| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.4, |
| "eval_steps": 100, |
| "global_step": 6000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.4866574574261904, |
| "epoch": 0.004, |
| "grad_norm": 616.0, |
| "learning_rate": 1.8e-07, |
| "loss": 33.2311, |
| "mean_token_accuracy": 0.6604987328872085, |
| "num_tokens": 51831.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.4859781404957175, |
| "epoch": 0.008, |
| "grad_norm": 676.0, |
| "learning_rate": 3.8e-07, |
| "loss": 33.1415, |
| "mean_token_accuracy": 0.6665693156421184, |
| "num_tokens": 103172.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.5029121663421392, |
| "epoch": 0.012, |
| "grad_norm": 824.0, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 34.7849, |
| "mean_token_accuracy": 0.6541365889832378, |
| "num_tokens": 147915.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.5095210419967771, |
| "epoch": 0.016, |
| "grad_norm": 588.0, |
| "learning_rate": 7.8e-07, |
| "loss": 33.2642, |
| "mean_token_accuracy": 0.6610883103683591, |
| "num_tokens": 195524.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.5093817379325628, |
| "epoch": 0.02, |
| "grad_norm": 544.0, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 32.1157, |
| "mean_token_accuracy": 0.6604763394221663, |
| "num_tokens": 243747.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.5527876092121005, |
| "epoch": 0.024, |
| "grad_norm": 628.0, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 31.9125, |
| "mean_token_accuracy": 0.6578346157446504, |
| "num_tokens": 296999.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.5809885170310736, |
| "epoch": 0.028, |
| "grad_norm": 498.0, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 32.3494, |
| "mean_token_accuracy": 0.6499254325404763, |
| "num_tokens": 342114.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.6243463078513741, |
| "epoch": 0.032, |
| "grad_norm": 430.0, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 29.4405, |
| "mean_token_accuracy": 0.6568279277533293, |
| "num_tokens": 390903.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.7070688692852855, |
| "epoch": 0.036, |
| "grad_norm": 422.0, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 27.9455, |
| "mean_token_accuracy": 0.6594765370711684, |
| "num_tokens": 440359.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.7858394492417574, |
| "epoch": 0.04, |
| "grad_norm": 334.0, |
| "learning_rate": 1.98e-06, |
| "loss": 27.3289, |
| "mean_token_accuracy": 0.6568881323561072, |
| "num_tokens": 488481.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04, |
| "eval_biology_entropy": 0.8150426388680935, |
| "eval_biology_loss": 1.601250410079956, |
| "eval_biology_mean_token_accuracy": 0.6667410479187965, |
| "eval_biology_num_tokens": 488481.0, |
| "eval_biology_runtime": 34.4305, |
| "eval_biology_samples_per_second": 14.522, |
| "eval_biology_steps_per_second": 14.522, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.04, |
| "eval_chemistry_entropy": 0.42218632949888707, |
| "eval_chemistry_loss": 0.7587577700614929, |
| "eval_chemistry_mean_token_accuracy": 0.8110456941127777, |
| "eval_chemistry_num_tokens": 488481.0, |
| "eval_chemistry_runtime": 39.1702, |
| "eval_chemistry_samples_per_second": 12.765, |
| "eval_chemistry_steps_per_second": 12.765, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.8809404099360109, |
| "epoch": 0.044, |
| "grad_norm": 246.0, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 25.34, |
| "mean_token_accuracy": 0.6623364068567753, |
| "num_tokens": 537471.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.9853538589552044, |
| "epoch": 0.048, |
| "grad_norm": 228.0, |
| "learning_rate": 2.38e-06, |
| "loss": 23.768, |
| "mean_token_accuracy": 0.6667981075122953, |
| "num_tokens": 589868.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 1.1170461051166058, |
| "epoch": 0.052, |
| "grad_norm": 171.0, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 23.5554, |
| "mean_token_accuracy": 0.6613909218460321, |
| "num_tokens": 639512.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 1.1676542518660427, |
| "epoch": 0.056, |
| "grad_norm": 151.0, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 22.6768, |
| "mean_token_accuracy": 0.6657767463475466, |
| "num_tokens": 690754.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 1.2725389588624239, |
| "epoch": 0.06, |
| "grad_norm": 132.0, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 23.1722, |
| "mean_token_accuracy": 0.6627637758851052, |
| "num_tokens": 737075.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 1.258068286627531, |
| "epoch": 0.064, |
| "grad_norm": 142.0, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 21.9045, |
| "mean_token_accuracy": 0.6725248593837023, |
| "num_tokens": 791190.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.306521786376834, |
| "epoch": 0.068, |
| "grad_norm": 112.0, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 22.1444, |
| "mean_token_accuracy": 0.6677160531282424, |
| "num_tokens": 841344.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 1.2990117866545916, |
| "epoch": 0.072, |
| "grad_norm": 141.0, |
| "learning_rate": 3.58e-06, |
| "loss": 22.1479, |
| "mean_token_accuracy": 0.6692137856036424, |
| "num_tokens": 890075.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.2258288636803627, |
| "epoch": 0.076, |
| "grad_norm": 99.0, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 20.8997, |
| "mean_token_accuracy": 0.6792997987940907, |
| "num_tokens": 940599.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.2281328312121331, |
| "epoch": 0.08, |
| "grad_norm": 123.0, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 20.9624, |
| "mean_token_accuracy": 0.6809809722006321, |
| "num_tokens": 989860.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_biology_entropy": 1.2447693885564803, |
| "eval_biology_loss": 1.31557035446167, |
| "eval_biology_mean_token_accuracy": 0.6810677705407142, |
| "eval_biology_num_tokens": 989860.0, |
| "eval_biology_runtime": 34.2892, |
| "eval_biology_samples_per_second": 14.582, |
| "eval_biology_steps_per_second": 14.582, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_chemistry_entropy": 0.6061330872625112, |
| "eval_chemistry_loss": 0.6521626710891724, |
| "eval_chemistry_mean_token_accuracy": 0.8165175079107284, |
| "eval_chemistry_num_tokens": 989860.0, |
| "eval_chemistry_runtime": 39.0885, |
| "eval_chemistry_samples_per_second": 12.791, |
| "eval_chemistry_steps_per_second": 12.791, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.2204063970595598, |
| "epoch": 0.084, |
| "grad_norm": 99.5, |
| "learning_rate": 4.18e-06, |
| "loss": 20.4735, |
| "mean_token_accuracy": 0.688131432235241, |
| "num_tokens": 1041199.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.3349104586988687, |
| "epoch": 0.088, |
| "grad_norm": 111.5, |
| "learning_rate": 4.38e-06, |
| "loss": 22.5142, |
| "mean_token_accuracy": 0.6627503650262951, |
| "num_tokens": 1087099.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.2557024333626032, |
| "epoch": 0.092, |
| "grad_norm": 96.5, |
| "learning_rate": 4.58e-06, |
| "loss": 20.9139, |
| "mean_token_accuracy": 0.6778965264558792, |
| "num_tokens": 1135090.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.280203009955585, |
| "epoch": 0.096, |
| "grad_norm": 105.0, |
| "learning_rate": 4.78e-06, |
| "loss": 21.3579, |
| "mean_token_accuracy": 0.6716440364718437, |
| "num_tokens": 1181777.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.276487385854125, |
| "epoch": 0.1, |
| "grad_norm": 100.5, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 20.8968, |
| "mean_token_accuracy": 0.6772379480302334, |
| "num_tokens": 1229242.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.2585501682013274, |
| "epoch": 0.104, |
| "grad_norm": 85.0, |
| "learning_rate": 5.18e-06, |
| "loss": 20.7316, |
| "mean_token_accuracy": 0.679420518502593, |
| "num_tokens": 1283881.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.248391578346491, |
| "epoch": 0.108, |
| "grad_norm": 91.0, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 20.4527, |
| "mean_token_accuracy": 0.682051457837224, |
| "num_tokens": 1332507.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.2360890079289675, |
| "epoch": 0.112, |
| "grad_norm": 84.5, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 20.2916, |
| "mean_token_accuracy": 0.6859552793204784, |
| "num_tokens": 1385513.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.2920903146266938, |
| "epoch": 0.116, |
| "grad_norm": 90.5, |
| "learning_rate": 5.78e-06, |
| "loss": 20.9268, |
| "mean_token_accuracy": 0.673469840362668, |
| "num_tokens": 1432251.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.2169338904321194, |
| "epoch": 0.12, |
| "grad_norm": 87.0, |
| "learning_rate": 5.98e-06, |
| "loss": 19.6752, |
| "mean_token_accuracy": 0.6875041898339986, |
| "num_tokens": 1483667.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.12, |
| "eval_biology_entropy": 1.2245318589806558, |
| "eval_biology_loss": 1.2560091018676758, |
| "eval_biology_mean_token_accuracy": 0.6886654210686683, |
| "eval_biology_num_tokens": 1483667.0, |
| "eval_biology_runtime": 34.2037, |
| "eval_biology_samples_per_second": 14.618, |
| "eval_biology_steps_per_second": 14.618, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.12, |
| "eval_chemistry_entropy": 0.6325840967297554, |
| "eval_chemistry_loss": 0.6409750580787659, |
| "eval_chemistry_mean_token_accuracy": 0.8179453414678574, |
| "eval_chemistry_num_tokens": 1483667.0, |
| "eval_chemistry_runtime": 39.0169, |
| "eval_chemistry_samples_per_second": 12.815, |
| "eval_chemistry_steps_per_second": 12.815, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.2429257260635496, |
| "epoch": 0.124, |
| "grad_norm": 79.5, |
| "learning_rate": 6.18e-06, |
| "loss": 20.6172, |
| "mean_token_accuracy": 0.6801770402118563, |
| "num_tokens": 1534286.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.2787830816581844, |
| "epoch": 0.128, |
| "grad_norm": 84.5, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 20.5543, |
| "mean_token_accuracy": 0.6807437939569354, |
| "num_tokens": 1582368.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.2588950738310813, |
| "epoch": 0.132, |
| "grad_norm": 89.0, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 20.3303, |
| "mean_token_accuracy": 0.6814694199711084, |
| "num_tokens": 1630878.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.2704457983374595, |
| "epoch": 0.136, |
| "grad_norm": 83.0, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 20.6418, |
| "mean_token_accuracy": 0.6803831990808249, |
| "num_tokens": 1677060.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.2321893341839314, |
| "epoch": 0.14, |
| "grad_norm": 87.5, |
| "learning_rate": 6.98e-06, |
| "loss": 19.8202, |
| "mean_token_accuracy": 0.6891451958566904, |
| "num_tokens": 1725186.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.2445756699889898, |
| "epoch": 0.144, |
| "grad_norm": 92.0, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 20.2527, |
| "mean_token_accuracy": 0.6785595057532191, |
| "num_tokens": 1773764.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.246081338264048, |
| "epoch": 0.148, |
| "grad_norm": 82.0, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 20.08, |
| "mean_token_accuracy": 0.6857875902205706, |
| "num_tokens": 1823706.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.2665387976914644, |
| "epoch": 0.152, |
| "grad_norm": 95.0, |
| "learning_rate": 7.58e-06, |
| "loss": 20.6091, |
| "mean_token_accuracy": 0.6813197182491422, |
| "num_tokens": 1873167.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.2548780847340821, |
| "epoch": 0.156, |
| "grad_norm": 85.0, |
| "learning_rate": 7.78e-06, |
| "loss": 20.406, |
| "mean_token_accuracy": 0.6822501845657826, |
| "num_tokens": 1921484.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.2407029550522566, |
| "epoch": 0.16, |
| "grad_norm": 89.0, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 19.8013, |
| "mean_token_accuracy": 0.6853796949610114, |
| "num_tokens": 1970077.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_biology_entropy": 1.214753457427025, |
| "eval_biology_loss": 1.2285655736923218, |
| "eval_biology_mean_token_accuracy": 0.6923919150233269, |
| "eval_biology_num_tokens": 1970077.0, |
| "eval_biology_runtime": 34.3823, |
| "eval_biology_samples_per_second": 14.542, |
| "eval_biology_steps_per_second": 14.542, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_chemistry_entropy": 0.6530047204196453, |
| "eval_chemistry_loss": 0.644757866859436, |
| "eval_chemistry_mean_token_accuracy": 0.8174828389883041, |
| "eval_chemistry_num_tokens": 1970077.0, |
| "eval_chemistry_runtime": 39.0828, |
| "eval_chemistry_samples_per_second": 12.793, |
| "eval_chemistry_steps_per_second": 12.793, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.2693706121295691, |
| "epoch": 0.164, |
| "grad_norm": 75.0, |
| "learning_rate": 8.18e-06, |
| "loss": 20.5934, |
| "mean_token_accuracy": 0.6737196171656251, |
| "num_tokens": 2019053.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.2044812317937612, |
| "epoch": 0.168, |
| "grad_norm": 72.5, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 19.2447, |
| "mean_token_accuracy": 0.6942890204489232, |
| "num_tokens": 2066815.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.219953553378582, |
| "epoch": 0.172, |
| "grad_norm": 76.5, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 20.0236, |
| "mean_token_accuracy": 0.6850357944145798, |
| "num_tokens": 2116402.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.1823513347655534, |
| "epoch": 0.176, |
| "grad_norm": 74.0, |
| "learning_rate": 8.78e-06, |
| "loss": 19.1286, |
| "mean_token_accuracy": 0.6948613569140434, |
| "num_tokens": 2168354.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.2351123623549938, |
| "epoch": 0.18, |
| "grad_norm": 75.0, |
| "learning_rate": 8.98e-06, |
| "loss": 19.847, |
| "mean_token_accuracy": 0.6850900270044804, |
| "num_tokens": 2219158.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.2182675819844007, |
| "epoch": 0.184, |
| "grad_norm": 83.0, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 19.9054, |
| "mean_token_accuracy": 0.6874489855021239, |
| "num_tokens": 2270617.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.2002981722354888, |
| "epoch": 0.188, |
| "grad_norm": 74.5, |
| "learning_rate": 9.38e-06, |
| "loss": 18.9485, |
| "mean_token_accuracy": 0.6958082247525453, |
| "num_tokens": 2322034.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.2613376142457127, |
| "epoch": 0.192, |
| "grad_norm": 98.5, |
| "learning_rate": 9.58e-06, |
| "loss": 20.7735, |
| "mean_token_accuracy": 0.6739435123279691, |
| "num_tokens": 2365822.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.205817231722176, |
| "epoch": 0.196, |
| "grad_norm": 89.5, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 19.1434, |
| "mean_token_accuracy": 0.6935921791940928, |
| "num_tokens": 2413853.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.2515719644725323, |
| "epoch": 0.2, |
| "grad_norm": 90.5, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 20.4938, |
| "mean_token_accuracy": 0.6777064654976129, |
| "num_tokens": 2459876.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_biology_entropy": 1.1872086516022682, |
| "eval_biology_loss": 1.2085970640182495, |
| "eval_biology_mean_token_accuracy": 0.6949169620871544, |
| "eval_biology_num_tokens": 2459876.0, |
| "eval_biology_runtime": 34.4, |
| "eval_biology_samples_per_second": 14.535, |
| "eval_biology_steps_per_second": 14.535, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_chemistry_entropy": 0.6488757311999798, |
| "eval_chemistry_loss": 0.6450773477554321, |
| "eval_chemistry_mean_token_accuracy": 0.8178821201324463, |
| "eval_chemistry_num_tokens": 2459876.0, |
| "eval_chemistry_runtime": 39.2328, |
| "eval_chemistry_samples_per_second": 12.744, |
| "eval_chemistry_steps_per_second": 12.744, |
| "step": 500 |
| }, |
| { |
| "entropy": 1.2209952611476182, |
| "epoch": 0.204, |
| "grad_norm": 73.0, |
| "learning_rate": 1.018e-05, |
| "loss": 19.6626, |
| "mean_token_accuracy": 0.6881283605471253, |
| "num_tokens": 2509187.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 1.2497611749917268, |
| "epoch": 0.208, |
| "grad_norm": 85.5, |
| "learning_rate": 1.038e-05, |
| "loss": 19.9518, |
| "mean_token_accuracy": 0.6810324152931571, |
| "num_tokens": 2558762.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 1.1757056951522826, |
| "epoch": 0.212, |
| "grad_norm": 78.0, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 19.3754, |
| "mean_token_accuracy": 0.6935973027721047, |
| "num_tokens": 2608616.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 1.2561812430620194, |
| "epoch": 0.216, |
| "grad_norm": 83.0, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 20.0513, |
| "mean_token_accuracy": 0.6874226013198494, |
| "num_tokens": 2654953.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 1.2020427081733942, |
| "epoch": 0.22, |
| "grad_norm": 78.5, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 19.5412, |
| "mean_token_accuracy": 0.688438569009304, |
| "num_tokens": 2704686.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 1.205025901272893, |
| "epoch": 0.224, |
| "grad_norm": 86.5, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 19.7851, |
| "mean_token_accuracy": 0.6885192114859819, |
| "num_tokens": 2755347.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 1.2252822391688825, |
| "epoch": 0.228, |
| "grad_norm": 74.5, |
| "learning_rate": 1.138e-05, |
| "loss": 19.3164, |
| "mean_token_accuracy": 0.6909766353666782, |
| "num_tokens": 2804340.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 1.1826427452266217, |
| "epoch": 0.232, |
| "grad_norm": 80.0, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 19.3402, |
| "mean_token_accuracy": 0.6917916279286146, |
| "num_tokens": 2852298.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 1.1695974227041006, |
| "epoch": 0.236, |
| "grad_norm": 89.0, |
| "learning_rate": 1.178e-05, |
| "loss": 19.0702, |
| "mean_token_accuracy": 0.6989134319126606, |
| "num_tokens": 2901227.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 1.2379697296768426, |
| "epoch": 0.24, |
| "grad_norm": 92.5, |
| "learning_rate": 1.198e-05, |
| "loss": 19.6279, |
| "mean_token_accuracy": 0.6916081016883254, |
| "num_tokens": 2947346.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_biology_entropy": 1.1614825129508972, |
| "eval_biology_loss": 1.197817087173462, |
| "eval_biology_mean_token_accuracy": 0.6965834299921989, |
| "eval_biology_num_tokens": 2947346.0, |
| "eval_biology_runtime": 34.3297, |
| "eval_biology_samples_per_second": 14.565, |
| "eval_biology_steps_per_second": 14.565, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_chemistry_entropy": 0.6421602384448052, |
| "eval_chemistry_loss": 0.6457399725914001, |
| "eval_chemistry_mean_token_accuracy": 0.8177337770462036, |
| "eval_chemistry_num_tokens": 2947346.0, |
| "eval_chemistry_runtime": 39.0493, |
| "eval_chemistry_samples_per_second": 12.804, |
| "eval_chemistry_steps_per_second": 12.804, |
| "step": 600 |
| }, |
| { |
| "entropy": 1.2068689458072186, |
| "epoch": 0.244, |
| "grad_norm": 83.0, |
| "learning_rate": 1.218e-05, |
| "loss": 19.9725, |
| "mean_token_accuracy": 0.6830457381904125, |
| "num_tokens": 2996175.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 1.2800738830119371, |
| "epoch": 0.248, |
| "grad_norm": 77.5, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 20.1671, |
| "mean_token_accuracy": 0.6834735475480557, |
| "num_tokens": 3043128.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 1.190559478290379, |
| "epoch": 0.252, |
| "grad_norm": 79.5, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 19.3606, |
| "mean_token_accuracy": 0.6899204226210713, |
| "num_tokens": 3090827.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 1.1722685299813747, |
| "epoch": 0.256, |
| "grad_norm": 81.5, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 18.8652, |
| "mean_token_accuracy": 0.6963184822350741, |
| "num_tokens": 3139957.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 1.2144603542983532, |
| "epoch": 0.26, |
| "grad_norm": 82.0, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 19.7174, |
| "mean_token_accuracy": 0.685292916931212, |
| "num_tokens": 3183188.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 1.2338291574269533, |
| "epoch": 0.264, |
| "grad_norm": 80.0, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 19.7757, |
| "mean_token_accuracy": 0.6882328724488616, |
| "num_tokens": 3232159.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 1.1896044220775366, |
| "epoch": 0.268, |
| "grad_norm": 79.5, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 19.1149, |
| "mean_token_accuracy": 0.6937362670898437, |
| "num_tokens": 3283908.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 1.1706442803144455, |
| "epoch": 0.272, |
| "grad_norm": 98.0, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 19.1832, |
| "mean_token_accuracy": 0.6947279252111912, |
| "num_tokens": 3335951.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 1.2103422859683632, |
| "epoch": 0.276, |
| "grad_norm": 88.0, |
| "learning_rate": 1.378e-05, |
| "loss": 19.5378, |
| "mean_token_accuracy": 0.6888058576732874, |
| "num_tokens": 3387755.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 1.1900839667767287, |
| "epoch": 0.28, |
| "grad_norm": 90.0, |
| "learning_rate": 1.398e-05, |
| "loss": 19.3154, |
| "mean_token_accuracy": 0.6907492563128471, |
| "num_tokens": 3437213.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.28, |
| "eval_biology_entropy": 1.1723266056776047, |
| "eval_biology_loss": 1.1935571432113647, |
| "eval_biology_mean_token_accuracy": 0.6974302505254746, |
| "eval_biology_num_tokens": 3437213.0, |
| "eval_biology_runtime": 34.4282, |
| "eval_biology_samples_per_second": 14.523, |
| "eval_biology_steps_per_second": 14.523, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.28, |
| "eval_chemistry_entropy": 0.6567539082467556, |
| "eval_chemistry_loss": 0.6476752758026123, |
| "eval_chemistry_mean_token_accuracy": 0.8169361530542374, |
| "eval_chemistry_num_tokens": 3437213.0, |
| "eval_chemistry_runtime": 39.1281, |
| "eval_chemistry_samples_per_second": 12.779, |
| "eval_chemistry_steps_per_second": 12.779, |
| "step": 700 |
| }, |
| { |
| "entropy": 1.1601122960448265, |
| "epoch": 0.284, |
| "grad_norm": 70.0, |
| "learning_rate": 1.418e-05, |
| "loss": 18.4856, |
| "mean_token_accuracy": 0.7002454694360495, |
| "num_tokens": 3489088.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 1.1836338724941016, |
| "epoch": 0.288, |
| "grad_norm": 74.0, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 19.4318, |
| "mean_token_accuracy": 0.6893038578331471, |
| "num_tokens": 3539731.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 1.215116088837385, |
| "epoch": 0.292, |
| "grad_norm": 76.5, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 19.2441, |
| "mean_token_accuracy": 0.6934854052960873, |
| "num_tokens": 3586433.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 1.2150515001267195, |
| "epoch": 0.296, |
| "grad_norm": 87.5, |
| "learning_rate": 1.478e-05, |
| "loss": 20.0166, |
| "mean_token_accuracy": 0.6854536816477775, |
| "num_tokens": 3633448.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 1.1540303017944098, |
| "epoch": 0.3, |
| "grad_norm": 77.5, |
| "learning_rate": 1.498e-05, |
| "loss": 18.4485, |
| "mean_token_accuracy": 0.705385484918952, |
| "num_tokens": 3685080.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 1.1504614014178514, |
| "epoch": 0.304, |
| "grad_norm": 79.0, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 18.9317, |
| "mean_token_accuracy": 0.6943564370274544, |
| "num_tokens": 3733488.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 1.227951274625957, |
| "epoch": 0.308, |
| "grad_norm": 81.0, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 19.6245, |
| "mean_token_accuracy": 0.69070417098701, |
| "num_tokens": 3779708.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 1.1914822444319726, |
| "epoch": 0.312, |
| "grad_norm": 91.5, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 19.3053, |
| "mean_token_accuracy": 0.691599304229021, |
