rate": 7.86888888888889e-06, |
| "loss": 0.1221, |
| "mean_token_accuracy": 0.9654671866446733, |
| "num_tokens": 107704886.0, |
| "step": 6460 |
| }, |
| { |
| "entropy": 0.1967856579925865, |
| "epoch": 5.176, |
| "grad_norm": 4.5, |
| "learning_rate": 7.846666666666667e-06, |
| "loss": 0.1283, |
| "mean_token_accuracy": 0.9639428824186325, |
| "num_tokens": 107867946.0, |
| "step": 6470 |
| }, |
| { |
| "entropy": 0.19449630728922784, |
| "epoch": 5.184, |
| "grad_norm": 3.953125, |
| "learning_rate": 7.824444444444445e-06, |
| "loss": 0.1274, |
| "mean_token_accuracy": 0.9638728551566601, |
| "num_tokens": 108033564.0, |
| "step": 6480 |
| }, |
| { |
| "entropy": 0.18775347215123475, |
| "epoch": 5.192, |
| "grad_norm": 4.09375, |
| "learning_rate": 7.802222222222222e-06, |
| "loss": 0.1227, |
| "mean_token_accuracy": 0.9655531663447618, |
| "num_tokens": 108202048.0, |
| "step": 6490 |
| }, |
| { |
| "entropy": 0.19767596102319657, |
| "epoch": 5.2, |
| "grad_norm": 4.21875, |
| "learning_rate": 7.78e-06, |
| "loss": 0.1293, |
| "mean_token_accuracy": 0.9626834951341152, |
| "num_tokens": 108368498.0, |
| "step": 6500 |
| }, |
| { |
| "epoch": 5.2, |
| "eval_coding_entropy": 0.8674574599266053, |
| "eval_coding_loss": 3.574704170227051, |
| "eval_coding_mean_token_accuracy": 0.4729505362510681, |
| "eval_coding_num_tokens": 108368498.0, |
| "eval_coding_runtime": 79.5644, |
| "eval_coding_samples_per_second": 6.284, |
| "eval_coding_steps_per_second": 3.142, |
| "step": 6500 |
| }, |
| { |
| "epoch": 5.2, |
| "eval_physics_entropy": 0.35313829815387726, |
| "eval_physics_loss": 1.3051433563232422, |
| "eval_physics_mean_token_accuracy": 0.7636888906955719, |
| "eval_physics_num_tokens": 108368498.0, |
| "eval_physics_runtime": 52.6391, |
| "eval_physics_samples_per_second": 9.499, |
| "eval_physics_steps_per_second": 4.749, |
| "step": 6500 |
| }, |
| { |
| "entropy": 0.19391954350285232, |
| "epoch": 5.208, |
| "grad_norm": 3.78125, |
| "learning_rate": 7.75777777777778e-06, |
| "loss": 0.1252, |
| "mean_token_accuracy": 0.9645942829549312, |
| "num_tokens": 108538030.0, |
| "step": 6510 |
| }, |
| { |
| "entropy": 0.19754811474122108, |
| "epoch": 5.216, |
| "grad_norm": 3.875, |
| "learning_rate": 7.735555555555557e-06, |
| "loss": 0.1238, |
| "mean_token_accuracy": 0.9637079253792763, |
| "num_tokens": 108699224.0, |
| "step": 6520 |
| }, |
| { |
| "entropy": 0.1891830234322697, |
| "epoch": 5.224, |
| "grad_norm": 3.921875, |
| "learning_rate": 7.713333333333335e-06, |
| "loss": 0.1216, |
| "mean_token_accuracy": 0.9655250731855631, |
| "num_tokens": 108868144.0, |
| "step": 6530 |
| }, |
| { |
| "entropy": 0.18879148364067078, |
| "epoch": 5.232, |
| "grad_norm": 4.125, |
| "learning_rate": 7.691111111111112e-06, |
| "loss": 0.1243, |
| "mean_token_accuracy": 0.9657261438667775, |
| "num_tokens": 109037108.0, |
| "step": 6540 |
| }, |
| { |
| "entropy": 0.20060481796972454, |
| "epoch": 5.24, |
| "grad_norm": 4.15625, |
| "learning_rate": 7.66888888888889e-06, |
| "loss": 0.1305, |
