| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 7.117509597729928, |
| "eval_steps": 100, |
| "global_step": 8000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 2.4347459591925142, |
| "epoch": 0.00890224225226729, |
| "grad_norm": 0.00732421875, |
| "learning_rate": 1.8e-07, |
| "loss": 3.3392, |
| "mean_token_accuracy": 0.4414398778229952, |
| "num_tokens": 128968.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 2.5189025782048704, |
| "epoch": 0.01780448450453458, |
| "grad_norm": 0.0098876953125, |
| "learning_rate": 3.8e-07, |
| "loss": 3.4461, |
| "mean_token_accuracy": 0.42950069066137075, |
| "num_tokens": 257189.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 2.4521907046437263, |
| "epoch": 0.026706726756801868, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 3.3929, |
| "mean_token_accuracy": 0.43435896690934894, |
| "num_tokens": 394654.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 2.4908889502286913, |
| "epoch": 0.03560896900906916, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 7.8e-07, |
| "loss": 3.3731, |
| "mean_token_accuracy": 0.4337937039323151, |
| "num_tokens": 528285.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 2.4656582184135916, |
| "epoch": 0.04451121126133645, |
| "grad_norm": 0.0078125, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 3.3705, |
| "mean_token_accuracy": 0.4305401614867151, |
| "num_tokens": 663126.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 2.5354390598833563, |
| "epoch": 0.053413453513603736, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 3.4166, |
| "mean_token_accuracy": 0.42836275026202203, |
| "num_tokens": 801635.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 2.496990965306759, |
| "epoch": 0.06231569576587103, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 3.4009, |
| "mean_token_accuracy": 0.43142524575814606, |
| "num_tokens": 931344.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 2.5406345002353192, |
| "epoch": 0.07121793801813832, |
| "grad_norm": 0.007171630859375, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 3.4506, |
| "mean_token_accuracy": 0.42618109108880164, |
| "num_tokens": 1060812.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 2.4690978921949864, |
| "epoch": 0.08012018027040561, |
| "grad_norm": 0.0111083984375, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 3.375, |
| "mean_token_accuracy": 0.4317979410290718, |
| "num_tokens": 1200650.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 2.537689981609583, |
| "epoch": 0.0890224225226729, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 1.98e-06, |
| "loss": 3.4588, |
| "mean_token_accuracy": 0.4282337296754122, |
| "num_tokens": 1327380.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.0890224225226729, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 1327380.0, |
| "eval_coding_runtime": 96.2988, |
| "eval_coding_samples_per_second": 5.192, |
| "eval_coding_steps_per_second": 2.596, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.0890224225226729, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 1327380.0, |
| "eval_chemistry_runtime": 53.689, |
| "eval_chemistry_samples_per_second": 9.313, |
| "eval_chemistry_steps_per_second": 4.656, |
| "step": 100 |
| }, |
| { |
| "entropy": 2.541736252605915, |
| "epoch": 0.09792466477494019, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 3.4778, |
| "mean_token_accuracy": 0.42182709854096173, |
| "num_tokens": 1460130.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 2.4891932401806116, |
| "epoch": 0.10682690702720747, |
| "grad_norm": 0.0091552734375, |
| "learning_rate": 2.38e-06, |
| "loss": 3.4123, |
| "mean_token_accuracy": 0.4318336697295308, |
| "num_tokens": 1597405.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 2.549714620411396, |
| "epoch": 0.11572914927947477, |
| "grad_norm": 0.013427734375, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 3.5233, |
| "mean_token_accuracy": 0.4176134932786226, |
| "num_tokens": 1728207.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 2.5017867423594, |
| "epoch": 0.12463139153174206, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 3.4252, |
| "mean_token_accuracy": 0.4325200604274869, |
| "num_tokens": 1859684.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 2.607860314100981, |
| "epoch": 0.13353363378400934, |
| "grad_norm": 0.0086669921875, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 3.5854, |
| "mean_token_accuracy": 0.41672497261315583, |
| "num_tokens": 1985766.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 2.546873392164707, |
| "epoch": 0.14243587603627664, |
| "grad_norm": 0.0096435546875, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 3.4451, |
| "mean_token_accuracy": 0.42558788452297447, |
| "num_tokens": 2118068.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 2.5062545992434027, |
| "epoch": 0.15133811828854393, |
| "grad_norm": 0.0125732421875, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 3.4525, |
| "mean_token_accuracy": 0.42451597433537247, |
| "num_tokens": 2252650.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 2.490347370505333, |
| "epoch": 0.16024036054081123, |
| "grad_norm": 0.0106201171875, |
| "learning_rate": 3.58e-06, |
| "loss": 3.398, |
| "mean_token_accuracy": 0.4340389359742403, |
| "num_tokens": 2388824.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 2.5069186124950646, |
| "epoch": 0.1691426027930785, |
| "grad_norm": 0.01055908203125, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 3.4995, |
| "mean_token_accuracy": 0.42545853555202484, |
| "num_tokens": 2515325.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 2.569493807852268, |
| "epoch": 0.1780448450453458, |
| "grad_norm": 0.0174560546875, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 3.5293, |
| "mean_token_accuracy": 0.4203047439455986, |
| "num_tokens": 2644330.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1780448450453458, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 2644330.0, |
| "eval_coding_runtime": 95.9249, |
| "eval_coding_samples_per_second": 5.212, |
| "eval_coding_steps_per_second": 2.606, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1780448450453458, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 2644330.0, |
| "eval_chemistry_runtime": 53.6456, |
| "eval_chemistry_samples_per_second": 9.32, |
| "eval_chemistry_steps_per_second": 4.66, |
| "step": 200 |
| }, |
| { |
| "entropy": 2.463035849481821, |
| "epoch": 0.18694708729761308, |
| "grad_norm": 0.01226806640625, |
| "learning_rate": 4.18e-06, |
| "loss": 3.3809, |
| "mean_token_accuracy": 0.43558520982041954, |
| "num_tokens": 2783261.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 2.5025305368006228, |
| "epoch": 0.19584932954988038, |
| "grad_norm": 0.007476806640625, |
| "learning_rate": 4.38e-06, |
| "loss": 3.4253, |
| "mean_token_accuracy": 0.4310411293059587, |
| "num_tokens": 2913700.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 2.455498095601797, |
| "epoch": 0.20475157180214767, |
| "grad_norm": 0.0068359375, |
| "learning_rate": 4.58e-06, |
| "loss": 3.3586, |
| "mean_token_accuracy": 0.43719491101801394, |
| "num_tokens": 3048973.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 2.5426308810710907, |
| "epoch": 0.21365381405441494, |
| "grad_norm": 0.007598876953125, |
| "learning_rate": 4.78e-06, |
| "loss": 3.406, |
| "mean_token_accuracy": 0.43309843018651006, |
| "num_tokens": 3185255.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 2.444308698922396, |
| "epoch": 0.22255605630668224, |
| "grad_norm": 0.01324462890625, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 3.337, |
| "mean_token_accuracy": 0.44262526389211415, |
| "num_tokens": 3322823.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 2.526992666721344, |
| "epoch": 0.23145829855894953, |
| "grad_norm": 0.0103759765625, |
| "learning_rate": 5.18e-06, |
| "loss": 3.478, |
| "mean_token_accuracy": 0.42772163953632114, |
| "num_tokens": 3454750.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 2.468804422393441, |
| "epoch": 0.24036054081121683, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 3.3899, |
| "mean_token_accuracy": 0.4310757084749639, |
| "num_tokens": 3589449.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 2.5963171511888503, |
| "epoch": 0.24926278306348412, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 3.5597, |
| "mean_token_accuracy": 0.4164831433445215, |
| "num_tokens": 3719113.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 2.6820507936179636, |
| "epoch": 0.2581650253157514, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 5.78e-06, |
| "loss": 3.6014, |
| "mean_token_accuracy": 0.4143718365579844, |
| "num_tokens": 3852276.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 2.5068675063550474, |
| "epoch": 0.2670672675680187, |
| "grad_norm": 0.01226806640625, |
| "learning_rate": 5.98e-06, |
| "loss": 3.4183, |
| "mean_token_accuracy": 0.4291886316612363, |
| "num_tokens": 3990505.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.2670672675680187, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 3990505.0, |
| "eval_coding_runtime": 95.8518, |
| "eval_coding_samples_per_second": 5.216, |
| "eval_coding_steps_per_second": 2.608, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.2670672675680187, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 3990505.0, |
| "eval_chemistry_runtime": 53.7283, |
| "eval_chemistry_samples_per_second": 9.306, |
| "eval_chemistry_steps_per_second": 4.653, |
| "step": 300 |
| }, |
| { |
| "entropy": 2.4614722445607184, |
| "epoch": 0.275969509820286, |
| "grad_norm": 0.0091552734375, |
| "learning_rate": 6.18e-06, |
| "loss": 3.3329, |
| "mean_token_accuracy": 0.4380343802273273, |
| "num_tokens": 4133119.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 2.4565507762134073, |
| "epoch": 0.2848717520725533, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 3.376, |
| "mean_token_accuracy": 0.4339581592008471, |
| "num_tokens": 4267403.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 2.4726619601249693, |
| "epoch": 0.29377399432482054, |
| "grad_norm": 0.007781982421875, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 3.3584, |
| "mean_token_accuracy": 0.43662677630782126, |
| "num_tokens": 4400242.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 2.4780636690557003, |
| "epoch": 0.30267623657708786, |
| "grad_norm": 0.009765625, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 3.3561, |
| "mean_token_accuracy": 0.4329976743087173, |
| "num_tokens": 4535458.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 2.4669046089053155, |
| "epoch": 0.31157847882935513, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 6.98e-06, |
| "loss": 3.4031, |
| "mean_token_accuracy": 0.43433742690831423, |
| "num_tokens": 4662903.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 2.5689725764095783, |
| "epoch": 0.32048072108162245, |
| "grad_norm": 0.00732421875, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 3.4889, |
| "mean_token_accuracy": 0.4287034338340163, |
| "num_tokens": 4796815.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 2.4410855643451215, |
| "epoch": 0.3293829633338897, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 3.3643, |
| "mean_token_accuracy": 0.43710872549563645, |
| "num_tokens": 4932671.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 2.5445352010428905, |
| "epoch": 0.338285205586157, |
| "grad_norm": 0.01129150390625, |
| "learning_rate": 7.58e-06, |
| "loss": 3.4525, |
| "mean_token_accuracy": 0.42816965784877536, |
| "num_tokens": 5066948.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 2.565714708715677, |
| "epoch": 0.3471874478384243, |
| "grad_norm": 0.01470947265625, |
| "learning_rate": 7.78e-06, |
| "loss": 3.4661, |
| "mean_token_accuracy": 0.4264904214069247, |
| "num_tokens": 5193519.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 2.516895131766796, |
| "epoch": 0.3560896900906916, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 3.4401, |
| "mean_token_accuracy": 0.4265883769840002, |
| "num_tokens": 5324751.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.3560896900906916, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 5324751.0, |
| "eval_coding_runtime": 95.8822, |
| "eval_coding_samples_per_second": 5.215, |
| "eval_coding_steps_per_second": 2.607, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.3560896900906916, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 5324751.0, |
| "eval_chemistry_runtime": 53.6645, |
| "eval_chemistry_samples_per_second": 9.317, |
| "eval_chemistry_steps_per_second": 4.659, |
| "step": 400 |
| }, |
| { |
| "entropy": 2.4508687026798723, |
| "epoch": 0.3649919323429589, |
| "grad_norm": 0.010986328125, |
| "learning_rate": 8.18e-06, |
| "loss": 3.3874, |
| "mean_token_accuracy": 0.4385754197835922, |
| "num_tokens": 5458519.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 2.4801586009562016, |
| "epoch": 0.37389417459522617, |
| "grad_norm": 0.00714111328125, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 3.4012, |
| "mean_token_accuracy": 0.43559422083199023, |
| "num_tokens": 5585508.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 2.4704861216247083, |
| "epoch": 0.38279641684749344, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 3.4353, |
| "mean_token_accuracy": 0.43275914546102284, |
| "num_tokens": 5716321.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 2.537934695184231, |
| "epoch": 0.39169865909976076, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 8.78e-06, |
| "loss": 3.4176, |
| "mean_token_accuracy": 0.4291856364347041, |
| "num_tokens": 5848889.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 2.513765400648117, |
| "epoch": 0.400600901352028, |
| "grad_norm": 0.00927734375, |
| "learning_rate": 8.98e-06, |
| "loss": 3.4106, |
| "mean_token_accuracy": 0.42827712278813124, |
| "num_tokens": 5980229.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 2.4634178556501864, |
| "epoch": 0.40950314360429535, |
| "grad_norm": 0.01068115234375, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 3.4007, |
| "mean_token_accuracy": 0.4337839787825942, |
| "num_tokens": 6114855.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 2.5320842534303667, |
| "epoch": 0.4184053858565626, |
| "grad_norm": 0.007049560546875, |
| "learning_rate": 9.38e-06, |
| "loss": 3.4216, |
| "mean_token_accuracy": 0.4268041457980871, |
| "num_tokens": 6244987.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 2.4959075897932053, |
| "epoch": 0.4273076281088299, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 9.58e-06, |
| "loss": 3.3715, |
| "mean_token_accuracy": 0.432720182929188, |
| "num_tokens": 6378152.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 2.503755620121956, |
| "epoch": 0.4362098703610972, |
| "grad_norm": 0.01708984375, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 3.4343, |
| "mean_token_accuracy": 0.43189559010788797, |
| "num_tokens": 6510231.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 2.539106211811304, |
| "epoch": 0.4451121126133645, |
| "grad_norm": 0.0115966796875, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 3.4747, |
| "mean_token_accuracy": 0.42332094311714175, |
| "num_tokens": 6637273.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4451121126133645, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 6637273.0, |
| "eval_coding_runtime": 95.8134, |
| "eval_coding_samples_per_second": 5.218, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4451121126133645, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 6637273.0, |
| "eval_chemistry_runtime": 53.3031, |
| "eval_chemistry_samples_per_second": 9.38, |
| "eval_chemistry_steps_per_second": 4.69, |
| "step": 500 |
| }, |
| { |
| "entropy": 2.536660289019346, |
| "epoch": 0.4540143548656318, |
| "grad_norm": 0.0146484375, |
| "learning_rate": 1.018e-05, |
| "loss": 3.4613, |
| "mean_token_accuracy": 0.4232974941842258, |
| "num_tokens": 6765798.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 2.459078107774258, |
| "epoch": 0.46291659711789906, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.038e-05, |
| "loss": 3.3655, |
| "mean_token_accuracy": 0.43400888703763485, |
| "num_tokens": 6896684.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 2.454593952000141, |
| "epoch": 0.4718188393701664, |
| "grad_norm": 0.014404296875, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 3.3614, |
| "mean_token_accuracy": 0.43178336657583716, |
| "num_tokens": 7033417.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 2.558536158502102, |
| "epoch": 0.48072108162243365, |
| "grad_norm": 0.018310546875, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 3.4999, |
| "mean_token_accuracy": 0.4232194619253278, |
| "num_tokens": 7166608.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 2.4503342963755133, |
| "epoch": 0.4896233238747009, |
| "grad_norm": 0.0150146484375, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 3.3626, |
| "mean_token_accuracy": 0.4358119173906744, |
| "num_tokens": 7311968.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 2.4068671748042108, |
| "epoch": 0.49852556612696824, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 3.2928, |
| "mean_token_accuracy": 0.44231050591915844, |
| "num_tokens": 7444923.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 2.5063270702958107, |
| "epoch": 0.5074278083792355, |
| "grad_norm": 0.007171630859375, |
| "learning_rate": 1.138e-05, |
| "loss": 3.4094, |
| "mean_token_accuracy": 0.42552004270255567, |
| "num_tokens": 7576456.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 2.5186563402414324, |
| "epoch": 0.5163300506315028, |
| "grad_norm": 0.00994873046875, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 3.4523, |
| "mean_token_accuracy": 0.42800150997936726, |
| "num_tokens": 7706502.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 2.500379876047373, |
| "epoch": 0.52523229288377, |
| "grad_norm": 0.016357421875, |
| "learning_rate": 1.178e-05, |
| "loss": 3.4366, |
| "mean_token_accuracy": 0.42906679548323157, |
| "num_tokens": 7838516.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 2.535025441646576, |
| "epoch": 0.5341345351360374, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.198e-05, |
| "loss": 3.5082, |
| "mean_token_accuracy": 0.42478432394564153, |
| "num_tokens": 7969704.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.5341345351360374, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 7969704.0, |
| "eval_coding_runtime": 95.8333, |
| "eval_coding_samples_per_second": 5.217, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.5341345351360374, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 7969704.0, |
| "eval_chemistry_runtime": 53.6736, |
| "eval_chemistry_samples_per_second": 9.316, |
| "eval_chemistry_steps_per_second": 4.658, |
| "step": 600 |
| }, |
| { |
| "entropy": 2.542660507559776, |
| "epoch": 0.5430367773883047, |
| "grad_norm": 0.009765625, |
| "learning_rate": 1.218e-05, |
| "loss": 3.4959, |
| "mean_token_accuracy": 0.42042196970433, |
| "num_tokens": 8103869.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 2.5036643177270888, |
| "epoch": 0.551939019640572, |
| "grad_norm": 0.007781982421875, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 3.4222, |
| "mean_token_accuracy": 0.43322109896689653, |
| "num_tokens": 8242162.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 2.5407710492610933, |
| "epoch": 0.5608412618928392, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 3.5112, |
| "mean_token_accuracy": 0.424757894128561, |
| "num_tokens": 8365617.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 2.4822314344346523, |
| "epoch": 0.5697435041451065, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 3.4362, |
| "mean_token_accuracy": 0.43353409245610236, |
| "num_tokens": 8497852.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 2.5407922022044658, |
| "epoch": 0.5786457463973739, |
| "grad_norm": 0.0172119140625, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 3.5164, |
| "mean_token_accuracy": 0.42170737963169813, |
| "num_tokens": 8622868.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 2.577139265835285, |
| "epoch": 0.5875479886496411, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 3.5058, |
| "mean_token_accuracy": 0.41903221048414707, |
| "num_tokens": 8757753.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 2.475165160000324, |