| "num_tokens": 3825212.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 1.2374459221959113, |
| "epoch": 0.316, |
| "grad_norm": 90.5, |
| "learning_rate": 1.578e-05, |
| "loss": 20.1615, |
| "mean_token_accuracy": 0.6793428828939796, |
| "num_tokens": 3870912.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 1.204637792892754, |
| "epoch": 0.32, |
| "grad_norm": 87.0, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 19.1533, |
| "mean_token_accuracy": 0.6921680744737386, |
| "num_tokens": 3920545.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 1.1595962298512459, |
| "eval_biology_loss": 1.1908730268478394, |
| "eval_biology_mean_token_accuracy": 0.6982393438220024, |
| "eval_biology_num_tokens": 3920545.0, |
| "eval_biology_runtime": 34.3027, |
| "eval_biology_samples_per_second": 14.576, |
| "eval_biology_steps_per_second": 14.576, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_chemistry_entropy": 0.6581202555298805, |
| "eval_chemistry_loss": 0.6543570756912231, |
| "eval_chemistry_mean_token_accuracy": 0.8150564895868302, |
| "eval_chemistry_num_tokens": 3920545.0, |
| "eval_chemistry_runtime": 39.1221, |
| "eval_chemistry_samples_per_second": 12.781, |
| "eval_chemistry_steps_per_second": 12.781, |
| "step": 800 |
| }, |
| { |
| "entropy": 1.1894057728350163, |
| "epoch": 0.324, |
| "grad_norm": 82.0, |
| "learning_rate": 1.618e-05, |
| "loss": 19.3278, |
| "mean_token_accuracy": 0.6910803083330392, |
| "num_tokens": 3967499.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 1.1689968170598148, |
| "epoch": 0.328, |
| "grad_norm": 75.5, |
| "learning_rate": 1.638e-05, |
| "loss": 18.6792, |
| "mean_token_accuracy": 0.6996502809226512, |
| "num_tokens": 4016673.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 1.1961608117446303, |
| "epoch": 0.332, |
| "grad_norm": 88.5, |
| "learning_rate": 1.658e-05, |
| "loss": 19.6175, |
| "mean_token_accuracy": 0.6912782397121191, |
| "num_tokens": 4063859.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 1.2022534670308231, |
| "epoch": 0.336, |
| "grad_norm": 75.5, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 19.3189, |
| "mean_token_accuracy": 0.6922560017555952, |
| "num_tokens": 4114077.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 1.2139138951897621, |
| "epoch": 0.34, |
| "grad_norm": 80.0, |
| "learning_rate": 1.698e-05, |
| "loss": 19.6407, |
| "mean_token_accuracy": 0.6891672356054187, |
| "num_tokens": 4160168.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 1.1922080295160413, |
| "epoch": 0.344, |
| "grad_norm": 76.5, |
| "learning_rate": 1.718e-05, |
| "loss": 19.3527, |
| "mean_token_accuracy": 0.6934164728969335, |
| "num_tokens": 4210403.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 1.177280587144196, |
| "epoch": 0.348, |
| "grad_norm": 76.0, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 18.7411, |
| "mean_token_accuracy": 0.6976979669183493, |
| "num_tokens": 4256944.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 1.2034492146223783, |
| "epoch": 0.352, |
| "grad_norm": 77.5, |
| "learning_rate": 1.758e-05, |
| "loss": 19.6262, |
| "mean_token_accuracy": 0.6897061046212911, |
| "num_tokens": 4306949.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 1.2024015821516514, |
| "epoch": 0.356, |
| "grad_norm": 70.5, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 19.4871, |
| "mean_token_accuracy": 0.6887454513460398, |
| "num_tokens": 4355070.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 1.1949962507933378, |
| "epoch": 0.36, |
| "grad_norm": 65.0, |
| "learning_rate": 1.798e-05, |
| "loss": 19.252, |
| "mean_token_accuracy": 0.6892035495489836, |
| "num_tokens": 4406104.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_biology_entropy": 1.149014304101467, |
| "eval_biology_loss": 1.191479206085205, |
| "eval_biology_mean_token_accuracy": 0.6979464264512062, |
| "eval_biology_num_tokens": 4406104.0, |
| "eval_biology_runtime": 34.5339, |
| "eval_biology_samples_per_second": 14.479, |
| "eval_biology_steps_per_second": 14.479, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_chemistry_entropy": 0.6591544916331769, |
| "eval_chemistry_loss": 0.6548386812210083, |
| "eval_chemistry_mean_token_accuracy": 0.8150800436735153, |
| "eval_chemistry_num_tokens": 4406104.0, |
| "eval_chemistry_runtime": 39.1161, |
| "eval_chemistry_samples_per_second": 12.782, |
| "eval_chemistry_steps_per_second": 12.782, |
| "step": 900 |
| }, |
| { |
| "entropy": 1.210719463415444, |
| "epoch": 0.364, |
| "grad_norm": 77.5, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 19.6229, |
| "mean_token_accuracy": 0.6920540288090706, |
| "num_tokens": 4455015.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 1.1651359049603343, |
| "epoch": 0.368, |
| "grad_norm": 85.0, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 18.965, |
| "mean_token_accuracy": 0.6945374647155405, |
| "num_tokens": 4504001.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 1.2309420805424451, |
| "epoch": 0.372, |
| "grad_norm": 79.0, |
| "learning_rate": 1.858e-05, |
| "loss": 19.4645, |
| "mean_token_accuracy": 0.6926089435815811, |
| "num_tokens": 4550269.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 1.1543804241344333, |
| "epoch": 0.376, |
| "grad_norm": 83.5, |
| "learning_rate": 1.878e-05, |
| "loss": 18.9949, |
| "mean_token_accuracy": 0.695966462418437, |
| "num_tokens": 4597559.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 1.1860109277069568, |
| "epoch": 0.38, |
| "grad_norm": 88.0, |
| "learning_rate": 1.898e-05, |
| "loss": 19.0122, |
| "mean_token_accuracy": 0.6945275101810694, |
| "num_tokens": 4646903.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 1.1854952551424502, |
| "epoch": 0.384, |
| "grad_norm": 72.0, |
| "learning_rate": 1.918e-05, |
| "loss": 19.3612, |
| "mean_token_accuracy": 0.6901288989931345, |
| "num_tokens": 4693812.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 1.2153874900192023, |
| "epoch": 0.388, |
| "grad_norm": 76.0, |
| "learning_rate": 1.938e-05, |
| "loss": 19.4325, |
| "mean_token_accuracy": 0.6911413293331862, |
| "num_tokens": 4740309.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 1.212491791881621, |
| "epoch": 0.392, |
| "grad_norm": 74.0, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 19.43, |
| "mean_token_accuracy": 0.6912798311561346, |
| "num_tokens": 4790965.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 1.1938600786030293, |
| "epoch": 0.396, |
| "grad_norm": 77.0, |
| "learning_rate": 1.978e-05, |
| "loss": 19.4531, |
| "mean_token_accuracy": 0.6905827283859253, |
| "num_tokens": 4839186.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 1.2035906264558434, |
| "epoch": 0.4, |
| "grad_norm": 82.5, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 19.4249, |
| "mean_token_accuracy": 0.6903344210237264, |
| "num_tokens": 4887094.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_biology_entropy": 1.1641007453799248, |
| "eval_biology_loss": 1.1950935125350952, |
| "eval_biology_mean_token_accuracy": 0.6969818140268326, |
| "eval_biology_num_tokens": 4887094.0, |
| "eval_biology_runtime": 34.3544, |
| "eval_biology_samples_per_second": 14.554, |
| "eval_biology_steps_per_second": 14.554, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_chemistry_entropy": 0.6638510838449001, |
| "eval_chemistry_loss": 0.662416934967041, |
| "eval_chemistry_mean_token_accuracy": 0.8127187831401825, |
| "eval_chemistry_num_tokens": 4887094.0, |
| "eval_chemistry_runtime": 38.9213, |
| "eval_chemistry_samples_per_second": 12.846, |
| "eval_chemistry_steps_per_second": 12.846, |
| "step": 1000 |
| }, |
| { |
| "entropy": 1.2332831894978882, |
| "epoch": 0.404, |
| "grad_norm": 88.5, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 19.9936, |
| "mean_token_accuracy": 0.6858718886971473, |
| "num_tokens": 4932228.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 1.1915043991059064, |
| "epoch": 0.408, |
| "grad_norm": 64.5, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 19.108, |
| "mean_token_accuracy": 0.6954168599098921, |
| "num_tokens": 4985394.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 1.2241652118042112, |
| "epoch": 0.412, |
| "grad_norm": 65.0, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 19.765, |
| "mean_token_accuracy": 0.6827985186129808, |
| "num_tokens": 5037274.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 1.207521514967084, |
| "epoch": 0.416, |
| "grad_norm": 65.0, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 19.6241, |
| "mean_token_accuracy": 0.6886004239320755, |
| "num_tokens": 5085369.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 1.2827957067638636, |
| "epoch": 0.42, |
| "grad_norm": 90.0, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 20.5645, |
| "mean_token_accuracy": 0.6782789714634418, |
| "num_tokens": 5129587.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 1.2159279894083739, |
| "epoch": 0.424, |
| "grad_norm": 70.5, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 19.7117, |
| "mean_token_accuracy": 0.6862207680940628, |
| "num_tokens": 5176083.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 1.1846632236614822, |
| "epoch": 0.428, |
| "grad_norm": 83.5, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 19.1506, |
| "mean_token_accuracy": 0.6937350874766708, |
| "num_tokens": 5225296.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 1.2120982160791756, |
| "epoch": 0.432, |
| "grad_norm": 81.0, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 19.3732, |
| "mean_token_accuracy": 0.6929085306823254, |
| "num_tokens": 5271275.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 1.216021078079939, |
| "epoch": 0.436, |
| "grad_norm": 74.0, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 19.6442, |
| "mean_token_accuracy": 0.6861835535615682, |
| "num_tokens": 5318781.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 1.2075613098219038, |
| "epoch": 0.44, |
| "grad_norm": 74.5, |
| "learning_rate": 1.978e-05, |
| "loss": 19.5288, |
| "mean_token_accuracy": 0.6872704658657313, |
| "num_tokens": 5362624.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.44, |
| "eval_biology_entropy": 1.2076813784837723, |
| "eval_biology_loss": 1.1966326236724854, |
| "eval_biology_mean_token_accuracy": 0.6968241932392121, |
| "eval_biology_num_tokens": 5362624.0, |
| "eval_biology_runtime": 34.4319, |
| "eval_biology_samples_per_second": 14.521, |
| "eval_biology_steps_per_second": 14.521, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.44, |
| "eval_chemistry_entropy": 0.6861711817383767, |
| "eval_chemistry_loss": 0.6706647872924805, |
| "eval_chemistry_mean_token_accuracy": 0.8111056753396988, |
| "eval_chemistry_num_tokens": 5362624.0, |
| "eval_chemistry_runtime": 39.1421, |
| "eval_chemistry_samples_per_second": 12.774, |
| "eval_chemistry_steps_per_second": 12.774, |
| "step": 1100 |
| }, |
| { |
| "entropy": 1.2112580770626664, |
| "epoch": 0.444, |
| "grad_norm": 70.5, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 19.3826, |
| "mean_token_accuracy": 0.6880253683775663, |
| "num_tokens": 5414444.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 1.202802482806146, |
| "epoch": 0.448, |
| "grad_norm": 73.5, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 19.4755, |
| "mean_token_accuracy": 0.6911316126585006, |
| "num_tokens": 5460559.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 1.1450790271162987, |
| "epoch": 0.452, |
| "grad_norm": 72.5, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 18.4276, |
| "mean_token_accuracy": 0.7056065019220114, |
| "num_tokens": 5510112.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 1.2133123382925988, |
| "epoch": 0.456, |
| "grad_norm": 71.5, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 19.3288, |
| "mean_token_accuracy": 0.6915038814768195, |
| "num_tokens": 5558471.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 1.1987600516527892, |
| "epoch": 0.46, |
| "grad_norm": 73.5, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 19.6063, |
| "mean_token_accuracy": 0.684781968779862, |
| "num_tokens": 5605237.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 1.2029788015410303, |
| "epoch": 0.464, |
| "grad_norm": 77.0, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 19.113, |
| "mean_token_accuracy": 0.6943607520312071, |
| "num_tokens": 5653809.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 1.1871271941810846, |
| "epoch": 0.468, |
| "grad_norm": 80.0, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 19.4077, |
| "mean_token_accuracy": 0.6932028122246265, |
| "num_tokens": 5699727.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 1.2316160146147013, |
| "epoch": 0.472, |
| "grad_norm": 67.0, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 19.4937, |
| "mean_token_accuracy": 0.6891589108854532, |
| "num_tokens": 5751529.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 1.1813869891688227, |
| "epoch": 0.476, |
| "grad_norm": 82.5, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 19.368, |
| "mean_token_accuracy": 0.6905937325209379, |
| "num_tokens": 5801167.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 1.2543733013793825, |
| "epoch": 0.48, |
| "grad_norm": 65.0, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 19.9054, |
| "mean_token_accuracy": 0.6872919265180826, |
| "num_tokens": 5850176.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_biology_entropy": 1.2115116498470306, |
| "eval_biology_loss": 1.1886382102966309, |
| "eval_biology_mean_token_accuracy": 0.6968740136623383, |
| "eval_biology_num_tokens": 5850176.0, |
| "eval_biology_runtime": 34.2937, |
| "eval_biology_samples_per_second": 14.58, |
| "eval_biology_steps_per_second": 14.58, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_chemistry_entropy": 0.7031495725512504, |
| "eval_chemistry_loss": 0.6727164387702942, |
| "eval_chemistry_mean_token_accuracy": 0.8113047035932541, |
| "eval_chemistry_num_tokens": 5850176.0, |
| "eval_chemistry_runtime": 39.0431, |
| "eval_chemistry_samples_per_second": 12.806, |
| "eval_chemistry_steps_per_second": 12.806, |
| "step": 1200 |
| }, |
| { |
| "entropy": 1.135433156043291, |
| "epoch": 0.484, |
| "grad_norm": 80.0, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 18.5087, |
| "mean_token_accuracy": 0.699444928765297, |
| "num_tokens": 5897377.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 1.1549700409173966, |
| "epoch": 0.488, |
| "grad_norm": 84.5, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 18.4892, |
| "mean_token_accuracy": 0.701468913629651, |
| "num_tokens": 5948898.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 1.154023634083569, |
| "epoch": 0.492, |
| "grad_norm": 80.5, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 18.485, |
| "mean_token_accuracy": 0.7046966258436441, |
| "num_tokens": 5997563.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 1.167623796314001, |
| "epoch": 0.496, |
| "grad_norm": 81.0, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 18.7743, |
| "mean_token_accuracy": 0.6952338460832834, |
| "num_tokens": 6046503.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 1.2217689264565705, |
| "epoch": 0.5, |
| "grad_norm": 62.75, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 19.5256, |
| "mean_token_accuracy": 0.6905383057892323, |
| "num_tokens": 6091587.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 1.1947737470269204, |
| "epoch": 0.504, |
| "grad_norm": 81.0, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 19.5094, |
| "mean_token_accuracy": 0.6902488011866807, |
| "num_tokens": 6138649.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 1.1597765712067485, |
| "epoch": 0.508, |
| "grad_norm": 64.0, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 18.4208, |
| "mean_token_accuracy": 0.7028776211664081, |
| "num_tokens": 6189122.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 1.2139740228652953, |
| "epoch": 0.512, |
| "grad_norm": 68.0, |
| "learning_rate": 1.938e-05, |
| "loss": 19.5968, |
| "mean_token_accuracy": 0.6862009439617396, |
| "num_tokens": 6240456.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 1.214512193016708, |
| "epoch": 0.516, |
| "grad_norm": 73.5, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 19.5979, |
| "mean_token_accuracy": 0.6916845444589853, |
| "num_tokens": 6286445.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 1.177449732273817, |
| "epoch": 0.52, |
| "grad_norm": 72.5, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 19.0899, |
| "mean_token_accuracy": 0.6944207726046443, |
| "num_tokens": 6333958.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.52, |
| "eval_biology_entropy": 1.1935186836719514, |
| "eval_biology_loss": 1.1775966882705688, |
| "eval_biology_mean_token_accuracy": 0.6992735857367516, |
| "eval_biology_num_tokens": 6333958.0, |
| "eval_biology_runtime": 34.2425, |
| "eval_biology_samples_per_second": 14.602, |
| "eval_biology_steps_per_second": 14.602, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.52, |
| "eval_chemistry_entropy": 0.6942039486765862, |
| "eval_chemistry_loss": 0.6730213165283203, |
| "eval_chemistry_mean_token_accuracy": 0.8110301185846329, |
| "eval_chemistry_num_tokens": 6333958.0, |
| "eval_chemistry_runtime": 39.0845, |
| "eval_chemistry_samples_per_second": 12.793, |
| "eval_chemistry_steps_per_second": 12.793, |
| "step": 1300 |
| }, |
| { |
| "entropy": 1.182255671173334, |
| "epoch": 0.524, |
| "grad_norm": 87.0, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 19.022, |
| "mean_token_accuracy": 0.6953186947852373, |
| "num_tokens": 6383679.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 1.225791597366333, |
| "epoch": 0.528, |
| "grad_norm": 66.0, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 19.8175, |
| "mean_token_accuracy": 0.6822869379073382, |
| "num_tokens": 6430555.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 1.1287783520296215, |
| "epoch": 0.532, |
| "grad_norm": 82.5, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 18.257, |
| "mean_token_accuracy": 0.7016044341027736, |
| "num_tokens": 6481759.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 1.229908336699009, |
| "epoch": 0.536, |
| "grad_norm": 72.0, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 19.7004, |
| "mean_token_accuracy": 0.6902735695242882, |
| "num_tokens": 6533033.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 1.1814987678080797, |
| "epoch": 0.54, |
| "grad_norm": 78.5, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 18.9379, |
| "mean_token_accuracy": 0.699426869675517, |
| "num_tokens": 6580914.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 1.1593723269179463, |
| "epoch": 0.544, |
| "grad_norm": 67.5, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 18.785, |
| "mean_token_accuracy": 0.6992750339210033, |
| "num_tokens": 6626006.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 1.1956233188509942, |
| "epoch": 0.548, |
| "grad_norm": 75.0, |
| "learning_rate": 1.918e-05, |
| "loss": 19.3378, |
| "mean_token_accuracy": 0.6945411875844002, |
| "num_tokens": 6672679.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 1.2158267103135585, |
| "epoch": 0.552, |
| "grad_norm": 74.5, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 19.4937, |
| "mean_token_accuracy": 0.6895309235900641, |
| "num_tokens": 6719864.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 1.1711071468889713, |
| "epoch": 0.556, |
| "grad_norm": 70.0, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 18.9627, |
| "mean_token_accuracy": 0.6955475777387619, |
| "num_tokens": 6768288.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 1.1587215423583985, |
| "epoch": 0.56, |
| "grad_norm": 70.0, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 18.6741, |
| "mean_token_accuracy": 0.6977546285837889, |
| "num_tokens": 6820754.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_biology_entropy": 1.1666600781083107, |
| "eval_biology_loss": 1.1740920543670654, |
| "eval_biology_mean_token_accuracy": 0.7002735708355904, |
| "eval_biology_num_tokens": 6820754.0, |
| "eval_biology_runtime": 34.4305, |
| "eval_biology_samples_per_second": 14.522, |
| "eval_biology_steps_per_second": 14.522, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_chemistry_entropy": 0.6812023703157902, |
| "eval_chemistry_loss": 0.6732838749885559, |
| "eval_chemistry_mean_token_accuracy": 0.8108260444402695, |
| "eval_chemistry_num_tokens": 6820754.0, |
| "eval_chemistry_runtime": 39.1375, |
| "eval_chemistry_samples_per_second": 12.775, |
| "eval_chemistry_steps_per_second": 12.775, |
| "step": 1400 |
| }, |
| { |
| "entropy": 1.1611128127202393, |
| "epoch": 0.564, |
| "grad_norm": 69.0, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 18.4737, |
| "mean_token_accuracy": 0.7005460750311613, |
| "num_tokens": 6875015.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 1.1703616185113788, |
| "epoch": 0.568, |
| "grad_norm": 55.0, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 18.9916, |
| "mean_token_accuracy": 0.6964723881334066, |
| "num_tokens": 6924942.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 1.2203245889395475, |
| "epoch": 0.572, |
| "grad_norm": 69.5, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 19.8421, |
| "mean_token_accuracy": 0.6847291864454746, |
| "num_tokens": 6969801.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 1.1570590244606138, |
| "epoch": 0.576, |
| "grad_norm": 79.5, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 18.3389, |
| "mean_token_accuracy": 0.7031517468392849, |
| "num_tokens": 7021844.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 1.1714151045307517, |
| "epoch": 0.58, |
| "grad_norm": 78.0, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 19.0768, |
| "mean_token_accuracy": 0.695703462511301, |
| "num_tokens": 7070185.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 1.1976398654282092, |
| "epoch": 0.584, |
| "grad_norm": 68.5, |
| "learning_rate": 1.898e-05, |
| "loss": 19.1845, |
| "mean_token_accuracy": 0.6903858676552772, |
| "num_tokens": 7118443.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 1.1526124764233827, |
| "epoch": 0.588, |
| "grad_norm": 79.0, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 18.7328, |
| "mean_token_accuracy": 0.6993513517081738, |
| "num_tokens": 7165265.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 1.152988604642451, |
| "epoch": 0.592, |
| "grad_norm": 68.0, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 18.4211, |
| "mean_token_accuracy": 0.7030052304267883, |
| "num_tokens": 7213951.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 1.1504094721749425, |
| "epoch": 0.596, |
| "grad_norm": 72.5, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 18.6603, |
| "mean_token_accuracy": 0.6974705925211311, |
| "num_tokens": 7265612.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 1.1832087337970734, |
| "epoch": 0.6, |
| "grad_norm": 82.0, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 18.9664, |
| "mean_token_accuracy": 0.6971027232706547, |
| "num_tokens": 7317984.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.6, |
| "eval_biology_entropy": 1.1557998611927032, |
| "eval_biology_loss": 1.1672067642211914, |
| "eval_biology_mean_token_accuracy": 0.7016319183707237, |
| "eval_biology_num_tokens": 7317984.0, |
| "eval_biology_runtime": 34.2829, |
| "eval_biology_samples_per_second": 14.585, |
| "eval_biology_steps_per_second": 14.585, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.6, |
| "eval_chemistry_entropy": 0.6806726072132587, |
| "eval_chemistry_loss": 0.670280396938324, |
| "eval_chemistry_mean_token_accuracy": 0.8115782773494721, |
| "eval_chemistry_num_tokens": 7317984.0, |
| "eval_chemistry_runtime": 39.0952, |
| "eval_chemistry_samples_per_second": 12.789, |
| "eval_chemistry_steps_per_second": 12.789, |
| "step": 1500 |
| }, |
| { |
| "entropy": 1.200187947973609, |
| "epoch": 0.604, |
| "grad_norm": 62.5, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 19.3371, |
| "mean_token_accuracy": 0.6898842452093958, |
| "num_tokens": 7366581.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 1.1842160675674678, |
| "epoch": 0.608, |
| "grad_norm": 73.0, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 19.0594, |
| "mean_token_accuracy": 0.6952343296259642, |
| "num_tokens": 7416773.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 1.1026831757277251, |
| "epoch": 0.612, |