| "mean_token_accuracy": 0.9635683462023735, |
| "num_tokens": 109207513.0, |
| "step": 6550 |
| }, |
| { |
| "entropy": 0.1975304267834872, |
| "epoch": 5.248, |
| "grad_norm": 4.0625, |
| "learning_rate": 7.646666666666667e-06, |
| "loss": 0.1289, |
| "mean_token_accuracy": 0.9641137260943651, |
| "num_tokens": 109368040.0, |
| "step": 6560 |
| }, |
| { |
| "entropy": 0.19977891994640232, |
| "epoch": 5.256, |
| "grad_norm": 4.34375, |
| "learning_rate": 7.624444444444445e-06, |
| "loss": 0.1299, |
| "mean_token_accuracy": 0.9634544603526592, |
| "num_tokens": 109526075.0, |
| "step": 6570 |
| }, |
| { |
| "entropy": 0.19655378069728613, |
| "epoch": 5.264, |
| "grad_norm": 4.1875, |
| "learning_rate": 7.602222222222223e-06, |
| "loss": 0.1312, |
| "mean_token_accuracy": 0.9628745052963495, |
| "num_tokens": 109691145.0, |
| "step": 6580 |
| }, |
| { |
| "entropy": 0.19395874086767434, |
| "epoch": 5.272, |
| "grad_norm": 4.0625, |
| "learning_rate": 7.58e-06, |
| "loss": 0.1252, |
| "mean_token_accuracy": 0.9642571996897459, |
| "num_tokens": 109850880.0, |
| "step": 6590 |
| }, |
| { |
| "entropy": 0.20438815425150098, |
| "epoch": 5.28, |
| "grad_norm": 4.09375, |
| "learning_rate": 7.557777777777779e-06, |
| "loss": 0.1364, |
| "mean_token_accuracy": 0.9609546218067407, |
| "num_tokens": 110019099.0, |
| "step": 6600 |
| }, |
| { |
| "epoch": 5.28, |
| "eval_coding_entropy": 0.8623562679290772, |
| "eval_coding_loss": 3.5841898918151855, |
| "eval_coding_mean_token_accuracy": 0.4725262248516083, |
| "eval_coding_num_tokens": 110019099.0, |
| "eval_coding_runtime": 79.145, |
| "eval_coding_samples_per_second": 6.318, |
| "eval_coding_steps_per_second": 3.159, |
| "step": 6600 |
| }, |
| { |
| "epoch": 5.28, |
| "eval_physics_entropy": 0.35105196392536164, |
| "eval_physics_loss": 1.311557412147522, |
| "eval_physics_mean_token_accuracy": 0.7634473481178283, |
| "eval_physics_num_tokens": 110019099.0, |
| "eval_physics_runtime": 52.0687, |
| "eval_physics_samples_per_second": 9.603, |
| "eval_physics_steps_per_second": 4.801, |
| "step": 6600 |
| }, |
| { |
| "entropy": 0.19830146143212915, |
| "epoch": 5.288, |
| "grad_norm": 4.71875, |
| "learning_rate": 7.535555555555556e-06, |
| "loss": 0.1302, |
| "mean_token_accuracy": 0.9621496997773648, |
| "num_tokens": 110177969.0, |
| "step": 6610 |
| }, |
| { |
| "entropy": 0.19456713651306928, |
| "epoch": 5.296, |
| "grad_norm": 4.28125, |
| "learning_rate": 7.513333333333334e-06, |
| "loss": 0.1257, |
| "mean_token_accuracy": 0.9633442517369986, |
| "num_tokens": 110349034.0, |
| "step": 6620 |
| }, |
| { |
| "entropy": 0.19574856031686066, |
| "epoch": 5.304, |
| "grad_norm": 4.375, |
| "learning_rate": 7.4911111111111116e-06, |
| "loss": 0.1302, |
| "mean_token_accuracy": 0.9632543411105872, |
| "num_tokens": 110513874.0, |
| "step": 6630 |
| }, |
| { |
| "entropy": 0.197614107420668, |
| "epoch": 5.312, |
| "grad_norm": 4.28125, |
| "learning_rate": 7.46888888888889e-06, |
| "loss": 0.1303, |
| "mean_token_accuracy": 0.9630021318793297, |
| "num_tokens": 110678547.0, |
| "step": 6640 |