| "epoch": 0.5964502309019084, |
| "grad_norm": 0.01422119140625, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 3.4273, |
| "mean_token_accuracy": 0.4251199818216264, |
| "num_tokens": 8886401.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 2.4363515108823774, |
| "epoch": 0.6053524731541757, |
| "grad_norm": 0.01300048828125, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 3.3275, |
| "mean_token_accuracy": 0.43763177264481784, |
| "num_tokens": 9024677.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 2.56791525632143, |
| "epoch": 0.614254715406443, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 1.378e-05, |
| "loss": 3.4703, |
| "mean_token_accuracy": 0.42369468677788974, |
| "num_tokens": 9153771.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 2.484988895058632, |
| "epoch": 0.6231569576587103, |
| "grad_norm": 0.0126953125, |
| "learning_rate": 1.398e-05, |
| "loss": 3.3769, |
| "mean_token_accuracy": 0.4326890670694411, |
| "num_tokens": 9291760.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.6231569576587103, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 9291760.0, |
| "eval_coding_runtime": 95.8189, |
| "eval_coding_samples_per_second": 5.218, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.6231569576587103, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 9291760.0, |
| "eval_chemistry_runtime": 53.6652, |
| "eval_chemistry_samples_per_second": 9.317, |
| "eval_chemistry_steps_per_second": 4.659, |
| "step": 700 |
| }, |
| { |
| "entropy": 2.4618755050003527, |
| "epoch": 0.6320591999109776, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 1.418e-05, |
| "loss": 3.4035, |
| "mean_token_accuracy": 0.4318504808470607, |
| "num_tokens": 9425022.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 2.4683789536356926, |
| "epoch": 0.6409614421632449, |
| "grad_norm": 0.0076904296875, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 3.3571, |
| "mean_token_accuracy": 0.43439894206821916, |
| "num_tokens": 9561091.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 2.4340669311583043, |
| "epoch": 0.6498636844155121, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 3.3348, |
| "mean_token_accuracy": 0.4359587837010622, |
| "num_tokens": 9695723.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 2.504717104136944, |
| "epoch": 0.6587659266677794, |
| "grad_norm": 0.0174560546875, |
| "learning_rate": 1.478e-05, |
| "loss": 3.4563, |
| "mean_token_accuracy": 0.4270874824374914, |
| "num_tokens": 9827272.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 2.4963896624743938, |
| "epoch": 0.6676681689200468, |
| "grad_norm": 0.010498046875, |
| "learning_rate": 1.498e-05, |
| "loss": 3.3969, |
| "mean_token_accuracy": 0.4328357223421335, |
| "num_tokens": 9955369.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 2.47789601534605, |
| "epoch": 0.676570411172314, |
| "grad_norm": 0.015869140625, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 3.3672, |
| "mean_token_accuracy": 0.43644896559417246, |
| "num_tokens": 10096065.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 2.4533426530659197, |
| "epoch": 0.6854726534245813, |
| "grad_norm": 0.00982666015625, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 3.3724, |
| "mean_token_accuracy": 0.43634699024260043, |
| "num_tokens": 10233806.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 2.5112246222794057, |
| "epoch": 0.6943748956768486, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 3.4341, |
| "mean_token_accuracy": 0.4313540508970618, |
| "num_tokens": 10364601.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 2.467985014617443, |
| "epoch": 0.703277137929116, |
| "grad_norm": 0.01171875, |
| "learning_rate": 1.578e-05, |
| "loss": 3.379, |
| "mean_token_accuracy": 0.4354724214412272, |
| "num_tokens": 10500891.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 2.4910646095871924, |
| "epoch": 0.7121793801813832, |
| "grad_norm": 0.0089111328125, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 3.4174, |
| "mean_token_accuracy": 0.4264952681958675, |
| "num_tokens": 10633325.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.7121793801813832, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 10633325.0, |
| "eval_coding_runtime": 95.7207, |
| "eval_coding_samples_per_second": 5.224, |
| "eval_coding_steps_per_second": 2.612, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.7121793801813832, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 10633325.0, |
| "eval_chemistry_runtime": 53.6314, |
| "eval_chemistry_samples_per_second": 9.323, |
| "eval_chemistry_steps_per_second": 4.661, |
| "step": 800 |
| }, |
| { |
| "entropy": 2.421719251573086, |
| "epoch": 0.7210816224336505, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 1.618e-05, |
| "loss": 3.373, |
| "mean_token_accuracy": 0.4373985629528761, |
| "num_tokens": 10764319.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 2.43545116186142, |
| "epoch": 0.7299838646859178, |
| "grad_norm": 0.01287841796875, |
| "learning_rate": 1.638e-05, |
| "loss": 3.3444, |
| "mean_token_accuracy": 0.43578007342293856, |
| "num_tokens": 10897916.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 2.5303331069648265, |
| "epoch": 0.738886106938185, |
| "grad_norm": 0.0125732421875, |
| "learning_rate": 1.658e-05, |
| "loss": 3.4252, |
| "mean_token_accuracy": 0.4248075334355235, |
| "num_tokens": 11030485.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 2.4731515564024447, |
| "epoch": 0.7477883491904523, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 3.4258, |
| "mean_token_accuracy": 0.43433472914621235, |
| "num_tokens": 11165356.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 2.4785347141325476, |
| "epoch": 0.7566905914427197, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.698e-05, |
| "loss": 3.4161, |
| "mean_token_accuracy": 0.4306844290345907, |
| "num_tokens": 11299223.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 2.4768461272120477, |
| "epoch": 0.7655928336949869, |
| "grad_norm": 0.01287841796875, |
| "learning_rate": 1.718e-05, |
| "loss": 3.3776, |
| "mean_token_accuracy": 0.4344117846339941, |
| "num_tokens": 11436799.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 2.5141974300146104, |
| "epoch": 0.7744950759472542, |
| "grad_norm": 0.00830078125, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 3.3958, |
| "mean_token_accuracy": 0.4330358326435089, |
| "num_tokens": 11567422.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 2.515311509370804, |
| "epoch": 0.7833973181995215, |
| "grad_norm": 0.009765625, |
| "learning_rate": 1.758e-05, |
| "loss": 3.4031, |
| "mean_token_accuracy": 0.4312752116471529, |
| "num_tokens": 11703496.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 2.453214881569147, |
| "epoch": 0.7922995604517888, |
| "grad_norm": 0.0093994140625, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 3.3834, |
| "mean_token_accuracy": 0.4389944301918149, |
| "num_tokens": 11840317.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 2.6043091461062433, |
| "epoch": 0.801201802704056, |
| "grad_norm": 0.01116943359375, |
| "learning_rate": 1.798e-05, |
| "loss": 3.5713, |
| "mean_token_accuracy": 0.4135642783716321, |
| "num_tokens": 11965530.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.801201802704056, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 11965530.0, |
| "eval_coding_runtime": 95.8462, |
| "eval_coding_samples_per_second": 5.217, |
| "eval_coding_steps_per_second": 2.608, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.801201802704056, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 11965530.0, |
| "eval_chemistry_runtime": 53.6177, |
| "eval_chemistry_samples_per_second": 9.325, |
| "eval_chemistry_steps_per_second": 4.663, |
| "step": 900 |
| }, |
| { |
| "entropy": 2.4632391795516013, |
| "epoch": 0.8101040449563234, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 3.4352, |
| "mean_token_accuracy": 0.42954989019781353, |
| "num_tokens": 12097917.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 2.468993365764618, |
| "epoch": 0.8190062872085907, |
| "grad_norm": 0.0089111328125, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 3.4071, |
| "mean_token_accuracy": 0.4282375952228904, |
| "num_tokens": 12224427.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 2.468296863883734, |
| "epoch": 0.8279085294608579, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.858e-05, |
| "loss": 3.3355, |
| "mean_token_accuracy": 0.43817838532850145, |
| "num_tokens": 12365332.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 2.427542605996132, |
| "epoch": 0.8368107717131252, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 1.878e-05, |
| "loss": 3.317, |
| "mean_token_accuracy": 0.44397469637915493, |
| "num_tokens": 12509124.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 2.4841547794640064, |
| "epoch": 0.8457130139653926, |
| "grad_norm": 0.018798828125, |
| "learning_rate": 1.898e-05, |
| "loss": 3.3654, |
| "mean_token_accuracy": 0.434748101234436, |
| "num_tokens": 12642651.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 2.449143522232771, |
| "epoch": 0.8546152562176598, |
| "grad_norm": 0.00738525390625, |
| "learning_rate": 1.918e-05, |
| "loss": 3.4035, |
| "mean_token_accuracy": 0.43319537229835986, |
| "num_tokens": 12778408.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 2.387574986368418, |
| "epoch": 0.8635174984699271, |
| "grad_norm": 0.0072021484375, |
| "learning_rate": 1.938e-05, |
| "loss": 3.2822, |
| "mean_token_accuracy": 0.4413015801459551, |
| "num_tokens": 12910630.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 2.5341439098119736, |
| "epoch": 0.8724197407221944, |
| "grad_norm": 0.01055908203125, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 3.4749, |
| "mean_token_accuracy": 0.4271688721142709, |
| "num_tokens": 13046473.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 2.5607902660965918, |
| "epoch": 0.8813219829744617, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 1.978e-05, |
| "loss": 3.4981, |
| "mean_token_accuracy": 0.4231617606244981, |
| "num_tokens": 13167768.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 2.4919896230101584, |
| "epoch": 0.890224225226729, |
| "grad_norm": 0.009765625, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 3.3681, |
| "mean_token_accuracy": 0.43273249492049215, |
| "num_tokens": 13301659.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.890224225226729, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 13301659.0, |
| "eval_coding_runtime": 95.788, |
| "eval_coding_samples_per_second": 5.22, |
| "eval_coding_steps_per_second": 2.61, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.890224225226729, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 13301659.0, |
| "eval_chemistry_runtime": 53.3313, |
| "eval_chemistry_samples_per_second": 9.375, |
| "eval_chemistry_steps_per_second": 4.688, |
| "step": 1000 |
| }, |
| { |
| "entropy": 2.482378314435482, |
| "epoch": 0.8991264674789963, |
| "grad_norm": 0.0098876953125, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 3.4139, |
| "mean_token_accuracy": 0.4351740827783942, |
| "num_tokens": 13434364.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 2.397843288630247, |
| "epoch": 0.9080287097312636, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 3.2598, |
| "mean_token_accuracy": 0.4428169864229858, |
| "num_tokens": 13570986.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 2.426850029826164, |
| "epoch": 0.9169309519835308, |
| "grad_norm": 0.0191650390625, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 3.3704, |
| "mean_token_accuracy": 0.43931040428578855, |
| "num_tokens": 13706343.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 2.5369828954339027, |
| "epoch": 0.9258331942357981, |
| "grad_norm": 0.013671875, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 3.4473, |
| "mean_token_accuracy": 0.42489275773987173, |
| "num_tokens": 13833901.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 2.495298910140991, |
| "epoch": 0.9347354364880655, |
| "grad_norm": 0.0186767578125, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 3.4641, |
| "mean_token_accuracy": 0.4274964226409793, |
| "num_tokens": 13954258.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 2.4657325744628906, |
| "epoch": 0.9436376787403328, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 3.3954, |
| "mean_token_accuracy": 0.4306732133962214, |
| "num_tokens": 14093888.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 2.533431290090084, |
| "epoch": 0.9525399209926, |
| "grad_norm": 0.0133056640625, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 3.4682, |
| "mean_token_accuracy": 0.4237683041021228, |
| "num_tokens": 14222626.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 2.535231586545706, |
| "epoch": 0.9614421632448673, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 3.4095, |
| "mean_token_accuracy": 0.4307479955255985, |
| "num_tokens": 14354918.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 2.507918354868889, |
| "epoch": 0.9703444054971346, |
| "grad_norm": 0.0076904296875, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 3.4265, |
| "mean_token_accuracy": 0.4296759738586843, |
| "num_tokens": 14493349.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 2.4675292283296586, |
| "epoch": 0.9792466477494018, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 1.978e-05, |
| "loss": 3.4234, |
| "mean_token_accuracy": 0.42840409390628337, |
| "num_tokens": 14620244.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.9792466477494018, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 14620244.0, |
| "eval_coding_runtime": 95.7872, |
| "eval_coding_samples_per_second": 5.22, |
| "eval_coding_steps_per_second": 2.61, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.9792466477494018, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 14620244.0, |
| "eval_chemistry_runtime": 53.6439, |
| "eval_chemistry_samples_per_second": 9.321, |
| "eval_chemistry_steps_per_second": 4.66, |
| "step": 1100 |
| }, |
| { |
| "entropy": 2.4005719617009165, |
| "epoch": 0.9881488900016692, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 3.3593, |
| "mean_token_accuracy": 0.4352086253464222, |
| "num_tokens": 14760951.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 2.558166432380676, |
| "epoch": 0.9970511322539365, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 3.5454, |
| "mean_token_accuracy": 0.42158968318253753, |
| "num_tokens": 14886593.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 2.456293193285897, |
| "epoch": 1.0053413453513604, |
| "grad_norm": 0.008544921875, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 3.329, |
| "mean_token_accuracy": 0.431678411744585, |
| "num_tokens": 15009964.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 2.3623578011989594, |
| "epoch": 1.0142435876036278, |
| "grad_norm": 0.01019287109375, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 3.2394, |
| "mean_token_accuracy": 0.4485295919701457, |
| "num_tokens": 15146469.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 2.537761741876602, |
| "epoch": 1.0231458298558949, |
| "grad_norm": 0.017578125, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 3.4544, |
| "mean_token_accuracy": 0.42700372273102405, |
| "num_tokens": 15274532.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 2.560138799250126, |
| "epoch": 1.0320480721081622, |
| "grad_norm": 0.01312255859375, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 3.5139, |
| "mean_token_accuracy": 0.42576750814914704, |
| "num_tokens": 15404569.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 2.4245593287050724, |
| "epoch": 1.0409503143604295, |
| "grad_norm": 0.00665283203125, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 3.3292, |
| "mean_token_accuracy": 0.4398623496294022, |
| "num_tokens": 15550449.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 2.447620178759098, |
| "epoch": 1.0498525566126968, |
| "grad_norm": 0.00872802734375, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 3.4233, |
| "mean_token_accuracy": 0.4298626586794853, |
| "num_tokens": 15681709.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 2.481611392647028, |
| "epoch": 1.0587547988649642, |
| "grad_norm": 0.01165771484375, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 3.3803, |
| "mean_token_accuracy": 0.43227153811603786, |
| "num_tokens": 15816502.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 2.441844770312309, |
| "epoch": 1.0676570411172315, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 3.3808, |
| "mean_token_accuracy": 0.4364687578752637, |
| "num_tokens": 15949421.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.0676570411172315, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 15949421.0, |
| "eval_coding_runtime": 95.6237, |
| "eval_coding_samples_per_second": 5.229, |
| "eval_coding_steps_per_second": 2.614, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.0676570411172315, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 15949421.0, |
| "eval_chemistry_runtime": 53.6265, |
| "eval_chemistry_samples_per_second": 9.324, |
| "eval_chemistry_steps_per_second": 4.662, |
| "step": 1200 |
| }, |
| { |
| "entropy": 2.4238644503057003, |
| "epoch": 1.0765592833694986, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 3.3419, |
| "mean_token_accuracy": 0.43726830892264845, |
| "num_tokens": 16088471.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 2.4736556738615034, |
| "epoch": 1.085461525621766, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 3.403, |
| "mean_token_accuracy": 0.43131555467844007, |
| "num_tokens": 16221916.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 2.5666570000350477, |
| "epoch": 1.0943637678740332, |
| "grad_norm": 0.0107421875, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 3.4806, |
| "mean_token_accuracy": 0.4240891897119582, |
| "num_tokens": 16353558.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 2.5181381188333036, |
| "epoch": 1.1032660101263005, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 3.4298, |
| "mean_token_accuracy": 0.4259687649086118, |
| "num_tokens": 16488154.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 2.514085105806589, |
| "epoch": 1.1121682523785679, |
| "grad_norm": 0.010498046875, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 3.4249, |
| "mean_token_accuracy": 0.42915988666936755, |
| "num_tokens": 16618151.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 2.4717010237276553, |
| "epoch": 1.1210704946308352, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 3.3908, |
| "mean_token_accuracy": 0.43228524839505555, |
| "num_tokens": 16756828.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 2.517887058854103, |
| "epoch": 1.1299727368831025, |
| "grad_norm": 0.01141357421875, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 3.4328, |
| "mean_token_accuracy": 0.42816000785678626, |
| "num_tokens": 16890026.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 2.4103298880159856, |
| "epoch": 1.1388749791353696, |
| "grad_norm": 0.01025390625, |
| "learning_rate": 1.938e-05, |
| "loss": 3.3448, |
| "mean_token_accuracy": 0.4348219852894545, |
| "num_tokens": 17023681.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 2.509786891192198, |
| "epoch": 1.147777221387637, |
| "grad_norm": 0.0179443359375, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 3.4512, |
| "mean_token_accuracy": 0.4285705531015992, |
| "num_tokens": 17155619.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 2.503946328163147, |
| "epoch": 1.1566794636399043, |
| "grad_norm": 0.01123046875, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 3.4103, |
| "mean_token_accuracy": 0.4285626333206892, |
| "num_tokens": 17286672.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.1566794636399043, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 17286672.0, |
| "eval_coding_runtime": 95.8416, |
| "eval_coding_samples_per_second": 5.217, |
| "eval_coding_steps_per_second": 2.608, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.1566794636399043, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 17286672.0, |
| "eval_chemistry_runtime": 53.6206, |
| "eval_chemistry_samples_per_second": 9.325, |
| "eval_chemistry_steps_per_second": 4.662, |
| "step": 1300 |
| }, |
| { |
| "entropy": 2.511377315223217, |