| "grad_norm": 89.0, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 17.8876, |
| "mean_token_accuracy": 0.7055454485118389, |
| "num_tokens": 7467438.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 1.2550879558548331, |
| "epoch": 0.616, |
| "grad_norm": 74.0, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 20.0007, |
| "mean_token_accuracy": 0.6839105855673552, |
| "num_tokens": 7514031.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 1.1557159405201674, |
| "epoch": 0.62, |
| "grad_norm": 79.0, |
| "learning_rate": 1.878e-05, |
| "loss": 18.6918, |
| "mean_token_accuracy": 0.6993272695690393, |
| "num_tokens": 7564846.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 1.1712292978540062, |
| "epoch": 0.624, |
| "grad_norm": 73.5, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 18.5986, |
| "mean_token_accuracy": 0.7031774617731571, |
| "num_tokens": 7612843.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 1.20854026786983, |
| "epoch": 0.628, |
| "grad_norm": 67.0, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 19.6133, |
| "mean_token_accuracy": 0.6873118534684182, |
| "num_tokens": 7658606.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 1.121881490200758, |
| "epoch": 0.632, |
| "grad_norm": 63.25, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 18.3274, |
| "mean_token_accuracy": 0.7022287994623184, |
| "num_tokens": 7710332.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 1.1943174842745066, |
| "epoch": 0.636, |
| "grad_norm": 77.0, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 18.9899, |
| "mean_token_accuracy": 0.6959004882723093, |
| "num_tokens": 7755346.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 1.1558701125904918, |
| "epoch": 0.64, |
| "grad_norm": 69.5, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 18.4954, |
| "mean_token_accuracy": 0.7063233058899641, |
| "num_tokens": 7801633.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 1.1487120282649994, |
| "eval_biology_loss": 1.1615265607833862, |
| "eval_biology_mean_token_accuracy": 0.70275445073843, |
| "eval_biology_num_tokens": 7801633.0, |
| "eval_biology_runtime": 34.4364, |
| "eval_biology_samples_per_second": 14.52, |
| "eval_biology_steps_per_second": 14.52, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_chemistry_entropy": 0.6864565768539905, |
| "eval_chemistry_loss": 0.6769028902053833, |
| "eval_chemistry_mean_token_accuracy": 0.8105848332643509, |
| "eval_chemistry_num_tokens": 7801633.0, |
| "eval_chemistry_runtime": 39.1899, |
| "eval_chemistry_samples_per_second": 12.758, |
| "eval_chemistry_steps_per_second": 12.758, |
| "step": 1600 |
| }, |
| { |
| "entropy": 1.153179651685059, |
| "epoch": 0.644, |
| "grad_norm": 74.0, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 18.6332, |
| "mean_token_accuracy": 0.7009639486670494, |
| "num_tokens": 7847413.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 1.1833068061619998, |
| "epoch": 0.648, |
| "grad_norm": 80.5, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 19.0619, |
| "mean_token_accuracy": 0.6930891156196595, |
| "num_tokens": 7899038.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 1.1857219353318214, |
| "epoch": 0.652, |
| "grad_norm": 65.5, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 19.5173, |
| "mean_token_accuracy": 0.6909411994740366, |
| "num_tokens": 7947565.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 1.2242064960300922, |
| "epoch": 0.656, |
| "grad_norm": 69.0, |
| "learning_rate": 1.858e-05, |
| "loss": 19.2696, |
| "mean_token_accuracy": 0.6927374888211488, |
| "num_tokens": 7995843.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 1.153996530920267, |
| "epoch": 0.66, |
| "grad_norm": 79.0, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 18.7819, |
| "mean_token_accuracy": 0.7004249170422554, |
| "num_tokens": 8042437.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 1.1605599666014315, |
| "epoch": 0.664, |
| "grad_norm": 60.5, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 18.8034, |
| "mean_token_accuracy": 0.6967503964900971, |
| "num_tokens": 8092148.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 1.2301644200459123, |
| "epoch": 0.668, |
| "grad_norm": 64.0, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 19.2725, |
| "mean_token_accuracy": 0.6914080068469047, |
| "num_tokens": 8138818.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 1.192318513803184, |
| "epoch": 0.672, |
| "grad_norm": 74.0, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 19.5569, |
| "mean_token_accuracy": 0.6895372271537781, |
| "num_tokens": 8183103.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 1.1792285278439523, |
| "epoch": 0.676, |
| "grad_norm": 68.0, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 18.8253, |
| "mean_token_accuracy": 0.6978292863816022, |
| "num_tokens": 8231901.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 1.1206298761069775, |
| "epoch": 0.68, |
| "grad_norm": 57.75, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 18.1249, |
| "mean_token_accuracy": 0.7071324445307254, |
| "num_tokens": 8283708.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.68, |
| "eval_biology_entropy": 1.1322392116189004, |
| "eval_biology_loss": 1.1545261144638062, |
| "eval_biology_mean_token_accuracy": 0.7039421809315681, |
| "eval_biology_num_tokens": 8283708.0, |
| "eval_biology_runtime": 34.409, |
| "eval_biology_samples_per_second": 14.531, |
| "eval_biology_steps_per_second": 14.531, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.68, |
| "eval_chemistry_entropy": 0.6786937348246574, |
| "eval_chemistry_loss": 0.6772208213806152, |
| "eval_chemistry_mean_token_accuracy": 0.8101196380853652, |
| "eval_chemistry_num_tokens": 8283708.0, |
| "eval_chemistry_runtime": 39.0374, |
| "eval_chemistry_samples_per_second": 12.808, |
| "eval_chemistry_steps_per_second": 12.808, |
| "step": 1700 |
| }, |
| { |
| "entropy": 1.1540476460009814, |
| "epoch": 0.684, |
| "grad_norm": 70.5, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 18.7848, |
| "mean_token_accuracy": 0.7017710149288178, |
| "num_tokens": 8334267.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 1.1785352129489184, |
| "epoch": 0.688, |
| "grad_norm": 76.5, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 18.6134, |
| "mean_token_accuracy": 0.6982114922255278, |
| "num_tokens": 8385976.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 1.1993559077382088, |
| "epoch": 0.692, |
| "grad_norm": 67.0, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 19.3967, |
| "mean_token_accuracy": 0.6916617508977652, |
| "num_tokens": 8431220.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 1.199912752211094, |
| "epoch": 0.696, |
| "grad_norm": 72.5, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 19.1362, |
| "mean_token_accuracy": 0.6968525558710098, |
| "num_tokens": 8480263.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 1.1471837723627687, |
| "epoch": 0.7, |
| "grad_norm": 73.0, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 18.8049, |
| "mean_token_accuracy": 0.6990611735731364, |
| "num_tokens": 8527244.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 1.1195857927203179, |
| "epoch": 0.704, |
| "grad_norm": 71.5, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 17.7489, |
| "mean_token_accuracy": 0.7118542682379484, |
| "num_tokens": 8578431.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 1.1418487200513483, |
| "epoch": 0.708, |
| "grad_norm": 72.0, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 18.4706, |
| "mean_token_accuracy": 0.7043117236346006, |
| "num_tokens": 8628349.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 1.1226626977324485, |
| "epoch": 0.712, |
| "grad_norm": 63.0, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 17.7757, |
| "mean_token_accuracy": 0.7101234398782253, |
| "num_tokens": 8679830.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 1.139358539134264, |
| "epoch": 0.716, |
| "grad_norm": 66.0, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 18.5864, |
| "mean_token_accuracy": 0.7063206866383552, |
| "num_tokens": 8730777.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 1.1311508536338806, |
| "epoch": 0.72, |
| "grad_norm": 62.5, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 18.1157, |
| "mean_token_accuracy": 0.7037663660943508, |
| "num_tokens": 8781342.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_biology_entropy": 1.119298170030117, |
| "eval_biology_loss": 1.1489436626434326, |
| "eval_biology_mean_token_accuracy": 0.7045390350222588, |
| "eval_biology_num_tokens": 8781342.0, |
| "eval_biology_runtime": 34.3095, |
| "eval_biology_samples_per_second": 14.573, |
| "eval_biology_steps_per_second": 14.573, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_chemistry_entropy": 0.6719300321638584, |
| "eval_chemistry_loss": 0.6753261685371399, |
| "eval_chemistry_mean_token_accuracy": 0.8107173951864243, |
| "eval_chemistry_num_tokens": 8781342.0, |
| "eval_chemistry_runtime": 39.0499, |
| "eval_chemistry_samples_per_second": 12.804, |
| "eval_chemistry_steps_per_second": 12.804, |
| "step": 1800 |
| }, |
| { |
| "entropy": 1.1399380365386604, |
| "epoch": 0.724, |
| "grad_norm": 70.0, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 18.4548, |
| "mean_token_accuracy": 0.704157155752182, |
| "num_tokens": 8827703.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 1.1481896413490176, |
| "epoch": 0.728, |
| "grad_norm": 64.5, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 18.4328, |
| "mean_token_accuracy": 0.7012046728283167, |
| "num_tokens": 8876240.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 1.1291126595810055, |
| "epoch": 0.732, |
| "grad_norm": 58.5, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 18.0903, |
| "mean_token_accuracy": 0.7058037102222443, |
| "num_tokens": 8928717.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 1.147701820358634, |
| "epoch": 0.736, |
| "grad_norm": 73.0, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 18.3147, |
| "mean_token_accuracy": 0.7024340128526092, |
| "num_tokens": 8977918.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 1.1560426784679294, |
| "epoch": 0.74, |
| "grad_norm": 68.0, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 18.6953, |
| "mean_token_accuracy": 0.7024053547531366, |
| "num_tokens": 9024698.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 1.165362073481083, |
| "epoch": 0.744, |
| "grad_norm": 74.5, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 18.9089, |
| "mean_token_accuracy": 0.7019367810338736, |
| "num_tokens": 9073936.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 1.2147667936980724, |
| "epoch": 0.748, |
| "grad_norm": 68.0, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 19.3236, |
| "mean_token_accuracy": 0.6906995607540012, |
| "num_tokens": 9120674.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 1.1410686820745468, |
| "epoch": 0.752, |
| "grad_norm": 58.0, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 18.3801, |
| "mean_token_accuracy": 0.704433299601078, |
| "num_tokens": 9169322.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 1.1186864526942373, |
| "epoch": 0.756, |
| "grad_norm": 75.0, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 18.0648, |
| "mean_token_accuracy": 0.7078946601599455, |
| "num_tokens": 9219250.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 1.1588866852223874, |
| "epoch": 0.76, |
| "grad_norm": 71.5, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 18.5677, |
| "mean_token_accuracy": 0.7036618825048209, |
| "num_tokens": 9267416.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.76, |
| "eval_biology_entropy": 1.139221048593521, |
| "eval_biology_loss": 1.1487153768539429, |
| "eval_biology_mean_token_accuracy": 0.7059274349808693, |
| "eval_biology_num_tokens": 9267416.0, |
| "eval_biology_runtime": 34.3656, |
| "eval_biology_samples_per_second": 14.549, |
| "eval_biology_steps_per_second": 14.549, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.76, |
| "eval_chemistry_entropy": 0.6905211313664913, |
| "eval_chemistry_loss": 0.6773802638053894, |
| "eval_chemistry_mean_token_accuracy": 0.8103563543558121, |
| "eval_chemistry_num_tokens": 9267416.0, |
| "eval_chemistry_runtime": 39.0486, |
| "eval_chemistry_samples_per_second": 12.805, |
| "eval_chemistry_steps_per_second": 12.805, |
| "step": 1900 |
| }, |
| { |
| "entropy": 1.119858698733151, |
| "epoch": 0.764, |
| "grad_norm": 80.5, |
| "learning_rate": 1.798e-05, |
| "loss": 17.8759, |
| "mean_token_accuracy": 0.7116206176578999, |
| "num_tokens": 9319054.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 1.1601388046517969, |
| "epoch": 0.768, |
| "grad_norm": 77.5, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 18.8759, |
| "mean_token_accuracy": 0.6984018977731467, |
| "num_tokens": 9368141.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 1.221377344056964, |
| "epoch": 0.772, |
| "grad_norm": 71.5, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 19.5676, |
| "mean_token_accuracy": 0.6873011328279972, |
| "num_tokens": 9411478.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 1.1033327596262097, |
| "epoch": 0.776, |
| "grad_norm": 68.0, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 17.6384, |
| "mean_token_accuracy": 0.7138782385736704, |
| "num_tokens": 9463691.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 1.1310196310281753, |
| "epoch": 0.78, |
| "grad_norm": 65.0, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 18.2027, |
| "mean_token_accuracy": 0.7050832413136959, |
| "num_tokens": 9514828.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 1.168931331858039, |
| "epoch": 0.784, |
| "grad_norm": 66.0, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 18.7762, |
| "mean_token_accuracy": 0.7014395378530025, |
| "num_tokens": 9565236.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 1.1767852138727903, |
| "epoch": 0.788, |
| "grad_norm": 63.25, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 18.9162, |
| "mean_token_accuracy": 0.6968433002009988, |
| "num_tokens": 9613803.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 1.1523513723164798, |
| "epoch": 0.792, |
| "grad_norm": 62.75, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 18.4318, |
| "mean_token_accuracy": 0.7035329312086105, |
| "num_tokens": 9664805.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 1.1664102960377931, |
| "epoch": 0.796, |
| "grad_norm": 72.0, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 18.7643, |
| "mean_token_accuracy": 0.6997625567018986, |
| "num_tokens": 9715064.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 1.1953407004475594, |
| "epoch": 0.8, |
| "grad_norm": 67.0, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 19.5221, |
| "mean_token_accuracy": 0.6904468927532434, |
| "num_tokens": 9761227.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_biology_entropy": 1.158657855272293, |
| "eval_biology_loss": 1.1407227516174316, |
| "eval_biology_mean_token_accuracy": 0.706549409031868, |
| "eval_biology_num_tokens": 9761227.0, |
| "eval_biology_runtime": 34.4156, |
| "eval_biology_samples_per_second": 14.528, |
| "eval_biology_steps_per_second": 14.528, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_chemistry_entropy": 0.7075672374665737, |
| "eval_chemistry_loss": 0.677842378616333, |
| "eval_chemistry_mean_token_accuracy": 0.809728811442852, |
| "eval_chemistry_num_tokens": 9761227.0, |
| "eval_chemistry_runtime": 38.9796, |
| "eval_chemistry_samples_per_second": 12.827, |
| "eval_chemistry_steps_per_second": 12.827, |
| "step": 2000 |
| }, |
| { |
| "entropy": 1.1363696370273828, |
| "epoch": 0.804, |
| "grad_norm": 62.0, |
| "learning_rate": 1.7757777777777777e-05, |
| "loss": 17.976, |
| "mean_token_accuracy": 0.7070415183901787, |
| "num_tokens": 9811589.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 1.1199781388044356, |
| "epoch": 0.808, |
| "grad_norm": 67.0, |
| "learning_rate": 1.7735555555555558e-05, |
| "loss": 17.9076, |
| "mean_token_accuracy": 0.7071813272312284, |
| "num_tokens": 9863410.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 1.1580626226961612, |
| "epoch": 0.812, |
| "grad_norm": 73.0, |
| "learning_rate": 1.7713333333333335e-05, |
| "loss": 18.5488, |
| "mean_token_accuracy": 0.7003198087215423, |
| "num_tokens": 9910509.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 1.1708898711949587, |
| "epoch": 0.816, |
| "grad_norm": 67.5, |
| "learning_rate": 1.7691111111111113e-05, |
| "loss": 18.893, |
| "mean_token_accuracy": 0.6949249973520637, |
| "num_tokens": 9958727.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 1.1190886072814465, |
| "epoch": 0.82, |
| "grad_norm": 68.0, |
| "learning_rate": 1.766888888888889e-05, |
| "loss": 18.03, |
| "mean_token_accuracy": 0.7062144052237272, |
| "num_tokens": 10008669.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 1.1490429989993571, |
| "epoch": 0.824, |
| "grad_norm": 77.0, |
| "learning_rate": 1.7646666666666668e-05, |
| "loss": 18.5414, |
| "mean_token_accuracy": 0.7021297257393598, |
| "num_tokens": 10059078.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 1.151954964734614, |
| "epoch": 0.828, |
| "grad_norm": 71.0, |
| "learning_rate": 1.7624444444444446e-05, |
| "loss": 18.6067, |
| "mean_token_accuracy": 0.6987083677202464, |
| "num_tokens": 10107666.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 1.1119983173906802, |
| "epoch": 0.832, |
| "grad_norm": 60.75, |
| "learning_rate": 1.7602222222222223e-05, |
| "loss": 17.6866, |
| "mean_token_accuracy": 0.7093611612915993, |
| "num_tokens": 10155771.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 1.1303761949762703, |
| "epoch": 0.836, |
| "grad_norm": 71.0, |
| "learning_rate": 1.758e-05, |
| "loss": 18.3625, |
| "mean_token_accuracy": 0.7020260445773602, |
| "num_tokens": 10205584.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 1.1609793374314905, |
| "epoch": 0.84, |
| "grad_norm": 69.5, |
| "learning_rate": 1.755777777777778e-05, |
| "loss": 18.7772, |
| "mean_token_accuracy": 0.6994457300752401, |
| "num_tokens": 10255927.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.84, |
| "eval_biology_entropy": 1.1735474869012832, |
| "eval_biology_loss": 1.1386175155639648, |
| "eval_biology_mean_token_accuracy": 0.7069284253120423, |
| "eval_biology_num_tokens": 10255927.0, |
| "eval_biology_runtime": 34.2834, |
| "eval_biology_samples_per_second": 14.584, |
| "eval_biology_steps_per_second": 14.584, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.84, |
| "eval_chemistry_entropy": 0.7099601251780987, |
| "eval_chemistry_loss": 0.6767250895500183, |
| "eval_chemistry_mean_token_accuracy": 0.8097227366566658, |
| "eval_chemistry_num_tokens": 10255927.0, |
| "eval_chemistry_runtime": 39.0454, |
| "eval_chemistry_samples_per_second": 12.806, |
| "eval_chemistry_steps_per_second": 12.806, |
| "step": 2100 |
| }, |
| { |
| "entropy": 1.1142319036647677, |
| "epoch": 0.844, |
| "grad_norm": 60.75, |
| "learning_rate": 1.7535555555555556e-05, |
| "loss": 17.6088, |
| "mean_token_accuracy": 0.711638954654336, |
| "num_tokens": 10308513.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 1.1440270494669675, |
| "epoch": 0.848, |
| "grad_norm": 73.0, |
| "learning_rate": 1.7513333333333334e-05, |
| "loss": 18.5026, |
| "mean_token_accuracy": 0.700833398476243, |
| "num_tokens": 10357721.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 1.1324728639796375, |
| "epoch": 0.852, |
| "grad_norm": 65.5, |
| "learning_rate": 1.749111111111111e-05, |
| "loss": 18.2524, |
| "mean_token_accuracy": 0.7033683439716697, |
| "num_tokens": 10405196.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 1.1318668030202388, |
| "epoch": 0.856, |
| "grad_norm": 73.0, |
| "learning_rate": 1.746888888888889e-05, |
| "loss": 17.8507, |
| "mean_token_accuracy": 0.7074570011347532, |
| "num_tokens": 10454335.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 1.1053960118442774, |
| "epoch": 0.86, |
| "grad_norm": 68.0, |
| "learning_rate": 1.7446666666666667e-05, |
| "loss": 18.3325, |
| "mean_token_accuracy": 0.7027432754635811, |
| "num_tokens": 10503987.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 1.21066178008914, |
| "epoch": 0.864, |
| "grad_norm": 69.5, |
| "learning_rate": 1.7424444444444444e-05, |
| "loss": 19.1575, |
| "mean_token_accuracy": 0.6939540956169367, |
| "num_tokens": 10552495.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 1.1965933687984944, |
| "epoch": 0.868, |
| "grad_norm": 58.75, |
| "learning_rate": 1.7402222222222222e-05, |
| "loss": 19.211, |
| "mean_token_accuracy": 0.6939904358237982, |
| "num_tokens": 10602903.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 1.1171547731384635, |
| "epoch": 0.872, |
| "grad_norm": 66.5, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 18.0543, |
| "mean_token_accuracy": 0.7061869710683822, |
| "num_tokens": 10651501.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 1.1249778434634208, |
| "epoch": 0.876, |
| "grad_norm": 62.25, |
| "learning_rate": 1.735777777777778e-05, |
| "loss": 18.0524, |
| "mean_token_accuracy": 0.7085037898272276, |
| "num_tokens": 10698431.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 1.1113598695024849, |
| "epoch": 0.88, |
| "grad_norm": 62.75, |
| "learning_rate": 1.7335555555555558e-05, |
| "loss": 18.0239, |
| "mean_token_accuracy": 0.7066719084978104, |
| "num_tokens": 10748749.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_biology_entropy": 1.1119387486577035, |
| "eval_biology_loss": 1.1339057683944702, |
| "eval_biology_mean_token_accuracy": 0.7078809962272644, |
| "eval_biology_num_tokens": 10748749.0, |
| "eval_biology_runtime": 34.5706, |
| "eval_biology_samples_per_second": 14.463, |
| "eval_biology_steps_per_second": 14.463, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_chemistry_entropy": 0.6606079009771347, |
| "eval_chemistry_loss": 0.6750172972679138, |
| "eval_chemistry_mean_token_accuracy": 0.8104267127513886, |
| "eval_chemistry_num_tokens": 10748749.0, |
| "eval_chemistry_runtime": 39.2002, |
| "eval_chemistry_samples_per_second": 12.755, |
| "eval_chemistry_steps_per_second": 12.755, |
| "step": 2200 |
| }, |
| { |
| "entropy": 1.1056056110188366, |
| "epoch": 0.884, |
| "grad_norm": 74.5, |
| "learning_rate": 1.7313333333333336e-05, |
| "loss": 17.7253, |
| "mean_token_accuracy": 0.7094237711280584, |
| "num_tokens": 10802228.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 1.1432907488197088, |
| "epoch": 0.888, |
| "grad_norm": 61.5, |
| "learning_rate": 1.7291111111111113e-05, |
| "loss": 18.3194, |
| "mean_token_accuracy": 0.7021823862567544, |
| "num_tokens": 10852932.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 1.1547889987006783, |
| "epoch": 0.892, |
| "grad_norm": 67.0, |
| "learning_rate": 1.726888888888889e-05, |
| "loss": 18.2762, |
| "mean_token_accuracy": 0.7085223745554685, |
| "num_tokens": 10899168.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 1.1641722086817026, |
| "epoch": 0.896, |
| "grad_norm": 76.0, |
| "learning_rate": 1.724666666666667e-05, |
| "loss": 18.8612, |
| "mean_token_accuracy": 0.6960871197283268, |
| "num_tokens": 10943319.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 1.1232851393520833, |
| "epoch": 0.9, |
| "grad_norm": 68.5, |
| "learning_rate": 1.7224444444444446e-05, |
| "loss": 18.3098, |
| "mean_token_accuracy": 0.7068704262375831, |
| "num_tokens": 10990879.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 1.1671414624899625, |
| "epoch": 0.904, |
| "grad_norm": 67.5, |
| "learning_rate": 1.7202222222222224e-05, |
| "loss": 18.5044, |
| "mean_token_accuracy": 0.7026478189975023, |
| "num_tokens": 11037833.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 1.139916703477502, |
| "epoch": 0.908, |
| "grad_norm": 69.5, |
| "learning_rate": 1.718e-05, |
| "loss": 18.4244, |
| "mean_token_accuracy": 0.7037582151591778, |
| "num_tokens": 11086725.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 1.127130401134491, |
| "epoch": 0.912, |
| "grad_norm": 61.25, |
| "learning_rate": 1.715777777777778e-05, |
| "loss": 17.9568, |
| "mean_token_accuracy": 0.7079363029450179, |
| "num_tokens": 11136935.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 1.1513679789379239, |
| "epoch": 0.916, |
| "grad_norm": 60.5, |
| "learning_rate": 1.7135555555555557e-05, |
| "loss": 18.5227, |