| }, |
| { |
| "entropy": 0.19904073821380736, |
| "epoch": 5.32, |
| "grad_norm": 4.53125, |
| "learning_rate": 7.446666666666668e-06, |
| "loss": 0.1348, |
| "mean_token_accuracy": 0.9616572860628366, |
| "num_tokens": 110845570.0, |
| "step": 6650 |
| }, |
| { |
| "entropy": 0.19347775275819004, |
| "epoch": 5.328, |
| "grad_norm": 4.3125, |
| "learning_rate": 7.424444444444445e-06, |
| "loss": 0.1273, |
| "mean_token_accuracy": 0.9638630177825689, |
| "num_tokens": 111013689.0, |
| "step": 6660 |
| }, |
| { |
| "entropy": 0.19356580921448768, |
| "epoch": 5.336, |
| "grad_norm": 4.34375, |
| "learning_rate": 7.402222222222223e-06, |
| "loss": 0.1301, |
| "mean_token_accuracy": 0.9627449579536915, |
| "num_tokens": 111180389.0, |
| "step": 6670 |
| }, |
| { |
| "entropy": 0.19895400274544955, |
| "epoch": 5.344, |
| "grad_norm": 3.90625, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 0.1269, |
| "mean_token_accuracy": 0.9627930507063865, |
| "num_tokens": 111344270.0, |
| "step": 6680 |
| }, |
| { |
| "entropy": 0.19328222065232695, |
| "epoch": 5.352, |
| "grad_norm": 4.78125, |
| "learning_rate": 7.357777777777778e-06, |
| "loss": 0.1289, |
| "mean_token_accuracy": 0.9631569210439921, |
| "num_tokens": 111516980.0, |
| "step": 6690 |
| }, |
| { |
| "entropy": 0.19710124852135777, |
| "epoch": 5.36, |
| "grad_norm": 4.65625, |
| "learning_rate": 7.335555555555556e-06, |
| "loss": 0.1257, |
| "mean_token_accuracy": 0.963890090957284, |
| "num_tokens": 111686051.0, |
| "step": 6700 |
| }, |
| { |
| "epoch": 5.36, |
| "eval_coding_entropy": 0.8623976769447327, |
| "eval_coding_loss": 3.585888385772705, |
| "eval_coding_mean_token_accuracy": 0.4722388235330582, |
| "eval_coding_num_tokens": 111686051.0, |
| "eval_coding_runtime": 79.1468, |
| "eval_coding_samples_per_second": 6.317, |
| "eval_coding_steps_per_second": 3.159, |
| "step": 6700 |
| }, |
| { |
| "epoch": 5.36, |
| "eval_physics_entropy": 0.35141410732269285, |
| "eval_physics_loss": 1.3126267194747925, |
| "eval_physics_mean_token_accuracy": 0.7636447312831879, |
| "eval_physics_num_tokens": 111686051.0, |
| "eval_physics_runtime": 52.2834, |
| "eval_physics_samples_per_second": 9.563, |
| "eval_physics_steps_per_second": 4.782, |
| "step": 6700 |
| }, |
| { |
| "entropy": 0.18666236158460378, |
| "epoch": 5.368, |
| "grad_norm": 4.1875, |
| "learning_rate": 7.313333333333333e-06, |
| "loss": 0.1247, |
| "mean_token_accuracy": 0.9651182435452939, |
| "num_tokens": 111861453.0, |
| "step": 6710 |
| }, |
| { |
| "entropy": 0.20253037605434657, |
| "epoch": 5.376, |
| "grad_norm": 4.375, |
| "learning_rate": 7.291111111111112e-06, |
| "loss": 0.1336, |
| "mean_token_accuracy": 0.9617640402168035, |
| "num_tokens": 112023874.0, |
| "step": 6720 |
| }, |
| { |
| "entropy": 0.2006529273930937, |
| "epoch": 5.384, |
| "grad_norm": 4.46875, |
| "learning_rate": 7.268888888888889e-06, |
| "loss": 0.1338, |
| "mean_token_accuracy": 0.961680406332016, |
| "num_tokens": 112190781.0, |
| "step": 6730 |
| }, |
| { |
| "entropy": 0.1951970740221441, |