| "epoch": 1.1655817058921716, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 3.493, |
| "mean_token_accuracy": 0.4217289647087455, |
| "num_tokens": 17418216.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 2.5961489491164684, |
| "epoch": 1.174483948144439, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 3.4532, |
| "mean_token_accuracy": 0.4226861286908388, |
| "num_tokens": 17551581.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 2.4733773186802863, |
| "epoch": 1.1833861903967062, |
| "grad_norm": 0.0078125, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 3.406, |
| "mean_token_accuracy": 0.43328642603009937, |
| "num_tokens": 17685328.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 2.504808610677719, |
| "epoch": 1.1922884326489736, |
| "grad_norm": 0.010986328125, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 3.3902, |
| "mean_token_accuracy": 0.43533999361097814, |
| "num_tokens": 17821205.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 2.554856265336275, |
| "epoch": 1.2011906749012407, |
| "grad_norm": 0.008544921875, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 3.5145, |
| "mean_token_accuracy": 0.4222357952967286, |
| "num_tokens": 17950081.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 2.4698183365166186, |
| "epoch": 1.210092917153508, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 3.3678, |
| "mean_token_accuracy": 0.4337294826284051, |
| "num_tokens": 18087759.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 2.491861944645643, |
| "epoch": 1.2189951594057753, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 1.918e-05, |
| "loss": 3.3956, |
| "mean_token_accuracy": 0.43155251136049627, |
| "num_tokens": 18229440.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 2.544880051165819, |
| "epoch": 1.2278974016580426, |
| "grad_norm": 0.01019287109375, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 3.4837, |
| "mean_token_accuracy": 0.4227764903567731, |
| "num_tokens": 18355732.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 2.608750829845667, |
| "epoch": 1.23679964391031, |
| "grad_norm": 0.0096435546875, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 3.5353, |
| "mean_token_accuracy": 0.42066937144845723, |
| "num_tokens": 18489654.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 2.4528128817677497, |
| "epoch": 1.2457018861625773, |
| "grad_norm": 0.0067138671875, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 3.3402, |
| "mean_token_accuracy": 0.43648717207834126, |
| "num_tokens": 18626245.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.2457018861625773, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 18626245.0, |
| "eval_coding_runtime": 95.8265, |
| "eval_coding_samples_per_second": 5.218, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.2457018861625773, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 18626245.0, |
| "eval_chemistry_runtime": 53.6283, |
| "eval_chemistry_samples_per_second": 9.323, |
| "eval_chemistry_steps_per_second": 4.662, |
| "step": 1400 |
| }, |
| { |
| "entropy": 2.4507749184966086, |
| "epoch": 1.2546041284148446, |
| "grad_norm": 0.00872802734375, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 3.3585, |
| "mean_token_accuracy": 0.43596437461674215, |
| "num_tokens": 18767785.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 2.432799159735441, |
| "epoch": 1.2635063706671117, |
| "grad_norm": 0.012451171875, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 3.309, |
| "mean_token_accuracy": 0.4405426589772105, |
| "num_tokens": 18899632.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 2.523366042226553, |
| "epoch": 1.272408612919379, |
| "grad_norm": 0.01068115234375, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 3.4447, |
| "mean_token_accuracy": 0.42682514479383826, |
| "num_tokens": 19039727.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 2.522237445414066, |
| "epoch": 1.2813108551716463, |
| "grad_norm": 0.01190185546875, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 3.4611, |
| "mean_token_accuracy": 0.4245713768526912, |
| "num_tokens": 19173635.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 2.4513756319880486, |
| "epoch": 1.2902130974239137, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 3.3933, |
| "mean_token_accuracy": 0.4320486688986421, |
| "num_tokens": 19317295.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 2.417076099663973, |
| "epoch": 1.299115339676181, |
| "grad_norm": 0.01165771484375, |
| "learning_rate": 1.898e-05, |
| "loss": 3.3298, |
| "mean_token_accuracy": 0.4392012171447277, |
| "num_tokens": 19453495.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 2.4171031497418882, |
| "epoch": 1.3080175819284483, |
| "grad_norm": 0.007720947265625, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 3.3321, |
| "mean_token_accuracy": 0.4389203391969204, |
| "num_tokens": 19588860.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 2.475263304263353, |
| "epoch": 1.3169198241807156, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 3.423, |
| "mean_token_accuracy": 0.42983661517500876, |
| "num_tokens": 19719086.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 2.4915241256356238, |
| "epoch": 1.3258220664329827, |
| "grad_norm": 0.01251220703125, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 3.3932, |
| "mean_token_accuracy": 0.4291253449395299, |
| "num_tokens": 19851321.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 2.515060818940401, |
| "epoch": 1.33472430868525, |
| "grad_norm": 0.01190185546875, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 3.4452, |
| "mean_token_accuracy": 0.4248286823742092, |
| "num_tokens": 19981272.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.33472430868525, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 19981272.0, |
| "eval_coding_runtime": 95.9024, |
| "eval_coding_samples_per_second": 5.214, |
| "eval_coding_steps_per_second": 2.607, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.33472430868525, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 19981272.0, |
| "eval_chemistry_runtime": 53.3916, |
| "eval_chemistry_samples_per_second": 9.365, |
| "eval_chemistry_steps_per_second": 4.682, |
| "step": 1500 |
| }, |
| { |
| "entropy": 2.537444035708904, |
| "epoch": 1.3436265509375174, |
| "grad_norm": 0.01397705078125, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 3.4621, |
| "mean_token_accuracy": 0.42645752560347316, |
| "num_tokens": 20112097.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 2.4578563667833806, |
| "epoch": 1.3525287931897847, |
| "grad_norm": 0.00701904296875, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 3.4011, |
| "mean_token_accuracy": 0.43212775196880104, |
| "num_tokens": 20243034.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 2.5289953693747522, |
| "epoch": 1.361431035442052, |
| "grad_norm": 0.015380859375, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 3.4629, |
| "mean_token_accuracy": 0.4234733306802809, |
| "num_tokens": 20372999.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 2.4936239890754224, |
| "epoch": 1.3703332776943191, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 3.4047, |
| "mean_token_accuracy": 0.43090203367173674, |
| "num_tokens": 20509028.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 2.6229711353778837, |
| "epoch": 1.3792355199465867, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.878e-05, |
| "loss": 3.6425, |
| "mean_token_accuracy": 0.41157348044216635, |
| "num_tokens": 20633479.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 2.507130479812622, |
| "epoch": 1.3881377621988538, |
| "grad_norm": 0.01025390625, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 3.4246, |
| "mean_token_accuracy": 0.42894090730696915, |
| "num_tokens": 20762668.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 2.56754671856761, |
| "epoch": 1.397040004451121, |
| "grad_norm": 0.00732421875, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 3.4484, |
| "mean_token_accuracy": 0.4255117969587445, |
| "num_tokens": 20894609.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 2.6108976893126963, |
| "epoch": 1.4059422467033884, |
| "grad_norm": 0.01123046875, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 3.5745, |
| "mean_token_accuracy": 0.4152322731912136, |
| "num_tokens": 21025277.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 2.575570172071457, |
| "epoch": 1.4148444889556557, |
| "grad_norm": 0.016357421875, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 3.5483, |
| "mean_token_accuracy": 0.4183428635820746, |
| "num_tokens": 21156848.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 2.4909443251788614, |
| "epoch": 1.423746731207923, |
| "grad_norm": 0.01080322265625, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 3.4169, |
| "mean_token_accuracy": 0.42875550473108887, |
| "num_tokens": 21292320.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.423746731207923, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 21292320.0, |
| "eval_coding_runtime": 95.8321, |
| "eval_coding_samples_per_second": 5.217, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.423746731207923, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 21292320.0, |
| "eval_chemistry_runtime": 53.6996, |
| "eval_chemistry_samples_per_second": 9.311, |
| "eval_chemistry_steps_per_second": 4.656, |
| "step": 1600 |
| }, |
| { |
| "entropy": 2.5857157841324807, |
| "epoch": 1.4326489734601902, |
| "grad_norm": 0.0120849609375, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 3.5036, |
| "mean_token_accuracy": 0.4268370010890067, |
| "num_tokens": 21426944.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 2.471120712906122, |
| "epoch": 1.4415512157124577, |
| "grad_norm": 0.01239013671875, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 3.4141, |
| "mean_token_accuracy": 0.43633277975022794, |
| "num_tokens": 21563668.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 2.572203493118286, |
| "epoch": 1.4504534579647248, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 3.5546, |
| "mean_token_accuracy": 0.416438902169466, |
| "num_tokens": 21687993.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 2.4553592331707477, |
| "epoch": 1.4593557002169921, |
| "grad_norm": 0.00927734375, |
| "learning_rate": 1.858e-05, |
| "loss": 3.42, |
| "mean_token_accuracy": 0.43670414835214616, |
| "num_tokens": 21817298.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 2.5596740052103995, |
| "epoch": 1.4682579424692594, |
| "grad_norm": 0.00897216796875, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 3.479, |
| "mean_token_accuracy": 0.42424469059333203, |
| "num_tokens": 21947593.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 2.558923949301243, |
| "epoch": 1.4771601847215268, |
| "grad_norm": 0.01104736328125, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 3.4637, |
| "mean_token_accuracy": 0.42529286816716194, |
| "num_tokens": 22083010.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 2.5305933080613614, |
| "epoch": 1.486062426973794, |
| "grad_norm": 0.01385498046875, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 3.4958, |
| "mean_token_accuracy": 0.4228696158155799, |
| "num_tokens": 22216759.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 2.5268216267228127, |
| "epoch": 1.4949646692260612, |
| "grad_norm": 0.010498046875, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 3.4217, |
| "mean_token_accuracy": 0.4293845698237419, |
| "num_tokens": 22350046.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 2.5090344458818437, |
| "epoch": 1.5038669114783287, |
| "grad_norm": 0.01373291015625, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 3.4432, |
| "mean_token_accuracy": 0.4325392559170723, |
| "num_tokens": 22485677.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 2.4932730071246625, |
| "epoch": 1.5127691537305958, |
| "grad_norm": 0.00860595703125, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 3.3879, |
| "mean_token_accuracy": 0.4324304299429059, |
| "num_tokens": 22615933.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.5127691537305958, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 22615933.0, |
| "eval_coding_runtime": 96.0736, |
| "eval_coding_samples_per_second": 5.204, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.5127691537305958, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 22615933.0, |
| "eval_chemistry_runtime": 53.6801, |
| "eval_chemistry_samples_per_second": 9.314, |
| "eval_chemistry_steps_per_second": 4.657, |
| "step": 1700 |
| }, |
| { |
| "entropy": 2.499319192022085, |
| "epoch": 1.5216713959828632, |
| "grad_norm": 0.01531982421875, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 3.4393, |
| "mean_token_accuracy": 0.42634991379454734, |
| "num_tokens": 22749756.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 2.3928098790347576, |
| "epoch": 1.5305736382351305, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 3.2777, |
| "mean_token_accuracy": 0.44273359794169664, |
| "num_tokens": 22886843.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 2.515210550278425, |
| "epoch": 1.5394758804873978, |
| "grad_norm": 0.0145263671875, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 3.4278, |
| "mean_token_accuracy": 0.430076424498111, |
| "num_tokens": 23013890.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 2.511288223415613, |
| "epoch": 1.5483781227396651, |
| "grad_norm": 0.0174560546875, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 3.4467, |
| "mean_token_accuracy": 0.42953401962295173, |
| "num_tokens": 23140244.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 2.5117117315530777, |
| "epoch": 1.5572803649919322, |
| "grad_norm": 0.01177978515625, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 3.437, |
| "mean_token_accuracy": 0.43029365949332715, |
| "num_tokens": 23268188.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 2.472087085992098, |
| "epoch": 1.5661826072441998, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 3.3994, |
| "mean_token_accuracy": 0.4351077421568334, |
| "num_tokens": 23405121.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 2.5334790728986265, |
| "epoch": 1.5750848494964669, |
| "grad_norm": 0.00897216796875, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 3.456, |
| "mean_token_accuracy": 0.42388113122433424, |
| "num_tokens": 23535787.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 2.5175728268921373, |
| "epoch": 1.5839870917487342, |
| "grad_norm": 0.007049560546875, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 3.4153, |
| "mean_token_accuracy": 0.426735052280128, |
| "num_tokens": 23671826.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 2.522493792325258, |
| "epoch": 1.5928893340010015, |
| "grad_norm": 0.007781982421875, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 3.4456, |
| "mean_token_accuracy": 0.42770035564899445, |
| "num_tokens": 23805612.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 2.5342708349227907, |
| "epoch": 1.6017915762532688, |
| "grad_norm": 0.0174560546875, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 3.4542, |
| "mean_token_accuracy": 0.4265250092372298, |
| "num_tokens": 23933374.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.6017915762532688, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 23933374.0, |
| "eval_coding_runtime": 95.8632, |
| "eval_coding_samples_per_second": 5.216, |
| "eval_coding_steps_per_second": 2.608, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.6017915762532688, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 23933374.0, |
| "eval_chemistry_runtime": 53.6651, |
| "eval_chemistry_samples_per_second": 9.317, |
| "eval_chemistry_steps_per_second": 4.659, |
| "step": 1800 |
| }, |
| { |
| "entropy": 2.4730284474790096, |
| "epoch": 1.6106938185055362, |
| "grad_norm": 0.01123046875, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 3.4249, |
| "mean_token_accuracy": 0.4301054811105132, |
| "num_tokens": 24070188.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 2.4333154663443564, |
| "epoch": 1.6195960607578033, |
| "grad_norm": 0.0162353515625, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 3.3849, |
| "mean_token_accuracy": 0.4358032012358308, |
| "num_tokens": 24210156.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 2.548785130679607, |
| "epoch": 1.6284983030100708, |
| "grad_norm": 0.00994873046875, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 3.4663, |
| "mean_token_accuracy": 0.4266028679907322, |
| "num_tokens": 24335746.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 2.493963433802128, |
| "epoch": 1.637400545262338, |
| "grad_norm": 0.0115966796875, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 3.4112, |
| "mean_token_accuracy": 0.43058306723833084, |
| "num_tokens": 24462726.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 2.4432963475584986, |
| "epoch": 1.6463027875146052, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 3.3659, |
| "mean_token_accuracy": 0.4354290099814534, |
| "num_tokens": 24597391.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 2.4352858766913412, |
| "epoch": 1.6552050297668726, |
| "grad_norm": 0.0150146484375, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 3.3265, |
| "mean_token_accuracy": 0.44209435898810623, |
| "num_tokens": 24732815.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 2.494772292673588, |
| "epoch": 1.6641072720191397, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 3.4225, |
| "mean_token_accuracy": 0.42676153369247916, |
| "num_tokens": 24869428.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 2.4837465301156043, |
| "epoch": 1.6730095142714072, |
| "grad_norm": 0.0181884765625, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 3.3856, |
| "mean_token_accuracy": 0.43438148498535156, |
| "num_tokens": 25002091.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 2.5113090477883815, |
| "epoch": 1.6819117565236743, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 3.4708, |
| "mean_token_accuracy": 0.426338127348572, |
| "num_tokens": 25133714.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 2.4073795929551123, |
| "epoch": 1.6908139987759419, |
| "grad_norm": 0.0106201171875, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 3.3643, |
| "mean_token_accuracy": 0.4404640804976225, |
| "num_tokens": 25267949.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.6908139987759419, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 25267949.0, |
| "eval_coding_runtime": 95.8242, |
| "eval_coding_samples_per_second": 5.218, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.6908139987759419, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 25267949.0, |
| "eval_chemistry_runtime": 53.6517, |
| "eval_chemistry_samples_per_second": 9.319, |
| "eval_chemistry_steps_per_second": 4.66, |
| "step": 1900 |
| }, |
| { |
| "entropy": 2.4964852534234523, |
| "epoch": 1.699716241028209, |
| "grad_norm": 0.01220703125, |
| "learning_rate": 1.798e-05, |
| "loss": 3.3746, |
| "mean_token_accuracy": 0.43742703888565304, |
| "num_tokens": 25401385.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 2.5503848932683466, |
| "epoch": 1.7086184832804763, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 3.4842, |
| "mean_token_accuracy": 0.42500705253332854, |
| "num_tokens": 25525772.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 2.5723555073142053, |
| "epoch": 1.7175207255327436, |
| "grad_norm": 0.01055908203125, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 3.4885, |
| "mean_token_accuracy": 0.4221608478575945, |
| "num_tokens": 25657946.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 2.5482451558113097, |
| "epoch": 1.7264229677850107, |
| "grad_norm": 0.01416015625, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 3.4748, |
| "mean_token_accuracy": 0.42012830562889575, |
| "num_tokens": 25787727.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 2.4831707894802095, |
| "epoch": 1.7353252100372782, |
| "grad_norm": 0.008544921875, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 3.402, |
| "mean_token_accuracy": 0.4355609222315252, |
| "num_tokens": 25925569.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 2.4309555172920225, |
| "epoch": 1.7442274522895453, |
| "grad_norm": 0.0072021484375, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 3.3031, |
| "mean_token_accuracy": 0.4397842838428915, |