| "mean_token_accuracy": 0.7013451868668199, |
| "num_tokens": 11188005.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 1.1256804373115301, |
| "epoch": 0.92, |
| "grad_norm": 57.0, |
| "learning_rate": 1.7113333333333334e-05, |
| "loss": 18.2868, |
| "mean_token_accuracy": 0.7067213650792837, |
| "num_tokens": 11238002.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.92, |
| "eval_biology_entropy": 1.1104887677431106, |
| "eval_biology_loss": 1.1293854713439941, |
| "eval_biology_mean_token_accuracy": 0.7086220408678054, |
| "eval_biology_num_tokens": 11238002.0, |
| "eval_biology_runtime": 34.405, |
| "eval_biology_samples_per_second": 14.533, |
| "eval_biology_steps_per_second": 14.533, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.92, |
| "eval_chemistry_entropy": 0.6767489816844463, |
| "eval_chemistry_loss": 0.675902247428894, |
| "eval_chemistry_mean_token_accuracy": 0.8104530355930328, |
| "eval_chemistry_num_tokens": 11238002.0, |
| "eval_chemistry_runtime": 39.1073, |
| "eval_chemistry_samples_per_second": 12.785, |
| "eval_chemistry_steps_per_second": 12.785, |
| "step": 2300 |
| }, |
| { |
| "entropy": 1.156477975845337, |
| "epoch": 0.924, |
| "grad_norm": 64.0, |
| "learning_rate": 1.7091111111111112e-05, |
| "loss": 18.5677, |
| "mean_token_accuracy": 0.6988354671746493, |
| "num_tokens": 11285759.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 1.1492901083081961, |
| "epoch": 0.928, |
| "grad_norm": 84.5, |
| "learning_rate": 1.706888888888889e-05, |
| "loss": 18.3243, |
| "mean_token_accuracy": 0.7017165068536997, |
| "num_tokens": 11331098.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 1.070316667854786, |
| "epoch": 0.932, |
| "grad_norm": 65.5, |
| "learning_rate": 1.704666666666667e-05, |
| "loss": 17.6875, |
| "mean_token_accuracy": 0.712903193384409, |
| "num_tokens": 11381267.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 1.092837950773537, |
| "epoch": 0.936, |
| "grad_norm": 57.25, |
| "learning_rate": 1.7024444444444445e-05, |
| "loss": 17.1894, |
| "mean_token_accuracy": 0.7153506379574537, |
| "num_tokens": 11435812.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 1.1070469096302986, |
| "epoch": 0.94, |
| "grad_norm": 66.5, |
| "learning_rate": 1.7002222222222226e-05, |
| "loss": 18.0288, |
| "mean_token_accuracy": 0.7079101942479611, |
| "num_tokens": 11482238.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 1.1188768319785596, |
| "epoch": 0.944, |
| "grad_norm": 69.0, |
| "learning_rate": 1.698e-05, |
| "loss": 18.0299, |
| "mean_token_accuracy": 0.7057675050571561, |
| "num_tokens": 11530550.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 1.1243172274902462, |
| "epoch": 0.948, |
| "grad_norm": 72.0, |
| "learning_rate": 1.695777777777778e-05, |
| "loss": 18.0313, |
| "mean_token_accuracy": 0.7121372889727354, |
| "num_tokens": 11577273.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 1.0865204200148582, |
| "epoch": 0.952, |
| "grad_norm": 58.0, |
| "learning_rate": 1.6935555555555555e-05, |
| "loss": 17.2998, |
| "mean_token_accuracy": 0.7168761320412159, |
| "num_tokens": 11629080.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 1.0908181142061948, |
| "epoch": 0.956, |
| "grad_norm": 65.0, |
| "learning_rate": 1.6913333333333336e-05, |
| "loss": 17.8917, |
| "mean_token_accuracy": 0.7086671654134988, |
| "num_tokens": 11681001.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 1.1510007094591856, |
| "epoch": 0.96, |
| "grad_norm": 64.5, |
| "learning_rate": 1.689111111111111e-05, |
| "loss": 18.2789, |
| "mean_token_accuracy": 0.7039446363225579, |
| "num_tokens": 11729645.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 1.1140780518651008, |
| "eval_biology_loss": 1.1259111166000366, |
| "eval_biology_mean_token_accuracy": 0.7096920711994171, |
| "eval_biology_num_tokens": 11729645.0, |
| "eval_biology_runtime": 34.5065, |
| "eval_biology_samples_per_second": 14.49, |
| "eval_biology_steps_per_second": 14.49, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_chemistry_entropy": 0.6870564540028572, |
| "eval_chemistry_loss": 0.6765609383583069, |
| "eval_chemistry_mean_token_accuracy": 0.8095985132455826, |
| "eval_chemistry_num_tokens": 11729645.0, |
| "eval_chemistry_runtime": 39.2452, |
| "eval_chemistry_samples_per_second": 12.74, |
| "eval_chemistry_steps_per_second": 12.74, |
| "step": 2400 |
| }, |
| { |
| "entropy": 1.0772071216255426, |
| "epoch": 0.964, |
| "grad_norm": 66.5, |
| "learning_rate": 1.686888888888889e-05, |
| "loss": 17.481, |
| "mean_token_accuracy": 0.7153752170503139, |
| "num_tokens": 11778772.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 1.1216032130643725, |
| "epoch": 0.968, |
| "grad_norm": 61.75, |
| "learning_rate": 1.684666666666667e-05, |
| "loss": 17.9555, |
| "mean_token_accuracy": 0.7085904907435179, |
| "num_tokens": 11823767.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 1.1032297348603606, |
| "epoch": 0.972, |
| "grad_norm": 69.0, |
| "learning_rate": 1.6824444444444447e-05, |
| "loss": 17.7725, |
| "mean_token_accuracy": 0.7095954738557338, |
| "num_tokens": 11875204.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 1.1178162148222328, |
| "epoch": 0.976, |
| "grad_norm": 67.0, |
| "learning_rate": 1.6802222222222224e-05, |
| "loss": 17.8673, |
| "mean_token_accuracy": 0.7087533060461283, |
| "num_tokens": 11924644.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 1.1170936610549689, |
| "epoch": 0.98, |
| "grad_norm": 68.0, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 18.0697, |
| "mean_token_accuracy": 0.7051723409444094, |
| "num_tokens": 11975359.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 1.146547881513834, |
| "epoch": 0.984, |
| "grad_norm": 63.5, |
| "learning_rate": 1.675777777777778e-05, |
| "loss": 18.2269, |
| "mean_token_accuracy": 0.7029001332819462, |
| "num_tokens": 12022090.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 1.115088721551001, |
| "epoch": 0.988, |
| "grad_norm": 64.0, |
| "learning_rate": 1.6735555555555557e-05, |
| "loss": 18.2689, |
| "mean_token_accuracy": 0.7032332498580217, |
| "num_tokens": 12071702.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 1.1118581231683493, |
| "epoch": 0.992, |
| "grad_norm": 65.5, |
| "learning_rate": 1.6713333333333335e-05, |
| "loss": 17.5517, |
| "mean_token_accuracy": 0.7128655791282654, |
| "num_tokens": 12123059.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 1.1398745885118842, |
| "epoch": 0.996, |
| "grad_norm": 68.5, |
| "learning_rate": 1.6691111111111112e-05, |
| "loss": 18.2919, |
| "mean_token_accuracy": 0.7025836504995823, |
| "num_tokens": 12170572.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 1.1277340853586793, |
| "epoch": 1.0, |
| "grad_norm": 63.0, |
| "learning_rate": 1.666888888888889e-05, |
| "loss": 18.4876, |
| "mean_token_accuracy": 0.703786114603281, |
| "num_tokens": 12221291.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_biology_entropy": 1.1159155780076981, |
| "eval_biology_loss": 1.1243422031402588, |
| "eval_biology_mean_token_accuracy": 0.7097521283626557, |
| "eval_biology_num_tokens": 12221291.0, |
| "eval_biology_runtime": 34.3645, |
| "eval_biology_samples_per_second": 14.55, |
| "eval_biology_steps_per_second": 14.55, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_chemistry_entropy": 0.6904440721571445, |
| "eval_chemistry_loss": 0.6775741577148438, |
| "eval_chemistry_mean_token_accuracy": 0.8101254143714904, |
| "eval_chemistry_num_tokens": 12221291.0, |
| "eval_chemistry_runtime": 39.0533, |
| "eval_chemistry_samples_per_second": 12.803, |
| "eval_chemistry_steps_per_second": 12.803, |
| "step": 2500 |
| }, |
| { |
| "entropy": 0.8578420992940664, |
| "epoch": 1.004, |
| "grad_norm": 79.5, |
| "learning_rate": 1.6646666666666668e-05, |
| "loss": 12.6487, |
| "mean_token_accuracy": 0.7860239744186401, |
| "num_tokens": 12266242.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 0.7053419223055244, |
| "epoch": 1.008, |
| "grad_norm": 75.0, |
| "learning_rate": 1.6624444444444445e-05, |
| "loss": 11.9323, |
| "mean_token_accuracy": 0.792425375059247, |
| "num_tokens": 12319366.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 0.7434766286984086, |
| "epoch": 1.012, |
| "grad_norm": 74.0, |
| "learning_rate": 1.6602222222222223e-05, |
| "loss": 11.6777, |
| "mean_token_accuracy": 0.7956793092191219, |
| "num_tokens": 12371275.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 0.7364071726799011, |
| "epoch": 1.016, |
| "grad_norm": 68.0, |
| "learning_rate": 1.658e-05, |
| "loss": 11.9129, |
| "mean_token_accuracy": 0.794680006429553, |
| "num_tokens": 12422323.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 0.7577000215649605, |
| "epoch": 1.02, |
| "grad_norm": 69.5, |
| "learning_rate": 1.6557777777777778e-05, |
| "loss": 12.4206, |
| "mean_token_accuracy": 0.7865326277911663, |
| "num_tokens": 12471781.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 0.7748347625136376, |
| "epoch": 1.024, |
| "grad_norm": 67.5, |
| "learning_rate": 1.6535555555555556e-05, |
| "loss": 12.0855, |
| "mean_token_accuracy": 0.78928126655519, |
| "num_tokens": 12524183.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 0.7415021266788244, |
| "epoch": 1.028, |
| "grad_norm": 78.5, |
| "learning_rate": 1.6513333333333333e-05, |
| "loss": 12.1399, |
| "mean_token_accuracy": 0.7882724687457084, |
| "num_tokens": 12573390.0, |
| "step": 2570 |
| }, |
| { |
| "entropy": 0.7520155070349575, |
| "epoch": 1.032, |
| "grad_norm": 73.5, |
| "learning_rate": 1.6491111111111114e-05, |
| "loss": 11.8195, |
| "mean_token_accuracy": 0.7939066126942634, |
| "num_tokens": 12621565.0, |
| "step": 2580 |
| }, |
| { |
| "entropy": 0.7552091253921389, |
| "epoch": 1.036, |
| "grad_norm": 77.0, |
| "learning_rate": 1.646888888888889e-05, |
| "loss": 12.3477, |
| "mean_token_accuracy": 0.7892356630414724, |
| "num_tokens": 12670328.0, |
| "step": 2590 |
| }, |
| { |
| "entropy": 0.785752167366445, |
| "epoch": 1.04, |
| "grad_norm": 76.0, |
| "learning_rate": 1.644666666666667e-05, |
| "loss": 12.31, |
| "mean_token_accuracy": 0.7897520136088133, |
| "num_tokens": 12718593.0, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_biology_entropy": 0.9385359708666802, |
| "eval_biology_loss": 1.152160882949829, |
| "eval_biology_mean_token_accuracy": 0.7080060958862304, |
| "eval_biology_num_tokens": 12718593.0, |
| "eval_biology_runtime": 34.5225, |
| "eval_biology_samples_per_second": 14.483, |
| "eval_biology_steps_per_second": 14.483, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_chemistry_entropy": 0.5707901727259159, |
| "eval_chemistry_loss": 0.6861657500267029, |
| "eval_chemistry_mean_token_accuracy": 0.8100856151580811, |
| "eval_chemistry_num_tokens": 12718593.0, |
| "eval_chemistry_runtime": 39.269, |
| "eval_chemistry_samples_per_second": 12.733, |
| "eval_chemistry_steps_per_second": 12.733, |
| "step": 2600 |
| }, |
| { |
| "entropy": 0.7259062381461263, |
| "epoch": 1.044, |
| "grad_norm": 75.0, |
| "learning_rate": 1.6424444444444444e-05, |
| "loss": 11.9342, |
| "mean_token_accuracy": 0.7963043585419655, |
| "num_tokens": 12768223.0, |
| "step": 2610 |
| }, |
| { |
| "entropy": 0.8006209449842572, |
| "epoch": 1.048, |
| "grad_norm": 68.0, |
| "learning_rate": 1.6402222222222225e-05, |
| "loss": 12.4858, |
| "mean_token_accuracy": 0.7876774240285158, |
| "num_tokens": 12817611.0, |
| "step": 2620 |
| }, |
| { |
| "entropy": 0.7280113430693745, |
| "epoch": 1.052, |
| "grad_norm": 84.5, |
| "learning_rate": 1.638e-05, |
| "loss": 11.9451, |
| "mean_token_accuracy": 0.7938677735626698, |
| "num_tokens": 12868298.0, |
| "step": 2630 |
| }, |
| { |
| "entropy": 0.7228842966258526, |
| "epoch": 1.056, |
| "grad_norm": 66.0, |
| "learning_rate": 1.635777777777778e-05, |
| "loss": 11.4312, |
| "mean_token_accuracy": 0.8008369661867618, |
| "num_tokens": 12917803.0, |
| "step": 2640 |
| }, |
| { |
| "entropy": 0.7347168253734708, |
| "epoch": 1.06, |
| "grad_norm": 83.0, |
| "learning_rate": 1.6335555555555558e-05, |
| "loss": 11.8974, |
| "mean_token_accuracy": 0.7960309438407421, |
| "num_tokens": 12963209.0, |
| "step": 2650 |
| }, |
| { |
| "entropy": 0.7789461050182581, |
| "epoch": 1.064, |
| "grad_norm": 72.5, |
| "learning_rate": 1.6313333333333335e-05, |
| "loss": 12.2428, |
| "mean_token_accuracy": 0.7872451439499855, |
| "num_tokens": 13013009.0, |
| "step": 2660 |
| }, |
| { |
| "entropy": 0.7627207360230386, |
| "epoch": 1.068, |
| "grad_norm": 67.0, |
| "learning_rate": 1.6291111111111113e-05, |
| "loss": 12.2857, |
| "mean_token_accuracy": 0.7906007189303637, |
| "num_tokens": 13058780.0, |
| "step": 2670 |
| }, |
| { |
| "entropy": 0.7417652688920497, |
| "epoch": 1.072, |
| "grad_norm": 85.0, |
| "learning_rate": 1.626888888888889e-05, |
| "loss": 11.9783, |
| "mean_token_accuracy": 0.7910834137350321, |
| "num_tokens": 13105826.0, |
| "step": 2680 |
| }, |
| { |
| "entropy": 0.7713781571015715, |
| "epoch": 1.076, |
| "grad_norm": 77.5, |
| "learning_rate": 1.6246666666666668e-05, |
| "loss": 12.4913, |
| "mean_token_accuracy": 0.7857859663665294, |
| "num_tokens": 13153681.0, |
| "step": 2690 |
| }, |
| { |
| "entropy": 0.7486217807978391, |
| "epoch": 1.08, |
| "grad_norm": 70.5, |
| "learning_rate": 1.6224444444444446e-05, |
| "loss": 11.6272, |
| "mean_token_accuracy": 0.79732346534729, |
| "num_tokens": 13201601.0, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.08, |
| "eval_biology_entropy": 0.9384198271036148, |
| "eval_biology_loss": 1.1512656211853027, |
| "eval_biology_mean_token_accuracy": 0.7077861483097077, |
| "eval_biology_num_tokens": 13201601.0, |
| "eval_biology_runtime": 34.5014, |
| "eval_biology_samples_per_second": 14.492, |
| "eval_biology_steps_per_second": 14.492, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.08, |
| "eval_chemistry_entropy": 0.5755563218891621, |
| "eval_chemistry_loss": 0.6884739995002747, |
| "eval_chemistry_mean_token_accuracy": 0.809848326921463, |
| "eval_chemistry_num_tokens": 13201601.0, |
| "eval_chemistry_runtime": 39.1338, |
| "eval_chemistry_samples_per_second": 12.777, |
| "eval_chemistry_steps_per_second": 12.777, |
| "step": 2700 |
| }, |
| { |
| "entropy": 0.7609818574041128, |
| "epoch": 1.084, |
| "grad_norm": 101.5, |
| "learning_rate": 1.6202222222222223e-05, |
| "loss": 12.3814, |
| "mean_token_accuracy": 0.7864615328609943, |
| "num_tokens": 13249788.0, |
| "step": 2710 |
| }, |
| { |
| "entropy": 0.7645085378549993, |
| "epoch": 1.088, |
| "grad_norm": 70.0, |
| "learning_rate": 1.618e-05, |
| "loss": 12.0776, |
| "mean_token_accuracy": 0.7907705314457416, |
| "num_tokens": 13298619.0, |
| "step": 2720 |
| }, |
| { |
| "entropy": 0.7094159284606576, |
| "epoch": 1.092, |
| "grad_norm": 66.0, |
| "learning_rate": 1.615777777777778e-05, |
| "loss": 11.5688, |
| "mean_token_accuracy": 0.7995603460818529, |
| "num_tokens": 13345926.0, |
| "step": 2730 |
| }, |
| { |
| "entropy": 0.7520682083442807, |
| "epoch": 1.096, |
| "grad_norm": 73.5, |
| "learning_rate": 1.6135555555555556e-05, |
| "loss": 11.847, |
| "mean_token_accuracy": 0.7953370809555054, |
| "num_tokens": 13390496.0, |
| "step": 2740 |
| }, |
| { |
| "entropy": 0.7643397770822048, |
| "epoch": 1.1, |
| "grad_norm": 72.5, |
| "learning_rate": 1.6113333333333334e-05, |
| "loss": 12.2348, |
| "mean_token_accuracy": 0.7897418607026339, |
| "num_tokens": 13441082.0, |
| "step": 2750 |
| }, |
| { |
| "entropy": 0.7417395184747875, |
| "epoch": 1.104, |
| "grad_norm": 79.5, |
| "learning_rate": 1.609111111111111e-05, |
| "loss": 11.9877, |
| "mean_token_accuracy": 0.7918722402304411, |
| "num_tokens": 13491486.0, |
| "step": 2760 |
| }, |
| { |
| "entropy": 0.7266374656930565, |
| "epoch": 1.108, |
| "grad_norm": 70.5, |
| "learning_rate": 1.606888888888889e-05, |
| "loss": 11.63, |
| "mean_token_accuracy": 0.7981215808540583, |
| "num_tokens": 13537164.0, |
| "step": 2770 |
| }, |
| { |
| "entropy": 0.7592888733372092, |
| "epoch": 1.112, |
| "grad_norm": 68.0, |
| "learning_rate": 1.6046666666666667e-05, |
| "loss": 12.0939, |
| "mean_token_accuracy": 0.7929974045604468, |
| "num_tokens": 13580102.0, |
| "step": 2780 |
| }, |
| { |
| "entropy": 0.7253062453120946, |
| "epoch": 1.116, |
| "grad_norm": 73.0, |
| "learning_rate": 1.6024444444444444e-05, |
| "loss": 11.6571, |
| "mean_token_accuracy": 0.797444849088788, |
| "num_tokens": 13628255.0, |
| "step": 2790 |
| }, |
| { |
| "entropy": 0.7494521101936698, |
| "epoch": 1.12, |
| "grad_norm": 74.5, |
| "learning_rate": 1.6002222222222222e-05, |
| "loss": 11.6976, |
| "mean_token_accuracy": 0.7954880796372891, |
| "num_tokens": 13674663.0, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_biology_entropy": 0.9074934704899787, |
| "eval_biology_loss": 1.153106451034546, |
| "eval_biology_mean_token_accuracy": 0.7099947838783264, |
| "eval_biology_num_tokens": 13674663.0, |
| "eval_biology_runtime": 34.3836, |
| "eval_biology_samples_per_second": 14.542, |
| "eval_biology_steps_per_second": 14.542, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_chemistry_entropy": 0.5607364977002144, |
| "eval_chemistry_loss": 0.6902413368225098, |
| "eval_chemistry_mean_token_accuracy": 0.8099774696826935, |
| "eval_chemistry_num_tokens": 13674663.0, |
| "eval_chemistry_runtime": 39.0759, |
| "eval_chemistry_samples_per_second": 12.796, |
| "eval_chemistry_steps_per_second": 12.796, |
| "step": 2800 |
| }, |
| { |
| "entropy": 0.7108782446011901, |
| "epoch": 1.124, |
| "grad_norm": 74.0, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 11.7852, |
| "mean_token_accuracy": 0.7942955374717713, |
| "num_tokens": 13726667.0, |
| "step": 2810 |
| }, |
| { |
| "entropy": 0.7647061448544263, |
| "epoch": 1.1280000000000001, |
| "grad_norm": 77.5, |
| "learning_rate": 1.5957777777777777e-05, |
| "loss": 12.0814, |
| "mean_token_accuracy": 0.789007980376482, |
| "num_tokens": 13774234.0, |
| "step": 2820 |
| }, |
| { |
| "entropy": 0.7714073432609438, |
| "epoch": 1.1320000000000001, |
| "grad_norm": 78.5, |
| "learning_rate": 1.5935555555555558e-05, |
| "loss": 12.5631, |
| "mean_token_accuracy": 0.7835530713200569, |
| "num_tokens": 13823081.0, |
| "step": 2830 |
| }, |
| { |
| "entropy": 0.7799764240160585, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 84.0, |
| "learning_rate": 1.5913333333333332e-05, |
| "loss": 12.4028, |
| "mean_token_accuracy": 0.7882769204676151, |
| "num_tokens": 13869134.0, |
| "step": 2840 |
| }, |
| { |
| "entropy": 0.7280555779114366, |
| "epoch": 1.1400000000000001, |
| "grad_norm": 71.5, |
| "learning_rate": 1.5891111111111113e-05, |
| "loss": 11.865, |
| "mean_token_accuracy": 0.7928812749683857, |
| "num_tokens": 13921188.0, |
| "step": 2850 |
| }, |
| { |
| "entropy": 0.7916178736835718, |
| "epoch": 1.144, |
| "grad_norm": 75.0, |
| "learning_rate": 1.5868888888888888e-05, |
| "loss": 12.5631, |
| "mean_token_accuracy": 0.7834193158894778, |
| "num_tokens": 13973498.0, |
| "step": 2860 |
| }, |
| { |
| "entropy": 0.7807808231562376, |
| "epoch": 1.148, |
| "grad_norm": 59.75, |
| "learning_rate": 1.584666666666667e-05, |
| "loss": 12.422, |
| "mean_token_accuracy": 0.7860423222184181, |
| "num_tokens": 14025537.0, |
| "step": 2870 |
| }, |
| { |
| "entropy": 0.7367991518229247, |
| "epoch": 1.152, |
| "grad_norm": 79.0, |
| "learning_rate": 1.5824444444444446e-05, |
| "loss": 11.8993, |
| "mean_token_accuracy": 0.7923643980175257, |
| "num_tokens": 14078365.0, |
| "step": 2880 |
| }, |
| { |
| "entropy": 0.7456745334900916, |
| "epoch": 1.156, |
| "grad_norm": 62.75, |
| "learning_rate": 1.5802222222222224e-05, |
| "loss": 12.0986, |
| "mean_token_accuracy": 0.7897498739883304, |
| "num_tokens": 14124490.0, |
| "step": 2890 |
| }, |
| { |
| "entropy": 0.7667106468230486, |
| "epoch": 1.16, |
| "grad_norm": 82.0, |
| "learning_rate": 1.578e-05, |
| "loss": 11.9812, |
| "mean_token_accuracy": 0.7933910895138979, |
| "num_tokens": 14171486.0, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.16, |
| "eval_biology_entropy": 0.9394930896759033, |
| "eval_biology_loss": 1.1460448503494263, |
| "eval_biology_mean_token_accuracy": 0.7106060327291489, |
| "eval_biology_num_tokens": 14171486.0, |
| "eval_biology_runtime": 34.2799, |
| "eval_biology_samples_per_second": 14.586, |
| "eval_biology_steps_per_second": 14.586, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.16, |
| "eval_chemistry_entropy": 0.5743012280762195, |
| "eval_chemistry_loss": 0.6900564432144165, |
| "eval_chemistry_mean_token_accuracy": 0.8098011220097542, |
| "eval_chemistry_num_tokens": 14171486.0, |
| "eval_chemistry_runtime": 39.039, |
| "eval_chemistry_samples_per_second": 12.808, |
| "eval_chemistry_steps_per_second": 12.808, |
| "step": 2900 |
| }, |
| { |
| "entropy": 0.7171152418479323, |
| "epoch": 1.164, |
| "grad_norm": 80.0, |
| "learning_rate": 1.575777777777778e-05, |
| "loss": 11.4539, |
| "mean_token_accuracy": 0.8005911096930504, |
| "num_tokens": 14215782.0, |
| "step": 2910 |
| }, |
| { |
| "entropy": 0.7433731539174915, |
| "epoch": 1.168, |
| "grad_norm": 80.0, |
| "learning_rate": 1.5735555555555557e-05, |
| "loss": 11.8894, |
| "mean_token_accuracy": 0.7935665342956781, |
| "num_tokens": 14266831.0, |
| "step": 2920 |
| }, |
| { |
| "entropy": 0.7337719526141881, |
| "epoch": 1.172, |
| "grad_norm": 82.0, |
| "learning_rate": 1.5713333333333334e-05, |
| "loss": 11.8059, |
| "mean_token_accuracy": 0.7947728868573904, |
| "num_tokens": 14318790.0, |
| "step": 2930 |
| }, |
| { |
| "entropy": 0.7904382789507508, |
| "epoch": 1.176, |
| "grad_norm": 68.0, |
| "learning_rate": 1.5691111111111112e-05, |
| "loss": 12.7704, |
| "mean_token_accuracy": 0.7790846366435289, |
| "num_tokens": 14370559.0, |
| "step": 2940 |
| }, |
| { |
| "entropy": 0.7636601082980633, |
| "epoch": 1.18, |
| "grad_norm": 72.0, |
| "learning_rate": 1.5668888888888893e-05, |
| "loss": 11.9856, |
| "mean_token_accuracy": 0.7933316480368375, |
| "num_tokens": 14417568.0, |
| "step": 2950 |
| }, |
| { |
| "entropy": 0.7631698185577989, |
| "epoch": 1.184, |
| "grad_norm": 79.0, |
| "learning_rate": 1.5646666666666667e-05, |
| "loss": 12.2466, |
| "mean_token_accuracy": 0.7882522609084844, |
| "num_tokens": 14465660.0, |
| "step": 2960 |
| }, |
| { |
| "entropy": 0.733278900757432, |
| "epoch": 1.188, |
| "grad_norm": 70.5, |
| "learning_rate": 1.5624444444444448e-05, |
| "loss": 11.8312, |
| "mean_token_accuracy": 0.7960432037711144, |
| "num_tokens": 14513118.0, |
| "step": 2970 |
| }, |
| { |
| "entropy": 0.738617195468396, |
| "epoch": 1.192, |
| "grad_norm": 76.0, |
| "learning_rate": 1.5602222222222222e-05, |
| "loss": 11.5544, |
| "mean_token_accuracy": 0.8030490033328533, |
| "num_tokens": 14559282.0, |
| "step": 2980 |
| }, |
| { |
| "entropy": 0.7442363886162638, |
| "epoch": 1.196, |
| "grad_norm": 79.0, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 12.3366, |
| "mean_token_accuracy": 0.7844108991324902, |
| "num_tokens": 14608488.0, |
| "step": 2990 |
| }, |
| { |
| "entropy": 0.7417133139446378, |
| "epoch": 1.2, |
| "grad_norm": 82.0, |
| "learning_rate": 1.5557777777777778e-05, |
| "loss": 11.4893, |
| "mean_token_accuracy": 0.7996674500405788, |
| "num_tokens": 14653228.0, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_biology_entropy": 0.9065682877898216, |
| "eval_biology_loss": 1.1520637273788452, |
| "eval_biology_mean_token_accuracy": 0.7110448713302613, |
| "eval_biology_num_tokens": 14653228.0, |
| "eval_biology_runtime": 34.3329, |
| "eval_biology_samples_per_second": 14.563, |
| "eval_biology_steps_per_second": 14.563, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_chemistry_entropy": 0.5510794932246208, |
| "eval_chemistry_loss": 0.6953387260437012, |
| "eval_chemistry_mean_token_accuracy": 0.8099227856397628, |
| "eval_chemistry_num_tokens": 14653228.0, |
| "eval_chemistry_runtime": 39.0125, |
| "eval_chemistry_samples_per_second": 12.816, |
| "eval_chemistry_steps_per_second": 12.816, |
| "step": 3000 |
| }, |
| { |
| "entropy": 0.7173154506832361, |
| "epoch": 1.204, |
| "grad_norm": 72.5, |
| "learning_rate": 1.553555555555556e-05, |
| "loss": 11.7422, |
| "mean_token_accuracy": 0.7940904635936021, |
| "num_tokens": 14704790.0, |
| "step": 3010 |
| }, |
| { |
| "entropy": 0.7326863499358296, |
| "epoch": 1.208, |
| "grad_norm": 77.0, |
| "learning_rate": 1.5513333333333333e-05, |
| "loss": 11.5618, |
| "mean_token_accuracy": 0.7965068083256484, |
| "num_tokens": 14756972.0, |
| "step": 3020 |
| }, |
| { |