| "epoch": 5.392, |
| "grad_norm": 4.28125, |
| "learning_rate": 7.246666666666667e-06, |
| "loss": 0.1299, |
| "mean_token_accuracy": 0.9624206535518169, |
| "num_tokens": 112358202.0, |
| "step": 6740 |
| }, |
| { |
| "entropy": 0.19778370643034576, |
| "epoch": 5.4, |
| "grad_norm": 4.71875, |
| "learning_rate": 7.224444444444445e-06, |
| "loss": 0.132, |
| "mean_token_accuracy": 0.9623905044049025, |
| "num_tokens": 112527115.0, |
| "step": 6750 |
| }, |
| { |
| "entropy": 0.19216167866252362, |
| "epoch": 5.408, |
| "grad_norm": 4.0625, |
| "learning_rate": 7.202222222222222e-06, |
| "loss": 0.1248, |
| "mean_token_accuracy": 0.9653086230158806, |
| "num_tokens": 112692972.0, |
| "step": 6760 |
| }, |
| { |
| "entropy": 0.195600410271436, |
| "epoch": 5.416, |
| "grad_norm": 4.03125, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 0.1273, |
| "mean_token_accuracy": 0.9631940271705389, |
| "num_tokens": 112854708.0, |
| "step": 6770 |
| }, |
| { |
| "entropy": 0.19283109321258962, |
| "epoch": 5.424, |
| "grad_norm": 4.34375, |
| "learning_rate": 7.157777777777778e-06, |
| "loss": 0.1307, |
| "mean_token_accuracy": 0.9628949403762818, |
| "num_tokens": 113025416.0, |
| "step": 6780 |
| }, |
| { |
| "entropy": 0.20051187574863433, |
| "epoch": 5.432, |
| "grad_norm": 4.375, |
| "learning_rate": 7.135555555555557e-06, |
| "loss": 0.1312, |
| "mean_token_accuracy": 0.9624277334660292, |
| "num_tokens": 113187923.0, |
| "step": 6790 |
| }, |
| { |
| "entropy": 0.19794103307649494, |
| "epoch": 5.44, |
| "grad_norm": 4.53125, |
| "learning_rate": 7.113333333333334e-06, |
| "loss": 0.1335, |
| "mean_token_accuracy": 0.9620528299361467, |
| "num_tokens": 113360632.0, |
| "step": 6800 |
| }, |
| { |
| "epoch": 5.44, |
| "eval_coding_entropy": 0.8600458872318267, |
| "eval_coding_loss": 3.589592695236206, |
| "eval_coding_mean_token_accuracy": 0.47239135658741, |
| "eval_coding_num_tokens": 113360632.0, |
| "eval_coding_runtime": 79.2453, |
| "eval_coding_samples_per_second": 6.31, |
| "eval_coding_steps_per_second": 3.155, |
| "step": 6800 |
| }, |
| { |
| "epoch": 5.44, |
| "eval_physics_entropy": 0.35010364663600924, |
| "eval_physics_loss": 1.3127825260162354, |
| "eval_physics_mean_token_accuracy": 0.7634428672790528, |
| "eval_physics_num_tokens": 113360632.0, |
| "eval_physics_runtime": 52.0766, |
| "eval_physics_samples_per_second": 9.601, |
| "eval_physics_steps_per_second": 4.801, |
| "step": 6800 |
| }, |
| { |
| "entropy": 0.1972166199237108, |
| "epoch": 5.448, |
| "grad_norm": 4.21875, |
| "learning_rate": 7.091111111111112e-06, |
| "loss": 0.1304, |
| "mean_token_accuracy": 0.962584475800395, |
| "num_tokens": 113522539.0, |
| "step": 6810 |
| }, |
| { |
| "entropy": 0.19796474049799145, |
| "epoch": 5.456, |
| "grad_norm": 4.34375, |
| "learning_rate": 7.06888888888889e-06, |
| "loss": 0.1315, |
| "mean_token_accuracy": 0.9625082284212112, |
| "num_tokens": 113687612.0, |
| "step": 6820 |
| }, |
| { |
| "entropy": 0.19926895541138948, |
| "epoch": 5.464, |
| "grad_norm": 4.4375, |
| "learning_rate": 7.046666666666667e-06, |
