| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 100, |
| "global_step": 2500, |
| "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 |
| } |
| ], |
| "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": 8.19490924203941e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