| "entropy": 0.796047899313271, |
| "epoch": 1.212, |
| "grad_norm": 74.5, |
| "learning_rate": 1.5491111111111114e-05, |
| "loss": 12.9122, |
| "mean_token_accuracy": 0.7754672013223172, |
| "num_tokens": 14807700.0, |
| "step": 3030 |
| }, |
| { |
| "entropy": 0.7484227666631341, |
| "epoch": 1.216, |
| "grad_norm": 101.0, |
| "learning_rate": 1.546888888888889e-05, |
| "loss": 12.0005, |
| "mean_token_accuracy": 0.7917307186871767, |
| "num_tokens": 14857782.0, |
| "step": 3040 |
| }, |
| { |
| "entropy": 0.7539288274943828, |
| "epoch": 1.22, |
| "grad_norm": 86.0, |
| "learning_rate": 1.544666666666667e-05, |
| "loss": 12.0567, |
| "mean_token_accuracy": 0.7941068332642317, |
| "num_tokens": 14903039.0, |
| "step": 3050 |
| }, |
| { |
| "entropy": 0.7474561285227537, |
| "epoch": 1.224, |
| "grad_norm": 86.5, |
| "learning_rate": 1.5424444444444447e-05, |
| "loss": 11.9094, |
| "mean_token_accuracy": 0.7954557746648788, |
| "num_tokens": 14950496.0, |
| "step": 3060 |
| }, |
| { |
| "entropy": 0.7144173635169864, |
| "epoch": 1.228, |
| "grad_norm": 76.0, |
| "learning_rate": 1.5402222222222224e-05, |
| "loss": 11.4602, |
| "mean_token_accuracy": 0.8011550813913345, |
| "num_tokens": 14997150.0, |
| "step": 3070 |
| }, |
| { |
| "entropy": 0.7608870290219784, |
| "epoch": 1.232, |
| "grad_norm": 64.5, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 12.1597, |
| "mean_token_accuracy": 0.7873735353350639, |
| "num_tokens": 15047356.0, |
| "step": 3080 |
| }, |
| { |
| "entropy": 0.734708639420569, |
| "epoch": 1.236, |
| "grad_norm": 70.0, |
| "learning_rate": 1.535777777777778e-05, |
| "loss": 11.8875, |
| "mean_token_accuracy": 0.7930993035435676, |
| "num_tokens": 15095292.0, |
| "step": 3090 |
| }, |
| { |
| "entropy": 0.7551441323012114, |
| "epoch": 1.24, |
| "grad_norm": 91.0, |
| "learning_rate": 1.5335555555555557e-05, |
| "loss": 12.041, |
| "mean_token_accuracy": 0.790062152966857, |
| "num_tokens": 15142592.0, |
| "step": 3100 |
| }, |
| { |
| "epoch": 1.24, |
| "eval_biology_entropy": 0.924182718038559, |
| "eval_biology_loss": 1.1492804288864136, |
| "eval_biology_mean_token_accuracy": 0.7098891415596008, |
| "eval_biology_num_tokens": 15142592.0, |
| "eval_biology_runtime": 34.371, |
| "eval_biology_samples_per_second": 14.547, |
| "eval_biology_steps_per_second": 14.547, |
| "step": 3100 |
| }, |
| { |
| "epoch": 1.24, |
| "eval_chemistry_entropy": 0.556536919593811, |
| "eval_chemistry_loss": 0.694181501865387, |
| "eval_chemistry_mean_token_accuracy": 0.8098216371536255, |
| "eval_chemistry_num_tokens": 15142592.0, |
| "eval_chemistry_runtime": 39.1099, |
| "eval_chemistry_samples_per_second": 12.784, |
| "eval_chemistry_steps_per_second": 12.784, |
| "step": 3100 |
| }, |
| { |
| "entropy": 0.7333076007664203, |
| "epoch": 1.244, |
| "grad_norm": 95.5, |
| "learning_rate": 1.5313333333333335e-05, |
| "loss": 11.7647, |
| "mean_token_accuracy": 0.7940379984676837, |
| "num_tokens": 15191931.0, |
| "step": 3110 |
| }, |
| { |
| "entropy": 0.767120277043432, |
| "epoch": 1.248, |
| "grad_norm": 79.0, |
| "learning_rate": 1.5291111111111112e-05, |
| "loss": 12.3582, |
| "mean_token_accuracy": 0.7837318338453769, |
| "num_tokens": 15241098.0, |
| "step": 3120 |
| }, |
| { |
| "entropy": 0.7576422481797636, |
| "epoch": 1.252, |
| "grad_norm": 78.0, |
| "learning_rate": 1.526888888888889e-05, |
| "loss": 12.1685, |
| "mean_token_accuracy": 0.7920863214880228, |
| "num_tokens": 15289709.0, |
| "step": 3130 |
| }, |
| { |
| "entropy": 0.8195444407872856, |
| "epoch": 1.256, |
| "grad_norm": 97.0, |
| "learning_rate": 1.5246666666666668e-05, |
| "loss": 13.396, |
| "mean_token_accuracy": 0.7758128978312016, |
| "num_tokens": 15335851.0, |
| "step": 3140 |
| }, |
| { |
| "entropy": 0.7622947560623288, |
| "epoch": 1.26, |
| "grad_norm": 82.5, |
| "learning_rate": 1.5224444444444447e-05, |
| "loss": 12.064, |
| "mean_token_accuracy": 0.7894668418914079, |
| "num_tokens": 15389881.0, |
| "step": 3150 |
| }, |
| { |
| "entropy": 0.7444451101124286, |
| "epoch": 1.264, |
| "grad_norm": 70.5, |
| "learning_rate": 1.5202222222222223e-05, |
| "loss": 11.6949, |
| "mean_token_accuracy": 0.7952967748045922, |
| "num_tokens": 15437657.0, |
| "step": 3160 |
| }, |
| { |
| "entropy": 0.7540538094937801, |
| "epoch": 1.268, |
| "grad_norm": 75.0, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 12.3735, |
| "mean_token_accuracy": 0.7883003624156117, |
| "num_tokens": 15484934.0, |
| "step": 3170 |
| }, |
| { |
| "entropy": 0.8269421357661486, |
| "epoch": 1.272, |
| "grad_norm": 72.5, |
| "learning_rate": 1.5157777777777778e-05, |
| "loss": 13.0126, |
| "mean_token_accuracy": 0.7773631617426873, |
| "num_tokens": 15530753.0, |
| "step": 3180 |
| }, |
| { |
| "entropy": 0.787413956783712, |
| "epoch": 1.276, |
| "grad_norm": 77.0, |
| "learning_rate": 1.5135555555555557e-05, |
| "loss": 12.7275, |
| "mean_token_accuracy": 0.7816727485507726, |
| "num_tokens": 15581515.0, |
| "step": 3190 |
| }, |
| { |
| "entropy": 0.7663712141104042, |
| "epoch": 1.28, |
| "grad_norm": 82.0, |
| "learning_rate": 1.5113333333333335e-05, |
| "loss": 12.1255, |
| "mean_token_accuracy": 0.788918101042509, |
| "num_tokens": 15630795.0, |
| "step": 3200 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_biology_entropy": 0.9539193958640099, |
| "eval_biology_loss": 1.1404262781143188, |
| "eval_biology_mean_token_accuracy": 0.7117404421567917, |
| "eval_biology_num_tokens": 15630795.0, |
| "eval_biology_runtime": 34.3839, |
| "eval_biology_samples_per_second": 14.542, |
| "eval_biology_steps_per_second": 14.542, |
| "step": 3200 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_chemistry_entropy": 0.5714361243247986, |
| "eval_chemistry_loss": 0.6895456910133362, |
| "eval_chemistry_mean_token_accuracy": 0.8099247926473617, |
| "eval_chemistry_num_tokens": 15630795.0, |
| "eval_chemistry_runtime": 39.0958, |
| "eval_chemistry_samples_per_second": 12.789, |
| "eval_chemistry_steps_per_second": 12.789, |
| "step": 3200 |
| }, |
| { |
| "entropy": 0.8007632769644261, |
| "epoch": 1.284, |
| "grad_norm": 86.5, |
| "learning_rate": 1.5091111111111113e-05, |
| "loss": 12.6561, |
| "mean_token_accuracy": 0.7856924194842577, |
| "num_tokens": 15676968.0, |
| "step": 3210 |
| }, |
| { |
| "entropy": 0.7722161913290619, |
| "epoch": 1.288, |
| "grad_norm": 65.0, |
| "learning_rate": 1.506888888888889e-05, |
| "loss": 12.677, |
| "mean_token_accuracy": 0.7808508489280939, |
| "num_tokens": 15725428.0, |
| "step": 3220 |
| }, |
| { |
| "entropy": 0.7684616031125188, |
| "epoch": 1.292, |
| "grad_norm": 72.5, |
| "learning_rate": 1.5046666666666668e-05, |
| "loss": 12.0247, |
| "mean_token_accuracy": 0.7897633019834757, |
| "num_tokens": 15776797.0, |
| "step": 3230 |
| }, |
| { |
| "entropy": 0.7860913366079331, |
| "epoch": 1.296, |
| "grad_norm": 70.5, |
| "learning_rate": 1.5024444444444445e-05, |
| "loss": 12.7967, |
| "mean_token_accuracy": 0.779197471216321, |
| "num_tokens": 15827105.0, |
| "step": 3240 |
| }, |
| { |
| "entropy": 0.7869927534833551, |
| "epoch": 1.3, |
| "grad_norm": 72.5, |
| "learning_rate": 1.5002222222222223e-05, |
| "loss": 12.3398, |
| "mean_token_accuracy": 0.7877550899982453, |
| "num_tokens": 15874120.0, |
| "step": 3250 |
| }, |
| { |
| "entropy": 0.7546575669199228, |
| "epoch": 1.304, |
| "grad_norm": 72.0, |
| "learning_rate": 1.498e-05, |
| "loss": 12.3919, |
| "mean_token_accuracy": 0.7849195256829262, |
| "num_tokens": 15923902.0, |
| "step": 3260 |
| }, |
| { |
| "entropy": 0.726429769769311, |
| "epoch": 1.308, |
| "grad_norm": 81.5, |
| "learning_rate": 1.495777777777778e-05, |
| "loss": 11.2365, |
| "mean_token_accuracy": 0.8021128010004759, |
| "num_tokens": 15972188.0, |
| "step": 3270 |
| }, |
| { |
| "entropy": 0.7616990976035595, |
| "epoch": 1.312, |
| "grad_norm": 87.0, |
| "learning_rate": 1.4935555555555556e-05, |
| "loss": 12.7199, |
| "mean_token_accuracy": 0.7811474315822124, |
| "num_tokens": 16019645.0, |
| "step": 3280 |
| }, |
| { |
| "entropy": 0.78043591119349, |
| "epoch": 1.316, |
| "grad_norm": 74.0, |
| "learning_rate": 1.4913333333333335e-05, |
| "loss": 12.1116, |
| "mean_token_accuracy": 0.7909189753234387, |
| "num_tokens": 16067720.0, |
| "step": 3290 |
| }, |
| { |
| "entropy": 0.7396340952254832, |
| "epoch": 1.32, |
| "grad_norm": 69.0, |
| "learning_rate": 1.4891111111111111e-05, |
| "loss": 11.8019, |
| "mean_token_accuracy": 0.7915644906461239, |
| "num_tokens": 16115487.0, |
| "step": 3300 |
| }, |
| { |
| "epoch": 1.32, |
| "eval_biology_entropy": 0.9063211707472801, |
| "eval_biology_loss": 1.1506577730178833, |
| "eval_biology_mean_token_accuracy": 0.7104792315363884, |
| "eval_biology_num_tokens": 16115487.0, |
| "eval_biology_runtime": 34.332, |
| "eval_biology_samples_per_second": 14.564, |
| "eval_biology_steps_per_second": 14.564, |
| "step": 3300 |
| }, |
| { |
| "epoch": 1.32, |
| "eval_chemistry_entropy": 0.5483866376578808, |
| "eval_chemistry_loss": 0.6969625353813171, |
| "eval_chemistry_mean_token_accuracy": 0.8089957860708237, |
| "eval_chemistry_num_tokens": 16115487.0, |
| "eval_chemistry_runtime": 39.0607, |
| "eval_chemistry_samples_per_second": 12.801, |
| "eval_chemistry_steps_per_second": 12.801, |
| "step": 3300 |
| }, |
| { |
| "entropy": 0.7483121111989022, |
| "epoch": 1.324, |
| "grad_norm": 71.5, |
| "learning_rate": 1.486888888888889e-05, |
| "loss": 12.5091, |
| "mean_token_accuracy": 0.7855912052094937, |
| "num_tokens": 16166663.0, |
| "step": 3310 |
| }, |
| { |
| "entropy": 0.800167062971741, |
| "epoch": 1.328, |
| "grad_norm": 72.0, |
| "learning_rate": 1.4846666666666666e-05, |
| "loss": 12.4624, |
| "mean_token_accuracy": 0.7831843540072441, |
| "num_tokens": 16221726.0, |
| "step": 3320 |
| }, |
| { |
| "entropy": 0.7649429302662611, |
| "epoch": 1.332, |
| "grad_norm": 74.5, |
| "learning_rate": 1.4824444444444446e-05, |
| "loss": 12.3579, |
| "mean_token_accuracy": 0.7824284922331571, |
| "num_tokens": 16274398.0, |
| "step": 3330 |
| }, |
| { |
| "entropy": 0.7414550604298711, |
| "epoch": 1.336, |
| "grad_norm": 87.5, |
| "learning_rate": 1.4802222222222225e-05, |
| "loss": 11.5613, |
| "mean_token_accuracy": 0.7962801866233349, |
| "num_tokens": 16322791.0, |
| "step": 3340 |
| }, |
| { |
| "entropy": 0.736030207760632, |
| "epoch": 1.34, |
| "grad_norm": 75.0, |
| "learning_rate": 1.478e-05, |
| "loss": 12.2629, |
| "mean_token_accuracy": 0.7882518745958805, |
| "num_tokens": 16372544.0, |
| "step": 3350 |
| }, |
| { |
| "entropy": 0.7831617563962936, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 69.5, |
| "learning_rate": 1.475777777777778e-05, |
| "loss": 12.291, |
| "mean_token_accuracy": 0.7873957261443139, |
| "num_tokens": 16427594.0, |
| "step": 3360 |
| }, |
| { |
| "entropy": 0.7723853016272187, |
| "epoch": 1.3479999999999999, |
| "grad_norm": 74.0, |
| "learning_rate": 1.4735555555555556e-05, |
| "loss": 12.1361, |
| "mean_token_accuracy": 0.7902800682932138, |
| "num_tokens": 16479616.0, |
| "step": 3370 |
| }, |
| { |
| "entropy": 0.7376682225614786, |
| "epoch": 1.3519999999999999, |
| "grad_norm": 79.5, |
| "learning_rate": 1.4713333333333335e-05, |
| "loss": 12.2972, |
| "mean_token_accuracy": 0.7924885641783476, |
| "num_tokens": 16527485.0, |
| "step": 3380 |
| }, |
| { |
| "entropy": 0.7550094440579415, |
| "epoch": 1.3559999999999999, |
| "grad_norm": 75.0, |
| "learning_rate": 1.4691111111111111e-05, |
| "loss": 11.626, |
| "mean_token_accuracy": 0.7963223490864039, |
| "num_tokens": 16574098.0, |
| "step": 3390 |
| }, |
| { |
| "entropy": 0.7329300165176391, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 87.5, |
| "learning_rate": 1.466888888888889e-05, |
| "loss": 11.8901, |
| "mean_token_accuracy": 0.7913072779774666, |
| "num_tokens": 16621605.0, |
| "step": 3400 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "eval_biology_entropy": 0.9236152126789093, |
| "eval_biology_loss": 1.1458607912063599, |
| "eval_biology_mean_token_accuracy": 0.7115779925584793, |
| "eval_biology_num_tokens": 16621605.0, |
| "eval_biology_runtime": 34.2893, |
| "eval_biology_samples_per_second": 14.582, |
| "eval_biology_steps_per_second": 14.582, |
| "step": 3400 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "eval_chemistry_entropy": 0.5512723879814148, |
| "eval_chemistry_loss": 0.6975026726722717, |
| "eval_chemistry_mean_token_accuracy": 0.8087222777605056, |
| "eval_chemistry_num_tokens": 16621605.0, |
| "eval_chemistry_runtime": 39.0577, |
| "eval_chemistry_samples_per_second": 12.802, |
| "eval_chemistry_steps_per_second": 12.802, |
| "step": 3400 |
| }, |
| { |
| "entropy": 0.7736082326620817, |
| "epoch": 1.3639999999999999, |
| "grad_norm": 92.0, |
| "learning_rate": 1.4646666666666666e-05, |
| "loss": 12.517, |
| "mean_token_accuracy": 0.7830415267497302, |
| "num_tokens": 16669983.0, |
| "step": 3410 |
| }, |
| { |
| "entropy": 0.7838009294122458, |
| "epoch": 1.3679999999999999, |
| "grad_norm": 71.0, |
| "learning_rate": 1.4624444444444446e-05, |
| "loss": 12.3122, |
| "mean_token_accuracy": 0.7880257587879896, |
| "num_tokens": 16719005.0, |
| "step": 3420 |
| }, |
| { |
| "entropy": 0.7837223557755351, |
| "epoch": 1.3719999999999999, |
| "grad_norm": 78.5, |
| "learning_rate": 1.4602222222222225e-05, |
| "loss": 12.8381, |
| "mean_token_accuracy": 0.7803949814289808, |
| "num_tokens": 16766964.0, |
| "step": 3430 |
| }, |
| { |
| "entropy": 0.791265351139009, |
| "epoch": 1.376, |
| "grad_norm": 82.0, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 12.3811, |
| "mean_token_accuracy": 0.7851585242897272, |
| "num_tokens": 16813444.0, |
| "step": 3440 |
| }, |
| { |
| "entropy": 0.7721597425639629, |
| "epoch": 1.38, |
| "grad_norm": 66.0, |
| "learning_rate": 1.455777777777778e-05, |
| "loss": 12.4334, |
| "mean_token_accuracy": 0.7842074308544398, |
| "num_tokens": 16859803.0, |
| "step": 3450 |
| }, |
| { |
| "entropy": 0.745121730491519, |
| "epoch": 1.384, |
| "grad_norm": 89.5, |
| "learning_rate": 1.4535555555555556e-05, |
| "loss": 12.0252, |
| "mean_token_accuracy": 0.7905770123004914, |
| "num_tokens": 16908557.0, |
| "step": 3460 |
| }, |
| { |
| "entropy": 0.7405239475890995, |
| "epoch": 1.388, |
| "grad_norm": 77.5, |
| "learning_rate": 1.4513333333333336e-05, |
| "loss": 11.9053, |
| "mean_token_accuracy": 0.7950120761990547, |
| "num_tokens": 16955915.0, |
| "step": 3470 |
| }, |
| { |
| "entropy": 0.7362896759063006, |
| "epoch": 1.392, |
| "grad_norm": 73.0, |
| "learning_rate": 1.4491111111111111e-05, |
| "loss": 11.9347, |
| "mean_token_accuracy": 0.7917971923947335, |
| "num_tokens": 17006509.0, |
| "step": 3480 |
| }, |
| { |
| "entropy": 0.7798717721365392, |
| "epoch": 1.396, |
| "grad_norm": 77.0, |
| "learning_rate": 1.446888888888889e-05, |
| "loss": 12.2187, |
| "mean_token_accuracy": 0.7846463203430176, |
| "num_tokens": 17055210.0, |
| "step": 3490 |
| }, |
| { |
| "entropy": 0.7319062616676092, |
| "epoch": 1.4, |
| "grad_norm": 84.0, |
| "learning_rate": 1.4446666666666668e-05, |
| "loss": 11.9169, |
| "mean_token_accuracy": 0.7976263463497162, |
| "num_tokens": 17097916.0, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.4, |
| "eval_biology_entropy": 0.9568568287491799, |
| "eval_biology_loss": 1.1350637674331665, |
| "eval_biology_mean_token_accuracy": 0.712535469353199, |
| "eval_biology_num_tokens": 17097916.0, |
| "eval_biology_runtime": 34.3522, |
| "eval_biology_samples_per_second": 14.555, |
| "eval_biology_steps_per_second": 14.555, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.4, |
| "eval_chemistry_entropy": 0.5817427493333817, |
| "eval_chemistry_loss": 0.691250741481781, |
| "eval_chemistry_mean_token_accuracy": 0.8094408006668091, |
| "eval_chemistry_num_tokens": 17097916.0, |
| "eval_chemistry_runtime": 38.9444, |
| "eval_chemistry_samples_per_second": 12.839, |
| "eval_chemistry_steps_per_second": 12.839, |
| "step": 3500 |
| }, |
| { |
| "entropy": 0.8049037892371416, |
| "epoch": 1.404, |
| "grad_norm": 72.5, |
| "learning_rate": 1.4424444444444446e-05, |
| "loss": 12.7158, |
| "mean_token_accuracy": 0.779417271912098, |
| "num_tokens": 17144457.0, |
| "step": 3510 |
| }, |
| { |
| "entropy": 0.7369912270456552, |
| "epoch": 1.408, |
| "grad_norm": 78.5, |
| "learning_rate": 1.4402222222222224e-05, |
| "loss": 11.9284, |
| "mean_token_accuracy": 0.7920796353369951, |
| "num_tokens": 17197870.0, |
| "step": 3520 |
| }, |
| { |
| "entropy": 0.7565042628906667, |
| "epoch": 1.412, |
| "grad_norm": 72.0, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 12.001, |
| "mean_token_accuracy": 0.7880884870886803, |
| "num_tokens": 17248845.0, |
| "step": 3530 |
| }, |
| { |
| "entropy": 0.7904578883200883, |
| "epoch": 1.416, |
| "grad_norm": 81.5, |
| "learning_rate": 1.4357777777777779e-05, |
| "loss": 12.6421, |
| "mean_token_accuracy": 0.783636374771595, |
| "num_tokens": 17296950.0, |
| "step": 3540 |
| }, |
| { |
| "entropy": 0.7670844430103898, |
| "epoch": 1.42, |
| "grad_norm": 74.5, |
| "learning_rate": 1.4335555555555556e-05, |
| "loss": 12.3219, |
| "mean_token_accuracy": 0.7852133817970752, |
| "num_tokens": 17345729.0, |
| "step": 3550 |
| }, |
| { |
| "entropy": 0.7857215842232108, |
| "epoch": 1.424, |
| "grad_norm": 74.5, |
| "learning_rate": 1.4313333333333334e-05, |
| "loss": 12.5051, |
| "mean_token_accuracy": 0.7824117984622717, |
| "num_tokens": 17394390.0, |
| "step": 3560 |
| }, |
| { |
| "entropy": 0.7532725343480706, |
| "epoch": 1.428, |
| "grad_norm": 71.0, |
| "learning_rate": 1.4291111111111113e-05, |
| "loss": 12.1841, |
| "mean_token_accuracy": 0.7876898244023323, |
| "num_tokens": 17445914.0, |
| "step": 3570 |
| }, |
| { |
| "entropy": 0.7417794173583389, |
| "epoch": 1.432, |
| "grad_norm": 57.75, |
| "learning_rate": 1.426888888888889e-05, |
| "loss": 11.6384, |
| "mean_token_accuracy": 0.7950344871729612, |
| "num_tokens": 17498287.0, |
| "step": 3580 |
| }, |
| { |
| "entropy": 0.7609250320121645, |
| "epoch": 1.436, |
| "grad_norm": 77.5, |
| "learning_rate": 1.4246666666666669e-05, |
| "loss": 12.4924, |
| "mean_token_accuracy": 0.7851621191948652, |
| "num_tokens": 17541465.0, |
| "step": 3590 |
| }, |
| { |
| "entropy": 0.7596838189288974, |
| "epoch": 1.44, |
| "grad_norm": 82.5, |
| "learning_rate": 1.4224444444444445e-05, |
| "loss": 12.1016, |
| "mean_token_accuracy": 0.7890080321580172, |
| "num_tokens": 17587777.0, |
| "step": 3600 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_biology_entropy": 0.9378208144903183, |
| "eval_biology_loss": 1.1354689598083496, |
| "eval_biology_mean_token_accuracy": 0.7127703504562378, |
| "eval_biology_num_tokens": 17587777.0, |
| "eval_biology_runtime": 34.3536, |
| "eval_biology_samples_per_second": 14.555, |
| "eval_biology_steps_per_second": 14.555, |
| "step": 3600 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_chemistry_entropy": 0.5693562558591366, |
| "eval_chemistry_loss": 0.6909820437431335, |
| "eval_chemistry_mean_token_accuracy": 0.8099824855327606, |
| "eval_chemistry_num_tokens": 17587777.0, |
| "eval_chemistry_runtime": 39.2129, |
| "eval_chemistry_samples_per_second": 12.751, |
| "eval_chemistry_steps_per_second": 12.751, |
| "step": 3600 |
| }, |
| { |
| "entropy": 0.7871572238393127, |
| "epoch": 1.444, |
| "grad_norm": 71.0, |
| "learning_rate": 1.4202222222222224e-05, |
| "loss": 12.7225, |
| "mean_token_accuracy": 0.7807210251688957, |
| "num_tokens": 17637532.0, |
| "step": 3610 |
| }, |
| { |
| "entropy": 0.7543072551488876, |
| "epoch": 1.448, |
| "grad_norm": 85.0, |
| "learning_rate": 1.418e-05, |
| "loss": 11.8324, |
| "mean_token_accuracy": 0.793193769082427, |
| "num_tokens": 17688697.0, |
| "step": 3620 |
| }, |
| { |
| "entropy": 0.739030692819506, |
| "epoch": 1.452, |
| "grad_norm": 78.0, |
| "learning_rate": 1.4157777777777779e-05, |
| "loss": 12.0811, |
| "mean_token_accuracy": 0.787498576566577, |
| "num_tokens": 17738855.0, |
| "step": 3630 |
| }, |
| { |
| "entropy": 0.7912683079019189, |
| "epoch": 1.456, |
| "grad_norm": 75.0, |
| "learning_rate": 1.4135555555555555e-05, |
| "loss": 12.6028, |
| "mean_token_accuracy": 0.7839001085609197, |
| "num_tokens": 17788456.0, |
| "step": 3640 |
| }, |
| { |
| "entropy": 0.7713869358412921, |
| "epoch": 1.46, |
| "grad_norm": 75.0, |
| "learning_rate": 1.4113333333333334e-05, |
| "loss": 12.4651, |
| "mean_token_accuracy": 0.7847725540399552, |
| "num_tokens": 17834237.0, |
| "step": 3650 |
| }, |
| { |
| "entropy": 0.7707416551187635, |
| "epoch": 1.464, |
| "grad_norm": 74.5, |
| "learning_rate": 1.4091111111111114e-05, |
| "loss": 12.2002, |
| "mean_token_accuracy": 0.7896449971944094, |
| "num_tokens": 17882731.0, |
| "step": 3660 |
| }, |
| { |
| "entropy": 0.742984014749527, |
| "epoch": 1.468, |
| "grad_norm": 71.0, |
| "learning_rate": 1.406888888888889e-05, |
| "loss": 12.1166, |
| "mean_token_accuracy": 0.7901315592229367, |
| "num_tokens": 17932183.0, |
| "step": 3670 |
| }, |
| { |
| "entropy": 0.7615336345508694, |
| "epoch": 1.472, |
| "grad_norm": 75.0, |
| "learning_rate": 1.4046666666666669e-05, |
| "loss": 11.668, |
| "mean_token_accuracy": 0.7966230910271406, |
| "num_tokens": 17984063.0, |
| "step": 3680 |
| }, |
| { |
| "entropy": 0.7279147446155548, |
| "epoch": 1.476, |
| "grad_norm": 74.0, |
| "learning_rate": 1.4024444444444445e-05, |
| "loss": 12.1293, |
| "mean_token_accuracy": 0.7888615641742944, |
| "num_tokens": 18032337.0, |
| "step": 3690 |
| }, |
| { |
| "entropy": 0.7246495549567044, |
| "epoch": 1.48, |
| "grad_norm": 74.0, |
| "learning_rate": 1.4002222222222224e-05, |
| "loss": 11.5785, |
| "mean_token_accuracy": 0.7983909532427788, |
| "num_tokens": 18080829.0, |
| "step": 3700 |
| }, |
| { |
| "epoch": 1.48, |
| "eval_biology_entropy": 0.9344790092706681, |
| "eval_biology_loss": 1.1373109817504883, |
| "eval_biology_mean_token_accuracy": 0.7122904448509216, |
| "eval_biology_num_tokens": 18080829.0, |
| "eval_biology_runtime": 34.3207, |
| "eval_biology_samples_per_second": 14.568, |
| "eval_biology_steps_per_second": 14.568, |
| "step": 3700 |
| }, |
| { |
| "epoch": 1.48, |
| "eval_chemistry_entropy": 0.5660818241536617, |
| "eval_chemistry_loss": 0.689975917339325, |
| "eval_chemistry_mean_token_accuracy": 0.8103064861297608, |
| "eval_chemistry_num_tokens": 18080829.0, |
| "eval_chemistry_runtime": 39.0696, |
| "eval_chemistry_samples_per_second": 12.798, |
| "eval_chemistry_steps_per_second": 12.798, |
| "step": 3700 |
| }, |
| { |
| "entropy": 0.7652681648731232, |
| "epoch": 1.484, |
| "grad_norm": 67.0, |
| "learning_rate": 1.398e-05, |
| "loss": 12.1816, |
| "mean_token_accuracy": 0.7897872734814883, |
| "num_tokens": 18127673.0, |
| "step": 3710 |
| }, |
| { |
| "entropy": 0.7631257550790906, |
| "epoch": 1.488, |
| "grad_norm": 77.5, |
| "learning_rate": 1.395777777777778e-05, |
| "loss": 12.284, |
| "mean_token_accuracy": 0.7913855224847793, |
| "num_tokens": 18175640.0, |
| "step": 3720 |
| }, |
| { |
| "entropy": 0.8054220713675022, |
| "epoch": 1.492, |
| "grad_norm": 73.5, |
| "learning_rate": 1.3935555555555557e-05, |
| "loss": 12.6798, |
| "mean_token_accuracy": 0.7794890634715557, |
| "num_tokens": 18222380.0, |
| "step": 3730 |
| }, |
| { |
| "entropy": 0.7389599463902414, |
| "epoch": 1.496, |
| "grad_norm": 78.5, |
| "learning_rate": 1.3913333333333335e-05, |
| "loss": 11.9253, |
| "mean_token_accuracy": 0.7916713930666447, |
| "num_tokens": 18268965.0, |
| "step": 3740 |
| }, |
| { |
| "entropy": 0.7846246063709259, |
| "epoch": 1.5, |
| "grad_norm": 72.5, |
| "learning_rate": 1.3891111111111114e-05, |
| "loss": 12.4666, |
| "mean_token_accuracy": 0.7847031205892563, |
| "num_tokens": 18320455.0, |
| "step": 3750 |
| }, |
| { |
| "entropy": 0.7640897994861007, |
| "epoch": 1.504, |
| "grad_norm": 76.5, |
| "learning_rate": 1.386888888888889e-05, |
| "loss": 12.275, |
| "mean_token_accuracy": 0.7885676752775908, |
| "num_tokens": 18367857.0, |
| "step": 3760 |
| }, |
| { |
| "entropy": 0.7773599276319146, |
| "epoch": 1.508, |
| "grad_norm": 72.5, |
| "learning_rate": 1.3846666666666669e-05, |
| "loss": 12.5182, |
| "mean_token_accuracy": 0.7810276404023171, |
| "num_tokens": 18419822.0, |
| "step": 3770 |
| }, |
| { |
| "entropy": 0.7846800515428185, |
| "epoch": 1.512, |
| "grad_norm": 73.5, |
| "learning_rate": 1.3824444444444445e-05, |
| "loss": 12.5421, |
| "mean_token_accuracy": 0.7808643691241741, |
| "num_tokens": 18468841.0, |
| "step": 3780 |
| }, |
| { |
| "entropy": 0.8059901356697082, |
| "epoch": 1.516, |
| "grad_norm": 76.5, |