| "loss": 0.1325, |
| "mean_token_accuracy": 0.9618335835635662, |
| "num_tokens": 113846052.0, |
| "step": 6830 |
| }, |
| { |
| "entropy": 0.19587073437869548, |
| "epoch": 5.4719999999999995, |
| "grad_norm": 4.375, |
| "learning_rate": 7.024444444444445e-06, |
| "loss": 0.1291, |
| "mean_token_accuracy": 0.9624919075518846, |
| "num_tokens": 114012825.0, |
| "step": 6840 |
| }, |
| { |
| "entropy": 0.19304803358390926, |
| "epoch": 5.48, |
| "grad_norm": 4.625, |
| "learning_rate": 7.0022222222222225e-06, |
| "loss": 0.127, |
| "mean_token_accuracy": 0.9630837731063366, |
| "num_tokens": 114176510.0, |
| "step": 6850 |
| }, |
| { |
| "entropy": 0.1884897816926241, |
| "epoch": 5.4879999999999995, |
| "grad_norm": 3.75, |
| "learning_rate": 6.98e-06, |
| "loss": 0.1257, |
| "mean_token_accuracy": 0.9642683621495962, |
| "num_tokens": 114352454.0, |
| "step": 6860 |
| }, |
| { |
| "entropy": 0.1957002142444253, |
| "epoch": 5.496, |
| "grad_norm": 4.46875, |
| "learning_rate": 6.9577777777777785e-06, |
| "loss": 0.1319, |
| "mean_token_accuracy": 0.962628673389554, |
| "num_tokens": 114514637.0, |
| "step": 6870 |
| }, |
| { |
| "entropy": 0.19551492873579263, |
| "epoch": 5.504, |
| "grad_norm": 4.25, |
| "learning_rate": 6.935555555555556e-06, |
| "loss": 0.1308, |
| "mean_token_accuracy": 0.9631638001650572, |
| "num_tokens": 114676433.0, |
| "step": 6880 |
| }, |
| { |
| "entropy": 0.19630830599926413, |
| "epoch": 5.5120000000000005, |
| "grad_norm": 4.28125, |
| "learning_rate": 6.913333333333334e-06, |
| "loss": 0.1298, |
| "mean_token_accuracy": 0.9623730078339576, |
| "num_tokens": 114847184.0, |
| "step": 6890 |
| }, |
| { |
| "entropy": 0.1939855834469199, |
| "epoch": 5.52, |
| "grad_norm": 4.375, |
| "learning_rate": 6.891111111111111e-06, |
| "loss": 0.1334, |
| "mean_token_accuracy": 0.9620220310986042, |
| "num_tokens": 115016974.0, |
| "step": 6900 |
| }, |
| { |
| "epoch": 5.52, |
| "eval_coding_entropy": 0.8589252479076386, |
| "eval_coding_loss": 3.589951276779175, |
| "eval_coding_mean_token_accuracy": 0.47215480303764346, |
| "eval_coding_num_tokens": 115016974.0, |
| "eval_coding_runtime": 79.1131, |
| "eval_coding_samples_per_second": 6.32, |
| "eval_coding_steps_per_second": 3.16, |
| "step": 6900 |
| }, |
| { |
| "epoch": 5.52, |
| "eval_physics_entropy": 0.3509551612734795, |
| "eval_physics_loss": 1.31234872341156, |
| "eval_physics_mean_token_accuracy": 0.7633123681545257, |
| "eval_physics_num_tokens": 115016974.0, |
| "eval_physics_runtime": 52.0606, |
| "eval_physics_samples_per_second": 9.604, |
| "eval_physics_steps_per_second": 4.802, |
| "step": 6900 |
| }, |
| { |
| "entropy": 0.19772629984654486, |
| "epoch": 5.5280000000000005, |
| "grad_norm": 4.09375, |
| "learning_rate": 6.868888888888889e-06, |
| "loss": 0.1299, |
| "mean_token_accuracy": 0.963387931510806, |
| "num_tokens": 115186270.0, |
| "step": 6910 |
| }, |
| { |
| "entropy": 0.19641432473436, |
| "epoch": 5.536, |
| "grad_norm": 4.03125, |
| "learning_rate": 6.846666666666667e-06, |
| "loss": 0.1278, |