| "num_tokens": 26058026.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 2.4790328785777094, |
| "epoch": 1.7531296945418127, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 3.3659, |
| "mean_token_accuracy": 0.4327216723002493, |
| "num_tokens": 26195268.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 2.3836306020617486, |
| "epoch": 1.76203193679408, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 3.3147, |
| "mean_token_accuracy": 0.4457951638847589, |
| "num_tokens": 26326631.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 2.5131599582731723, |
| "epoch": 1.7709341790463473, |
| "grad_norm": 0.0086669921875, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 3.4339, |
| "mean_token_accuracy": 0.42837768141180277, |
| "num_tokens": 26462379.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 2.4848237335681915, |
| "epoch": 1.7798364212986146, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 3.3823, |
| "mean_token_accuracy": 0.4355517864227295, |
| "num_tokens": 26591735.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.7798364212986146, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 26591735.0, |
| "eval_coding_runtime": 95.8115, |
| "eval_coding_samples_per_second": 5.219, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.7798364212986146, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 26591735.0, |
| "eval_chemistry_runtime": 53.3564, |
| "eval_chemistry_samples_per_second": 9.371, |
| "eval_chemistry_steps_per_second": 4.685, |
| "step": 2000 |
| }, |
| { |
| "entropy": 2.512414899468422, |
| "epoch": 1.7887386635508817, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.7757777777777777e-05, |
| "loss": 3.461, |
| "mean_token_accuracy": 0.42785221468657253, |
| "num_tokens": 26722192.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 2.49338650777936, |
| "epoch": 1.7976409058031493, |
| "grad_norm": 0.0078125, |
| "learning_rate": 1.7735555555555558e-05, |
| "loss": 3.4375, |
| "mean_token_accuracy": 0.43186294184997676, |
| "num_tokens": 26857071.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 2.5387425035238267, |
| "epoch": 1.8065431480554164, |
| "grad_norm": 0.00775146484375, |
| "learning_rate": 1.7713333333333335e-05, |
| "loss": 3.5114, |
| "mean_token_accuracy": 0.42104734824970363, |
| "num_tokens": 26985670.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 2.5376450821757315, |
| "epoch": 1.8154453903076837, |
| "grad_norm": 0.006744384765625, |
| "learning_rate": 1.7691111111111113e-05, |
| "loss": 3.4528, |
| "mean_token_accuracy": 0.42794880773872135, |
| "num_tokens": 27120617.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 2.4720287576317785, |
| "epoch": 1.824347632559951, |
| "grad_norm": 0.008544921875, |
| "learning_rate": 1.766888888888889e-05, |
| "loss": 3.4072, |
| "mean_token_accuracy": 0.43193903416395185, |
| "num_tokens": 27256091.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 2.4498261220753195, |
| "epoch": 1.8332498748122183, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.7646666666666668e-05, |
| "loss": 3.3817, |
| "mean_token_accuracy": 0.4341492637991905, |
| "num_tokens": 27391174.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 2.539250801503658, |
| "epoch": 1.8421521170644857, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 1.7624444444444446e-05, |
| "loss": 3.4641, |
| "mean_token_accuracy": 0.42349757589399817, |
| "num_tokens": 27520594.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 2.4508532620966434, |
| "epoch": 1.8510543593167528, |
| "grad_norm": 0.01080322265625, |
| "learning_rate": 1.7602222222222223e-05, |
| "loss": 3.3258, |
| "mean_token_accuracy": 0.43899478651583196, |
| "num_tokens": 27656941.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 2.5394405357539656, |
| "epoch": 1.8599566015690203, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 1.758e-05, |
| "loss": 3.4321, |
| "mean_token_accuracy": 0.4316885365173221, |
| "num_tokens": 27789788.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 2.4399430051445963, |
| "epoch": 1.8688588438212874, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 1.755777777777778e-05, |
| "loss": 3.3817, |
| "mean_token_accuracy": 0.4349720967933536, |
| "num_tokens": 27927118.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.8688588438212874, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 27927118.0, |
| "eval_coding_runtime": 95.8336, |
| "eval_coding_samples_per_second": 5.217, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.8688588438212874, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 27927118.0, |
| "eval_chemistry_runtime": 53.6675, |
| "eval_chemistry_samples_per_second": 9.317, |
| "eval_chemistry_steps_per_second": 4.658, |
| "step": 2100 |
| }, |
| { |
| "entropy": 2.4159545600414276, |
| "epoch": 1.8777610860735547, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.7535555555555556e-05, |
| "loss": 3.3056, |
| "mean_token_accuracy": 0.44210368683561685, |
| "num_tokens": 28059537.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 2.441173566132784, |
| "epoch": 1.886663328325822, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 1.7513333333333334e-05, |
| "loss": 3.3488, |
| "mean_token_accuracy": 0.4367098089307547, |
| "num_tokens": 28192620.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 2.5123432710766793, |
| "epoch": 1.8955655705780894, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.749111111111111e-05, |
| "loss": 3.408, |
| "mean_token_accuracy": 0.43012905353680253, |
| "num_tokens": 28322688.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 2.453406761586666, |
| "epoch": 1.9044678128303567, |
| "grad_norm": 0.01361083984375, |
| "learning_rate": 1.746888888888889e-05, |
| "loss": 3.3969, |
| "mean_token_accuracy": 0.42973558623343705, |
| "num_tokens": 28458168.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 2.4366604276001453, |
| "epoch": 1.9133700550826238, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 1.7446666666666667e-05, |
| "loss": 3.3541, |
| "mean_token_accuracy": 0.4435499781742692, |
| "num_tokens": 28592965.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 2.559662352502346, |
| "epoch": 1.9222722973348914, |
| "grad_norm": 0.007476806640625, |
| "learning_rate": 1.7424444444444444e-05, |
| "loss": 3.478, |
| "mean_token_accuracy": 0.42056706119328735, |
| "num_tokens": 28722756.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 2.5161592070013286, |
| "epoch": 1.9311745395871585, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.7402222222222222e-05, |
| "loss": 3.4081, |
| "mean_token_accuracy": 0.43007858814671635, |
| "num_tokens": 28851608.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 2.5472119063138963, |
| "epoch": 1.9400767818394258, |
| "grad_norm": 0.012939453125, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 3.4278, |
| "mean_token_accuracy": 0.43115096278488635, |
| "num_tokens": 28978375.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 2.5229906484484674, |
| "epoch": 1.948979024091693, |
| "grad_norm": 0.011962890625, |
| "learning_rate": 1.735777777777778e-05, |
| "loss": 3.4578, |
| "mean_token_accuracy": 0.42716494342312217, |
| "num_tokens": 29102319.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 2.476004705578089, |
| "epoch": 1.9578812663439604, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 1.7335555555555558e-05, |
| "loss": 3.3686, |
| "mean_token_accuracy": 0.4350167758762836, |
| "num_tokens": 29236865.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.9578812663439604, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 29236865.0, |
| "eval_coding_runtime": 95.8605, |
| "eval_coding_samples_per_second": 5.216, |
| "eval_coding_steps_per_second": 2.608, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.9578812663439604, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 29236865.0, |
| "eval_chemistry_runtime": 53.6705, |
| "eval_chemistry_samples_per_second": 9.316, |
| "eval_chemistry_steps_per_second": 4.658, |
| "step": 2200 |
| }, |
| { |
| "entropy": 2.536533968895674, |
| "epoch": 1.9667835085962277, |
| "grad_norm": 0.01263427734375, |
| "learning_rate": 1.7313333333333336e-05, |
| "loss": 3.4119, |
| "mean_token_accuracy": 0.42778037320822476, |
| "num_tokens": 29366663.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 2.449043183401227, |
| "epoch": 1.9756857508484948, |
| "grad_norm": 0.00927734375, |
| "learning_rate": 1.7291111111111113e-05, |
| "loss": 3.3386, |
| "mean_token_accuracy": 0.43799091819673774, |
| "num_tokens": 29500428.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 2.4487401939928533, |
| "epoch": 1.9845879931007624, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 1.726888888888889e-05, |
| "loss": 3.3163, |
| "mean_token_accuracy": 0.4401108752004802, |
| "num_tokens": 29633749.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 2.48353983014822, |
| "epoch": 1.9934902353530295, |
| "grad_norm": 0.00872802734375, |
| "learning_rate": 1.724666666666667e-05, |
| "loss": 3.4529, |
| "mean_token_accuracy": 0.431061902269721, |
| "num_tokens": 29766739.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 2.4969849810504274, |
| "epoch": 2.0017804484504533, |
| "grad_norm": 0.0186767578125, |
| "learning_rate": 1.7224444444444446e-05, |
| "loss": 3.4145, |
| "mean_token_accuracy": 0.43463202230882325, |
| "num_tokens": 29890084.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 2.4849643670022488, |
| "epoch": 2.010682690702721, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 1.7202222222222224e-05, |
| "loss": 3.4236, |
| "mean_token_accuracy": 0.42919157091528176, |
| "num_tokens": 30025427.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 2.5458383657038213, |
| "epoch": 2.019584932954988, |
| "grad_norm": 0.01458740234375, |
| "learning_rate": 1.718e-05, |
| "loss": 3.4659, |
| "mean_token_accuracy": 0.4249839408323169, |
| "num_tokens": 30157741.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 2.43228619992733, |
| "epoch": 2.0284871752072555, |
| "grad_norm": 0.007049560546875, |
| "learning_rate": 1.715777777777778e-05, |
| "loss": 3.3168, |
| "mean_token_accuracy": 0.44042068161070347, |
| "num_tokens": 30286651.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 2.463851699233055, |
| "epoch": 2.0373894174595226, |
| "grad_norm": 0.012939453125, |
| "learning_rate": 1.7135555555555557e-05, |
| "loss": 3.4278, |
| "mean_token_accuracy": 0.4308917077258229, |
| "num_tokens": 30420976.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 2.520537596195936, |
| "epoch": 2.0462916597117897, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.7113333333333334e-05, |
| "loss": 3.4402, |
| "mean_token_accuracy": 0.426590237673372, |
| "num_tokens": 30552624.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 2.0462916597117897, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 30552624.0, |
| "eval_coding_runtime": 96.1075, |
| "eval_coding_samples_per_second": 5.203, |
| "eval_coding_steps_per_second": 2.601, |
| "step": 2300 |
| }, |
| { |
| "epoch": 2.0462916597117897, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 30552624.0, |
| "eval_chemistry_runtime": 53.7785, |
| "eval_chemistry_samples_per_second": 9.297, |
| "eval_chemistry_steps_per_second": 4.649, |
| "step": 2300 |
| }, |
| { |
| "entropy": 2.4775105245411395, |
| "epoch": 2.0551939019640573, |
| "grad_norm": 0.007415771484375, |
| "learning_rate": 1.7091111111111112e-05, |
| "loss": 3.3684, |
| "mean_token_accuracy": 0.4364167733117938, |
| "num_tokens": 30691265.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 2.4963763944804667, |
| "epoch": 2.0640961442163244, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.706888888888889e-05, |
| "loss": 3.4127, |
| "mean_token_accuracy": 0.4294741300866008, |
| "num_tokens": 30823746.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 2.666494831442833, |
| "epoch": 2.072998386468592, |
| "grad_norm": 0.008544921875, |
| "learning_rate": 1.704666666666667e-05, |
| "loss": 3.6227, |
| "mean_token_accuracy": 0.40998966582119467, |
| "num_tokens": 30949206.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 2.529926239699125, |
| "epoch": 2.081900628720859, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.7024444444444445e-05, |
| "loss": 3.4504, |
| "mean_token_accuracy": 0.4259375176392496, |
| "num_tokens": 31076307.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 2.55358622521162, |
| "epoch": 2.0908028709731266, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 1.7002222222222226e-05, |
| "loss": 3.4705, |
| "mean_token_accuracy": 0.42103142719715836, |
| "num_tokens": 31209110.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 2.5771147549152373, |
| "epoch": 2.0997051132253937, |
| "grad_norm": 0.0093994140625, |
| "learning_rate": 1.698e-05, |
| "loss": 3.5117, |
| "mean_token_accuracy": 0.42043230924755337, |
| "num_tokens": 31340763.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 2.4816695883870126, |
| "epoch": 2.1086073554776608, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.695777777777778e-05, |
| "loss": 3.4333, |
| "mean_token_accuracy": 0.4332979379221797, |
| "num_tokens": 31474578.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 2.4634600311517714, |
| "epoch": 2.1175095977299283, |
| "grad_norm": 0.007080078125, |
| "learning_rate": 1.6935555555555555e-05, |
| "loss": 3.3682, |
| "mean_token_accuracy": 0.4341349918395281, |
| "num_tokens": 31607200.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 2.52581527531147, |
| "epoch": 2.1264118399821954, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 1.6913333333333336e-05, |
| "loss": 3.4431, |
| "mean_token_accuracy": 0.4261651573702693, |
| "num_tokens": 31745285.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 2.585964485257864, |
| "epoch": 2.135314082234463, |
| "grad_norm": 0.01434326171875, |
| "learning_rate": 1.689111111111111e-05, |
| "loss": 3.5199, |
| "mean_token_accuracy": 0.41684299279004333, |
| "num_tokens": 31874392.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 2.135314082234463, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 31874392.0, |
| "eval_coding_runtime": 96.0172, |
| "eval_coding_samples_per_second": 5.207, |
| "eval_coding_steps_per_second": 2.604, |
| "step": 2400 |
| }, |
| { |
| "epoch": 2.135314082234463, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 31874392.0, |
| "eval_chemistry_runtime": 53.7772, |
| "eval_chemistry_samples_per_second": 9.298, |
| "eval_chemistry_steps_per_second": 4.649, |
| "step": 2400 |
| }, |
| { |
| "entropy": 2.5983475029468535, |
| "epoch": 2.14421632448673, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 1.686888888888889e-05, |
| "loss": 3.4957, |
| "mean_token_accuracy": 0.42263941895216706, |
| "num_tokens": 32009011.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 2.369788056612015, |
| "epoch": 2.153118566738997, |
| "grad_norm": 0.0098876953125, |
| "learning_rate": 1.684666666666667e-05, |
| "loss": 3.2562, |
| "mean_token_accuracy": 0.4482082508504391, |
| "num_tokens": 32144016.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 2.455921271443367, |
| "epoch": 2.1620208089912647, |
| "grad_norm": 0.016357421875, |
| "learning_rate": 1.6824444444444447e-05, |
| "loss": 3.4044, |
| "mean_token_accuracy": 0.4395538216456771, |
| "num_tokens": 32278381.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 2.5169711895287037, |
| "epoch": 2.170923051243532, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 1.6802222222222224e-05, |
| "loss": 3.4661, |
| "mean_token_accuracy": 0.42682260293513535, |
| "num_tokens": 32410902.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 2.5512830149382353, |
| "epoch": 2.1798252934957993, |
| "grad_norm": 0.00762939453125, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 3.4623, |
| "mean_token_accuracy": 0.4289132535457611, |
| "num_tokens": 32542717.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 2.4463692352175714, |
| "epoch": 2.1887275357480664, |
| "grad_norm": 0.0093994140625, |
| "learning_rate": 1.675777777777778e-05, |
| "loss": 3.3254, |
| "mean_token_accuracy": 0.4344227942638099, |
| "num_tokens": 32674338.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 2.4979622453451156, |
| "epoch": 2.197629778000334, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 1.6735555555555557e-05, |
| "loss": 3.3992, |
| "mean_token_accuracy": 0.42839485611766576, |
| "num_tokens": 32795691.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 2.487682148069143, |
| "epoch": 2.206532020252601, |
| "grad_norm": 0.0115966796875, |
| "learning_rate": 1.6713333333333335e-05, |
| "loss": 3.4191, |
| "mean_token_accuracy": 0.42853680457919835, |
| "num_tokens": 32930219.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 2.5642348393797874, |
| "epoch": 2.2154342625048686, |
| "grad_norm": 0.009765625, |
| "learning_rate": 1.6691111111111112e-05, |
| "loss": 3.5499, |
| "mean_token_accuracy": 0.4216444227844477, |
| "num_tokens": 33060606.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 2.464671529084444, |
| "epoch": 2.2243365047571357, |
| "grad_norm": 0.007476806640625, |
| "learning_rate": 1.666888888888889e-05, |
| "loss": 3.3666, |
| "mean_token_accuracy": 0.4348222777247429, |
| "num_tokens": 33194318.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.2243365047571357, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 33194318.0, |
| "eval_coding_runtime": 96.001, |
| "eval_coding_samples_per_second": 5.208, |
| "eval_coding_steps_per_second": 2.604, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.2243365047571357, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 33194318.0, |
| "eval_chemistry_runtime": 53.4749, |
| "eval_chemistry_samples_per_second": 9.35, |
| "eval_chemistry_steps_per_second": 4.675, |
| "step": 2500 |
| }, |
| { |
| "entropy": 2.5238001853227616, |
| "epoch": 2.233238747009403, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.6646666666666668e-05, |
| "loss": 3.4723, |
| "mean_token_accuracy": 0.42478870414197445, |
| "num_tokens": 33324345.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 2.431478377431631, |
| "epoch": 2.2421409892616704, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 1.6624444444444445e-05, |
| "loss": 3.3323, |
| "mean_token_accuracy": 0.43992548789829017, |
| "num_tokens": 33453292.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 2.413362357020378, |
| "epoch": 2.2510432315139375, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 1.6602222222222223e-05, |
| "loss": 3.289, |
| "mean_token_accuracy": 0.4414545498788357, |
| "num_tokens": 33590568.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 2.4422323763370515, |
| "epoch": 2.259945473766205, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.658e-05, |
| "loss": 3.3433, |
| "mean_token_accuracy": 0.43745585475116966, |
| "num_tokens": 33728875.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 2.4755229219794272, |
| "epoch": 2.268847716018472, |
| "grad_norm": 0.01153564453125, |
| "learning_rate": 1.6557777777777778e-05, |
| "loss": 3.4154, |
| "mean_token_accuracy": 0.43391959443688394, |
| "num_tokens": 33864335.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 2.5827997736632824, |
| "epoch": 2.2777499582707392, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 1.6535555555555556e-05, |
| "loss": 3.497, |
| "mean_token_accuracy": 0.4233642414212227, |
| "num_tokens": 33999305.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 2.5560081459581854, |
| "epoch": 2.2866522005230068, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.6513333333333333e-05, |
| "loss": 3.5117, |
| "mean_token_accuracy": 0.42018666174262764, |
| "num_tokens": 34134199.0, |
| "step": 2570 |
| }, |
| { |
| "entropy": 2.4955980271100997, |
| "epoch": 2.295554442775274, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.6491111111111114e-05, |
| "loss": 3.4138, |
| "mean_token_accuracy": 0.42663040682673453, |
| "num_tokens": 34269727.0, |
| "step": 2580 |
| }, |
| { |
| "entropy": 2.4141868740320205, |
| "epoch": 2.3044566850275414, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 1.646888888888889e-05, |
| "loss": 3.3136, |
| "mean_token_accuracy": 0.44276040960103275, |
| "num_tokens": 34401207.0, |
| "step": 2590 |
| }, |
| { |
| "entropy": 2.4291097052395343, |
| "epoch": 2.3133589272798085, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 1.644666666666667e-05, |
| "loss": 3.3374, |