| "learning_rate": 1.3802222222222224e-05, |
| "loss": 13.0769, |
| "mean_token_accuracy": 0.7769437693059444, |
| "num_tokens": 18517838.0, |
| "step": 3790 |
| }, |
| { |
| "entropy": 0.7683778650127351, |
| "epoch": 1.52, |
| "grad_norm": 82.0, |
| "learning_rate": 1.378e-05, |
| "loss": 12.053, |
| "mean_token_accuracy": 0.7904394565150141, |
| "num_tokens": 18569146.0, |
| "step": 3800 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_biology_entropy": 0.9397989763021469, |
| "eval_biology_loss": 1.1334812641143799, |
| "eval_biology_mean_token_accuracy": 0.7136930270195008, |
| "eval_biology_num_tokens": 18569146.0, |
| "eval_biology_runtime": 34.3392, |
| "eval_biology_samples_per_second": 14.561, |
| "eval_biology_steps_per_second": 14.561, |
| "step": 3800 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_chemistry_entropy": 0.5740775786340236, |
| "eval_chemistry_loss": 0.6921254992485046, |
| "eval_chemistry_mean_token_accuracy": 0.8090926426053047, |
| "eval_chemistry_num_tokens": 18569146.0, |
| "eval_chemistry_runtime": 39.1088, |
| "eval_chemistry_samples_per_second": 12.785, |
| "eval_chemistry_steps_per_second": 12.785, |
| "step": 3800 |
| }, |
| { |
| "entropy": 0.7832599593326449, |
| "epoch": 1.524, |
| "grad_norm": 80.5, |
| "learning_rate": 1.375777777777778e-05, |
| "loss": 12.6259, |
| "mean_token_accuracy": 0.7822577580809593, |
| "num_tokens": 18614461.0, |
| "step": 3810 |
| }, |
| { |
| "entropy": 0.7467021363787353, |
| "epoch": 1.528, |
| "grad_norm": 75.5, |
| "learning_rate": 1.3735555555555557e-05, |
| "loss": 11.8916, |
| "mean_token_accuracy": 0.7923420272767544, |
| "num_tokens": 18661379.0, |
| "step": 3820 |
| }, |
| { |
| "entropy": 0.7401287262327969, |
| "epoch": 1.532, |
| "grad_norm": 85.0, |
| "learning_rate": 1.3713333333333335e-05, |
| "loss": 12.0491, |
| "mean_token_accuracy": 0.7901012677699327, |
| "num_tokens": 18708652.0, |
| "step": 3830 |
| }, |
| { |
| "entropy": 0.7989715838804841, |
| "epoch": 1.536, |
| "grad_norm": 76.5, |
| "learning_rate": 1.3691111111111112e-05, |
| "loss": 12.6193, |
| "mean_token_accuracy": 0.784449428319931, |
| "num_tokens": 18755079.0, |
| "step": 3840 |
| }, |
| { |
| "entropy": 0.7329860411584377, |
| "epoch": 1.54, |
| "grad_norm": 70.5, |
| "learning_rate": 1.366888888888889e-05, |
| "loss": 11.8698, |
| "mean_token_accuracy": 0.7935557685792446, |
| "num_tokens": 18805765.0, |
| "step": 3850 |
| }, |
| { |
| "entropy": 0.7509209958836436, |
| "epoch": 1.544, |
| "grad_norm": 73.0, |
| "learning_rate": 1.3646666666666668e-05, |
| "loss": 11.9954, |
| "mean_token_accuracy": 0.7929477907717228, |
| "num_tokens": 18855275.0, |
| "step": 3860 |
| }, |
| { |
| "entropy": 0.7660559002310038, |
| "epoch": 1.548, |
| "grad_norm": 65.5, |
| "learning_rate": 1.3624444444444445e-05, |
| "loss": 12.2408, |
| "mean_token_accuracy": 0.7859826818108558, |
| "num_tokens": 18907327.0, |
| "step": 3870 |
| }, |
| { |
| "entropy": 0.7669518861919642, |
| "epoch": 1.552, |
| "grad_norm": 82.5, |
| "learning_rate": 1.3602222222222223e-05, |
| "loss": 12.1744, |
| "mean_token_accuracy": 0.7876989297568798, |
| "num_tokens": 18956248.0, |
| "step": 3880 |
| }, |
| { |
| "entropy": 0.708898575976491, |
| "epoch": 1.556, |
| "grad_norm": 77.5, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 11.3492, |
| "mean_token_accuracy": 0.8000110641121865, |
| "num_tokens": 19004227.0, |
| "step": 3890 |
| }, |
| { |
| "entropy": 0.7402307221665978, |
| "epoch": 1.56, |
| "grad_norm": 65.0, |
| "learning_rate": 1.3557777777777778e-05, |
| "loss": 12.0553, |
| "mean_token_accuracy": 0.790681303665042, |
| "num_tokens": 19051957.0, |
| "step": 3900 |
| }, |
| { |
| "epoch": 1.56, |
| "eval_biology_entropy": 0.9643143783807755, |
| "eval_biology_loss": 1.1253645420074463, |
| "eval_biology_mean_token_accuracy": 0.7130407758355141, |
| "eval_biology_num_tokens": 19051957.0, |
| "eval_biology_runtime": 34.3329, |
| "eval_biology_samples_per_second": 14.563, |
| "eval_biology_steps_per_second": 14.563, |
| "step": 3900 |
| }, |
| { |
| "epoch": 1.56, |
| "eval_chemistry_entropy": 0.5940712922215462, |
| "eval_chemistry_loss": 0.6880068182945251, |
| "eval_chemistry_mean_token_accuracy": 0.8089482567310333, |
| "eval_chemistry_num_tokens": 19051957.0, |
| "eval_chemistry_runtime": 39.0493, |
| "eval_chemistry_samples_per_second": 12.804, |
| "eval_chemistry_steps_per_second": 12.804, |
| "step": 3900 |
| }, |
| { |
| "entropy": 0.8025386364199221, |
| "epoch": 1.564, |
| "grad_norm": 73.5, |
| "learning_rate": 1.3535555555555557e-05, |
| "loss": 12.7007, |
| "mean_token_accuracy": 0.7816927436739206, |
| "num_tokens": 19100798.0, |
| "step": 3910 |
| }, |
| { |
| "entropy": 0.7244784448295831, |
| "epoch": 1.568, |
| "grad_norm": 76.5, |
| "learning_rate": 1.3513333333333333e-05, |
| "loss": 11.6952, |
| "mean_token_accuracy": 0.7933868307620286, |
| "num_tokens": 19150315.0, |
| "step": 3920 |
| }, |
| { |
| "entropy": 0.7755278231576085, |
| "epoch": 1.572, |
| "grad_norm": 70.5, |
| "learning_rate": 1.3491111111111113e-05, |
| "loss": 12.2287, |
| "mean_token_accuracy": 0.785486176237464, |
| "num_tokens": 19203607.0, |
| "step": 3930 |
| }, |
| { |
| "entropy": 0.7854612486436963, |
| "epoch": 1.576, |
| "grad_norm": 94.0, |
| "learning_rate": 1.3468888888888888e-05, |
| "loss": 12.6133, |
| "mean_token_accuracy": 0.7813806410878896, |
| "num_tokens": 19248041.0, |
| "step": 3940 |
| }, |
| { |
| "entropy": 0.7251294530928135, |
| "epoch": 1.58, |
| "grad_norm": 76.0, |
| "learning_rate": 1.3446666666666668e-05, |
| "loss": 11.4986, |
| "mean_token_accuracy": 0.798180129379034, |
| "num_tokens": 19296596.0, |
| "step": 3950 |
| }, |
| { |
| "entropy": 0.7431621631607414, |
| "epoch": 1.584, |
| "grad_norm": 81.0, |
| "learning_rate": 1.3424444444444447e-05, |
| "loss": 12.0534, |
| "mean_token_accuracy": 0.7893318895250558, |
| "num_tokens": 19344260.0, |
| "step": 3960 |
| }, |
| { |
| "entropy": 0.7591244658455253, |
| "epoch": 1.588, |
| "grad_norm": 76.5, |
| "learning_rate": 1.3402222222222223e-05, |
| "loss": 11.9359, |
| "mean_token_accuracy": 0.7914566043764353, |
| "num_tokens": 19391465.0, |
| "step": 3970 |
| }, |
| { |
| "entropy": 0.7654343968257308, |
| "epoch": 1.592, |
| "grad_norm": 76.5, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 12.3681, |
| "mean_token_accuracy": 0.7888023316860199, |
| "num_tokens": 19436040.0, |
| "step": 3980 |
| }, |
| { |
| "entropy": 0.7819727979600429, |
| "epoch": 1.596, |
| "grad_norm": 71.0, |
| "learning_rate": 1.3357777777777778e-05, |
| "loss": 12.5551, |
| "mean_token_accuracy": 0.7826974406838417, |
| "num_tokens": 19485036.0, |
| "step": 3990 |
| }, |
| { |
| "entropy": 0.7443074682727456, |
| "epoch": 1.6, |
| "grad_norm": 69.0, |
| "learning_rate": 1.3335555555555558e-05, |
| "loss": 11.989, |
| "mean_token_accuracy": 0.7934192854911088, |
| "num_tokens": 19532552.0, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_biology_entropy": 0.9328739583492279, |
| "eval_biology_loss": 1.1316684484481812, |
| "eval_biology_mean_token_accuracy": 0.7133588783144951, |
| "eval_biology_num_tokens": 19532552.0, |
| "eval_biology_runtime": 34.3678, |
| "eval_biology_samples_per_second": 14.549, |
| "eval_biology_steps_per_second": 14.549, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_chemistry_entropy": 0.5615170063972473, |
| "eval_chemistry_loss": 0.6929484605789185, |
| "eval_chemistry_mean_token_accuracy": 0.8095834488868714, |
| "eval_chemistry_num_tokens": 19532552.0, |
| "eval_chemistry_runtime": 38.9798, |
| "eval_chemistry_samples_per_second": 12.827, |
| "eval_chemistry_steps_per_second": 12.827, |
| "step": 4000 |
| }, |
| { |
| "entropy": 0.7438844283111393, |
| "epoch": 1.604, |
| "grad_norm": 70.0, |
| "learning_rate": 1.3313333333333333e-05, |
| "loss": 11.9578, |
| "mean_token_accuracy": 0.7924849327653647, |
| "num_tokens": 19582559.0, |
| "step": 4010 |
| }, |
| { |
| "entropy": 0.7544092898257077, |
| "epoch": 1.608, |
| "grad_norm": 75.5, |
| "learning_rate": 1.3291111111111113e-05, |
| "loss": 12.0837, |
| "mean_token_accuracy": 0.7906231626868248, |
| "num_tokens": 19631506.0, |
| "step": 4020 |
| }, |
| { |
| "entropy": 0.7679222824983298, |
| "epoch": 1.612, |
| "grad_norm": 68.5, |
| "learning_rate": 1.3268888888888889e-05, |
| "loss": 12.1314, |
| "mean_token_accuracy": 0.7900735452771187, |
| "num_tokens": 19683563.0, |
| "step": 4030 |
| }, |
| { |
| "entropy": 0.7877044372260571, |
| "epoch": 1.616, |
| "grad_norm": 77.0, |
| "learning_rate": 1.3246666666666668e-05, |
| "loss": 12.5062, |
| "mean_token_accuracy": 0.7822695929557085, |
| "num_tokens": 19732719.0, |
| "step": 4040 |
| }, |
| { |
| "entropy": 0.7412923231720925, |
| "epoch": 1.62, |
| "grad_norm": 86.5, |
| "learning_rate": 1.3224444444444446e-05, |
| "loss": 11.7625, |
| "mean_token_accuracy": 0.792553286626935, |
| "num_tokens": 19784215.0, |
| "step": 4050 |
| }, |
| { |
| "entropy": 0.7919308325275779, |
| "epoch": 1.624, |
| "grad_norm": 77.0, |
| "learning_rate": 1.3202222222222223e-05, |
| "loss": 12.9325, |
| "mean_token_accuracy": 0.778983835503459, |
| "num_tokens": 19834441.0, |
| "step": 4060 |
| }, |
| { |
| "entropy": 0.8048484846949577, |
| "epoch": 1.6280000000000001, |
| "grad_norm": 74.0, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 12.5222, |
| "mean_token_accuracy": 0.7825064171105623, |
| "num_tokens": 19882788.0, |
| "step": 4070 |
| }, |
| { |
| "entropy": 0.7687547641806305, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 74.0, |
| "learning_rate": 1.3157777777777778e-05, |
| "loss": 12.6721, |
| "mean_token_accuracy": 0.7845367696136236, |
| "num_tokens": 19926830.0, |
| "step": 4080 |
| }, |
| { |
| "entropy": 0.803656572662294, |
| "epoch": 1.6360000000000001, |
| "grad_norm": 71.0, |
| "learning_rate": 1.3135555555555558e-05, |
| "loss": 12.6392, |
| "mean_token_accuracy": 0.7795426450669766, |
| "num_tokens": 19971998.0, |
| "step": 4090 |
| }, |
| { |
| "entropy": 0.772695763129741, |
| "epoch": 1.6400000000000001, |
| "grad_norm": 77.5, |
| "learning_rate": 1.3113333333333334e-05, |
| "loss": 12.1783, |
| "mean_token_accuracy": 0.7921535439789296, |
| "num_tokens": 20017708.0, |
| "step": 4100 |
| }, |
| { |
| "epoch": 1.6400000000000001, |
| "eval_biology_entropy": 0.9123168474435807, |
| "eval_biology_loss": 1.1304035186767578, |
| "eval_biology_mean_token_accuracy": 0.7135367204546929, |
| "eval_biology_num_tokens": 20017708.0, |
| "eval_biology_runtime": 34.3408, |
| "eval_biology_samples_per_second": 14.56, |
| "eval_biology_steps_per_second": 14.56, |
| "step": 4100 |
| }, |
| { |
| "epoch": 1.6400000000000001, |
| "eval_chemistry_entropy": 0.5566242938041687, |
| "eval_chemistry_loss": 0.6916686296463013, |
| "eval_chemistry_mean_token_accuracy": 0.8106292953491211, |
| "eval_chemistry_num_tokens": 20017708.0, |
| "eval_chemistry_runtime": 39.1101, |
| "eval_chemistry_samples_per_second": 12.784, |
| "eval_chemistry_steps_per_second": 12.784, |
| "step": 4100 |
| }, |
| { |
| "entropy": 0.7303692138753831, |
| "epoch": 1.6440000000000001, |
| "grad_norm": 79.0, |
| "learning_rate": 1.3091111111111113e-05, |
| "loss": 11.8662, |
| "mean_token_accuracy": 0.7949420250952244, |
| "num_tokens": 20062624.0, |
| "step": 4110 |
| }, |
| { |
| "entropy": 0.7385077781043947, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 76.0, |
| "learning_rate": 1.306888888888889e-05, |
| "loss": 11.775, |
| "mean_token_accuracy": 0.7933512518182397, |
| "num_tokens": 20118800.0, |
| "step": 4120 |
| }, |
| { |
| "entropy": 0.7466002348810434, |
| "epoch": 1.6520000000000001, |
| "grad_norm": 72.5, |
| "learning_rate": 1.3046666666666668e-05, |
| "loss": 11.8535, |
| "mean_token_accuracy": 0.7943586148321629, |
| "num_tokens": 20169740.0, |
| "step": 4130 |
| }, |
| { |
| "entropy": 0.7688881213776767, |
| "epoch": 1.6560000000000001, |
| "grad_norm": 65.5, |
| "learning_rate": 1.3024444444444446e-05, |
| "loss": 12.3287, |
| "mean_token_accuracy": 0.7854791436344385, |
| "num_tokens": 20219215.0, |
| "step": 4140 |
| }, |
| { |
| "entropy": 0.7554127791896462, |
| "epoch": 1.6600000000000001, |
| "grad_norm": 77.5, |
| "learning_rate": 1.3002222222222223e-05, |
| "loss": 12.2316, |
| "mean_token_accuracy": 0.7927513282746077, |
| "num_tokens": 20268553.0, |
| "step": 4150 |
| }, |
| { |
| "entropy": 0.770198050327599, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 75.0, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 12.307, |
| "mean_token_accuracy": 0.7872584585100413, |
| "num_tokens": 20320511.0, |
| "step": 4160 |
| }, |
| { |
| "entropy": 0.7530902106314897, |
| "epoch": 1.6680000000000001, |
| "grad_norm": 82.0, |
| "learning_rate": 1.2957777777777779e-05, |
| "loss": 11.7547, |
| "mean_token_accuracy": 0.7961310651153326, |
| "num_tokens": 20369169.0, |
| "step": 4170 |
| }, |
| { |
| "entropy": 0.7120116321370006, |
| "epoch": 1.6720000000000002, |
| "grad_norm": 72.5, |
| "learning_rate": 1.2935555555555556e-05, |
| "loss": 11.4351, |
| "mean_token_accuracy": 0.8031725943088531, |
| "num_tokens": 20414832.0, |
| "step": 4180 |
| }, |
| { |
| "entropy": 0.7409597878344357, |
| "epoch": 1.6760000000000002, |
| "grad_norm": 75.5, |
| "learning_rate": 1.2913333333333336e-05, |
| "loss": 11.7781, |
| "mean_token_accuracy": 0.7940782386809587, |
| "num_tokens": 20461161.0, |
| "step": 4190 |
| }, |
| { |
| "entropy": 0.7605687515810132, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 73.0, |
| "learning_rate": 1.2891111111111112e-05, |
| "loss": 12.3596, |
| "mean_token_accuracy": 0.7855475205928087, |
| "num_tokens": 20513393.0, |
| "step": 4200 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "eval_biology_entropy": 0.9375361059308052, |
| "eval_biology_loss": 1.1245396137237549, |
| "eval_biology_mean_token_accuracy": 0.7145720510482788, |
| "eval_biology_num_tokens": 20513393.0, |
| "eval_biology_runtime": 34.2686, |
| "eval_biology_samples_per_second": 14.591, |
| "eval_biology_steps_per_second": 14.591, |
| "step": 4200 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "eval_chemistry_entropy": 0.5727805478870869, |
| "eval_chemistry_loss": 0.6867600083351135, |
| "eval_chemistry_mean_token_accuracy": 0.8108530569076539, |
| "eval_chemistry_num_tokens": 20513393.0, |
| "eval_chemistry_runtime": 39.0971, |
| "eval_chemistry_samples_per_second": 12.789, |
| "eval_chemistry_steps_per_second": 12.789, |
| "step": 4200 |
| }, |
| { |
| "entropy": 0.804387055709958, |
| "epoch": 1.6840000000000002, |
| "grad_norm": 81.0, |
| "learning_rate": 1.2868888888888891e-05, |
| "loss": 12.6459, |
| "mean_token_accuracy": 0.7842802103608847, |
| "num_tokens": 20559535.0, |
| "step": 4210 |
| }, |
| { |
| "entropy": 0.7875384515151381, |
| "epoch": 1.688, |
| "grad_norm": 86.0, |
| "learning_rate": 1.2846666666666667e-05, |
| "loss": 12.4777, |
| "mean_token_accuracy": 0.7840598132461309, |
| "num_tokens": 20604938.0, |
| "step": 4220 |
| }, |
| { |
| "entropy": 0.7277499793097377, |
| "epoch": 1.692, |
| "grad_norm": 69.0, |
| "learning_rate": 1.2824444444444446e-05, |
| "loss": 11.8097, |
| "mean_token_accuracy": 0.7936189398169518, |
| "num_tokens": 20655254.0, |
| "step": 4230 |
| }, |
| { |
| "entropy": 0.7605584662407636, |
| "epoch": 1.696, |
| "grad_norm": 75.0, |
| "learning_rate": 1.2802222222222222e-05, |
| "loss": 12.1278, |
| "mean_token_accuracy": 0.7934175331145525, |
| "num_tokens": 20707237.0, |
| "step": 4240 |
| }, |
| { |
| "entropy": 0.7750071801245213, |
| "epoch": 1.7, |
| "grad_norm": 75.5, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 12.4739, |
| "mean_token_accuracy": 0.7827725235372782, |
| "num_tokens": 20760161.0, |
| "step": 4250 |
| }, |
| { |
| "entropy": 0.739550247695297, |
| "epoch": 1.704, |
| "grad_norm": 81.0, |
| "learning_rate": 1.2757777777777777e-05, |
| "loss": 11.8344, |
| "mean_token_accuracy": 0.7931854896247387, |
| "num_tokens": 20812088.0, |
| "step": 4260 |
| }, |
| { |
| "entropy": 0.7947698433883488, |
| "epoch": 1.708, |
| "grad_norm": 74.0, |
| "learning_rate": 1.2735555555555557e-05, |
| "loss": 12.8991, |
| "mean_token_accuracy": 0.7821339972317218, |
| "num_tokens": 20861269.0, |
| "step": 4270 |
| }, |
| { |
| "entropy": 0.8027881752699614, |
| "epoch": 1.712, |
| "grad_norm": 75.5, |
| "learning_rate": 1.2713333333333336e-05, |
| "loss": 12.489, |
| "mean_token_accuracy": 0.7851102694869041, |
| "num_tokens": 20910419.0, |
| "step": 4280 |
| }, |
| { |
| "entropy": 0.7524696609005332, |
| "epoch": 1.716, |
| "grad_norm": 76.0, |
| "learning_rate": 1.2691111111111112e-05, |
| "loss": 11.9732, |
| "mean_token_accuracy": 0.7946411196142436, |
| "num_tokens": 20959115.0, |
| "step": 4290 |
| }, |
| { |
| "entropy": 0.7871804080903531, |
| "epoch": 1.72, |
| "grad_norm": 77.5, |
| "learning_rate": 1.2668888888888891e-05, |
| "loss": 12.9838, |
| "mean_token_accuracy": 0.7770213726907969, |
| "num_tokens": 21009172.0, |
| "step": 4300 |
| }, |
| { |
| "epoch": 1.72, |
| "eval_biology_entropy": 0.9545542849898339, |
| "eval_biology_loss": 1.1189427375793457, |
| "eval_biology_mean_token_accuracy": 0.7158340885639191, |
| "eval_biology_num_tokens": 21009172.0, |
| "eval_biology_runtime": 34.5549, |
| "eval_biology_samples_per_second": 14.47, |
| "eval_biology_steps_per_second": 14.47, |
| "step": 4300 |
| }, |
| { |
| "epoch": 1.72, |
| "eval_chemistry_entropy": 0.5831989452838898, |
| "eval_chemistry_loss": 0.6852427124977112, |
| "eval_chemistry_mean_token_accuracy": 0.8102431092262268, |
| "eval_chemistry_num_tokens": 21009172.0, |
| "eval_chemistry_runtime": 39.1505, |
| "eval_chemistry_samples_per_second": 12.771, |
| "eval_chemistry_steps_per_second": 12.771, |
| "step": 4300 |
| }, |
| { |
| "entropy": 0.7638085896149278, |
| "epoch": 1.724, |
| "grad_norm": 77.0, |
| "learning_rate": 1.2646666666666667e-05, |
| "loss": 12.0052, |
| "mean_token_accuracy": 0.7911715507507324, |
| "num_tokens": 21058632.0, |
| "step": 4310 |
| }, |
| { |
| "entropy": 0.7821283914148808, |
| "epoch": 1.728, |
| "grad_norm": 86.0, |
| "learning_rate": 1.2624444444444446e-05, |
| "loss": 12.5315, |
| "mean_token_accuracy": 0.7821295309811831, |
| "num_tokens": 21107498.0, |
| "step": 4320 |
| }, |
| { |
| "entropy": 0.7628530548885465, |
| "epoch": 1.732, |
| "grad_norm": 80.5, |
| "learning_rate": 1.2602222222222222e-05, |
| "loss": 12.341, |
| "mean_token_accuracy": 0.7852707453072071, |
| "num_tokens": 21158620.0, |
| "step": 4330 |
| }, |
| { |
| "entropy": 0.7746778642758727, |
| "epoch": 1.736, |
| "grad_norm": 80.5, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 12.1484, |
| "mean_token_accuracy": 0.7889657273888588, |
| "num_tokens": 21204959.0, |
| "step": 4340 |
| }, |
| { |
| "entropy": 0.7576858004555106, |
| "epoch": 1.74, |
| "grad_norm": 70.5, |
| "learning_rate": 1.2557777777777779e-05, |
| "loss": 12.3318, |
| "mean_token_accuracy": 0.7863867662847042, |
| "num_tokens": 21254219.0, |
| "step": 4350 |
| }, |
| { |
| "entropy": 0.7718432117253542, |
| "epoch": 1.744, |
| "grad_norm": 77.0, |
| "learning_rate": 1.2535555555555557e-05, |
| "loss": 11.9385, |
| "mean_token_accuracy": 0.7914053495973349, |
| "num_tokens": 21303955.0, |
| "step": 4360 |
| }, |
| { |
| "entropy": 0.7469873716123402, |
| "epoch": 1.748, |
| "grad_norm": 71.0, |
| "learning_rate": 1.2513333333333334e-05, |
| "loss": 12.2858, |
| "mean_token_accuracy": 0.7867584012448787, |
| "num_tokens": 21356640.0, |
| "step": 4370 |
| }, |
| { |
| "entropy": 0.7525212218053638, |
| "epoch": 1.752, |
| "grad_norm": 70.0, |
| "learning_rate": 1.2491111111111112e-05, |
| "loss": 12.0364, |
| "mean_token_accuracy": 0.7897812929004431, |
| "num_tokens": 21408903.0, |
| "step": 4380 |
| }, |
| { |
| "entropy": 0.7888920472934842, |
| "epoch": 1.756, |
| "grad_norm": 89.0, |
| "learning_rate": 1.246888888888889e-05, |
| "loss": 12.5184, |
| "mean_token_accuracy": 0.7839067671447992, |
| "num_tokens": 21454071.0, |
| "step": 4390 |
| }, |
| { |
| "entropy": 0.7525685601867735, |
| "epoch": 1.76, |
| "grad_norm": 81.0, |
| "learning_rate": 1.2446666666666667e-05, |
| "loss": 12.0802, |
| "mean_token_accuracy": 0.7955807693302631, |
| "num_tokens": 21499572.0, |
| "step": 4400 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_biology_entropy": 0.9342851682901382, |
| "eval_biology_loss": 1.1234320402145386, |
| "eval_biology_mean_token_accuracy": 0.7157456163167953, |
| "eval_biology_num_tokens": 21499572.0, |
| "eval_biology_runtime": 34.4321, |
| "eval_biology_samples_per_second": 14.521, |
| "eval_biology_steps_per_second": 14.521, |
| "step": 4400 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_chemistry_entropy": 0.5653623167574405, |
| "eval_chemistry_loss": 0.6908249258995056, |
| "eval_chemistry_mean_token_accuracy": 0.8106068837642669, |
| "eval_chemistry_num_tokens": 21499572.0, |
| "eval_chemistry_runtime": 39.1506, |
| "eval_chemistry_samples_per_second": 12.771, |
| "eval_chemistry_steps_per_second": 12.771, |
| "step": 4400 |
| }, |
| { |
| "entropy": 0.7559589648619294, |
| "epoch": 1.764, |
| "grad_norm": 78.5, |
| "learning_rate": 1.2424444444444445e-05, |
| "loss": 12.225, |
| "mean_token_accuracy": 0.7869325697422027, |
| "num_tokens": 21545496.0, |
| "step": 4410 |
| }, |
| { |
| "entropy": 0.7354301054961979, |
| "epoch": 1.768, |
| "grad_norm": 75.5, |
| "learning_rate": 1.2402222222222222e-05, |
| "loss": 11.491, |
| "mean_token_accuracy": 0.7959539409726858, |
| "num_tokens": 21596260.0, |
| "step": 4420 |
| }, |
| { |
| "entropy": 0.7255675679072737, |
| "epoch": 1.772, |
| "grad_norm": 84.0, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 11.5179, |
| "mean_token_accuracy": 0.7998525436967612, |
| "num_tokens": 21643178.0, |
| "step": 4430 |
| }, |
| { |
| "entropy": 0.759679536614567, |
| "epoch": 1.776, |
| "grad_norm": 82.0, |
| "learning_rate": 1.235777777777778e-05, |
| "loss": 12.3985, |
| "mean_token_accuracy": 0.7848724327981472, |
| "num_tokens": 21692804.0, |
| "step": 4440 |
| }, |
| { |
| "entropy": 0.7558694992214441, |
| "epoch": 1.78, |
| "grad_norm": 74.5, |
| "learning_rate": 1.2335555555555557e-05, |
| "loss": 12.1135, |
| "mean_token_accuracy": 0.7914426643401384, |
| "num_tokens": 21744916.0, |
| "step": 4450 |
| }, |
| { |
| "entropy": 0.6852248509414494, |
| "epoch": 1.784, |
| "grad_norm": 68.0, |
| "learning_rate": 1.2313333333333335e-05, |
| "loss": 10.7765, |
| "mean_token_accuracy": 0.8085872050374746, |
| "num_tokens": 21796414.0, |
| "step": 4460 |
| }, |
| { |
| "entropy": 0.7381771637126804, |
| "epoch": 1.788, |
| "grad_norm": 79.5, |
| "learning_rate": 1.2291111111111112e-05, |
| "loss": 11.933, |
| "mean_token_accuracy": 0.7943452958017587, |
| "num_tokens": 21845501.0, |
| "step": 4470 |
| }, |
| { |
| "entropy": 0.7799038235098124, |
| "epoch": 1.792, |
| "grad_norm": 75.5, |
| "learning_rate": 1.226888888888889e-05, |
| "loss": 12.4321, |
| "mean_token_accuracy": 0.7859773490577936, |
| "num_tokens": 21894218.0, |
| "step": 4480 |
| }, |
| { |
| "entropy": 0.7611330568790435, |
| "epoch": 1.796, |
| "grad_norm": 90.0, |
| "learning_rate": 1.2246666666666667e-05, |
| "loss": 12.3269, |
| "mean_token_accuracy": 0.7898826003074646, |
| "num_tokens": 21938255.0, |
| "step": 4490 |
| }, |
| { |
| "entropy": 0.7597966071218252, |
| "epoch": 1.8, |
| "grad_norm": 78.5, |
| "learning_rate": 1.2224444444444445e-05, |
| "loss": 12.0074, |
| "mean_token_accuracy": 0.7932655941694975, |
| "num_tokens": 21985664.0, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.8, |
| "eval_biology_entropy": 0.9412995492815971, |
| "eval_biology_loss": 1.1186201572418213, |
| "eval_biology_mean_token_accuracy": 0.7161767404079438, |
| "eval_biology_num_tokens": 21985664.0, |
| "eval_biology_runtime": 34.5556, |
| "eval_biology_samples_per_second": 14.469, |
| "eval_biology_steps_per_second": 14.469, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.8, |
| "eval_chemistry_entropy": 0.5702413580715656, |
| "eval_chemistry_loss": 0.6902027130126953, |
| "eval_chemistry_mean_token_accuracy": 0.8090504283905029, |
| "eval_chemistry_num_tokens": 21985664.0, |
| "eval_chemistry_runtime": 38.929, |
| "eval_chemistry_samples_per_second": 12.844, |
| "eval_chemistry_steps_per_second": 12.844, |
| "step": 4500 |
| }, |
| { |
| "entropy": 0.7250780941918492, |
| "epoch": 1.804, |
| "grad_norm": 83.5, |
| "learning_rate": 1.2202222222222224e-05, |
| "loss": 11.6549, |
| "mean_token_accuracy": 0.7970849990844726, |
| "num_tokens": 22034099.0, |