| "mean_token_accuracy": 0.9629920769482851, |
| "num_tokens": 115349329.0, |
| "step": 6920 |
| }, |
| { |
| "entropy": 0.19159218021668495, |
| "epoch": 5.5440000000000005, |
| "grad_norm": 4.40625, |
| "learning_rate": 6.824444444444444e-06, |
| "loss": 0.1289, |
| "mean_token_accuracy": 0.9623739127069711, |
| "num_tokens": 115516501.0, |
| "step": 6930 |
| }, |
| { |
| "entropy": 0.18864571540616454, |
| "epoch": 5.552, |
| "grad_norm": 4.46875, |
| "learning_rate": 6.802222222222223e-06, |
| "loss": 0.1231, |
| "mean_token_accuracy": 0.9642074469476938, |
| "num_tokens": 115685578.0, |
| "step": 6940 |
| }, |
| { |
| "entropy": 0.19846407412551342, |
| "epoch": 5.5600000000000005, |
| "grad_norm": 4.40625, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 0.1308, |
| "mean_token_accuracy": 0.9620319828391075, |
| "num_tokens": 115846509.0, |
| "step": 6950 |
| }, |
| { |
| "entropy": 0.19245594800449908, |
| "epoch": 5.568, |
| "grad_norm": 4.1875, |
| "learning_rate": 6.757777777777779e-06, |
| "loss": 0.1286, |
| "mean_token_accuracy": 0.9636559452861547, |
| "num_tokens": 116015903.0, |
| "step": 6960 |
| }, |
| { |
| "entropy": 0.1946714937221259, |
| "epoch": 5.576, |
| "grad_norm": 4.09375, |
| "learning_rate": 6.735555555555556e-06, |
| "loss": 0.1267, |
| "mean_token_accuracy": 0.9637411668896675, |
| "num_tokens": 116183927.0, |
| "step": 6970 |
| }, |
| { |
| "entropy": 0.19161317558027804, |
| "epoch": 5.584, |
| "grad_norm": 3.65625, |
| "learning_rate": 6.713333333333334e-06, |
| "loss": 0.1282, |
| "mean_token_accuracy": 0.9632056415081024, |
| "num_tokens": 116353789.0, |
| "step": 6980 |
| }, |
| { |
| "entropy": 0.19859318835660816, |
| "epoch": 5.592, |
| "grad_norm": 3.890625, |
| "learning_rate": 6.691111111111112e-06, |
| "loss": 0.1321, |
| "mean_token_accuracy": 0.9618345819413662, |
| "num_tokens": 116523921.0, |
| "step": 6990 |
| }, |
| { |
| "entropy": 0.20267009483650328, |
| "epoch": 5.6, |
| "grad_norm": 4.34375, |
| "learning_rate": 6.668888888888889e-06, |
| "loss": 0.1336, |
| "mean_token_accuracy": 0.9616450056433677, |
| "num_tokens": 116691024.0, |
| "step": 7000 |
| }, |
| { |
| "epoch": 5.6, |
| "eval_coding_entropy": 0.8538971610069275, |
| "eval_coding_loss": 3.5922751426696777, |
| "eval_coding_mean_token_accuracy": 0.47259270763397215, |
| "eval_coding_num_tokens": 116691024.0, |
| "eval_coding_runtime": 79.1703, |
| "eval_coding_samples_per_second": 6.316, |
| "eval_coding_steps_per_second": 3.158, |
| "step": 7000 |
| }, |
| { |
| "epoch": 5.6, |
| "eval_physics_entropy": 0.3496944042444229, |
| "eval_physics_loss": 1.3129007816314697, |
| "eval_physics_mean_token_accuracy": 0.7638417301177979, |
| "eval_physics_num_tokens": 116691024.0, |
| "eval_physics_runtime": 51.8041, |
| "eval_physics_samples_per_second": 9.652, |
| "eval_physics_steps_per_second": 4.826, |
| "step": 7000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 8, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 7.32129390465612e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