| "mean_token_accuracy": 0.4356713238172233, |
| "num_tokens": 34537714.0, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.3133589272798085, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 34537714.0, |
| "eval_coding_runtime": 95.9565, |
| "eval_coding_samples_per_second": 5.211, |
| "eval_coding_steps_per_second": 2.605, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.3133589272798085, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 34537714.0, |
| "eval_chemistry_runtime": 53.8638, |
| "eval_chemistry_samples_per_second": 9.283, |
| "eval_chemistry_steps_per_second": 4.641, |
| "step": 2600 |
| }, |
| { |
| "entropy": 2.476061158627272, |
| "epoch": 2.322261169532076, |
| "grad_norm": 0.0142822265625, |
| "learning_rate": 1.6424444444444444e-05, |
| "loss": 3.3664, |
| "mean_token_accuracy": 0.4312489262782037, |
| "num_tokens": 34670848.0, |
| "step": 2610 |
| }, |
| { |
| "entropy": 2.5185798734426497, |
| "epoch": 2.331163411784343, |
| "grad_norm": 0.007781982421875, |
| "learning_rate": 1.6402222222222225e-05, |
| "loss": 3.4559, |
| "mean_token_accuracy": 0.4257050332613289, |
| "num_tokens": 34806647.0, |
| "step": 2620 |
| }, |
| { |
| "entropy": 2.4210305772721767, |
| "epoch": 2.3400656540366107, |
| "grad_norm": 0.010009765625, |
| "learning_rate": 1.638e-05, |
| "loss": 3.3581, |
| "mean_token_accuracy": 0.4404707044363022, |
| "num_tokens": 34938626.0, |
| "step": 2630 |
| }, |
| { |
| "entropy": 2.5381113201379777, |
| "epoch": 2.348967896288878, |
| "grad_norm": 0.007232666015625, |
| "learning_rate": 1.635777777777778e-05, |
| "loss": 3.4599, |
| "mean_token_accuracy": 0.4281400805339217, |
| "num_tokens": 35076423.0, |
| "step": 2640 |
| }, |
| { |
| "entropy": 2.586879266053438, |
| "epoch": 2.357870138541145, |
| "grad_norm": 0.01214599609375, |
| "learning_rate": 1.6335555555555558e-05, |
| "loss": 3.5017, |
| "mean_token_accuracy": 0.4232019340619445, |
| "num_tokens": 35207405.0, |
| "step": 2650 |
| }, |
| { |
| "entropy": 2.4670156836509705, |
| "epoch": 2.3667723807934125, |
| "grad_norm": 0.0069580078125, |
| "learning_rate": 1.6313333333333335e-05, |
| "loss": 3.3556, |
| "mean_token_accuracy": 0.43554529137909415, |
| "num_tokens": 35343494.0, |
| "step": 2660 |
| }, |
| { |
| "entropy": 2.492999928444624, |
| "epoch": 2.3756746230456796, |
| "grad_norm": 0.0140380859375, |
| "learning_rate": 1.6291111111111113e-05, |
| "loss": 3.4064, |
| "mean_token_accuracy": 0.4312469597905874, |
| "num_tokens": 35479230.0, |
| "step": 2670 |
| }, |
| { |
| "entropy": 2.533026181906462, |
| "epoch": 2.384576865297947, |
| "grad_norm": 0.01171875, |
| "learning_rate": 1.626888888888889e-05, |
| "loss": 3.444, |
| "mean_token_accuracy": 0.42157922238111495, |
| "num_tokens": 35614731.0, |
| "step": 2680 |
| }, |
| { |
| "entropy": 2.515670567750931, |
| "epoch": 2.393479107550214, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.6246666666666668e-05, |
| "loss": 3.4599, |
| "mean_token_accuracy": 0.4264078289270401, |
| "num_tokens": 35744776.0, |
| "step": 2690 |
| }, |
| { |
| "entropy": 2.5209390692412854, |
| "epoch": 2.4023813498024813, |
| "grad_norm": 0.0101318359375, |
| "learning_rate": 1.6224444444444446e-05, |
| "loss": 3.4605, |
| "mean_token_accuracy": 0.4283377526327968, |
| "num_tokens": 35877264.0, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.4023813498024813, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 35877264.0, |
| "eval_coding_runtime": 95.892, |
| "eval_coding_samples_per_second": 5.214, |
| "eval_coding_steps_per_second": 2.607, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.4023813498024813, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 35877264.0, |
| "eval_chemistry_runtime": 53.8472, |
| "eval_chemistry_samples_per_second": 9.286, |
| "eval_chemistry_steps_per_second": 4.643, |
| "step": 2700 |
| }, |
| { |
| "entropy": 2.5011694349348543, |
| "epoch": 2.411283592054749, |
| "grad_norm": 0.008544921875, |
| "learning_rate": 1.6202222222222223e-05, |
| "loss": 3.4186, |
| "mean_token_accuracy": 0.4300513813272119, |
| "num_tokens": 36008909.0, |
| "step": 2710 |
| }, |
| { |
| "entropy": 2.3889233499765394, |
| "epoch": 2.420185834307016, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 1.618e-05, |
| "loss": 3.284, |
| "mean_token_accuracy": 0.4462329100817442, |
| "num_tokens": 36147587.0, |
| "step": 2720 |
| }, |
| { |
| "entropy": 2.4801433332264424, |
| "epoch": 2.4290880765592835, |
| "grad_norm": 0.0108642578125, |
| "learning_rate": 1.615777777777778e-05, |
| "loss": 3.4046, |
| "mean_token_accuracy": 0.4334236603230238, |
| "num_tokens": 36284720.0, |
| "step": 2730 |
| }, |
| { |
| "entropy": 2.5423569701611997, |
| "epoch": 2.4379903188115506, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 1.6135555555555556e-05, |
| "loss": 3.4457, |
| "mean_token_accuracy": 0.43008503001183274, |
| "num_tokens": 36416125.0, |
| "step": 2740 |
| }, |
| { |
| "entropy": 2.497436165064573, |
| "epoch": 2.446892561063818, |
| "grad_norm": 0.00830078125, |
| "learning_rate": 1.6113333333333334e-05, |
| "loss": 3.4253, |
| "mean_token_accuracy": 0.42939351154491306, |
| "num_tokens": 36547915.0, |
| "step": 2750 |
| }, |
| { |
| "entropy": 2.5412722565233707, |
| "epoch": 2.4557948033160852, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 1.609111111111111e-05, |
| "loss": 3.4513, |
| "mean_token_accuracy": 0.42643848303705456, |
| "num_tokens": 36680383.0, |
| "step": 2760 |
| }, |
| { |
| "entropy": 2.4999127708375455, |
| "epoch": 2.4646970455683523, |
| "grad_norm": 0.01531982421875, |
| "learning_rate": 1.606888888888889e-05, |
| "loss": 3.4181, |
| "mean_token_accuracy": 0.42936620451509955, |
| "num_tokens": 36810342.0, |
| "step": 2770 |
| }, |
| { |
| "entropy": 2.5146854370832443, |
| "epoch": 2.47359928782062, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.6046666666666667e-05, |
| "loss": 3.416, |
| "mean_token_accuracy": 0.4320184450596571, |
| "num_tokens": 36946349.0, |
| "step": 2780 |
| }, |
| { |
| "entropy": 2.4285895794630052, |
| "epoch": 2.482501530072887, |
| "grad_norm": 0.00994873046875, |
| "learning_rate": 1.6024444444444444e-05, |
| "loss": 3.3828, |
| "mean_token_accuracy": 0.43509705597534776, |
| "num_tokens": 37084455.0, |
| "step": 2790 |
| }, |
| { |
| "entropy": 2.433745351433754, |
| "epoch": 2.4914037723251545, |
| "grad_norm": 0.007720947265625, |
| "learning_rate": 1.6002222222222222e-05, |
| "loss": 3.3568, |
| "mean_token_accuracy": 0.4401464581489563, |
| "num_tokens": 37219532.0, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.4914037723251545, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 37219532.0, |
| "eval_coding_runtime": 96.0836, |
| "eval_coding_samples_per_second": 5.204, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.4914037723251545, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 37219532.0, |
| "eval_chemistry_runtime": 54.1629, |
| "eval_chemistry_samples_per_second": 9.231, |
| "eval_chemistry_steps_per_second": 4.616, |
| "step": 2800 |
| }, |
| { |
| "entropy": 2.552928839996457, |
| "epoch": 2.5003060145774216, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 3.4724, |
| "mean_token_accuracy": 0.4233877120539546, |
| "num_tokens": 37344129.0, |
| "step": 2810 |
| }, |
| { |
| "entropy": 2.4043555468320847, |
| "epoch": 2.509208256829689, |
| "grad_norm": 0.01251220703125, |
| "learning_rate": 1.5957777777777777e-05, |
| "loss": 3.3213, |
| "mean_token_accuracy": 0.44005600484088064, |
| "num_tokens": 37477914.0, |
| "step": 2820 |
| }, |
| { |
| "entropy": 2.471666166931391, |
| "epoch": 2.5181104990819563, |
| "grad_norm": 0.00714111328125, |
| "learning_rate": 1.5935555555555558e-05, |
| "loss": 3.3936, |
| "mean_token_accuracy": 0.43629008531570435, |
| "num_tokens": 37607781.0, |
| "step": 2830 |
| }, |
| { |
| "entropy": 2.4896231681108474, |
| "epoch": 2.5270127413342234, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.5913333333333332e-05, |
| "loss": 3.3637, |
| "mean_token_accuracy": 0.4352514324709773, |
| "num_tokens": 37746690.0, |
| "step": 2840 |
| }, |
| { |
| "entropy": 2.571011830866337, |
| "epoch": 2.535914983586491, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.5891111111111113e-05, |
| "loss": 3.4708, |
| "mean_token_accuracy": 0.42845321968197825, |
| "num_tokens": 37879134.0, |
| "step": 2850 |
| }, |
| { |
| "entropy": 2.4915307737886905, |
| "epoch": 2.544817225838758, |
| "grad_norm": 0.0120849609375, |
| "learning_rate": 1.5868888888888888e-05, |
| "loss": 3.4299, |
| "mean_token_accuracy": 0.4302468568086624, |
| "num_tokens": 38005792.0, |
| "step": 2860 |
| }, |
| { |
| "entropy": 2.5264852702617646, |
| "epoch": 2.5537194680910256, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.584666666666667e-05, |
| "loss": 3.4387, |
| "mean_token_accuracy": 0.42782908510416745, |
| "num_tokens": 38139995.0, |
| "step": 2870 |
| }, |
| { |
| "entropy": 2.5440791875123976, |
| "epoch": 2.5626217103432927, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.5824444444444446e-05, |
| "loss": 3.4712, |
| "mean_token_accuracy": 0.4231386434286833, |
| "num_tokens": 38269021.0, |
| "step": 2880 |
| }, |
| { |
| "entropy": 2.504351659864187, |
| "epoch": 2.5715239525955598, |
| "grad_norm": 0.01324462890625, |
| "learning_rate": 1.5802222222222224e-05, |
| "loss": 3.3877, |
| "mean_token_accuracy": 0.4293781638145447, |
| "num_tokens": 38406316.0, |
| "step": 2890 |
| }, |
| { |
| "entropy": 2.5397102061659096, |
| "epoch": 2.5804261948478273, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.578e-05, |
| "loss": 3.44, |
| "mean_token_accuracy": 0.4284314359538257, |
| "num_tokens": 38545224.0, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.5804261948478273, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 38545224.0, |
| "eval_coding_runtime": 96.0156, |
| "eval_coding_samples_per_second": 5.207, |
| "eval_coding_steps_per_second": 2.604, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.5804261948478273, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 38545224.0, |
| "eval_chemistry_runtime": 53.7768, |
| "eval_chemistry_samples_per_second": 9.298, |
| "eval_chemistry_steps_per_second": 4.649, |
| "step": 2900 |
| }, |
| { |
| "entropy": 2.4839721962809564, |
| "epoch": 2.589328437100095, |
| "grad_norm": 0.01708984375, |
| "learning_rate": 1.575777777777778e-05, |
| "loss": 3.4748, |
| "mean_token_accuracy": 0.42800207175314425, |
| "num_tokens": 38670842.0, |
| "step": 2910 |
| }, |
| { |
| "entropy": 2.5038020804524423, |
| "epoch": 2.598230679352362, |
| "grad_norm": 0.00689697265625, |
| "learning_rate": 1.5735555555555557e-05, |
| "loss": 3.4185, |
| "mean_token_accuracy": 0.43284653443843124, |
| "num_tokens": 38802290.0, |
| "step": 2920 |
| }, |
| { |
| "entropy": 2.4623313575983046, |
| "epoch": 2.607132921604629, |
| "grad_norm": 0.01397705078125, |
| "learning_rate": 1.5713333333333334e-05, |
| "loss": 3.4017, |
| "mean_token_accuracy": 0.4338153407908976, |
| "num_tokens": 38932277.0, |
| "step": 2930 |
| }, |
| { |
| "entropy": 2.4417869858443737, |
| "epoch": 2.6160351638568966, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 1.5691111111111112e-05, |
| "loss": 3.347, |
| "mean_token_accuracy": 0.4401353685185313, |
| "num_tokens": 39065787.0, |
| "step": 2940 |
| }, |
| { |
| "entropy": 2.4268119893968105, |
| "epoch": 2.6249374061091637, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.5668888888888893e-05, |
| "loss": 3.3138, |
| "mean_token_accuracy": 0.4408662669360638, |
| "num_tokens": 39196987.0, |
| "step": 2950 |
| }, |
| { |
| "entropy": 2.5124353006482125, |
| "epoch": 2.6338396483614313, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 1.5646666666666667e-05, |
| "loss": 3.4596, |
| "mean_token_accuracy": 0.4245289133861661, |
| "num_tokens": 39328517.0, |
| "step": 2960 |
| }, |
| { |
| "entropy": 2.4018856599926948, |
| "epoch": 2.6427418906136984, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.5624444444444448e-05, |
| "loss": 3.2497, |
| "mean_token_accuracy": 0.4455626096576452, |
| "num_tokens": 39469764.0, |
| "step": 2970 |
| }, |
| { |
| "entropy": 2.4757705137133597, |
| "epoch": 2.6516441328659655, |
| "grad_norm": 0.010009765625, |
| "learning_rate": 1.5602222222222222e-05, |
| "loss": 3.4335, |
| "mean_token_accuracy": 0.43123262878507373, |
| "num_tokens": 39600641.0, |
| "step": 2980 |
| }, |
| { |
| "entropy": 2.5936547361314295, |
| "epoch": 2.660546375118233, |
| "grad_norm": 0.0115966796875, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 3.5659, |
| "mean_token_accuracy": 0.41880734991282226, |
| "num_tokens": 39729058.0, |
| "step": 2990 |
| }, |
| { |
| "entropy": 2.521015575528145, |
| "epoch": 2.6694486173705, |
| "grad_norm": 0.010009765625, |
| "learning_rate": 1.5557777777777778e-05, |
| "loss": 3.4316, |
| "mean_token_accuracy": 0.42712221732363104, |
| "num_tokens": 39857917.0, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.6694486173705, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 39857917.0, |
| "eval_coding_runtime": 96.0078, |
| "eval_coding_samples_per_second": 5.208, |
| "eval_coding_steps_per_second": 2.604, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.6694486173705, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 39857917.0, |
| "eval_chemistry_runtime": 53.8905, |
| "eval_chemistry_samples_per_second": 9.278, |
| "eval_chemistry_steps_per_second": 4.639, |
| "step": 3000 |
| }, |
| { |
| "entropy": 2.514738455414772, |
| "epoch": 2.6783508596227676, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.553555555555556e-05, |
| "loss": 3.4919, |
| "mean_token_accuracy": 0.4237719819881022, |
| "num_tokens": 39992435.0, |
| "step": 3010 |
| }, |
| { |
| "entropy": 2.5164104022085665, |
| "epoch": 2.6872531018750347, |
| "grad_norm": 0.00860595703125, |
| "learning_rate": 1.5513333333333333e-05, |
| "loss": 3.4537, |
| "mean_token_accuracy": 0.4282899649813771, |
| "num_tokens": 40119174.0, |
| "step": 3020 |
| }, |
| { |
| "entropy": 2.5460123769938945, |
| "epoch": 2.696155344127302, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 1.5491111111111114e-05, |
| "loss": 3.5083, |
| "mean_token_accuracy": 0.42434458900243044, |
| "num_tokens": 40248906.0, |
| "step": 3030 |
| }, |
| { |
| "entropy": 2.5710619531571863, |
| "epoch": 2.7050575863795694, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.546888888888889e-05, |
| "loss": 3.4819, |
| "mean_token_accuracy": 0.42083613742142917, |
| "num_tokens": 40379054.0, |
| "step": 3040 |
| }, |
| { |
| "entropy": 2.432519157975912, |
| "epoch": 2.7139598286318365, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.544666666666667e-05, |
| "loss": 3.3553, |
| "mean_token_accuracy": 0.43840309549123047, |
| "num_tokens": 40517497.0, |
| "step": 3050 |
| }, |
| { |
| "entropy": 2.5490108489990235, |
| "epoch": 2.722862070884104, |
| "grad_norm": 0.01171875, |
| "learning_rate": 1.5424444444444447e-05, |
| "loss": 3.48, |
| "mean_token_accuracy": 0.42548066675662993, |
| "num_tokens": 40650970.0, |
| "step": 3060 |
| }, |
| { |
| "entropy": 2.4773500584065915, |
| "epoch": 2.731764313136371, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.5402222222222224e-05, |
| "loss": 3.4586, |
| "mean_token_accuracy": 0.42619508197531103, |
| "num_tokens": 40781902.0, |
| "step": 3070 |
| }, |
| { |
| "entropy": 2.520800007134676, |
| "epoch": 2.7406665553886382, |
| "grad_norm": 0.01409912109375, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 3.4498, |
| "mean_token_accuracy": 0.43010157672688365, |
| "num_tokens": 40914339.0, |
| "step": 3080 |
| }, |
| { |
| "entropy": 2.5275987461209297, |
| "epoch": 2.749568797640906, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.535777777777778e-05, |
| "loss": 3.4496, |
| "mean_token_accuracy": 0.43131065648049116, |
| "num_tokens": 41046899.0, |
| "step": 3090 |
| }, |
| { |
| "entropy": 2.465614350885153, |
| "epoch": 2.7584710398931733, |
| "grad_norm": 0.0106201171875, |
| "learning_rate": 1.5335555555555557e-05, |
| "loss": 3.4002, |
| "mean_token_accuracy": 0.4297482826747, |
| "num_tokens": 41181827.0, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.7584710398931733, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 41181827.0, |
| "eval_coding_runtime": 95.9395, |
| "eval_coding_samples_per_second": 5.212, |
| "eval_coding_steps_per_second": 2.606, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.7584710398931733, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 41181827.0, |
| "eval_chemistry_runtime": 53.7755, |
| "eval_chemistry_samples_per_second": 9.298, |
| "eval_chemistry_steps_per_second": 4.649, |
| "step": 3100 |
| }, |
| { |
| "entropy": 2.43465124592185, |
| "epoch": 2.7673732821454404, |
| "grad_norm": 0.00738525390625, |
| "learning_rate": 1.5313333333333335e-05, |
| "loss": 3.3302, |
| "mean_token_accuracy": 0.4404412625357509, |
| "num_tokens": 41318568.0, |
| "step": 3110 |
| }, |
| { |
| "entropy": 2.476666798442602, |
| "epoch": 2.7762755243977075, |
| "grad_norm": 0.0086669921875, |
| "learning_rate": 1.5291111111111112e-05, |
| "loss": 3.402, |
| "mean_token_accuracy": 0.429670218937099, |
| "num_tokens": 41448961.0, |
| "step": 3120 |
| }, |
| { |
| "entropy": 2.4710974212735892, |
| "epoch": 2.785177766649975, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 1.526888888888889e-05, |
| "loss": 3.4179, |
| "mean_token_accuracy": 0.4358305299654603, |
| "num_tokens": 41577277.0, |
| "step": 3130 |
| }, |
| { |
| "entropy": 2.5007879421114922, |
| "epoch": 2.794080008902242, |
| "grad_norm": 0.00982666015625, |
| "learning_rate": 1.5246666666666668e-05, |
| "loss": 3.4352, |
| "mean_token_accuracy": 0.43269760962575676, |
| "num_tokens": 41712243.0, |
| "step": 3140 |
| }, |
| { |
| "entropy": 2.4640229254961015, |
| "epoch": 2.8029822511545097, |
| "grad_norm": 0.01043701171875, |
| "learning_rate": 1.5224444444444447e-05, |
| "loss": 3.4041, |
| "mean_token_accuracy": 0.42871168088167905, |
| "num_tokens": 41842853.0, |
| "step": 3150 |
| }, |
| { |
| "entropy": 2.4800624772906303, |
| "epoch": 2.811884493406777, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 1.5202222222222223e-05, |
| "loss": 3.3829, |
| "mean_token_accuracy": 0.4388408211991191, |
| "num_tokens": 41981755.0, |
| "step": 3160 |
| }, |
| { |
| "entropy": 2.4440943859517574, |
| "epoch": 2.820786735659044, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 3.3477, |
| "mean_token_accuracy": 0.4349145981483161, |
| "num_tokens": 42118503.0, |
| "step": 3170 |
| }, |
| { |
| "entropy": 2.4008580282330514, |
| "epoch": 2.8296889779113115, |
| "grad_norm": 0.01190185546875, |
| "learning_rate": 1.5157777777777778e-05, |
| "loss": 3.2913, |
| "mean_token_accuracy": 0.44715584106743334, |
| "num_tokens": 42258270.0, |
| "step": 3180 |
| }, |
| { |
| "entropy": 2.4839060604572296, |
| "epoch": 2.8385912201635786, |
| "grad_norm": 0.00732421875, |
| "learning_rate": 1.5135555555555557e-05, |
| "loss": 3.4047, |
| "mean_token_accuracy": 0.42959455121308565, |
| "num_tokens": 42389715.0, |
| "step": 3190 |
| }, |
| { |
| "entropy": 2.4769830726087094, |
| "epoch": 2.847493462415846, |
| "grad_norm": 0.00994873046875, |
| "learning_rate": 1.5113333333333335e-05, |
| "loss": 3.4001, |
| "mean_token_accuracy": 0.4322687996551394, |
| "num_tokens": 42529321.0, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.847493462415846, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 42529321.0, |
| "eval_coding_runtime": 96.1003, |
| "eval_coding_samples_per_second": 5.203, |
| "eval_coding_steps_per_second": 2.601, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.847493462415846, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 42529321.0, |
| "eval_chemistry_runtime": 53.8649, |
| "eval_chemistry_samples_per_second": 9.282, |
| "eval_chemistry_steps_per_second": 4.641, |
| "step": 3200 |
| }, |
| { |
| "entropy": 2.5920967496931553, |
| "epoch": 2.856395704668113, |
| "grad_norm": 0.00738525390625, |
| "learning_rate": 1.5091111111111113e-05, |
| "loss": 3.4768, |
| "mean_token_accuracy": 0.42277047960087655, |
| "num_tokens": 42661653.0, |
| "step": 3210 |
| }, |
| { |
| "entropy": 2.5145366221666334, |
| "epoch": 2.8652979469203803, |
| "grad_norm": 0.0089111328125, |
| "learning_rate": 1.506888888888889e-05, |
| "loss": 3.467, |
| "mean_token_accuracy": 0.4248306507244706, |
| "num_tokens": 42793729.0, |
| "step": 3220 |
| }, |
| { |
| "entropy": 2.508847577124834, |
| "epoch": 2.874200189172648, |