| "step": 4510 |
| }, |
| { |
| "entropy": 0.7555977020412683, |
| "epoch": 1.808, |
| "grad_norm": 84.0, |
| "learning_rate": 1.218e-05, |
| "loss": 11.9896, |
| "mean_token_accuracy": 0.7945190455764533, |
| "num_tokens": 22082522.0, |
| "step": 4520 |
| }, |
| { |
| "entropy": 0.7932162513956428, |
| "epoch": 1.812, |
| "grad_norm": 77.5, |
| "learning_rate": 1.215777777777778e-05, |
| "loss": 12.5764, |
| "mean_token_accuracy": 0.784709844738245, |
| "num_tokens": 22130155.0, |
| "step": 4530 |
| }, |
| { |
| "entropy": 0.759617529809475, |
| "epoch": 1.8159999999999998, |
| "grad_norm": 74.5, |
| "learning_rate": 1.2135555555555556e-05, |
| "loss": 12.0311, |
| "mean_token_accuracy": 0.7915454138070345, |
| "num_tokens": 22180616.0, |
| "step": 4540 |
| }, |
| { |
| "entropy": 0.7576798176392913, |
| "epoch": 1.8199999999999998, |
| "grad_norm": 76.0, |
| "learning_rate": 1.2113333333333335e-05, |
| "loss": 12.1648, |
| "mean_token_accuracy": 0.786707516387105, |
| "num_tokens": 22233349.0, |
| "step": 4550 |
| }, |
| { |
| "entropy": 0.7596826974302531, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 79.0, |
| "learning_rate": 1.209111111111111e-05, |
| "loss": 12.1929, |
| "mean_token_accuracy": 0.7874425649642944, |
| "num_tokens": 22278933.0, |
| "step": 4560 |
| }, |
| { |
| "entropy": 0.8124894758686423, |
| "epoch": 1.8279999999999998, |
| "grad_norm": 76.5, |
| "learning_rate": 1.206888888888889e-05, |
| "loss": 13.1482, |
| "mean_token_accuracy": 0.7792382642626763, |
| "num_tokens": 22327862.0, |
| "step": 4570 |
| }, |
| { |
| "entropy": 0.7856536395847797, |
| "epoch": 1.8319999999999999, |
| "grad_norm": 83.0, |
| "learning_rate": 1.204666666666667e-05, |
| "loss": 12.5191, |
| "mean_token_accuracy": 0.7838058475404978, |
| "num_tokens": 22377874.0, |
| "step": 4580 |
| }, |
| { |
| "entropy": 0.8021014945581555, |
| "epoch": 1.8359999999999999, |
| "grad_norm": 79.0, |
| "learning_rate": 1.2024444444444445e-05, |
| "loss": 12.9831, |
| "mean_token_accuracy": 0.7790773436427116, |
| "num_tokens": 22427163.0, |
| "step": 4590 |
| }, |
| { |
| "entropy": 0.7926059128716588, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 79.0, |
| "learning_rate": 1.2002222222222225e-05, |
| "loss": 12.5852, |
| "mean_token_accuracy": 0.7842131458222866, |
| "num_tokens": 22473801.0, |
| "step": 4600 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "eval_biology_entropy": 0.9561592536568642, |
| "eval_biology_loss": 1.1144318580627441, |
| "eval_biology_mean_token_accuracy": 0.7163329623937607, |
| "eval_biology_num_tokens": 22473801.0, |
| "eval_biology_runtime": 34.4474, |
| "eval_biology_samples_per_second": 14.515, |
| "eval_biology_steps_per_second": 14.515, |
| "step": 4600 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "eval_chemistry_entropy": 0.5752664268910885, |
| "eval_chemistry_loss": 0.6889669895172119, |
| "eval_chemistry_mean_token_accuracy": 0.809889578461647, |
| "eval_chemistry_num_tokens": 22473801.0, |
| "eval_chemistry_runtime": 39.1555, |
| "eval_chemistry_samples_per_second": 12.77, |
| "eval_chemistry_steps_per_second": 12.77, |
| "step": 4600 |
| }, |
| { |
| "entropy": 0.7702212953940034, |
| "epoch": 1.8439999999999999, |
| "grad_norm": 91.0, |
| "learning_rate": 1.198e-05, |
| "loss": 12.1281, |
| "mean_token_accuracy": 0.7855017110705376, |
| "num_tokens": 22528230.0, |
| "step": 4610 |
| }, |
| { |
| "entropy": 0.7323052290827036, |
| "epoch": 1.8479999999999999, |
| "grad_norm": 66.0, |
| "learning_rate": 1.195777777777778e-05, |
| "loss": 11.8185, |
| "mean_token_accuracy": 0.7914682097733021, |
| "num_tokens": 22582148.0, |
| "step": 4620 |
| }, |
| { |
| "entropy": 0.7609475309029221, |
| "epoch": 1.8519999999999999, |
| "grad_norm": 67.5, |
| "learning_rate": 1.1935555555555556e-05, |
| "loss": 12.1881, |
| "mean_token_accuracy": 0.786286699399352, |
| "num_tokens": 22628809.0, |
| "step": 4630 |
| }, |
| { |
| "entropy": 0.7941112034954131, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 76.5, |
| "learning_rate": 1.1913333333333335e-05, |
| "loss": 12.8678, |
| "mean_token_accuracy": 0.7777332227677107, |
| "num_tokens": 22677853.0, |
| "step": 4640 |
| }, |
| { |
| "entropy": 0.7736134089529514, |
| "epoch": 1.8599999999999999, |
| "grad_norm": 76.0, |
| "learning_rate": 1.1891111111111111e-05, |
| "loss": 12.307, |
| "mean_token_accuracy": 0.7884436974301934, |
| "num_tokens": 22728933.0, |
| "step": 4650 |
| }, |
| { |
| "entropy": 0.7835546713322401, |
| "epoch": 1.8639999999999999, |
| "grad_norm": 85.0, |
| "learning_rate": 1.186888888888889e-05, |
| "loss": 12.4031, |
| "mean_token_accuracy": 0.78635428622365, |
| "num_tokens": 22775707.0, |
| "step": 4660 |
| }, |
| { |
| "entropy": 0.7035165684297681, |
| "epoch": 1.8679999999999999, |
| "grad_norm": 88.0, |
| "learning_rate": 1.1846666666666668e-05, |
| "loss": 11.3401, |
| "mean_token_accuracy": 0.8006389241665601, |
| "num_tokens": 22825010.0, |
| "step": 4670 |
| }, |
| { |
| "entropy": 0.721017730049789, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 84.5, |
| "learning_rate": 1.1824444444444445e-05, |
| "loss": 11.4571, |
| "mean_token_accuracy": 0.8009402837604285, |
| "num_tokens": 22874965.0, |
| "step": 4680 |
| }, |
| { |
| "entropy": 0.7480081206187605, |
| "epoch": 1.876, |
| "grad_norm": 72.0, |
| "learning_rate": 1.1802222222222223e-05, |
| "loss": 12.2063, |
| "mean_token_accuracy": 0.7898276366293431, |
| "num_tokens": 22924587.0, |
| "step": 4690 |
| }, |
| { |
| "entropy": 0.7527163816615939, |
| "epoch": 1.88, |
| "grad_norm": 80.5, |
| "learning_rate": 1.178e-05, |
| "loss": 12.03, |
| "mean_token_accuracy": 0.7898941207677126, |
| "num_tokens": 22972554.0, |
| "step": 4700 |
| }, |
| { |
| "epoch": 1.88, |
| "eval_biology_entropy": 0.943822091460228, |
| "eval_biology_loss": 1.1152360439300537, |
| "eval_biology_mean_token_accuracy": 0.7161377201080322, |
| "eval_biology_num_tokens": 22972554.0, |
| "eval_biology_runtime": 34.4054, |
| "eval_biology_samples_per_second": 14.533, |
| "eval_biology_steps_per_second": 14.533, |
| "step": 4700 |
| }, |
| { |
| "epoch": 1.88, |
| "eval_chemistry_entropy": 0.564557602763176, |
| "eval_chemistry_loss": 0.6891387104988098, |
| "eval_chemistry_mean_token_accuracy": 0.8101301358938218, |
| "eval_chemistry_num_tokens": 22972554.0, |
| "eval_chemistry_runtime": 39.1372, |
| "eval_chemistry_samples_per_second": 12.776, |
| "eval_chemistry_steps_per_second": 12.776, |
| "step": 4700 |
| }, |
| { |
| "entropy": 0.8030757773667574, |
| "epoch": 1.884, |
| "grad_norm": 75.0, |
| "learning_rate": 1.1757777777777778e-05, |
| "loss": 12.6717, |
| "mean_token_accuracy": 0.781605789065361, |
| "num_tokens": 23020747.0, |
| "step": 4710 |
| }, |
| { |
| "entropy": 0.7514794295653701, |
| "epoch": 1.888, |
| "grad_norm": 88.0, |
| "learning_rate": 1.1735555555555556e-05, |
| "loss": 11.9206, |
| "mean_token_accuracy": 0.7918515712022781, |
| "num_tokens": 23068892.0, |
| "step": 4720 |
| }, |
| { |
| "entropy": 0.7259659945964814, |
| "epoch": 1.892, |
| "grad_norm": 72.0, |
| "learning_rate": 1.1713333333333334e-05, |
| "loss": 11.871, |
| "mean_token_accuracy": 0.7941229268908501, |
| "num_tokens": 23120446.0, |
| "step": 4730 |
| }, |
| { |
| "entropy": 0.7660525813698769, |
| "epoch": 1.896, |
| "grad_norm": 78.5, |
| "learning_rate": 1.1691111111111113e-05, |
| "loss": 11.7163, |
| "mean_token_accuracy": 0.7948390446603298, |
| "num_tokens": 23168907.0, |
| "step": 4740 |
| }, |
| { |
| "entropy": 0.7565810011699796, |
| "epoch": 1.9, |
| "grad_norm": 84.5, |
| "learning_rate": 1.1668888888888889e-05, |
| "loss": 12.5257, |
| "mean_token_accuracy": 0.7865264303982258, |
| "num_tokens": 23217329.0, |
| "step": 4750 |
| }, |
| { |
| "entropy": 0.795278568379581, |
| "epoch": 1.904, |
| "grad_norm": 78.0, |
| "learning_rate": 1.1646666666666668e-05, |
| "loss": 12.8965, |
| "mean_token_accuracy": 0.78194130808115, |
| "num_tokens": 23263827.0, |
| "step": 4760 |
| }, |
| { |
| "entropy": 0.8595306348986924, |
| "epoch": 1.908, |
| "grad_norm": 82.0, |
| "learning_rate": 1.1624444444444446e-05, |
| "loss": 13.5751, |
| "mean_token_accuracy": 0.7697425246238708, |
| "num_tokens": 23310898.0, |
| "step": 4770 |
| }, |
| { |
| "entropy": 0.7707650167867541, |
| "epoch": 1.912, |
| "grad_norm": 73.5, |
| "learning_rate": 1.1602222222222223e-05, |
| "loss": 12.4017, |
| "mean_token_accuracy": 0.7873672746121884, |
| "num_tokens": 23361737.0, |
| "step": 4780 |
| }, |
| { |
| "entropy": 0.7786004858091473, |
| "epoch": 1.916, |
| "grad_norm": 73.5, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 12.5604, |
| "mean_token_accuracy": 0.7827336050570011, |
| "num_tokens": 23411869.0, |
| "step": 4790 |
| }, |
| { |
| "entropy": 0.7274590061977506, |
| "epoch": 1.92, |
| "grad_norm": 75.0, |
| "learning_rate": 1.1557777777777779e-05, |
| "loss": 11.4684, |
| "mean_token_accuracy": 0.7989695444703102, |
| "num_tokens": 23463627.0, |
| "step": 4800 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_biology_entropy": 0.9175975527763367, |
| "eval_biology_loss": 1.1146620512008667, |
| "eval_biology_mean_token_accuracy": 0.7175298974514007, |
| "eval_biology_num_tokens": 23463627.0, |
| "eval_biology_runtime": 34.4346, |
| "eval_biology_samples_per_second": 14.52, |
| "eval_biology_steps_per_second": 14.52, |
| "step": 4800 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_chemistry_entropy": 0.5578563950359822, |
| "eval_chemistry_loss": 0.6878622770309448, |
| "eval_chemistry_mean_token_accuracy": 0.8101852288246155, |
| "eval_chemistry_num_tokens": 23463627.0, |
| "eval_chemistry_runtime": 39.1868, |
| "eval_chemistry_samples_per_second": 12.759, |
| "eval_chemistry_steps_per_second": 12.759, |
| "step": 4800 |
| }, |
| { |
| "entropy": 0.7224604841321707, |
| "epoch": 1.924, |
| "grad_norm": 84.5, |
| "learning_rate": 1.1535555555555556e-05, |
| "loss": 11.7315, |
| "mean_token_accuracy": 0.7972664751112462, |
| "num_tokens": 23511485.0, |
| "step": 4810 |
| }, |
| { |
| "entropy": 0.7550310818478465, |
| "epoch": 1.928, |
| "grad_norm": 75.0, |
| "learning_rate": 1.1513333333333334e-05, |
| "loss": 12.0074, |
| "mean_token_accuracy": 0.7929213266819716, |
| "num_tokens": 23558006.0, |
| "step": 4820 |
| }, |
| { |
| "entropy": 0.7386426834389568, |
| "epoch": 1.932, |
| "grad_norm": 69.0, |
| "learning_rate": 1.1491111111111113e-05, |
| "loss": 11.7443, |
| "mean_token_accuracy": 0.7930562358349562, |
| "num_tokens": 23609539.0, |
| "step": 4830 |
| }, |
| { |
| "entropy": 0.7318497506901622, |
| "epoch": 1.936, |
| "grad_norm": 79.5, |
| "learning_rate": 1.1468888888888889e-05, |
| "loss": 11.7037, |
| "mean_token_accuracy": 0.7959757678210735, |
| "num_tokens": 23660418.0, |
| "step": 4840 |
| }, |
| { |
| "entropy": 0.7448375387117266, |
| "epoch": 1.94, |
| "grad_norm": 77.0, |
| "learning_rate": 1.1446666666666668e-05, |
| "loss": 12.0927, |
| "mean_token_accuracy": 0.7925669450312853, |
| "num_tokens": 23710867.0, |
| "step": 4850 |
| }, |
| { |
| "entropy": 0.7261123044416309, |
| "epoch": 1.944, |
| "grad_norm": 75.5, |
| "learning_rate": 1.1424444444444444e-05, |
| "loss": 11.6087, |
| "mean_token_accuracy": 0.7991454791277647, |
| "num_tokens": 23764848.0, |
| "step": 4860 |
| }, |
| { |
| "entropy": 0.7884697087109089, |
| "epoch": 1.948, |
| "grad_norm": 79.0, |
| "learning_rate": 1.1402222222222224e-05, |
| "loss": 12.5079, |
| "mean_token_accuracy": 0.7828566361218691, |
| "num_tokens": 23809822.0, |
| "step": 4870 |
| }, |
| { |
| "entropy": 0.7381803538650274, |
| "epoch": 1.952, |
| "grad_norm": 73.0, |
| "learning_rate": 1.138e-05, |
| "loss": 11.9708, |
| "mean_token_accuracy": 0.792114881053567, |
| "num_tokens": 23858145.0, |
| "step": 4880 |
| }, |
| { |
| "entropy": 0.7878429546952248, |
| "epoch": 1.956, |
| "grad_norm": 82.5, |
| "learning_rate": 1.1357777777777779e-05, |
| "loss": 12.3501, |
| "mean_token_accuracy": 0.7866133779287339, |
| "num_tokens": 23908031.0, |
| "step": 4890 |
| }, |
| { |
| "entropy": 0.7213960700668395, |
| "epoch": 1.96, |
| "grad_norm": 80.0, |
| "learning_rate": 1.1335555555555558e-05, |
| "loss": 11.429, |
| "mean_token_accuracy": 0.7975703597068786, |
| "num_tokens": 23957580.0, |
| "step": 4900 |
| }, |
| { |
| "epoch": 1.96, |
| "eval_biology_entropy": 0.9123851335048676, |
| "eval_biology_loss": 1.1188366413116455, |
| "eval_biology_mean_token_accuracy": 0.7178824002742767, |
| "eval_biology_num_tokens": 23957580.0, |
| "eval_biology_runtime": 34.4307, |
| "eval_biology_samples_per_second": 14.522, |
| "eval_biology_steps_per_second": 14.522, |
| "step": 4900 |
| }, |
| { |
| "epoch": 1.96, |
| "eval_chemistry_entropy": 0.5569564428329468, |
| "eval_chemistry_loss": 0.6892187595367432, |
| "eval_chemistry_mean_token_accuracy": 0.8106620668172836, |
| "eval_chemistry_num_tokens": 23957580.0, |
| "eval_chemistry_runtime": 39.0312, |
| "eval_chemistry_samples_per_second": 12.81, |
| "eval_chemistry_steps_per_second": 12.81, |
| "step": 4900 |
| }, |
| { |
| "entropy": 0.7475189819931984, |
| "epoch": 1.964, |
| "grad_norm": 88.5, |
| "learning_rate": 1.1313333333333334e-05, |
| "loss": 12.5745, |
| "mean_token_accuracy": 0.7836092554032803, |
| "num_tokens": 24005876.0, |
| "step": 4910 |
| }, |
| { |
| "entropy": 0.7916459109634161, |
| "epoch": 1.968, |
| "grad_norm": 84.0, |
| "learning_rate": 1.1291111111111113e-05, |
| "loss": 11.9758, |
| "mean_token_accuracy": 0.7930448938161134, |
| "num_tokens": 24053364.0, |
| "step": 4920 |
| }, |
| { |
| "entropy": 0.7299448670819402, |
| "epoch": 1.972, |
| "grad_norm": 79.0, |
| "learning_rate": 1.126888888888889e-05, |
| "loss": 12.0603, |
| "mean_token_accuracy": 0.7917775686830282, |
| "num_tokens": 24102031.0, |
| "step": 4930 |
| }, |
| { |
| "entropy": 0.7498288176953792, |
| "epoch": 1.976, |
| "grad_norm": 81.0, |
| "learning_rate": 1.1246666666666669e-05, |
| "loss": 11.7521, |
| "mean_token_accuracy": 0.7941938649863005, |
| "num_tokens": 24151503.0, |
| "step": 4940 |
| }, |
| { |
| "entropy": 0.7158438642509282, |
| "epoch": 1.98, |
| "grad_norm": 83.5, |
| "learning_rate": 1.1224444444444444e-05, |
| "loss": 11.6032, |
| "mean_token_accuracy": 0.7980798907577992, |
| "num_tokens": 24202125.0, |
| "step": 4950 |
| }, |
| { |
| "entropy": 0.7748325534164906, |
| "epoch": 1.984, |
| "grad_norm": 85.0, |
| "learning_rate": 1.1202222222222224e-05, |
| "loss": 12.4189, |
| "mean_token_accuracy": 0.7855555981397628, |
| "num_tokens": 24249465.0, |
| "step": 4960 |
| }, |
| { |
| "entropy": 0.7879916010424495, |
| "epoch": 1.988, |
| "grad_norm": 92.0, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 12.6032, |
| "mean_token_accuracy": 0.7843513961881399, |
| "num_tokens": 24297612.0, |
| "step": 4970 |
| }, |
| { |
| "entropy": 0.7670526288449764, |
| "epoch": 1.992, |
| "grad_norm": 83.0, |
| "learning_rate": 1.1157777777777779e-05, |
| "loss": 12.3352, |
| "mean_token_accuracy": 0.7907863549888134, |
| "num_tokens": 24346850.0, |
| "step": 4980 |
| }, |
| { |
| "entropy": 0.7388068053871393, |
| "epoch": 1.996, |
| "grad_norm": 86.0, |
| "learning_rate": 1.1135555555555557e-05, |
| "loss": 11.8284, |
| "mean_token_accuracy": 0.7938032135367393, |
| "num_tokens": 24394646.0, |
| "step": 4990 |
| }, |
| { |
| "entropy": 0.7698258339427412, |
| "epoch": 2.0, |
| "grad_norm": 68.0, |
| "learning_rate": 1.1113333333333334e-05, |
| "loss": 12.4126, |
| "mean_token_accuracy": 0.7893036972731352, |
| "num_tokens": 24442582.0, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_biology_entropy": 0.9551792715191841, |
| "eval_biology_loss": 1.1066296100616455, |
| "eval_biology_mean_token_accuracy": 0.7174452912807464, |
| "eval_biology_num_tokens": 24442582.0, |
| "eval_biology_runtime": 34.2635, |
| "eval_biology_samples_per_second": 14.593, |
| "eval_biology_steps_per_second": 14.593, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_chemistry_entropy": 0.5818323932886124, |
| "eval_chemistry_loss": 0.6834538578987122, |
| "eval_chemistry_mean_token_accuracy": 0.8109288862943649, |
| "eval_chemistry_num_tokens": 24442582.0, |
| "eval_chemistry_runtime": 39.0048, |
| "eval_chemistry_samples_per_second": 12.819, |
| "eval_chemistry_steps_per_second": 12.819, |
| "step": 5000 |
| }, |
| { |
| "entropy": 0.4419955566525459, |
| "epoch": 2.004, |
| "grad_norm": 74.5, |
| "learning_rate": 1.1091111111111112e-05, |
| "loss": 5.839, |
| "mean_token_accuracy": 0.8963859099894762, |
| "num_tokens": 24491247.0, |
| "step": 5010 |
| }, |
| { |
| "entropy": 0.3450068977661431, |
| "epoch": 2.008, |
| "grad_norm": 109.5, |
| "learning_rate": 1.106888888888889e-05, |
| "loss": 6.1503, |
| "mean_token_accuracy": 0.8918702431023121, |
| "num_tokens": 24536380.0, |
| "step": 5020 |
| }, |
| { |
| "entropy": 0.39934638468548656, |
| "epoch": 2.012, |
| "grad_norm": 83.5, |
| "learning_rate": 1.1046666666666667e-05, |
| "loss": 5.969, |
| "mean_token_accuracy": 0.8938017319887877, |
| "num_tokens": 24584860.0, |
| "step": 5030 |
| }, |
| { |
| "entropy": 0.31468719872646034, |
| "epoch": 2.016, |
| "grad_norm": 76.0, |
| "learning_rate": 1.1024444444444445e-05, |
| "loss": 5.1345, |
| "mean_token_accuracy": 0.9070979334414006, |
| "num_tokens": 24632353.0, |
| "step": 5040 |
| }, |
| { |
| "entropy": 0.3367378649301827, |
| "epoch": 2.02, |
| "grad_norm": 94.0, |
| "learning_rate": 1.1002222222222222e-05, |
| "loss": 5.2509, |
| "mean_token_accuracy": 0.9052438508719206, |
| "num_tokens": 24678106.0, |
| "step": 5050 |
| }, |
| { |
| "entropy": 0.3646087530069053, |
| "epoch": 2.024, |
| "grad_norm": 85.5, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 5.8247, |
| "mean_token_accuracy": 0.8937901120632887, |
| "num_tokens": 24728292.0, |
| "step": 5060 |
| }, |
| { |
| "entropy": 0.3838879710063338, |
| "epoch": 2.028, |
| "grad_norm": 80.0, |
| "learning_rate": 1.0957777777777778e-05, |
| "loss": 6.0327, |
| "mean_token_accuracy": 0.8932748179882765, |
| "num_tokens": 24774172.0, |
| "step": 5070 |
| }, |
| { |
| "entropy": 0.35468394025228916, |
| "epoch": 2.032, |
| "grad_norm": 89.5, |
| "learning_rate": 1.0935555555555557e-05, |
| "loss": 5.6878, |
| "mean_token_accuracy": 0.8971023190766573, |
| "num_tokens": 24822715.0, |
| "step": 5080 |
| }, |
| { |
| "entropy": 0.3482358773704618, |
| "epoch": 2.036, |
| "grad_norm": 71.0, |
| "learning_rate": 1.0913333333333333e-05, |
| "loss": 5.5808, |
| "mean_token_accuracy": 0.9003458511084318, |
| "num_tokens": 24874325.0, |
| "step": 5090 |
| }, |
| { |
| "entropy": 0.38926079203374686, |
| "epoch": 2.04, |
| "grad_norm": 86.0, |
| "learning_rate": 1.0891111111111112e-05, |
| "loss": 5.9003, |
| "mean_token_accuracy": 0.8921462755650282, |
| "num_tokens": 24922642.0, |
| "step": 5100 |
| }, |
| { |
| "epoch": 2.04, |
| "eval_biology_entropy": 0.714579436302185, |
| "eval_biology_loss": 1.2661364078521729, |
| "eval_biology_mean_token_accuracy": 0.7092267655730248, |
| "eval_biology_num_tokens": 24922642.0, |
| "eval_biology_runtime": 34.3409, |
| "eval_biology_samples_per_second": 14.56, |
| "eval_biology_steps_per_second": 14.56, |
| "step": 5100 |
| }, |
| { |
| "epoch": 2.04, |
| "eval_chemistry_entropy": 0.4432766874581575, |
| "eval_chemistry_loss": 0.7682344317436218, |
| "eval_chemistry_mean_token_accuracy": 0.8067204010486603, |
| "eval_chemistry_num_tokens": 24922642.0, |
| "eval_chemistry_runtime": 39.1729, |
| "eval_chemistry_samples_per_second": 12.764, |
| "eval_chemistry_steps_per_second": 12.764, |
| "step": 5100 |
| }, |
| { |
| "entropy": 0.34190642829053103, |
| "epoch": 2.044, |
| "grad_norm": 94.5, |
| "learning_rate": 1.0868888888888888e-05, |
| "loss": 5.6165, |
| "mean_token_accuracy": 0.8974835887551308, |
| "num_tokens": 24976140.0, |
| "step": 5110 |
| }, |
| { |
| "entropy": 0.3614497453439981, |
| "epoch": 2.048, |
| "grad_norm": 104.5, |
| "learning_rate": 1.0846666666666667e-05, |
| "loss": 5.658, |
| "mean_token_accuracy": 0.8976224355399609, |
| "num_tokens": 25023118.0, |
| "step": 5120 |
| }, |
| { |
| "entropy": 0.3339924947358668, |
| "epoch": 2.052, |
| "grad_norm": 76.5, |
| "learning_rate": 1.0824444444444447e-05, |
| "loss": 5.2369, |
| "mean_token_accuracy": 0.90715386942029, |
| "num_tokens": 25068732.0, |
| "step": 5130 |
| }, |
| { |
| "entropy": 0.3395977844949812, |
| "epoch": 2.056, |
| "grad_norm": 79.5, |
| "learning_rate": 1.0802222222222223e-05, |
| "loss": 5.5205, |
| "mean_token_accuracy": 0.8981464941054582, |
| "num_tokens": 25120290.0, |
| "step": 5140 |
| }, |
| { |
| "entropy": 0.35286668543703853, |
| "epoch": 2.06, |
| "grad_norm": 90.5, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 5.5479, |
| "mean_token_accuracy": 0.8987288679927588, |
| "num_tokens": 25167016.0, |
| "step": 5150 |
| }, |
| { |
| "entropy": 0.3560471310280263, |
| "epoch": 2.064, |
| "grad_norm": 90.5, |
| "learning_rate": 1.0757777777777778e-05, |
| "loss": 5.7488, |
| "mean_token_accuracy": 0.8962458807975053, |
| "num_tokens": 25217414.0, |
| "step": 5160 |
| }, |
| { |
| "entropy": 0.3478313543833792, |
| "epoch": 2.068, |
| "grad_norm": 73.5, |
| "learning_rate": 1.0735555555555557e-05, |
| "loss": 5.4593, |
| "mean_token_accuracy": 0.8999776981770993, |
| "num_tokens": 25264158.0, |
| "step": 5170 |
| }, |
| { |
| "entropy": 0.34340923740528523, |
| "epoch": 2.072, |
| "grad_norm": 81.0, |
| "learning_rate": 1.0713333333333333e-05, |
| "loss": 5.2888, |
| "mean_token_accuracy": 0.9052454613149166, |
| "num_tokens": 25311278.0, |
| "step": 5180 |
| }, |
| { |
| "entropy": 0.33875691071152686, |
| "epoch": 2.076, |
| "grad_norm": 82.0, |
| "learning_rate": 1.0691111111111112e-05, |
| "loss": 5.4946, |
| "mean_token_accuracy": 0.9005961749702692, |
| "num_tokens": 25358607.0, |
| "step": 5190 |
| }, |
| { |
| "entropy": 0.3712239449378103, |
| "epoch": 2.08, |
| "grad_norm": 86.0, |
| "learning_rate": 1.0668888888888892e-05, |
| "loss": 5.8881, |
| "mean_token_accuracy": 0.8934778086841106, |
| "num_tokens": 25408961.0, |
| "step": 5200 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_biology_entropy": 0.7216597673594951, |
| "eval_biology_loss": 1.2753515243530273, |
| "eval_biology_mean_token_accuracy": 0.7089424991607666, |
| "eval_biology_num_tokens": 25408961.0, |
| "eval_biology_runtime": 34.3307, |
| "eval_biology_samples_per_second": 14.564, |
| "eval_biology_steps_per_second": 14.564, |
| "step": 5200 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_chemistry_entropy": 0.44941052079200744, |
| "eval_chemistry_loss": 0.7696194648742676, |
| "eval_chemistry_mean_token_accuracy": 0.806272873878479, |
| "eval_chemistry_num_tokens": 25408961.0, |
| "eval_chemistry_runtime": 39.1271, |
| "eval_chemistry_samples_per_second": 12.779, |
| "eval_chemistry_steps_per_second": 12.779, |
| "step": 5200 |
| }, |
| { |
| "entropy": 0.35078061311505737, |
| "epoch": 2.084, |
| "grad_norm": 75.5, |
| "learning_rate": 1.0646666666666668e-05, |
| "loss": 5.5293, |
| "mean_token_accuracy": 0.900812690705061, |
| "num_tokens": 25456010.0, |
| "step": 5210 |
| }, |
| { |
| "entropy": 0.35190070238895715, |
| "epoch": 2.088, |
| "grad_norm": 89.0, |
| "learning_rate": 1.0624444444444447e-05, |
| "loss": 5.6173, |
| "mean_token_accuracy": 0.8986608624458313, |
| "num_tokens": 25508270.0, |
| "step": 5220 |
| }, |
| { |
| "entropy": 0.3442512787412852, |
| "epoch": 2.092, |
| "grad_norm": 77.5, |
| "learning_rate": 1.0602222222222223e-05, |
| "loss": 5.5055, |
| "mean_token_accuracy": 0.8994646199047566, |
| "num_tokens": 25555165.0, |
| "step": 5230 |
| }, |
| { |
| "entropy": 0.3520056060049683, |
| "epoch": 2.096, |
| "grad_norm": 105.0, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 5.5616, |
| "mean_token_accuracy": 0.9001151539385319, |
| "num_tokens": 25600511.0, |
| "step": 5240 |
| }, |
| { |
| "entropy": 0.3468271298799664, |
| "epoch": 2.1, |
| "grad_norm": 93.0, |
| "learning_rate": 1.0557777777777778e-05, |
| "loss": 5.5238, |
| "mean_token_accuracy": 0.902195344492793, |
| "num_tokens": 25647398.0, |
| "step": 5250 |
| }, |
| { |
| "entropy": 0.3376512651331723, |
| "epoch": 2.104, |
| "grad_norm": 90.0, |
| "learning_rate": 1.0535555555555557e-05, |
| "loss": 5.3063, |
| "mean_token_accuracy": 0.9044670958071948, |
| "num_tokens": 25700023.0, |
| "step": 5260 |
| }, |
| { |
| "entropy": 0.31951607335358856, |
| "epoch": 2.108, |
| "grad_norm": 83.5, |
| "learning_rate": 1.0513333333333333e-05, |
| "loss": 5.0937, |
| "mean_token_accuracy": 0.9072355292737484, |
| "num_tokens": 25747545.0, |
| "step": 5270 |
| }, |
| { |
| "entropy": 0.3573747973423451, |
| "epoch": 2.112, |
| "grad_norm": 72.5, |
| "learning_rate": 1.0491111111111112e-05, |
| "loss": 5.6665, |
| "mean_token_accuracy": 0.895047577098012, |
| "num_tokens": 25796565.0, |
| "step": 5280 |
| }, |
| { |
| "entropy": 0.3414256284479052, |
| "epoch": 2.116, |
| "grad_norm": 78.0, |
| "learning_rate": 1.046888888888889e-05, |
| "loss": 5.4252, |
| "mean_token_accuracy": 0.9007425580173731, |
| "num_tokens": 25844713.0, |
| "step": 5290 |
| }, |
| { |
| "entropy": 0.3637283912394196, |
| "epoch": 2.12, |
| "grad_norm": 87.0, |
| "learning_rate": 1.0446666666666668e-05, |
| "loss": 5.819, |
| "mean_token_accuracy": 0.8972341198474169, |
| "num_tokens": 25892925.0, |
| "step": 5300 |
| }, |
| { |
| "epoch": 2.12, |
| "eval_biology_entropy": 0.6809677287042141, |
| "eval_biology_loss": 1.3074564933776855, |
| "eval_biology_mean_token_accuracy": 0.7081616702079773, |
| "eval_biology_num_tokens": 25892925.0, |
| "eval_biology_runtime": 34.4705, |
| "eval_biology_samples_per_second": 14.505, |
| "eval_biology_steps_per_second": 14.505, |
| "step": 5300 |
| }, |
| { |
| "epoch": 2.12, |
| "eval_chemistry_entropy": 0.43393542151153086, |
| "eval_chemistry_loss": 0.7827609777450562, |
| "eval_chemistry_mean_token_accuracy": 0.8060798108577728, |
| "eval_chemistry_num_tokens": 25892925.0, |
| "eval_chemistry_runtime": 39.1978, |
| "eval_chemistry_samples_per_second": 12.756, |
| "eval_chemistry_steps_per_second": 12.756, |
| "step": 5300 |
| }, |
| { |
| "entropy": 0.35925020920112727, |
| "epoch": 2.124, |
| "grad_norm": 80.0, |
| "learning_rate": 1.0424444444444445e-05, |
| "loss": 5.6587, |
| "mean_token_accuracy": 0.8991015840321779, |
| "num_tokens": 25941779.0, |
| "step": 5310 |
| }, |
| { |
| "entropy": 0.3487645898712799, |
| "epoch": 2.128, |
| "grad_norm": 82.5, |
| "learning_rate": 1.0402222222222223e-05, |
| "loss": 5.5031, |
| "mean_token_accuracy": 0.89953579492867, |
| "num_tokens": 25991156.0, |
| "step": 5320 |
| }, |
| { |
| "entropy": 0.3543167905882001, |
| "epoch": 2.132, |
| "grad_norm": 75.5, |
| "learning_rate": 1.038e-05, |
| "loss": 5.8372, |
| "mean_token_accuracy": 0.8936107587069273, |
| "num_tokens": 26046051.0, |
| "step": 5330 |
| }, |
| { |
| "entropy": 0.34262634366750716, |
| "epoch": 2.136, |
| "grad_norm": 84.0, |
| "learning_rate": 1.0357777777777778e-05, |
| "loss": 5.3034, |
| "mean_token_accuracy": 0.9032966937869787, |
| "num_tokens": 26095179.0, |
| "step": 5340 |
| }, |
| { |
| "entropy": 0.34080974087119104, |
| "epoch": 2.14, |
| "grad_norm": 86.0, |
| "learning_rate": 1.0335555555555556e-05, |
| "loss": 5.4823, |
| "mean_token_accuracy": 0.9005962569266558, |
| "num_tokens": 26146785.0, |
| "step": 5350 |
| }, |
| { |
| "entropy": 0.35654561570845544, |
| "epoch": 2.144, |
| "grad_norm": 89.0, |
| "learning_rate": 1.0313333333333335e-05, |
| "loss": 5.6659, |
| "mean_token_accuracy": 0.8971914123743773, |
| "num_tokens": 26193237.0, |
| "step": 5360 |
| }, |
| { |
| "entropy": 0.3564099142793566, |
| "epoch": 2.148, |
| "grad_norm": 78.0, |
| "learning_rate": 1.0291111111111111e-05, |
| "loss": 5.6224, |
| "mean_token_accuracy": 0.8963177409023046, |
| "num_tokens": 26245116.0, |
| "step": 5370 |
| }, |
| { |
| "entropy": 0.3443096862407401, |
| "epoch": 2.152, |
| "grad_norm": 77.0, |
| "learning_rate": 1.026888888888889e-05, |
| "loss": 5.6184, |
| "mean_token_accuracy": 0.8987428039312363, |
| "num_tokens": 26291850.0, |
| "step": 5380 |
| }, |
| { |
| "entropy": 0.3495043867267668, |
| "epoch": 2.156, |
| "grad_norm": 97.5, |
| "learning_rate": 1.0246666666666666e-05, |
| "loss": 5.6272, |
| "mean_token_accuracy": 0.8992205079644918, |
| "num_tokens": 26342863.0, |
| "step": 5390 |
| }, |
| { |
| "entropy": 0.3661334573291242, |
| "epoch": 2.16, |
| "grad_norm": 70.0, |
| "learning_rate": 1.0224444444444446e-05, |
| "loss": 5.747, |
| "mean_token_accuracy": 0.8980169851332903, |
| "num_tokens": 26387993.0, |
| "step": 5400 |
| }, |
| { |
| "epoch": 2.16, |
| "eval_biology_entropy": 0.7138346272706986, |
| "eval_biology_loss": 1.2817533016204834, |
| "eval_biology_mean_token_accuracy": 0.7087586787939072, |
| "eval_biology_num_tokens": 26387993.0, |
| "eval_biology_runtime": 34.4106, |
| "eval_biology_samples_per_second": 14.53, |
| "eval_biology_steps_per_second": 14.53, |
| "step": 5400 |
| }, |
| { |
| "epoch": 2.16, |
| "eval_chemistry_entropy": 0.4493626266568899, |
| "eval_chemistry_loss": 0.7716115713119507, |
| "eval_chemistry_mean_token_accuracy": 0.8059261897802353, |
| "eval_chemistry_num_tokens": 26387993.0, |
| "eval_chemistry_runtime": 39.1755, |
| "eval_chemistry_samples_per_second": 12.763, |
| "eval_chemistry_steps_per_second": 12.763, |
| "step": 5400 |
| }, |
| { |
| "entropy": 0.37763382866978645, |
| "epoch": 2.164, |
| "grad_norm": 84.5, |
| "learning_rate": 1.0202222222222221e-05, |
| "loss": 5.9653, |
| "mean_token_accuracy": 0.8920822974294425, |
| "num_tokens": 26437019.0, |
| "step": 5410 |
| }, |
| { |
| "entropy": 0.3422889744862914, |
| "epoch": 2.168, |
| "grad_norm": 80.5, |
| "learning_rate": 1.018e-05, |
| "loss": 5.2912, |
| "mean_token_accuracy": 0.9047667905688286, |
| "num_tokens": 26485327.0, |
| "step": 5420 |
| }, |
| { |
| "entropy": 0.3824759030714631, |
| "epoch": 2.172, |
| "grad_norm": 97.5, |
| "learning_rate": 1.0157777777777777e-05, |
| "loss": 6.2331, |
| "mean_token_accuracy": 0.8875840537250042, |
| "num_tokens": 26537684.0, |
| "step": 5430 |
| }, |
| { |
| "entropy": 0.3783214149996638, |
| "epoch": 2.176, |
| "grad_norm": 141.0, |
| "learning_rate": 1.0135555555555556e-05, |
| "loss": 6.0392, |
| "mean_token_accuracy": 0.8905686359852553, |
| "num_tokens": 26585269.0, |
| "step": 5440 |
| }, |
| { |
| "entropy": 0.35017434218898413, |
| "epoch": 2.18, |
| "grad_norm": 86.0, |
| "learning_rate": 1.0113333333333335e-05, |
| "loss": 5.4687, |
| "mean_token_accuracy": 0.9012813758105039, |
| "num_tokens": 26635624.0, |
| "step": 5450 |
| }, |
| { |
| "entropy": 0.3150952396448702, |
| "epoch": 2.184, |
| "grad_norm": 85.0, |
| "learning_rate": 1.0091111111111111e-05, |
| "loss": 5.0165, |
| "mean_token_accuracy": 0.907957437261939, |
| "num_tokens": 26683997.0, |
| "step": 5460 |
| }, |
| { |
| "entropy": 0.3295361045282334, |
| "epoch": 2.188, |
| "grad_norm": 80.5, |
| "learning_rate": 1.006888888888889e-05, |
| "loss": 5.3016, |
| "mean_token_accuracy": 0.9037402000278234, |
| "num_tokens": 26729106.0, |
| "step": 5470 |
| }, |
| { |
| "entropy": 0.34348361208103595, |
| "epoch": 2.192, |
| "grad_norm": 96.5, |
| "learning_rate": 1.0046666666666666e-05, |
| "loss": 5.4882, |
| "mean_token_accuracy": 0.9010420013219118, |
| "num_tokens": 26776318.0, |
| "step": 5480 |
| }, |
| { |
| "entropy": 0.3404245681595057, |
| "epoch": 2.196, |
| "grad_norm": 89.0, |
| "learning_rate": 1.0024444444444446e-05, |
| "loss": 5.3721, |
| "mean_token_accuracy": 0.902960704639554, |
| "num_tokens": 26825875.0, |
| "step": 5490 |
| }, |
| { |
| "entropy": 0.36795946648344396, |
| "epoch": 2.2, |
| "grad_norm": 74.5, |
| "learning_rate": 1.0002222222222222e-05, |
| "loss": 5.8561, |
| "mean_token_accuracy": 0.8949945516884327, |
| "num_tokens": 26876705.0, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.2, |
| "eval_biology_entropy": 0.692384944587946, |
| "eval_biology_loss": 1.2900505065917969, |
| "eval_biology_mean_token_accuracy": 0.7090325899720192, |
| "eval_biology_num_tokens": 26876705.0, |
| "eval_biology_runtime": 34.5176, |
| "eval_biology_samples_per_second": 14.485, |
| "eval_biology_steps_per_second": 14.485, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.2, |
| "eval_chemistry_entropy": 0.43885773393511773, |
| "eval_chemistry_loss": 0.7766873240470886, |
| "eval_chemistry_mean_token_accuracy": 0.8052822499275207, |
| "eval_chemistry_num_tokens": 26876705.0, |
| "eval_chemistry_runtime": 39.0541, |
| "eval_chemistry_samples_per_second": 12.803, |
| "eval_chemistry_steps_per_second": 12.803, |
| "step": 5500 |
| }, |
| { |
| "entropy": 0.3372880739159882, |
| "epoch": 2.204, |
| "grad_norm": 78.5, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 5.3749, |
| "mean_token_accuracy": 0.9033362224698067, |
| "num_tokens": 26924361.0, |
| "step": 5510 |
| }, |
| { |
| "entropy": 0.3626332323532552, |
| "epoch": 2.208, |
| "grad_norm": 74.5, |
| "learning_rate": 9.957777777777779e-06, |
| "loss": 5.7476, |
| "mean_token_accuracy": 0.8974151767790317, |
| "num_tokens": 26974420.0, |
| "step": 5520 |
| }, |
| { |
| "entropy": 0.3565032765269279, |
| "epoch": 2.212, |
| "grad_norm": 106.0, |
| "learning_rate": 9.935555555555556e-06, |
| "loss": 5.8661, |
| "mean_token_accuracy": 0.8928722627460957, |
| "num_tokens": 27022334.0, |
| "step": 5530 |
| }, |
| { |
| "entropy": 0.34673582690302285, |
| "epoch": 2.216, |
| "grad_norm": 86.5, |
| "learning_rate": 9.913333333333334e-06, |
| "loss": 5.4139, |
| "mean_token_accuracy": 0.9004093270748854, |
| "num_tokens": 27072242.0, |
| "step": 5540 |
| }, |
| { |
| "entropy": 0.3444731017574668, |
| "epoch": 2.22, |
| "grad_norm": 84.5, |
| "learning_rate": 9.891111111111113e-06, |
| "loss": 5.4524, |
| "mean_token_accuracy": 0.900804677978158, |
| "num_tokens": 27120624.0, |
| "step": 5550 |
| }, |
| { |
| "entropy": 0.3635823673568666, |
| "epoch": 2.224, |
| "grad_norm": 84.0, |
| "learning_rate": 9.86888888888889e-06, |
| "loss": 5.6865, |
| "mean_token_accuracy": 0.8966283708810806, |
| "num_tokens": 27167577.0, |
| "step": 5560 |
| }, |
| { |
| "entropy": 0.33280278779566286, |
| "epoch": 2.228, |
| "grad_norm": 83.5, |
| "learning_rate": 9.846666666666668e-06, |
| "loss": 5.5317, |
| "mean_token_accuracy": 0.8989233128726483, |
| "num_tokens": 27221846.0, |
| "step": 5570 |
| }, |
| { |
| "entropy": 0.37154059871099887, |
| "epoch": 2.232, |
| "grad_norm": 94.5, |
| "learning_rate": 9.824444444444446e-06, |
| "loss": 5.8831, |
| "mean_token_accuracy": 0.8943906705826521, |
| "num_tokens": 27267457.0, |
| "step": 5580 |
| }, |
| { |
| "entropy": 0.3668776168487966, |
| "epoch": 2.2359999999999998, |
| "grad_norm": 86.0, |
| "learning_rate": 9.802222222222224e-06, |
| "loss": 5.6025, |
| "mean_token_accuracy": 0.8962242476642132, |
| "num_tokens": 27315962.0, |
| "step": 5590 |
| }, |
| { |
| "entropy": 0.328641360392794, |
| "epoch": 2.24, |
| "grad_norm": 75.5, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 5.3663, |
| "mean_token_accuracy": 0.9018039170652628, |
| "num_tokens": 27364189.0, |
| "step": 5600 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_biology_entropy": 0.6802926383018494, |
| "eval_biology_loss": 1.3054646253585815, |
| "eval_biology_mean_token_accuracy": 0.7086836161613465, |
| "eval_biology_num_tokens": 27364189.0, |
| "eval_biology_runtime": 34.4789, |
| "eval_biology_samples_per_second": 14.502, |
| "eval_biology_steps_per_second": 14.502, |
| "step": 5600 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_chemistry_entropy": 0.42942048321664333, |
| "eval_chemistry_loss": 0.7878117561340332, |
| "eval_chemistry_mean_token_accuracy": 0.8056240211725235, |
| "eval_chemistry_num_tokens": 27364189.0, |
| "eval_chemistry_runtime": 39.2294, |
| "eval_chemistry_samples_per_second": 12.746, |
| "eval_chemistry_steps_per_second": 12.746, |
| "step": 5600 |
| }, |
| { |
| "entropy": 0.32891160105355083, |
| "epoch": 2.2439999999999998, |
| "grad_norm": 83.5, |
| "learning_rate": 9.757777777777779e-06, |
| "loss": 5.1989, |
| "mean_token_accuracy": 0.9064031202346087, |
| "num_tokens": 27410356.0, |
| "step": 5610 |
| }, |
| { |
| "entropy": 0.34276762397494165, |
| "epoch": 2.248, |
| "grad_norm": 71.5, |
| "learning_rate": 9.735555555555556e-06, |
| "loss": 5.3333, |
| "mean_token_accuracy": 0.9020731687545777, |
| "num_tokens": 27456862.0, |
| "step": 5620 |
| }, |
| { |
| "entropy": 0.33199566826224325, |
| "epoch": 2.252, |
| "grad_norm": 92.5, |
| "learning_rate": 9.713333333333334e-06, |
| "loss": 5.302, |
| "mean_token_accuracy": 0.902812210097909, |
| "num_tokens": 27508193.0, |
| "step": 5630 |
| }, |
| { |
| "entropy": 0.33429202530533075, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 74.5, |
| "learning_rate": 9.691111111111112e-06, |
| "loss": 5.3909, |
| "mean_token_accuracy": 0.9017354410141707, |
| "num_tokens": 27557387.0, |
| "step": 5640 |
| }, |
| { |
| "entropy": 0.34626029995270075, |
| "epoch": 2.26, |
| "grad_norm": 82.0, |
| "learning_rate": 9.66888888888889e-06, |
| "loss": 5.5302, |
| "mean_token_accuracy": 0.9014634776860475, |
| "num_tokens": 27604472.0, |
| "step": 5650 |
| }, |
| { |
| "entropy": 0.3552477842196822, |
| "epoch": 2.2640000000000002, |
| "grad_norm": 82.0, |
| "learning_rate": 9.646666666666667e-06, |
| "loss": 5.5513, |
| "mean_token_accuracy": 0.8960914067924023, |
| "num_tokens": 27655988.0, |
| "step": 5660 |
| }, |
| { |
| "entropy": 0.3381340900901705, |
| "epoch": 2.268, |
| "grad_norm": 73.5, |
| "learning_rate": 9.624444444444445e-06, |
| "loss": 5.3368, |
| "mean_token_accuracy": 0.9032924998551607, |
| "num_tokens": 27704024.0, |
| "step": 5670 |
| }, |
| { |
| "entropy": 0.3595274162944406, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 82.5, |
| "learning_rate": 9.602222222222222e-06, |
| "loss": 5.8296, |
| "mean_token_accuracy": 0.8925424657762051, |
| "num_tokens": 27755725.0, |
| "step": 5680 |
| }, |
| { |
| "entropy": 0.35308567942120134, |
| "epoch": 2.276, |
| "grad_norm": 87.0, |
| "learning_rate": 9.58e-06, |
| "loss": 5.5848, |
| "mean_token_accuracy": 0.8981157165020704, |
| "num_tokens": 27807954.0, |
| "step": 5690 |
| }, |
| { |
| "entropy": 0.3619683378841728, |
| "epoch": 2.2800000000000002, |
| "grad_norm": 82.5, |
| "learning_rate": 9.557777777777777e-06, |
| "loss": 5.8017, |
| "mean_token_accuracy": 0.8936318770051003, |
| "num_tokens": 27854353.0, |
| "step": 5700 |
| }, |
| { |
| "epoch": 2.2800000000000002, |
| "eval_biology_entropy": 0.6790939620435238, |
| "eval_biology_loss": 1.3060617446899414, |
| "eval_biology_mean_token_accuracy": 0.7090682356357575, |
| "eval_biology_num_tokens": 27854353.0, |
| "eval_biology_runtime": 34.5223, |
| "eval_biology_samples_per_second": 14.483, |
| "eval_biology_steps_per_second": 14.483, |
| "step": 5700 |
| }, |
| { |
| "epoch": 2.2800000000000002, |
| "eval_chemistry_entropy": 0.4305421074181795, |
| "eval_chemistry_loss": 0.7880338430404663, |
| "eval_chemistry_mean_token_accuracy": 0.805678306221962, |
| "eval_chemistry_num_tokens": 27854353.0, |
| "eval_chemistry_runtime": 39.2281, |
| "eval_chemistry_samples_per_second": 12.746, |
| "eval_chemistry_steps_per_second": 12.746, |
| "step": 5700 |
| }, |
| { |
| "entropy": 0.33629751061089336, |
| "epoch": 2.284, |
| "grad_norm": 82.0, |
| "learning_rate": 9.535555555555557e-06, |
| "loss": 5.4636, |
| "mean_token_accuracy": 0.9047619588673115, |
| "num_tokens": 27899614.0, |
| "step": 5710 |
| }, |
| { |
| "entropy": 0.3628993091639131, |
| "epoch": 2.288, |
| "grad_norm": 82.0, |
| "learning_rate": 9.513333333333334e-06, |
| "loss": 5.6917, |
| "mean_token_accuracy": 0.8948325637727976, |
| "num_tokens": 27950694.0, |
| "step": 5720 |
| }, |
| { |
| "entropy": 0.33335186340846124, |
| "epoch": 2.292, |
| "grad_norm": 95.5, |
| "learning_rate": 9.491111111111112e-06, |
| "loss": 5.4098, |
| "mean_token_accuracy": 0.901162626966834, |
| "num_tokens": 28001094.0, |
| "step": 5730 |
| }, |
| { |
| "entropy": 0.358959797443822, |
| "epoch": 2.296, |
| "grad_norm": 96.5, |
| "learning_rate": 9.46888888888889e-06, |
| "loss": 5.6004, |
| "mean_token_accuracy": 0.900016188248992, |
| "num_tokens": 28048799.0, |
| "step": 5740 |
| }, |
| { |
| "entropy": 0.3797079413663596, |
| "epoch": 2.3, |
| "grad_norm": 84.5, |
| "learning_rate": 9.446666666666667e-06, |
| "loss": 6.1567, |
| "mean_token_accuracy": 0.887949476018548, |
| "num_tokens": 28099315.0, |
| "step": 5750 |
| }, |
| { |
| "entropy": 0.3543362025171518, |
| "epoch": 2.304, |
| "grad_norm": 77.0, |
| "learning_rate": 9.424444444444445e-06, |
| "loss": 5.6625, |
| "mean_token_accuracy": 0.8960063226521016, |
| "num_tokens": 28150719.0, |
| "step": 5760 |
| }, |
| { |
| "entropy": 0.3434872718062252, |
| "epoch": 2.308, |
| "grad_norm": 88.0, |
| "learning_rate": 9.402222222222222e-06, |
| "loss": 5.3404, |
| "mean_token_accuracy": 0.9025525059551001, |
| "num_tokens": 28199787.0, |
| "step": 5770 |
| }, |
| { |
| "entropy": 0.3476636452134699, |
| "epoch": 2.312, |
| "grad_norm": 99.0, |
| "learning_rate": 9.38e-06, |
| "loss": 5.6001, |
| "mean_token_accuracy": 0.8989078853279352, |
| "num_tokens": 28248071.0, |
| "step": 5780 |
| }, |
| { |
| "entropy": 0.3676468168850988, |
| "epoch": 2.316, |
| "grad_norm": 91.5, |
| "learning_rate": 9.35777777777778e-06, |
| "loss": 5.7236, |
| "mean_token_accuracy": 0.8957135591655969, |
| "num_tokens": 28297970.0, |
| "step": 5790 |
| }, |
| { |
| "entropy": 0.36766890343278646, |
| "epoch": 2.32, |
| "grad_norm": 85.0, |
| "learning_rate": 9.335555555555557e-06, |
| "loss": 5.8665, |
| "mean_token_accuracy": 0.8931352898478508, |
| "num_tokens": 28348652.0, |
| "step": 5800 |
| }, |
| { |
| "epoch": 2.32, |
| "eval_biology_entropy": 0.6775494822263718, |
| "eval_biology_loss": 1.3076176643371582, |
| "eval_biology_mean_token_accuracy": 0.7095211740136147, |
| "eval_biology_num_tokens": 28348652.0, |
| "eval_biology_runtime": 34.3708, |
| "eval_biology_samples_per_second": 14.547, |
| "eval_biology_steps_per_second": 14.547, |
| "step": 5800 |
| }, |
| { |
| "epoch": 2.32, |
| "eval_chemistry_entropy": 0.42541772104799747, |
| "eval_chemistry_loss": 0.7929761409759521, |
| "eval_chemistry_mean_token_accuracy": 0.8055232439041138, |
| "eval_chemistry_num_tokens": 28348652.0, |
| "eval_chemistry_runtime": 39.1216, |
| "eval_chemistry_samples_per_second": 12.781, |
| "eval_chemistry_steps_per_second": 12.781, |
| "step": 5800 |
| }, |
| { |
| "entropy": 0.36581603963859377, |
| "epoch": 2.324, |
| "grad_norm": 109.5, |
| "learning_rate": 9.313333333333335e-06, |
| "loss": 5.8539, |
| "mean_token_accuracy": 0.8923335958272218, |
| "num_tokens": 28397613.0, |
| "step": 5810 |
| }, |
| { |
| "entropy": 0.33900672756135464, |
| "epoch": 2.328, |
| "grad_norm": 77.5, |
| "learning_rate": 9.291111111111112e-06, |
| "loss": 5.379, |
| "mean_token_accuracy": 0.9033070992678404, |
| "num_tokens": 28446430.0, |
| "step": 5820 |
| }, |
| { |
| "entropy": 0.3558291880879551, |
| "epoch": 2.332, |
| "grad_norm": 76.0, |
| "learning_rate": 9.26888888888889e-06, |
| "loss": 5.633, |
| "mean_token_accuracy": 0.896314574033022, |
| "num_tokens": 28496520.0, |
| "step": 5830 |
| }, |
| { |
| "entropy": 0.33307430893182755, |
| "epoch": 2.336, |
| "grad_norm": 90.0, |
| "learning_rate": 9.246666666666667e-06, |
| "loss": 5.2189, |
| "mean_token_accuracy": 0.9069060478359461, |
| "num_tokens": 28542945.0, |
| "step": 5840 |
| }, |
| { |
| "entropy": 0.3442193596623838, |
| "epoch": 2.34, |
| "grad_norm": 90.0, |
| "learning_rate": 9.224444444444445e-06, |
| "loss": 5.5058, |
| "mean_token_accuracy": 0.899949275329709, |
| "num_tokens": 28588842.0, |
| "step": 5850 |
| }, |
| { |
| "entropy": 0.3473319072742015, |
| "epoch": 2.344, |
| "grad_norm": 106.5, |
| "learning_rate": 9.202222222222224e-06, |
| "loss": 5.5765, |
| "mean_token_accuracy": 0.8980496440082788, |
| "num_tokens": 28636050.0, |
| "step": 5860 |
| }, |
| { |
| "entropy": 0.3654441684484482, |
| "epoch": 2.348, |
| "grad_norm": 97.5, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 5.9377, |
| "mean_token_accuracy": 0.8941320039331913, |
| "num_tokens": 28684299.0, |
| "step": 5870 |
| }, |
| { |
| "entropy": 0.3462490413803607, |
| "epoch": 2.352, |
| "grad_norm": 84.5, |
| "learning_rate": 9.15777777777778e-06, |
| "loss": 5.3802, |
| "mean_token_accuracy": 0.9020478032529354, |
| "num_tokens": 28731927.0, |
| "step": 5880 |
| }, |
| { |
| "entropy": 0.3592828899621964, |
| "epoch": 2.356, |
| "grad_norm": 76.5, |
| "learning_rate": 9.135555555555557e-06, |
| "loss": 5.6701, |
| "mean_token_accuracy": 0.8949876382946969, |
| "num_tokens": 28780720.0, |
| "step": 5890 |
| }, |
| { |
| "entropy": 0.3423851332627237, |
| "epoch": 2.36, |
| "grad_norm": 94.0, |
| "learning_rate": 9.113333333333335e-06, |
| "loss": 5.5412, |
| "mean_token_accuracy": 0.9002565760165453, |
| "num_tokens": 28829775.0, |
| "step": 5900 |
| }, |
| { |
| "epoch": 2.36, |
| "eval_biology_entropy": 0.6726023618280887, |
| "eval_biology_loss": 1.3191787004470825, |
| "eval_biology_mean_token_accuracy": 0.7086760124564171, |
| "eval_biology_num_tokens": 28829775.0, |
| "eval_biology_runtime": 34.4662, |
| "eval_biology_samples_per_second": 14.507, |
| "eval_biology_steps_per_second": 14.507, |
| "step": 5900 |
| }, |
| { |
| "epoch": 2.36, |
| "eval_chemistry_entropy": 0.4267604645192623, |
| "eval_chemistry_loss": 0.7931302785873413, |
| "eval_chemistry_mean_token_accuracy": 0.8050788148641587, |
| "eval_chemistry_num_tokens": 28829775.0, |
| "eval_chemistry_runtime": 39.174, |
| "eval_chemistry_samples_per_second": 12.764, |
| "eval_chemistry_steps_per_second": 12.764, |
| "step": 5900 |
| }, |
| { |
| "entropy": 0.32639851826243105, |
| "epoch": 2.364, |
| "grad_norm": 67.5, |
| "learning_rate": 9.091111111111112e-06, |
| "loss": 5.2265, |
| "mean_token_accuracy": 0.9054272320121527, |
| "num_tokens": 28878516.0, |
| "step": 5910 |
| }, |
| { |
| "entropy": 0.35729803815484046, |
| "epoch": 2.368, |
| "grad_norm": 89.5, |
| "learning_rate": 9.06888888888889e-06, |
| "loss": 5.5114, |
| "mean_token_accuracy": 0.9008417587727309, |
| "num_tokens": 28929298.0, |
| "step": 5920 |
| }, |
| { |
| "entropy": 0.3330498012248427, |
| "epoch": 2.372, |
| "grad_norm": 86.5, |
| "learning_rate": 9.046666666666668e-06, |
| "loss": 5.3957, |
| "mean_token_accuracy": 0.9011346828192472, |
| "num_tokens": 28981670.0, |
| "step": 5930 |
| }, |
| { |
| "entropy": 0.3523591878823936, |
| "epoch": 2.376, |
| "grad_norm": 77.5, |
| "learning_rate": 9.024444444444445e-06, |
| "loss": 5.6407, |
| "mean_token_accuracy": 0.8988840639591217, |
| "num_tokens": 29027827.0, |
| "step": 5940 |
| }, |
| { |
| "entropy": 0.33195446808822454, |
| "epoch": 2.38, |
| "grad_norm": 89.5, |
| "learning_rate": 9.002222222222223e-06, |
| "loss": 5.3825, |
| "mean_token_accuracy": 0.9034351270645857, |
| "num_tokens": 29074413.0, |
| "step": 5950 |
| }, |
| { |
| "entropy": 0.34116955525241793, |
| "epoch": 2.384, |
| "grad_norm": 85.5, |
| "learning_rate": 8.98e-06, |
| "loss": 5.3772, |
| "mean_token_accuracy": 0.9002278726547956, |
| "num_tokens": 29121142.0, |
| "step": 5960 |
| }, |
| { |
| "entropy": 0.353854395262897, |
| "epoch": 2.388, |
| "grad_norm": 93.0, |
| "learning_rate": 8.957777777777778e-06, |
| "loss": 5.6522, |
| "mean_token_accuracy": 0.898543794080615, |
| "num_tokens": 29169695.0, |
| "step": 5970 |
| }, |
| { |
| "entropy": 0.35660858717747035, |
| "epoch": 2.392, |
| "grad_norm": 78.0, |
| "learning_rate": 8.935555555555556e-06, |
| "loss": 5.6549, |
| "mean_token_accuracy": 0.8972622275352478, |
| "num_tokens": 29218220.0, |
| "step": 5980 |
| }, |
| { |
| "entropy": 0.34393711425364015, |
| "epoch": 2.396, |
| "grad_norm": 76.0, |
| "learning_rate": 8.913333333333333e-06, |
| "loss": 5.5077, |
| "mean_token_accuracy": 0.8999034509062767, |
| "num_tokens": 29266851.0, |
| "step": 5990 |
| }, |
| { |
| "entropy": 0.33578765713609754, |
| "epoch": 2.4, |
| "grad_norm": 97.0, |
| "learning_rate": 8.891111111111111e-06, |
| "loss": 5.2672, |
| "mean_token_accuracy": 0.9027582634240389, |
| "num_tokens": 29317919.0, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_biology_entropy": 0.6759352746009827, |
| "eval_biology_loss": 1.3154243230819702, |
| "eval_biology_mean_token_accuracy": 0.7083351585865021, |
| "eval_biology_num_tokens": 29317919.0, |
| "eval_biology_runtime": 34.578, |
| "eval_biology_samples_per_second": 14.46, |
| "eval_biology_steps_per_second": 14.46, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_chemistry_entropy": 0.42978759948909284, |
| "eval_chemistry_loss": 0.7914813160896301, |
| "eval_chemistry_mean_token_accuracy": 0.8055099663734436, |
| "eval_chemistry_num_tokens": 29317919.0, |
| "eval_chemistry_runtime": 39.0167, |
| "eval_chemistry_samples_per_second": 12.815, |
| "eval_chemistry_steps_per_second": 12.815, |
| "step": 6000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "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": 1.9658944817733494e+18, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