| "grad_norm": 0.0126953125, |
| "learning_rate": 1.5046666666666668e-05, |
| "loss": 3.4596, |
| "mean_token_accuracy": 0.4254114419221878, |
| "num_tokens": 42932339.0, |
| "step": 3230 |
| }, |
| { |
| "entropy": 2.5004635378718376, |
| "epoch": 2.8831024314249154, |
| "grad_norm": 0.0086669921875, |
| "learning_rate": 1.5024444444444445e-05, |
| "loss": 3.415, |
| "mean_token_accuracy": 0.42922934256494044, |
| "num_tokens": 43067006.0, |
| "step": 3240 |
| }, |
| { |
| "entropy": 2.4226531594991685, |
| "epoch": 2.8920046736771825, |
| "grad_norm": 0.0118408203125, |
| "learning_rate": 1.5002222222222223e-05, |
| "loss": 3.3478, |
| "mean_token_accuracy": 0.4390448907390237, |
| "num_tokens": 43204892.0, |
| "step": 3250 |
| }, |
| { |
| "entropy": 2.5149272002279757, |
| "epoch": 2.9009069159294496, |
| "grad_norm": 0.0228271484375, |
| "learning_rate": 1.498e-05, |
| "loss": 3.47, |
| "mean_token_accuracy": 0.4275529269129038, |
| "num_tokens": 43333868.0, |
| "step": 3260 |
| }, |
| { |
| "entropy": 2.4623923763632773, |
| "epoch": 2.909809158181717, |
| "grad_norm": 0.01251220703125, |
| "learning_rate": 1.495777777777778e-05, |
| "loss": 3.3621, |
| "mean_token_accuracy": 0.43572554057464, |
| "num_tokens": 43469937.0, |
| "step": 3270 |
| }, |
| { |
| "entropy": 2.50618307441473, |
| "epoch": 2.9187114004339842, |
| "grad_norm": 0.0101318359375, |
| "learning_rate": 1.4935555555555556e-05, |
| "loss": 3.4544, |
| "mean_token_accuracy": 0.42696905247867106, |
| "num_tokens": 43605151.0, |
| "step": 3280 |
| }, |
| { |
| "entropy": 2.4748622618615626, |
| "epoch": 2.927613642686252, |
| "grad_norm": 0.00927734375, |
| "learning_rate": 1.4913333333333335e-05, |
| "loss": 3.3707, |
| "mean_token_accuracy": 0.43523391522467136, |
| "num_tokens": 43735123.0, |
| "step": 3290 |
| }, |
| { |
| "entropy": 2.429765923321247, |
| "epoch": 2.936515884938519, |
| "grad_norm": 0.00726318359375, |
| "learning_rate": 1.4891111111111111e-05, |
| "loss": 3.3219, |
| "mean_token_accuracy": 0.4356573619879782, |
| "num_tokens": 43863034.0, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.936515884938519, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 43863034.0, |
| "eval_coding_runtime": 95.998, |
| "eval_coding_samples_per_second": 5.208, |
| "eval_coding_steps_per_second": 2.604, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.936515884938519, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 43863034.0, |
| "eval_chemistry_runtime": 53.7557, |
| "eval_chemistry_samples_per_second": 9.301, |
| "eval_chemistry_steps_per_second": 4.651, |
| "step": 3300 |
| }, |
| { |
| "entropy": 2.6052315428853037, |
| "epoch": 2.945418127190786, |
| "grad_norm": 0.00738525390625, |
| "learning_rate": 1.486888888888889e-05, |
| "loss": 3.5086, |
| "mean_token_accuracy": 0.4217910974286497, |
| "num_tokens": 43996070.0, |
| "step": 3310 |
| }, |
| { |
| "entropy": 2.4676881857216357, |
| "epoch": 2.9543203694430535, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.4846666666666666e-05, |
| "loss": 3.3646, |
| "mean_token_accuracy": 0.4361693524755538, |
| "num_tokens": 44126796.0, |
| "step": 3320 |
| }, |
| { |
| "entropy": 2.569188876450062, |
| "epoch": 2.9632226116953206, |
| "grad_norm": 0.01153564453125, |
| "learning_rate": 1.4824444444444446e-05, |
| "loss": 3.4793, |
| "mean_token_accuracy": 0.42434552125632763, |
| "num_tokens": 44254900.0, |
| "step": 3330 |
| }, |
| { |
| "entropy": 2.533152540773153, |
| "epoch": 2.972124853947588, |
| "grad_norm": 0.010498046875, |
| "learning_rate": 1.4802222222222225e-05, |
| "loss": 3.4794, |
| "mean_token_accuracy": 0.42293534940108657, |
| "num_tokens": 44382724.0, |
| "step": 3340 |
| }, |
| { |
| "entropy": 2.452186634391546, |
| "epoch": 2.9810270961998553, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 1.478e-05, |
| "loss": 3.3787, |
| "mean_token_accuracy": 0.43849403914064167, |
| "num_tokens": 44509601.0, |
| "step": 3350 |
| }, |
| { |
| "entropy": 2.4548177666962148, |
| "epoch": 2.9899293384521224, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.475777777777778e-05, |
| "loss": 3.3342, |
| "mean_token_accuracy": 0.4381516374647617, |
| "num_tokens": 44651471.0, |
| "step": 3360 |
| }, |
| { |
| "entropy": 2.5030359968543054, |
| "epoch": 2.99883158070439, |
| "grad_norm": 0.015380859375, |
| "learning_rate": 1.4735555555555556e-05, |
| "loss": 3.3836, |
| "mean_token_accuracy": 0.43045337554067375, |
| "num_tokens": 44780243.0, |
| "step": 3370 |
| }, |
| { |
| "entropy": 2.4356001191491248, |
| "epoch": 3.0071217938018138, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 1.4713333333333335e-05, |
| "loss": 3.3374, |
| "mean_token_accuracy": 0.43639238418748716, |
| "num_tokens": 44909981.0, |
| "step": 3380 |
| }, |
| { |
| "entropy": 2.5676138043403625, |
| "epoch": 3.0160240360540813, |
| "grad_norm": 0.01318359375, |
| "learning_rate": 1.4691111111111111e-05, |
| "loss": 3.4993, |
| "mean_token_accuracy": 0.4237755025736988, |
| "num_tokens": 45042617.0, |
| "step": 3390 |
| }, |
| { |
| "entropy": 2.6604440554976465, |
| "epoch": 3.0249262783063484, |
| "grad_norm": 0.00830078125, |
| "learning_rate": 1.466888888888889e-05, |
| "loss": 3.5598, |
| "mean_token_accuracy": 0.41665924787521363, |
| "num_tokens": 45176547.0, |
| "step": 3400 |
| }, |
| { |
| "epoch": 3.0249262783063484, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 45176547.0, |
| "eval_coding_runtime": 95.8912, |
| "eval_coding_samples_per_second": 5.214, |
| "eval_coding_steps_per_second": 2.607, |
| "step": 3400 |
| }, |
| { |
| "epoch": 3.0249262783063484, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 45176547.0, |
| "eval_chemistry_runtime": 53.6774, |
| "eval_chemistry_samples_per_second": 9.315, |
| "eval_chemistry_steps_per_second": 4.657, |
| "step": 3400 |
| }, |
| { |
| "entropy": 2.439664166420698, |
| "epoch": 3.0338285205586155, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 1.4646666666666666e-05, |
| "loss": 3.3854, |
| "mean_token_accuracy": 0.42942666802555324, |
| "num_tokens": 45303236.0, |
| "step": 3410 |
| }, |
| { |
| "entropy": 2.4116016045212745, |
| "epoch": 3.042730762810883, |
| "grad_norm": 0.016357421875, |
| "learning_rate": 1.4624444444444446e-05, |
| "loss": 3.3243, |
| "mean_token_accuracy": 0.438990593329072, |
| "num_tokens": 45437181.0, |
| "step": 3420 |
| }, |
| { |
| "entropy": 2.504232635349035, |
| "epoch": 3.05163300506315, |
| "grad_norm": 0.019775390625, |
| "learning_rate": 1.4602222222222225e-05, |
| "loss": 3.4323, |
| "mean_token_accuracy": 0.4301702308468521, |
| "num_tokens": 45570054.0, |
| "step": 3430 |
| }, |
| { |
| "entropy": 2.501502574980259, |
| "epoch": 3.0605352473154177, |
| "grad_norm": 0.00982666015625, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 3.4388, |
| "mean_token_accuracy": 0.4273995775729418, |
| "num_tokens": 45702431.0, |
| "step": 3440 |
| }, |
| { |
| "entropy": 2.4722267992794515, |
| "epoch": 3.069437489567685, |
| "grad_norm": 0.01348876953125, |
| "learning_rate": 1.455777777777778e-05, |
| "loss": 3.3632, |
| "mean_token_accuracy": 0.4381282507441938, |
| "num_tokens": 45845103.0, |
| "step": 3450 |
| }, |
| { |
| "entropy": 2.480802757292986, |
| "epoch": 3.0783397318199524, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.4535555555555556e-05, |
| "loss": 3.393, |
| "mean_token_accuracy": 0.4303450108505785, |
| "num_tokens": 45975788.0, |
| "step": 3460 |
| }, |
| { |
| "entropy": 2.3935215406119825, |
| "epoch": 3.0872419740722195, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 1.4513333333333336e-05, |
| "loss": 3.2972, |
| "mean_token_accuracy": 0.4492803938686848, |
| "num_tokens": 46116162.0, |
| "step": 3470 |
| }, |
| { |
| "entropy": 2.596017763018608, |
| "epoch": 3.0961442163244866, |
| "grad_norm": 0.01141357421875, |
| "learning_rate": 1.4491111111111111e-05, |
| "loss": 3.5173, |
| "mean_token_accuracy": 0.4214028215035796, |
| "num_tokens": 46251349.0, |
| "step": 3480 |
| }, |
| { |
| "entropy": 2.495695576071739, |
| "epoch": 3.105046458576754, |
| "grad_norm": 0.01116943359375, |
| "learning_rate": 1.446888888888889e-05, |
| "loss": 3.4303, |
| "mean_token_accuracy": 0.4294706436805427, |
| "num_tokens": 46382955.0, |
| "step": 3490 |
| }, |
| { |
| "entropy": 2.5789878353476525, |
| "epoch": 3.113948700829021, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.4446666666666668e-05, |
| "loss": 3.5014, |
| "mean_token_accuracy": 0.42273485716432335, |
| "num_tokens": 46516802.0, |
| "step": 3500 |
| }, |
| { |
| "epoch": 3.113948700829021, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 46516802.0, |
| "eval_coding_runtime": 95.9822, |
| "eval_coding_samples_per_second": 5.209, |
| "eval_coding_steps_per_second": 2.605, |
| "step": 3500 |
| }, |
| { |
| "epoch": 3.113948700829021, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 46516802.0, |
| "eval_chemistry_runtime": 53.7305, |
| "eval_chemistry_samples_per_second": 9.306, |
| "eval_chemistry_steps_per_second": 4.653, |
| "step": 3500 |
| }, |
| { |
| "entropy": 2.4588116995990275, |
| "epoch": 3.1228509430812887, |
| "grad_norm": 0.007049560546875, |
| "learning_rate": 1.4424444444444446e-05, |
| "loss": 3.3785, |
| "mean_token_accuracy": 0.4377431318163872, |
| "num_tokens": 46651257.0, |
| "step": 3510 |
| }, |
| { |
| "entropy": 2.4462937697768212, |
| "epoch": 3.131753185333556, |
| "grad_norm": 0.0111083984375, |
| "learning_rate": 1.4402222222222224e-05, |
| "loss": 3.3334, |
| "mean_token_accuracy": 0.4411364750936627, |
| "num_tokens": 46790537.0, |
| "step": 3520 |
| }, |
| { |
| "entropy": 2.502244517952204, |
| "epoch": 3.1406554275858234, |
| "grad_norm": 0.01397705078125, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 3.3835, |
| "mean_token_accuracy": 0.4341385864652693, |
| "num_tokens": 46924520.0, |
| "step": 3530 |
| }, |
| { |
| "entropy": 2.445697668939829, |
| "epoch": 3.1495576698380905, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 1.4357777777777779e-05, |
| "loss": 3.3572, |
| "mean_token_accuracy": 0.4375282542780042, |
| "num_tokens": 47052877.0, |
| "step": 3540 |
| }, |
| { |
| "entropy": 2.4848343193531037, |
| "epoch": 3.1584599120903576, |
| "grad_norm": 0.01141357421875, |
| "learning_rate": 1.4335555555555556e-05, |
| "loss": 3.3995, |
| "mean_token_accuracy": 0.43292110990732907, |
| "num_tokens": 47183255.0, |
| "step": 3550 |
| }, |
| { |
| "entropy": 2.5108699925243854, |
| "epoch": 3.167362154342625, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.4313333333333334e-05, |
| "loss": 3.4398, |
| "mean_token_accuracy": 0.4308509321883321, |
| "num_tokens": 47315587.0, |
| "step": 3560 |
| }, |
| { |
| "entropy": 2.506462197750807, |
| "epoch": 3.1762643965948922, |
| "grad_norm": 0.01080322265625, |
| "learning_rate": 1.4291111111111113e-05, |
| "loss": 3.4125, |
| "mean_token_accuracy": 0.42895674575120213, |
| "num_tokens": 47449226.0, |
| "step": 3570 |
| }, |
| { |
| "entropy": 2.423535010963678, |
| "epoch": 3.18516663884716, |
| "grad_norm": 0.01092529296875, |
| "learning_rate": 1.426888888888889e-05, |
| "loss": 3.3495, |
| "mean_token_accuracy": 0.43755780160427094, |
| "num_tokens": 47578352.0, |
| "step": 3580 |
| }, |
| { |
| "entropy": 2.463869632780552, |
| "epoch": 3.194068881099427, |
| "grad_norm": 0.009765625, |
| "learning_rate": 1.4246666666666669e-05, |
| "loss": 3.3931, |
| "mean_token_accuracy": 0.42939086267724635, |
| "num_tokens": 47712245.0, |
| "step": 3590 |
| }, |
| { |
| "entropy": 2.5216509342193603, |
| "epoch": 3.202971123351694, |
| "grad_norm": 0.0137939453125, |
| "learning_rate": 1.4224444444444445e-05, |
| "loss": 3.469, |
| "mean_token_accuracy": 0.4243164682760835, |
| "num_tokens": 47844623.0, |
| "step": 3600 |
| }, |
| { |
| "epoch": 3.202971123351694, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 47844623.0, |
| "eval_coding_runtime": 96.1664, |
| "eval_coding_samples_per_second": 5.199, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 3600 |
| }, |
| { |
| "epoch": 3.202971123351694, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 47844623.0, |
| "eval_chemistry_runtime": 53.9399, |
| "eval_chemistry_samples_per_second": 9.27, |
| "eval_chemistry_steps_per_second": 4.635, |
| "step": 3600 |
| }, |
| { |
| "entropy": 2.397772777825594, |
| "epoch": 3.2118733656039615, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 1.4202222222222224e-05, |
| "loss": 3.2856, |
| "mean_token_accuracy": 0.4466150883585215, |
| "num_tokens": 47980987.0, |
| "step": 3610 |
| }, |
| { |
| "entropy": 2.4860513128340243, |
| "epoch": 3.2207756078562286, |
| "grad_norm": 0.01220703125, |
| "learning_rate": 1.418e-05, |
| "loss": 3.4216, |
| "mean_token_accuracy": 0.42931431215256455, |
| "num_tokens": 48112262.0, |
| "step": 3620 |
| }, |
| { |
| "entropy": 2.5791078358888626, |
| "epoch": 3.229677850108496, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.4157777777777779e-05, |
| "loss": 3.4846, |
| "mean_token_accuracy": 0.41979573853313923, |
| "num_tokens": 48240383.0, |
| "step": 3630 |
| }, |
| { |
| "entropy": 2.5925147205591204, |
| "epoch": 3.2385800923607633, |
| "grad_norm": 0.0164794921875, |
| "learning_rate": 1.4135555555555555e-05, |
| "loss": 3.5448, |
| "mean_token_accuracy": 0.41899388274177907, |
| "num_tokens": 48369097.0, |
| "step": 3640 |
| }, |
| { |
| "entropy": 2.4554927572607994, |
| "epoch": 3.247482334613031, |
| "grad_norm": 0.0106201171875, |
| "learning_rate": 1.4113333333333334e-05, |
| "loss": 3.3668, |
| "mean_token_accuracy": 0.4345109906047583, |
| "num_tokens": 48502179.0, |
| "step": 3650 |
| }, |
| { |
| "entropy": 2.527767989784479, |
| "epoch": 3.256384576865298, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 1.4091111111111114e-05, |
| "loss": 3.4956, |
| "mean_token_accuracy": 0.42610255610197784, |
| "num_tokens": 48628165.0, |
| "step": 3660 |
| }, |
| { |
| "entropy": 2.472541968524456, |
| "epoch": 3.2652868191175655, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 1.406888888888889e-05, |
| "loss": 3.377, |
| "mean_token_accuracy": 0.42942148055881263, |
| "num_tokens": 48760755.0, |
| "step": 3670 |
| }, |
| { |
| "entropy": 2.4852148868143558, |
| "epoch": 3.2741890613698326, |
| "grad_norm": 0.00927734375, |
| "learning_rate": 1.4046666666666669e-05, |
| "loss": 3.3921, |
| "mean_token_accuracy": 0.43031275290995835, |
| "num_tokens": 48898634.0, |
| "step": 3680 |
| }, |
| { |
| "entropy": 2.4387554459273817, |
| "epoch": 3.2830913036220997, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.4024444444444445e-05, |
| "loss": 3.3804, |
| "mean_token_accuracy": 0.43460222873836757, |
| "num_tokens": 49034606.0, |
| "step": 3690 |
| }, |
| { |
| "entropy": 2.4800694681704045, |
| "epoch": 3.291993545874367, |
| "grad_norm": 0.00872802734375, |
| "learning_rate": 1.4002222222222224e-05, |
| "loss": 3.4086, |
| "mean_token_accuracy": 0.43042036704719067, |
| "num_tokens": 49167839.0, |
| "step": 3700 |
| }, |
| { |
| "epoch": 3.291993545874367, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 49167839.0, |
| "eval_coding_runtime": 96.1789, |
| "eval_coding_samples_per_second": 5.199, |
| "eval_coding_steps_per_second": 2.599, |
| "step": 3700 |
| }, |
| { |
| "epoch": 3.291993545874367, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 49167839.0, |
| "eval_chemistry_runtime": 53.7123, |
| "eval_chemistry_samples_per_second": 9.309, |
| "eval_chemistry_steps_per_second": 4.654, |
| "step": 3700 |
| }, |
| { |
| "entropy": 2.5049490705132484, |
| "epoch": 3.3008957881266343, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.398e-05, |
| "loss": 3.4338, |
| "mean_token_accuracy": 0.42748601287603377, |
| "num_tokens": 49297881.0, |
| "step": 3710 |
| }, |
| { |
| "entropy": 2.4800257585942744, |
| "epoch": 3.309798030378902, |
| "grad_norm": 0.01116943359375, |
| "learning_rate": 1.395777777777778e-05, |
| "loss": 3.3637, |
| "mean_token_accuracy": 0.43758540898561477, |
| "num_tokens": 49431378.0, |
| "step": 3720 |
| }, |
| { |
| "entropy": 2.4624336190521716, |
| "epoch": 3.318700272631169, |
| "grad_norm": 0.0076904296875, |
| "learning_rate": 1.3935555555555557e-05, |
| "loss": 3.3715, |
| "mean_token_accuracy": 0.430605623498559, |
| "num_tokens": 49566679.0, |
| "step": 3730 |
| }, |
| { |
| "entropy": 2.4866604186594485, |
| "epoch": 3.327602514883436, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.3913333333333335e-05, |
| "loss": 3.4275, |
| "mean_token_accuracy": 0.42791981585323813, |
| "num_tokens": 49699309.0, |
| "step": 3740 |
| }, |
| { |
| "entropy": 2.5161143854260444, |
| "epoch": 3.3365047571357036, |
| "grad_norm": 0.01025390625, |
| "learning_rate": 1.3891111111111114e-05, |
| "loss": 3.4586, |
| "mean_token_accuracy": 0.4291952922940254, |
| "num_tokens": 49839280.0, |
| "step": 3750 |
| }, |
| { |
| "entropy": 2.513023091852665, |
| "epoch": 3.3454069993879707, |
| "grad_norm": 0.01300048828125, |
| "learning_rate": 1.386888888888889e-05, |
| "loss": 3.4235, |
| "mean_token_accuracy": 0.4273349129594862, |
| "num_tokens": 49972032.0, |
| "step": 3760 |
| }, |
| { |
| "entropy": 2.453387563303113, |
| "epoch": 3.3543092416402382, |
| "grad_norm": 0.0205078125, |
| "learning_rate": 1.3846666666666669e-05, |
| "loss": 3.371, |
| "mean_token_accuracy": 0.433937793970108, |
| "num_tokens": 50110880.0, |
| "step": 3770 |
| }, |
| { |
| "entropy": 2.5347550593316557, |
| "epoch": 3.3632114838925053, |
| "grad_norm": 0.023681640625, |
| "learning_rate": 1.3824444444444445e-05, |
| "loss": 3.485, |
| "mean_token_accuracy": 0.4235278295353055, |
| "num_tokens": 50241367.0, |
| "step": 3780 |
| }, |
| { |
| "entropy": 2.5014845818281173, |
| "epoch": 3.372113726144773, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.3802222222222224e-05, |
| "loss": 3.432, |
| "mean_token_accuracy": 0.4295598682947457, |
| "num_tokens": 50377432.0, |
| "step": 3790 |
| }, |
| { |
| "entropy": 2.564122501015663, |
| "epoch": 3.38101596839704, |
| "grad_norm": 0.007354736328125, |
| "learning_rate": 1.378e-05, |
| "loss": 3.5331, |
| "mean_token_accuracy": 0.4206165148876607, |
| "num_tokens": 50511167.0, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.38101596839704, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 50511167.0, |
| "eval_coding_runtime": 95.9706, |
| "eval_coding_samples_per_second": 5.21, |
| "eval_coding_steps_per_second": 2.605, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.38101596839704, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 50511167.0, |
| "eval_chemistry_runtime": 53.7899, |
| "eval_chemistry_samples_per_second": 9.295, |
| "eval_chemistry_steps_per_second": 4.648, |
| "step": 3800 |
| }, |
| { |
| "entropy": 2.5211972281336785, |
| "epoch": 3.3899182106493075, |
| "grad_norm": 0.00775146484375, |
| "learning_rate": 1.375777777777778e-05, |
| "loss": 3.4843, |
| "mean_token_accuracy": 0.4271978460252285, |
| "num_tokens": 50643244.0, |
| "step": 3810 |
| }, |
| { |
| "entropy": 2.547449840605259, |
| "epoch": 3.3988204529015746, |
| "grad_norm": 0.0107421875, |
| "learning_rate": 1.3735555555555557e-05, |
| "loss": 3.4586, |
| "mean_token_accuracy": 0.42623377256095407, |
| "num_tokens": 50776842.0, |
| "step": 3820 |
| }, |
| { |
| "entropy": 2.4848173283040524, |
| "epoch": 3.4077226951538417, |
| "grad_norm": 0.01312255859375, |
| "learning_rate": 1.3713333333333335e-05, |
| "loss": 3.3897, |
| "mean_token_accuracy": 0.42863423377275467, |
| "num_tokens": 50910635.0, |
| "step": 3830 |
| }, |
| { |
| "entropy": 2.544556252658367, |
| "epoch": 3.4166249374061093, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.3691111111111112e-05, |
| "loss": 3.4951, |
| "mean_token_accuracy": 0.42494107950478793, |
| "num_tokens": 51040514.0, |
| "step": 3840 |
| }, |
| { |
| "entropy": 2.56239742487669, |
| "epoch": 3.4255271796583764, |
| "grad_norm": 0.0113525390625, |
| "learning_rate": 1.366888888888889e-05, |
| "loss": 3.4645, |
| "mean_token_accuracy": 0.41969831828027965, |
| "num_tokens": 51169529.0, |
| "step": 3850 |
| }, |
| { |
| "entropy": 2.537663085013628, |
| "epoch": 3.434429421910644, |
| "grad_norm": 0.007781982421875, |
| "learning_rate": 1.3646666666666668e-05, |
| "loss": 3.449, |
| "mean_token_accuracy": 0.42752351220697166, |
| "num_tokens": 51300696.0, |
| "step": 3860 |
| }, |
| { |
| "entropy": 2.482908922433853, |
| "epoch": 3.443331664162911, |
| "grad_norm": 0.007598876953125, |
| "learning_rate": 1.3624444444444445e-05, |
| "loss": 3.3884, |
| "mean_token_accuracy": 0.429613190703094, |
| "num_tokens": 51435474.0, |
| "step": 3870 |
| }, |
| { |
| "entropy": 2.578912515938282, |
| "epoch": 3.452233906415178, |
| "grad_norm": 0.0101318359375, |
| "learning_rate": 1.3602222222222223e-05, |
| "loss": 3.5873, |
| "mean_token_accuracy": 0.41274937335401773, |
| "num_tokens": 51558606.0, |
| "step": 3880 |
| }, |
| { |
| "entropy": 2.544483095407486, |
| "epoch": 3.4611361486674457, |
| "grad_norm": 0.01068115234375, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 3.4514, |
| "mean_token_accuracy": 0.4252054486423731, |
| "num_tokens": 51691071.0, |
| "step": 3890 |
| }, |
| { |
| "entropy": 2.483997346460819, |
| "epoch": 3.470038390919713, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 1.3557777777777778e-05, |
| "loss": 3.3723, |
| "mean_token_accuracy": 0.43083390276879074, |
| "num_tokens": 51827834.0, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.470038390919713, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 51827834.0, |
| "eval_coding_runtime": 95.9457, |
| "eval_coding_samples_per_second": 5.211, |
| "eval_coding_steps_per_second": 2.606, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.470038390919713, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 51827834.0, |
| "eval_chemistry_runtime": 53.8009, |
| "eval_chemistry_samples_per_second": 9.294, |
| "eval_chemistry_steps_per_second": 4.647, |
| "step": 3900 |
| }, |
| { |
| "entropy": 2.427941220998764, |
| "epoch": 3.4789406331719803, |
| "grad_norm": 0.01123046875, |
| "learning_rate": 1.3535555555555557e-05, |
| "loss": 3.3789, |
| "mean_token_accuracy": 0.4396441696211696, |
| "num_tokens": 51964206.0, |
| "step": 3910 |
| }, |
| { |
| "entropy": 2.495053059607744, |
| "epoch": 3.4878428754242474, |
| "grad_norm": 0.00946044921875, |
| "learning_rate": 1.3513333333333333e-05, |
| "loss": 3.3946, |
| "mean_token_accuracy": 0.43167120348662136, |
| "num_tokens": 52100179.0, |
| "step": 3920 |
| }, |
| { |
| "entropy": 2.4900404512882233, |
| "epoch": 3.4967451176765145, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 1.3491111111111113e-05, |
| "loss": 3.4278, |
| "mean_token_accuracy": 0.4292658809572458, |
| "num_tokens": 52228798.0, |
| "step": 3930 |
| }, |
| { |
| "entropy": 2.5086374834179876, |
| "epoch": 3.505647359928782, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.3468888888888888e-05, |
| "loss": 3.3802, |
| "mean_token_accuracy": 0.43472402803599836, |
| "num_tokens": 52368055.0, |
| "step": 3940 |
| }, |
| { |
| "entropy": 2.4841456547379495, |
| "epoch": 3.5145496021810496, |
| "grad_norm": 0.007781982421875, |
| "learning_rate": 1.3446666666666668e-05, |
| "loss": 3.4326, |
| "mean_token_accuracy": 0.4296139331534505, |
| "num_tokens": 52504822.0, |
| "step": 3950 |
| }, |
| { |
| "entropy": 2.4040485605597497, |
| "epoch": 3.5234518444333167, |
| "grad_norm": 0.00714111328125, |
| "learning_rate": 1.3424444444444447e-05, |
| "loss": 3.3708, |
| "mean_token_accuracy": 0.43940948490053416, |
| "num_tokens": 52637505.0, |
| "step": 3960 |
| }, |
| { |
| "entropy": 2.4827254422008993, |
| "epoch": 3.532354086685584, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.3402222222222223e-05, |
| "loss": 3.3893, |
| "mean_token_accuracy": 0.43419276354834435, |
| "num_tokens": 52769075.0, |
| "step": 3970 |
| }, |
| { |
| "entropy": 2.501670014113188, |
| "epoch": 3.5412563289378514, |
| "grad_norm": 0.01239013671875, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 3.3996, |
| "mean_token_accuracy": 0.4322089777328074, |
| "num_tokens": 52906194.0, |
| "step": 3980 |
| }, |
| { |
| "entropy": 2.409384577721357, |
| "epoch": 3.5501585711901185, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 1.3357777777777778e-05, |
| "loss": 3.2791, |
| "mean_token_accuracy": 0.4397667655721307, |
| "num_tokens": 53037673.0, |
| "step": 3990 |
| }, |
| { |
| "entropy": 2.4114706970751287, |
| "epoch": 3.559060813442386, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 1.3335555555555558e-05, |
| "loss": 3.3609, |
| "mean_token_accuracy": 0.43898537997156384, |
| "num_tokens": 53172376.0, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.559060813442386, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 53172376.0, |
| "eval_coding_runtime": 96.1127, |
| "eval_coding_samples_per_second": 5.202, |
| "eval_coding_steps_per_second": 2.601, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.559060813442386, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 53172376.0, |
| "eval_chemistry_runtime": 53.5193, |
| "eval_chemistry_samples_per_second": 9.342, |
| "eval_chemistry_steps_per_second": 4.671, |
| "step": 4000 |
| }, |
| { |
| "entropy": 2.4349915616214277, |
| "epoch": 3.567963055694653, |
| "grad_norm": 0.00860595703125, |
| "learning_rate": 1.3313333333333333e-05, |
| "loss": 3.3335, |
| "mean_token_accuracy": 0.43587681353092195, |
| "num_tokens": 53304873.0, |
| "step": 4010 |
| }, |
| { |
| "entropy": 2.4979785762727262, |
| "epoch": 3.57686529794692, |
| "grad_norm": 0.0107421875, |
| "learning_rate": 1.3291111111111113e-05, |
| "loss": 3.412, |
| "mean_token_accuracy": 0.4285335911437869, |
| "num_tokens": 53443036.0, |
| "step": 4020 |
| }, |
| { |
| "entropy": 2.443399702757597, |
| "epoch": 3.5857675401991878, |
| "grad_norm": 0.00860595703125, |
| "learning_rate": 1.3268888888888889e-05, |
| "loss": 3.3346, |
| "mean_token_accuracy": 0.4366750448010862, |
| "num_tokens": 53575313.0, |
| "step": 4030 |
| }, |
| { |
| "entropy": 2.533459421247244, |
| "epoch": 3.594669782451455, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.3246666666666668e-05, |
| "loss": 3.4473, |
| "mean_token_accuracy": 0.4297458892688155, |
| "num_tokens": 53706113.0, |
| "step": 4040 |
| }, |
| { |
| "entropy": 2.5755886659026146, |
| "epoch": 3.6035720247037224, |
| "grad_norm": 0.01141357421875, |
| "learning_rate": 1.3224444444444446e-05, |
| "loss": 3.5037, |
| "mean_token_accuracy": 0.4269995864480734, |
| "num_tokens": 53837721.0, |
| "step": 4050 |
| }, |
| { |
| "entropy": 2.4343385607004167, |
| "epoch": 3.6124742669559895, |
| "grad_norm": 0.007049560546875, |
| "learning_rate": 1.3202222222222223e-05, |
| "loss": 3.3276, |
| "mean_token_accuracy": 0.44015026167035104, |
| "num_tokens": 53964233.0, |
| "step": 4060 |
| }, |
| { |
| "entropy": 2.499518417567015, |
| "epoch": 3.6213765092082566, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 3.3613, |
| "mean_token_accuracy": 0.4333439216017723, |
| "num_tokens": 54096340.0, |
| "step": 4070 |
| }, |
| { |
| "entropy": 2.472881755232811, |
| "epoch": 3.630278751460524, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.3157777777777778e-05, |
| "loss": 3.3714, |
| "mean_token_accuracy": 0.43354098405689, |
| "num_tokens": 54234732.0, |
| "step": 4080 |
| }, |
| { |
| "entropy": 2.415682143718004, |
| "epoch": 3.6391809937127912, |
| "grad_norm": 0.009521484375, |
| "learning_rate": 1.3135555555555558e-05, |
| "loss": 3.3391, |
| "mean_token_accuracy": 0.4409346355125308, |
| "num_tokens": 54371392.0, |
| "step": 4090 |
| }, |
| { |
| "entropy": 2.468610843271017, |
| "epoch": 3.648083235965059, |
| "grad_norm": 0.0125732421875, |
| "learning_rate": 1.3113333333333334e-05, |
| "loss": 3.4369, |
| "mean_token_accuracy": 0.4306518994271755, |
| "num_tokens": 54501459.0, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.648083235965059, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 54501459.0, |
| "eval_coding_runtime": 95.8619, |
| "eval_coding_samples_per_second": 5.216, |
| "eval_coding_steps_per_second": 2.608, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.648083235965059, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 54501459.0, |
| "eval_chemistry_runtime": 53.6404, |
| "eval_chemistry_samples_per_second": 9.321, |
| "eval_chemistry_steps_per_second": 4.661, |
| "step": 4100 |
| }, |
| { |
| "entropy": 2.4158985540270805, |
| "epoch": 3.656985478217326, |
| "grad_norm": 0.0135498046875, |
| "learning_rate": 1.3091111111111113e-05, |
| "loss": 3.3393, |
| "mean_token_accuracy": 0.43676703665405514, |
| "num_tokens": 54633856.0, |
| "step": 4110 |
| }, |
| { |
| "entropy": 2.488046611845493, |
| "epoch": 3.6658877204695934, |
| "grad_norm": 0.007720947265625, |
| "learning_rate": 1.306888888888889e-05, |
| "loss": 3.4112, |
| "mean_token_accuracy": 0.4344742437824607, |
| "num_tokens": 54767963.0, |
| "step": 4120 |
| }, |
| { |
| "entropy": 2.5141491636633875, |
| "epoch": 3.6747899627218605, |
| "grad_norm": 0.01025390625, |
| "learning_rate": 1.3046666666666668e-05, |
| "loss": 3.4335, |
| "mean_token_accuracy": 0.427042343840003, |
| "num_tokens": 54900905.0, |
| "step": 4130 |
| }, |
| { |
| "entropy": 2.5171223916113377, |
| "epoch": 3.683692204974128, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.3024444444444446e-05, |
| "loss": 3.4559, |
| "mean_token_accuracy": 0.429931721277535, |
| "num_tokens": 55030717.0, |
| "step": 4140 |
| }, |
| { |
| "entropy": 2.5547206729650496, |
| "epoch": 3.692594447226395, |
| "grad_norm": 0.00860595703125, |
| "learning_rate": 1.3002222222222223e-05, |
| "loss": 3.46, |
| "mean_token_accuracy": 0.42631878620013597, |
| "num_tokens": 55166713.0, |
| "step": 4150 |
| }, |
| { |
| "entropy": 2.4785854801535607, |
| "epoch": 3.7014966894786623, |
| "grad_norm": 0.01470947265625, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 3.418, |
| "mean_token_accuracy": 0.4313357323408127, |
| "num_tokens": 55305577.0, |
| "step": 4160 |
| }, |
| { |
| "entropy": 2.4793594792485236, |
| "epoch": 3.71039893173093, |
| "grad_norm": 0.010986328125, |
| "learning_rate": 1.2957777777777779e-05, |
| "loss": 3.4206, |
| "mean_token_accuracy": 0.42865511756390334, |
| "num_tokens": 55438696.0, |
| "step": 4170 |
| }, |
| { |
| "entropy": 2.5034429252147676, |
| "epoch": 3.719301173983197, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.2935555555555556e-05, |
| "loss": 3.4431, |
| "mean_token_accuracy": 0.4283999059349298, |
| "num_tokens": 55571633.0, |
| "step": 4180 |
| }, |
| { |
| "entropy": 2.4811120800673963, |
| "epoch": 3.7282034162354645, |
| "grad_norm": 0.0091552734375, |
| "learning_rate": 1.2913333333333336e-05, |
| "loss": 3.4019, |
| "mean_token_accuracy": 0.43121442943811417, |
| "num_tokens": 55705219.0, |
| "step": 4190 |
| }, |
| { |
| "entropy": 2.481932933628559, |
| "epoch": 3.7371056584877316, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.2891111111111112e-05, |
| "loss": 3.3719, |
| "mean_token_accuracy": 0.4327344296500087, |
| "num_tokens": 55838629.0, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.7371056584877316, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 55838629.0, |
| "eval_coding_runtime": 96.0466, |
| "eval_coding_samples_per_second": 5.206, |
| "eval_coding_steps_per_second": 2.603, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.7371056584877316, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 55838629.0, |
| "eval_chemistry_runtime": 53.7837, |
| "eval_chemistry_samples_per_second": 9.297, |
| "eval_chemistry_steps_per_second": 4.648, |
| "step": 4200 |
| }, |
| { |
| "entropy": 2.4805993296206, |
| "epoch": 3.7460079007399987, |
| "grad_norm": 0.00726318359375, |
| "learning_rate": 1.2868888888888891e-05, |
| "loss": 3.4077, |
| "mean_token_accuracy": 0.4316241913475096, |
| "num_tokens": 55967304.0, |
| "step": 4210 |
| }, |
| { |
| "entropy": 2.4968034610152245, |
| "epoch": 3.754910142992266, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.2846666666666667e-05, |
| "loss": 3.4359, |
| "mean_token_accuracy": 0.42965076919645073, |
| "num_tokens": 56095994.0, |
| "step": 4220 |
| }, |
| { |
| "entropy": 2.4632013380527495, |
| "epoch": 3.7638123852445333, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 1.2824444444444446e-05, |
| "loss": 3.3984, |
| "mean_token_accuracy": 0.4318878747522831, |
| "num_tokens": 56228180.0, |
| "step": 4230 |
| }, |
| { |
| "entropy": 2.5331189803779126, |
| "epoch": 3.772714627496801, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 1.2802222222222222e-05, |
| "loss": 3.4526, |
| "mean_token_accuracy": 0.42790951747447253, |
| "num_tokens": 56353653.0, |
| "step": 4240 |
| }, |
| { |
| "entropy": 2.434956595301628, |
| "epoch": 3.781616869749068, |
| "grad_norm": 0.01190185546875, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 3.3329, |
| "mean_token_accuracy": 0.4386392910033464, |
| "num_tokens": 56483264.0, |
| "step": 4250 |
| }, |
| { |
| "entropy": 2.492247222363949, |
| "epoch": 3.790519112001335, |
| "grad_norm": 0.011962890625, |
| "learning_rate": 1.2757777777777777e-05, |
| "loss": 3.4726, |
| "mean_token_accuracy": 0.42826817985624077, |
| "num_tokens": 56613510.0, |
| "step": 4260 |
| }, |
| { |
| "entropy": 2.492461886256933, |
| "epoch": 3.7994213542536026, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.2735555555555557e-05, |
| "loss": 3.3993, |
| "mean_token_accuracy": 0.43280363716185094, |
| "num_tokens": 56737398.0, |
| "step": 4270 |
| }, |
| { |
| "entropy": 2.5369254499673843, |
| "epoch": 3.80832359650587, |
| "grad_norm": 0.01239013671875, |
| "learning_rate": 1.2713333333333336e-05, |
| "loss": 3.4826, |
| "mean_token_accuracy": 0.4216884272173047, |
| "num_tokens": 56867014.0, |
| "step": 4280 |
| }, |
| { |
| "entropy": 2.56393309161067, |
| "epoch": 3.8172258387581373, |
| "grad_norm": 0.0184326171875, |
| "learning_rate": 1.2691111111111112e-05, |
| "loss": 3.4847, |
| "mean_token_accuracy": 0.42533156666904687, |
| "num_tokens": 57002779.0, |
| "step": 4290 |
| }, |
| { |
| "entropy": 2.5075097613036634, |
| "epoch": 3.8261280810104044, |
| "grad_norm": 0.007537841796875, |
| "learning_rate": 1.2668888888888891e-05, |
| "loss": 3.4056, |
| "mean_token_accuracy": 0.4334170328453183, |
| "num_tokens": 57140038.0, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.8261280810104044, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 57140038.0, |
| "eval_coding_runtime": 96.1699, |
| "eval_coding_samples_per_second": 5.199, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.8261280810104044, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 57140038.0, |
| "eval_chemistry_runtime": 54.0103, |
| "eval_chemistry_samples_per_second": 9.257, |
| "eval_chemistry_steps_per_second": 4.629, |
| "step": 4300 |
| }, |
| { |
| "entropy": 2.4285498566925527, |
| "epoch": 3.835030323262672, |
| "grad_norm": 0.00726318359375, |
| "learning_rate": 1.2646666666666667e-05, |
| "loss": 3.3478, |
| "mean_token_accuracy": 0.439479798078537, |
| "num_tokens": 57270175.0, |
| "step": 4310 |
| }, |
| { |
| "entropy": 2.5225548088550567, |
| "epoch": 3.843932565514939, |
| "grad_norm": 0.01177978515625, |
| "learning_rate": 1.2624444444444446e-05, |
| "loss": 3.437, |
| "mean_token_accuracy": 0.42834057565778494, |
| "num_tokens": 57403931.0, |
| "step": 4320 |
| }, |
| { |
| "entropy": 2.5161063991487027, |
| "epoch": 3.8528348077672065, |
| "grad_norm": 0.0093994140625, |
| "learning_rate": 1.2602222222222222e-05, |
| "loss": 3.4614, |
| "mean_token_accuracy": 0.4289934171363711, |
| "num_tokens": 57536163.0, |
| "step": 4330 |
| }, |
| { |
| "entropy": 2.572610738128424, |
| "epoch": 3.8617370500194736, |
| "grad_norm": 0.007476806640625, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 3.5261, |
| "mean_token_accuracy": 0.4227509105578065, |
| "num_tokens": 57663118.0, |
| "step": 4340 |
| }, |
| { |
| "entropy": 2.514560434222221, |
| "epoch": 3.8706392922717408, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.2557777777777779e-05, |
| "loss": 3.4228, |
| "mean_token_accuracy": 0.42896437644958496, |
| "num_tokens": 57795299.0, |
| "step": 4350 |
| }, |
| { |
| "entropy": 2.5517844907939433, |
| "epoch": 3.8795415345240083, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.2535555555555557e-05, |
| "loss": 3.4515, |
| "mean_token_accuracy": 0.4274668062105775, |
| "num_tokens": 57930933.0, |
| "step": 4360 |
| }, |
| { |
| "entropy": 2.5404103457927705, |
| "epoch": 3.8884437767762754, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 1.2513333333333334e-05, |
| "loss": 3.4551, |
| "mean_token_accuracy": 0.4282092021778226, |
| "num_tokens": 58065117.0, |
| "step": 4370 |
| }, |
| { |
| "entropy": 2.5318206906318665, |
| "epoch": 3.897346019028543, |
| "grad_norm": 0.01043701171875, |
| "learning_rate": 1.2491111111111112e-05, |
| "loss": 3.4849, |
| "mean_token_accuracy": 0.42346208794042467, |
| "num_tokens": 58204350.0, |
| "step": 4380 |
| }, |
| { |
| "entropy": 2.4756961673498155, |
| "epoch": 3.90624826128081, |
| "grad_norm": 0.0093994140625, |
| "learning_rate": 1.246888888888889e-05, |
| "loss": 3.3981, |
| "mean_token_accuracy": 0.42993527110666035, |
| "num_tokens": 58333935.0, |
| "step": 4390 |
| }, |
| { |
| "entropy": 2.535651671886444, |
| "epoch": 3.915150503533077, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 1.2446666666666667e-05, |
| "loss": 3.5174, |
| "mean_token_accuracy": 0.42215079814195633, |
| "num_tokens": 58455198.0, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.915150503533077, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 58455198.0, |
| "eval_coding_runtime": 96.1114, |
| "eval_coding_samples_per_second": 5.202, |
| "eval_coding_steps_per_second": 2.601, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.915150503533077, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 58455198.0, |
| "eval_chemistry_runtime": 53.8417, |
| "eval_chemistry_samples_per_second": 9.286, |
| "eval_chemistry_steps_per_second": 4.643, |
| "step": 4400 |
| }, |
| { |
| "entropy": 2.553132354468107, |
| "epoch": 3.9240527457853447, |
| "grad_norm": 0.0064697265625, |
| "learning_rate": 1.2424444444444445e-05, |
| "loss": 3.4833, |
| "mean_token_accuracy": 0.4280298611149192, |
| "num_tokens": 58587605.0, |
| "step": 4410 |
| }, |
| { |
| "entropy": 2.50293924510479, |
| "epoch": 3.9329549880376122, |
| "grad_norm": 0.0107421875, |
| "learning_rate": 1.2402222222222222e-05, |
| "loss": 3.4184, |
| "mean_token_accuracy": 0.4321668451651931, |
| "num_tokens": 58728723.0, |
| "step": 4420 |
| }, |
| { |
| "entropy": 2.454167349636555, |
| "epoch": 3.9418572302898793, |
| "grad_norm": 0.009521484375, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 3.274, |
| "mean_token_accuracy": 0.43636299669742584, |
| "num_tokens": 58866751.0, |
| "step": 4430 |
| }, |
| { |
| "entropy": 2.5441963128745555, |
| "epoch": 3.9507594725421464, |
| "grad_norm": 0.0098876953125, |
| "learning_rate": 1.235777777777778e-05, |
| "loss": 3.4859, |
| "mean_token_accuracy": 0.42529882956296206, |
| "num_tokens": 58995707.0, |
| "step": 4440 |
| }, |
| { |
| "entropy": 2.4839250311255454, |
| "epoch": 3.959661714794414, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 1.2335555555555557e-05, |
| "loss": 3.4174, |
| "mean_token_accuracy": 0.43717300305143, |
| "num_tokens": 59134278.0, |
| "step": 4450 |
| }, |
| { |
| "entropy": 2.6635706923902034, |
| "epoch": 3.968563957046681, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.2313333333333335e-05, |
| "loss": 3.545, |
| "mean_token_accuracy": 0.4157288433983922, |
| "num_tokens": 59266644.0, |
| "step": 4460 |
| }, |
| { |
| "entropy": 2.4297814324498175, |
| "epoch": 3.9774661992989486, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.2291111111111112e-05, |
| "loss": 3.3672, |
| "mean_token_accuracy": 0.43812910076230765, |
| "num_tokens": 59397546.0, |
| "step": 4470 |
| }, |
| { |
| "entropy": 2.5034111022949217, |
| "epoch": 3.9863684415512157, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 1.226888888888889e-05, |
| "loss": 3.4156, |
| "mean_token_accuracy": 0.42928003706038, |
| "num_tokens": 59527189.0, |
| "step": 4480 |
| }, |
| { |
| "entropy": 2.465388122946024, |
| "epoch": 3.995270683803483, |
| "grad_norm": 0.0164794921875, |
| "learning_rate": 1.2246666666666667e-05, |
| "loss": 3.3632, |
| "mean_token_accuracy": 0.4356782435439527, |
| "num_tokens": 59660631.0, |
| "step": 4490 |
| }, |
| { |
| "entropy": 2.4827641264704248, |
| "epoch": 4.003560896900907, |
| "grad_norm": 0.0078125, |
| "learning_rate": 1.2224444444444445e-05, |
| "loss": 3.3256, |
| "mean_token_accuracy": 0.4338601073562699, |
| "num_tokens": 59787590.0, |
| "step": 4500 |
| }, |
| { |
| "epoch": 4.003560896900907, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 59787590.0, |
| "eval_coding_runtime": 96.1551, |
| "eval_coding_samples_per_second": 5.2, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 4500 |
| }, |
| { |
| "epoch": 4.003560896900907, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 59787590.0, |
| "eval_chemistry_runtime": 53.8373, |
| "eval_chemistry_samples_per_second": 9.287, |
| "eval_chemistry_steps_per_second": 4.644, |
| "step": 4500 |
| }, |
| { |
| "entropy": 2.4753748916089533, |
| "epoch": 4.012463139153174, |
| "grad_norm": 0.01165771484375, |
| "learning_rate": 1.2202222222222224e-05, |
| "loss": 3.3593, |
| "mean_token_accuracy": 0.4394760325551033, |
| "num_tokens": 59924553.0, |
| "step": 4510 |
| }, |
| { |
| "entropy": 2.456591346859932, |
| "epoch": 4.021365381405442, |
| "grad_norm": 0.01165771484375, |
| "learning_rate": 1.218e-05, |
| "loss": 3.2794, |
| "mean_token_accuracy": 0.439447841886431, |
| "num_tokens": 60060387.0, |
| "step": 4520 |
| }, |
| { |
| "entropy": 2.515965834259987, |
| "epoch": 4.030267623657709, |
| "grad_norm": 0.0111083984375, |
| "learning_rate": 1.215777777777778e-05, |
| "loss": 3.4906, |
| "mean_token_accuracy": 0.42838661205023526, |
| "num_tokens": 60192061.0, |
| "step": 4530 |
| }, |
| { |
| "entropy": 2.4712025128304957, |
| "epoch": 4.039169865909976, |
| "grad_norm": 0.013427734375, |
| "learning_rate": 1.2135555555555556e-05, |
| "loss": 3.3715, |
| "mean_token_accuracy": 0.43209693282842637, |
| "num_tokens": 60323476.0, |
| "step": 4540 |
| }, |
| { |
| "entropy": 2.4202633328735828, |
| "epoch": 4.048072108162243, |
| "grad_norm": 0.00946044921875, |
| "learning_rate": 1.2113333333333335e-05, |
| "loss": 3.3467, |
| "mean_token_accuracy": 0.4408137839287519, |
| "num_tokens": 60458450.0, |
| "step": 4550 |
| }, |
| { |
| "entropy": 2.439309500157833, |
| "epoch": 4.056974350414511, |
| "grad_norm": 0.00830078125, |
| "learning_rate": 1.209111111111111e-05, |
| "loss": 3.3439, |
| "mean_token_accuracy": 0.43894658228382466, |
| "num_tokens": 60596504.0, |
| "step": 4560 |
| }, |
| { |
| "entropy": 2.465762722492218, |
| "epoch": 4.065876592666778, |
| "grad_norm": 0.0205078125, |
| "learning_rate": 1.206888888888889e-05, |
| "loss": 3.4424, |
| "mean_token_accuracy": 0.4322014456614852, |
| "num_tokens": 60726443.0, |
| "step": 4570 |
| }, |
| { |
| "entropy": 2.45337098762393, |
| "epoch": 4.074778834919045, |
| "grad_norm": 0.0103759765625, |
| "learning_rate": 1.204666666666667e-05, |
| "loss": 3.3805, |
| "mean_token_accuracy": 0.43537741936743257, |
| "num_tokens": 60865204.0, |
| "step": 4580 |
| }, |
| { |
| "entropy": 2.535358273237944, |
| "epoch": 4.083681077171312, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 1.2024444444444445e-05, |
| "loss": 3.4461, |
| "mean_token_accuracy": 0.42631899528205397, |
| "num_tokens": 60999002.0, |
| "step": 4590 |
| }, |
| { |
| "entropy": 2.464350725710392, |
| "epoch": 4.0925833194235794, |
| "grad_norm": 0.01025390625, |
| "learning_rate": 1.2002222222222225e-05, |
| "loss": 3.3813, |
| "mean_token_accuracy": 0.4342010916210711, |
| "num_tokens": 61136524.0, |
| "step": 4600 |
| }, |
| { |
| "epoch": 4.0925833194235794, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 61136524.0, |
| "eval_coding_runtime": 95.8044, |
| "eval_coding_samples_per_second": 5.219, |
| "eval_coding_steps_per_second": 2.609, |
| "step": 4600 |
| }, |
| { |
| "epoch": 4.0925833194235794, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 61136524.0, |
| "eval_chemistry_runtime": 53.908, |
| "eval_chemistry_samples_per_second": 9.275, |
| "eval_chemistry_steps_per_second": 4.638, |
| "step": 4600 |
| }, |
| { |
| "entropy": 2.528155630081892, |
| "epoch": 4.101485561675847, |
| "grad_norm": 0.01513671875, |
| "learning_rate": 1.198e-05, |
| "loss": 3.4777, |
| "mean_token_accuracy": 0.42658356465399266, |
| "num_tokens": 61268769.0, |
| "step": 4610 |
| }, |
| { |
| "entropy": 2.5526251435279845, |
| "epoch": 4.1103878039281145, |
| "grad_norm": 0.009765625, |
| "learning_rate": 1.195777777777778e-05, |
| "loss": 3.477, |
| "mean_token_accuracy": 0.42346252147108315, |
| "num_tokens": 61400311.0, |
| "step": 4620 |
| }, |
| { |
| "entropy": 2.517058031260967, |
| "epoch": 4.119290046180382, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.1935555555555556e-05, |
| "loss": 3.4104, |
| "mean_token_accuracy": 0.4348121965304017, |
| "num_tokens": 61535353.0, |
| "step": 4630 |
| }, |
| { |
| "entropy": 2.478480698913336, |
| "epoch": 4.128192288432649, |
| "grad_norm": 0.011962890625, |
| "learning_rate": 1.1913333333333335e-05, |
| "loss": 3.4077, |
| "mean_token_accuracy": 0.42657945957034826, |
| "num_tokens": 61671052.0, |
| "step": 4640 |
| }, |
| { |
| "entropy": 2.497458207607269, |
| "epoch": 4.137094530684916, |
| "grad_norm": 0.0123291015625, |
| "learning_rate": 1.1891111111111111e-05, |
| "loss": 3.4027, |
| "mean_token_accuracy": 0.4296461737714708, |
| "num_tokens": 61805997.0, |
| "step": 4650 |
| }, |
| { |
| "entropy": 2.485201074928045, |
| "epoch": 4.145996772937184, |
| "grad_norm": 0.007537841796875, |
| "learning_rate": 1.186888888888889e-05, |
| "loss": 3.4022, |
| "mean_token_accuracy": 0.42965454775840045, |
| "num_tokens": 61938772.0, |
| "step": 4660 |
| }, |
| { |
| "entropy": 2.4816596046090127, |
| "epoch": 4.154899015189451, |
| "grad_norm": 0.01068115234375, |
| "learning_rate": 1.1846666666666668e-05, |
| "loss": 3.4073, |
| "mean_token_accuracy": 0.43170268591493366, |
| "num_tokens": 62069519.0, |
| "step": 4670 |
| }, |
| { |
| "entropy": 2.4487347967922686, |
| "epoch": 4.163801257441718, |
| "grad_norm": 0.0096435546875, |
| "learning_rate": 1.1824444444444445e-05, |
| "loss": 3.3327, |
| "mean_token_accuracy": 0.43628135407343505, |
| "num_tokens": 62212548.0, |
| "step": 4680 |
| }, |
| { |
| "entropy": 2.4895458199083804, |
| "epoch": 4.172703499693985, |
| "grad_norm": 0.0111083984375, |
| "learning_rate": 1.1802222222222223e-05, |
| "loss": 3.4081, |
| "mean_token_accuracy": 0.43173519475385547, |
| "num_tokens": 62343324.0, |
| "step": 4690 |
| }, |
| { |
| "entropy": 2.5754899688065054, |
| "epoch": 4.181605741946253, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 1.178e-05, |
| "loss": 3.5512, |
| "mean_token_accuracy": 0.4174567579291761, |
| "num_tokens": 62469444.0, |
| "step": 4700 |
| }, |
| { |
| "epoch": 4.181605741946253, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 62469444.0, |
| "eval_coding_runtime": 96.0602, |
| "eval_coding_samples_per_second": 5.205, |
| "eval_coding_steps_per_second": 2.603, |
| "step": 4700 |
| }, |
| { |
| "epoch": 4.181605741946253, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 62469444.0, |
| "eval_chemistry_runtime": 53.898, |
| "eval_chemistry_samples_per_second": 9.277, |
| "eval_chemistry_steps_per_second": 4.638, |
| "step": 4700 |
| }, |
| { |
| "entropy": 2.4956134453415872, |
| "epoch": 4.19050798419852, |
| "grad_norm": 0.0174560546875, |
| "learning_rate": 1.1757777777777778e-05, |
| "loss": 3.3521, |
| "mean_token_accuracy": 0.4332751795649529, |
| "num_tokens": 62610289.0, |
| "step": 4710 |
| }, |
| { |
| "entropy": 2.471109450608492, |
| "epoch": 4.199410226450787, |
| "grad_norm": 0.01251220703125, |
| "learning_rate": 1.1735555555555556e-05, |
| "loss": 3.405, |
| "mean_token_accuracy": 0.43353689890354874, |
| "num_tokens": 62746184.0, |
| "step": 4720 |
| }, |
| { |
| "entropy": 2.514505376666784, |
| "epoch": 4.208312468703054, |
| "grad_norm": 0.011962890625, |
| "learning_rate": 1.1713333333333334e-05, |
| "loss": 3.4738, |
| "mean_token_accuracy": 0.4285815415903926, |
| "num_tokens": 62884720.0, |
| "step": 4730 |
| }, |
| { |
| "entropy": 2.5430411636829375, |
| "epoch": 4.2172147109553215, |
| "grad_norm": 0.009521484375, |
| "learning_rate": 1.1691111111111113e-05, |
| "loss": 3.5175, |
| "mean_token_accuracy": 0.4227385140955448, |
| "num_tokens": 63014904.0, |
| "step": 4740 |
| }, |
| { |
| "entropy": 2.421365215629339, |
| "epoch": 4.2261169532075895, |
| "grad_norm": 0.01153564453125, |
| "learning_rate": 1.1668888888888889e-05, |
| "loss": 3.3123, |
| "mean_token_accuracy": 0.44363965913653375, |
| "num_tokens": 63153305.0, |
| "step": 4750 |
| }, |
| { |
| "entropy": 2.4818004824221136, |
| "epoch": 4.235019195459857, |
| "grad_norm": 0.01220703125, |
| "learning_rate": 1.1646666666666668e-05, |
| "loss": 3.4731, |
| "mean_token_accuracy": 0.4281780956313014, |
| "num_tokens": 63286580.0, |
| "step": 4760 |
| }, |
| { |
| "entropy": 2.547441676259041, |
| "epoch": 4.243921437712124, |
| "grad_norm": 0.00897216796875, |
| "learning_rate": 1.1624444444444446e-05, |
| "loss": 3.5136, |
| "mean_token_accuracy": 0.42384677920490504, |
| "num_tokens": 63412248.0, |
| "step": 4770 |
| }, |
| { |
| "entropy": 2.507705695182085, |
| "epoch": 4.252823679964391, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.1602222222222223e-05, |
| "loss": 3.4108, |
| "mean_token_accuracy": 0.42969165947288274, |
| "num_tokens": 63544361.0, |
| "step": 4780 |
| }, |
| { |
| "entropy": 2.4510108195245266, |
| "epoch": 4.261725922216658, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 3.3388, |
| "mean_token_accuracy": 0.4395698314532638, |
| "num_tokens": 63683325.0, |
| "step": 4790 |
| }, |
| { |
| "entropy": 2.446478872746229, |
| "epoch": 4.270628164468926, |
| "grad_norm": 0.0107421875, |
| "learning_rate": 1.1557777777777779e-05, |
| "loss": 3.4045, |
| "mean_token_accuracy": 0.4308119174093008, |
| "num_tokens": 63813469.0, |
| "step": 4800 |
| }, |
| { |
| "epoch": 4.270628164468926, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 63813469.0, |
| "eval_coding_runtime": 96.1615, |
| "eval_coding_samples_per_second": 5.2, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 4800 |
| }, |
| { |
| "epoch": 4.270628164468926, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 63813469.0, |
| "eval_chemistry_runtime": 53.6895, |
| "eval_chemistry_samples_per_second": 9.313, |
| "eval_chemistry_steps_per_second": 4.656, |
| "step": 4800 |
| }, |
| { |
| "entropy": 2.494009657949209, |
| "epoch": 4.279530406721193, |
| "grad_norm": 0.009521484375, |
| "learning_rate": 1.1535555555555556e-05, |
| "loss": 3.436, |
| "mean_token_accuracy": 0.42943568248301744, |
| "num_tokens": 63942245.0, |
| "step": 4810 |
| }, |
| { |
| "entropy": 2.4311224967241287, |
| "epoch": 4.28843264897346, |
| "grad_norm": 0.0103759765625, |
| "learning_rate": 1.1513333333333334e-05, |
| "loss": 3.3351, |
| "mean_token_accuracy": 0.44133828617632387, |
| "num_tokens": 64074240.0, |
| "step": 4820 |
| }, |
| { |
| "entropy": 2.4826130270957947, |
| "epoch": 4.297334891225727, |
| "grad_norm": 0.01275634765625, |
| "learning_rate": 1.1491111111111113e-05, |
| "loss": 3.3785, |
| "mean_token_accuracy": 0.43262541918084024, |
| "num_tokens": 64199776.0, |
| "step": 4830 |
| }, |
| { |
| "entropy": 2.5273991763591765, |
| "epoch": 4.306237133477994, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 1.1468888888888889e-05, |
| "loss": 3.3793, |
| "mean_token_accuracy": 0.4331956706941128, |
| "num_tokens": 64338220.0, |
| "step": 4840 |
| }, |
| { |
| "entropy": 2.4641799479722977, |
| "epoch": 4.315139375730262, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 1.1446666666666668e-05, |
| "loss": 3.3709, |
| "mean_token_accuracy": 0.43348179180175067, |
| "num_tokens": 64474251.0, |
| "step": 4850 |
| }, |
| { |
| "entropy": 2.477324938774109, |
| "epoch": 4.324041617982529, |
| "grad_norm": 0.0133056640625, |
| "learning_rate": 1.1424444444444444e-05, |
| "loss": 3.416, |
| "mean_token_accuracy": 0.4315943540073931, |
| "num_tokens": 64612457.0, |
| "step": 4860 |
| }, |
| { |
| "entropy": 2.4926598593592644, |
| "epoch": 4.3329438602347965, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 1.1402222222222224e-05, |
| "loss": 3.4054, |
| "mean_token_accuracy": 0.43276866618543863, |
| "num_tokens": 64745805.0, |
| "step": 4870 |
| }, |
| { |
| "entropy": 2.396423150599003, |
| "epoch": 4.341846102487064, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.138e-05, |
| "loss": 3.3393, |
| "mean_token_accuracy": 0.43795022219419477, |
| "num_tokens": 64876908.0, |
| "step": 4880 |
| }, |
| { |
| "entropy": 2.4940830834209917, |
| "epoch": 4.350748344739332, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.1357777777777779e-05, |
| "loss": 3.4439, |
| "mean_token_accuracy": 0.42853713911026714, |
| "num_tokens": 65008394.0, |
| "step": 4890 |
| }, |
| { |
| "entropy": 2.446301008015871, |
| "epoch": 4.359650586991599, |
| "grad_norm": 0.01165771484375, |
| "learning_rate": 1.1335555555555558e-05, |
| "loss": 3.3475, |
| "mean_token_accuracy": 0.4348562330007553, |
| "num_tokens": 65147596.0, |
| "step": 4900 |
| }, |
| { |
| "epoch": 4.359650586991599, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 65147596.0, |
| "eval_coding_runtime": 96.1524, |
| "eval_coding_samples_per_second": 5.2, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 4900 |
| }, |
| { |
| "epoch": 4.359650586991599, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 65147596.0, |
| "eval_chemistry_runtime": 53.8825, |
| "eval_chemistry_samples_per_second": 9.279, |
| "eval_chemistry_steps_per_second": 4.64, |
| "step": 4900 |
| }, |
| { |
| "entropy": 2.539576554298401, |
| "epoch": 4.368552829243866, |
| "grad_norm": 0.0101318359375, |
| "learning_rate": 1.1313333333333334e-05, |
| "loss": 3.4722, |
| "mean_token_accuracy": 0.42030207607895137, |
| "num_tokens": 65283664.0, |
| "step": 4910 |
| }, |
| { |
| "entropy": 2.5522966988384725, |
| "epoch": 4.377455071496133, |
| "grad_norm": 0.0098876953125, |
| "learning_rate": 1.1291111111111113e-05, |
| "loss": 3.5002, |
| "mean_token_accuracy": 0.41999158738180997, |
| "num_tokens": 65418480.0, |
| "step": 4920 |
| }, |
| { |
| "entropy": 2.517087984830141, |
| "epoch": 4.3863573137484, |
| "grad_norm": 0.006927490234375, |
| "learning_rate": 1.126888888888889e-05, |
| "loss": 3.4595, |
| "mean_token_accuracy": 0.4256157610565424, |
| "num_tokens": 65543684.0, |
| "step": 4930 |
| }, |
| { |
| "entropy": 2.4599628783762455, |
| "epoch": 4.395259556000668, |
| "grad_norm": 0.007049560546875, |
| "learning_rate": 1.1246666666666669e-05, |
| "loss": 3.3771, |
| "mean_token_accuracy": 0.43713085670024154, |
| "num_tokens": 65680301.0, |
| "step": 4940 |
| }, |
| { |
| "entropy": 2.497390619665384, |
| "epoch": 4.404161798252935, |
| "grad_norm": 0.01080322265625, |
| "learning_rate": 1.1224444444444444e-05, |
| "loss": 3.4455, |
| "mean_token_accuracy": 0.4305057739838958, |
| "num_tokens": 65815093.0, |
| "step": 4950 |
| }, |
| { |
| "entropy": 2.504514559358358, |
| "epoch": 4.413064040505202, |
| "grad_norm": 0.00872802734375, |
| "learning_rate": 1.1202222222222224e-05, |
| "loss": 3.4347, |
| "mean_token_accuracy": 0.4267650000751019, |
| "num_tokens": 65945233.0, |
| "step": 4960 |
| }, |
| { |
| "entropy": 2.484743408858776, |
| "epoch": 4.421966282757469, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 3.4323, |
| "mean_token_accuracy": 0.43043280243873594, |
| "num_tokens": 66079056.0, |
| "step": 4970 |
| }, |
| { |
| "entropy": 2.4898355223238466, |
| "epoch": 4.430868525009737, |
| "grad_norm": 0.012451171875, |
| "learning_rate": 1.1157777777777779e-05, |
| "loss": 3.424, |
| "mean_token_accuracy": 0.43193226046860217, |
| "num_tokens": 66209932.0, |
| "step": 4980 |
| }, |
| { |
| "entropy": 2.5051794923841952, |
| "epoch": 4.439770767262004, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 1.1135555555555557e-05, |
| "loss": 3.3859, |
| "mean_token_accuracy": 0.4320788636803627, |
| "num_tokens": 66337780.0, |
| "step": 4990 |
| }, |
| { |
| "entropy": 2.5868745751678945, |
| "epoch": 4.4486730095142715, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.1113333333333334e-05, |
| "loss": 3.5462, |
| "mean_token_accuracy": 0.41811424903571603, |
| "num_tokens": 66463918.0, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.4486730095142715, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 66463918.0, |
| "eval_coding_runtime": 95.9479, |
| "eval_coding_samples_per_second": 5.211, |
| "eval_coding_steps_per_second": 2.606, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.4486730095142715, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 66463918.0, |
| "eval_chemistry_runtime": 53.9091, |
| "eval_chemistry_samples_per_second": 9.275, |
| "eval_chemistry_steps_per_second": 4.637, |
| "step": 5000 |
| }, |
| { |
| "entropy": 2.5006047405302523, |
| "epoch": 4.457575251766539, |
| "grad_norm": 0.01263427734375, |
| "learning_rate": 1.1091111111111112e-05, |
| "loss": 3.472, |
| "mean_token_accuracy": 0.42307986905798317, |
| "num_tokens": 66596903.0, |
| "step": 5010 |
| }, |
| { |
| "entropy": 2.512437270581722, |
| "epoch": 4.466477494018806, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 1.106888888888889e-05, |
| "loss": 3.4207, |
| "mean_token_accuracy": 0.4291929883882403, |
| "num_tokens": 66729042.0, |
| "step": 5020 |
| }, |
| { |
| "entropy": 2.512176238000393, |
| "epoch": 4.475379736271074, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 1.1046666666666667e-05, |
| "loss": 3.4536, |
| "mean_token_accuracy": 0.42558202855288985, |
| "num_tokens": 66861608.0, |
| "step": 5030 |
| }, |
| { |
| "entropy": 2.5115060031414034, |
| "epoch": 4.484281978523341, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 1.1024444444444445e-05, |
| "loss": 3.4401, |
| "mean_token_accuracy": 0.4291109647601843, |
| "num_tokens": 66997591.0, |
| "step": 5040 |
| }, |
| { |
| "entropy": 2.5043463915586472, |
| "epoch": 4.493184220775608, |
| "grad_norm": 0.0118408203125, |
| "learning_rate": 1.1002222222222222e-05, |
| "loss": 3.4293, |
| "mean_token_accuracy": 0.430062410235405, |
| "num_tokens": 67127383.0, |
| "step": 5050 |
| }, |
| { |
| "entropy": 2.485823146253824, |
| "epoch": 4.502086463027875, |
| "grad_norm": 0.01458740234375, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 3.4319, |
| "mean_token_accuracy": 0.43265841212123635, |
| "num_tokens": 67259835.0, |
| "step": 5060 |
| }, |
| { |
| "entropy": 2.4754792124032976, |
| "epoch": 4.510988705280142, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.0957777777777778e-05, |
| "loss": 3.3701, |
| "mean_token_accuracy": 0.433258808311075, |
| "num_tokens": 67389630.0, |
| "step": 5070 |
| }, |
| { |
| "entropy": 2.526766945421696, |
| "epoch": 4.51989094753241, |
| "grad_norm": 0.007598876953125, |
| "learning_rate": 1.0935555555555557e-05, |
| "loss": 3.4787, |
| "mean_token_accuracy": 0.4227756954729557, |
| "num_tokens": 67521050.0, |
| "step": 5080 |
| }, |
| { |
| "entropy": 2.468719993531704, |
| "epoch": 4.528793189784677, |
| "grad_norm": 0.0155029296875, |
| "learning_rate": 1.0913333333333333e-05, |
| "loss": 3.3837, |
| "mean_token_accuracy": 0.4352194467559457, |
| "num_tokens": 67655872.0, |
| "step": 5090 |
| }, |
| { |
| "entropy": 2.474339720606804, |
| "epoch": 4.537695432036944, |
| "grad_norm": 0.0101318359375, |
| "learning_rate": 1.0891111111111112e-05, |
| "loss": 3.4164, |
| "mean_token_accuracy": 0.43226807564496994, |
| "num_tokens": 67791072.0, |
| "step": 5100 |
| }, |
| { |
| "epoch": 4.537695432036944, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 67791072.0, |
| "eval_coding_runtime": 95.9268, |
| "eval_coding_samples_per_second": 5.212, |
| "eval_coding_steps_per_second": 2.606, |
| "step": 5100 |
| }, |
| { |
| "epoch": 4.537695432036944, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 67791072.0, |
| "eval_chemistry_runtime": 53.832, |
| "eval_chemistry_samples_per_second": 9.288, |
| "eval_chemistry_steps_per_second": 4.644, |
| "step": 5100 |
| }, |
| { |
| "entropy": 2.453265733271837, |
| "epoch": 4.546597674289211, |
| "grad_norm": 0.0140380859375, |
| "learning_rate": 1.0868888888888888e-05, |
| "loss": 3.3566, |
| "mean_token_accuracy": 0.4371098093688488, |
| "num_tokens": 67930560.0, |
| "step": 5110 |
| }, |
| { |
| "entropy": 2.5135868452489376, |
| "epoch": 4.5554999165414785, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.0846666666666667e-05, |
| "loss": 3.3859, |
| "mean_token_accuracy": 0.4341611025854945, |
| "num_tokens": 68065726.0, |
| "step": 5120 |
| }, |
| { |
| "entropy": 2.535103219747543, |
| "epoch": 4.5644021587937464, |
| "grad_norm": 0.00830078125, |
| "learning_rate": 1.0824444444444447e-05, |
| "loss": 3.431, |
| "mean_token_accuracy": 0.4322080958634615, |
| "num_tokens": 68205422.0, |
| "step": 5130 |
| }, |
| { |
| "entropy": 2.549791398644447, |
| "epoch": 4.5733044010460135, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 1.0802222222222223e-05, |
| "loss": 3.4594, |
| "mean_token_accuracy": 0.42580029759556054, |
| "num_tokens": 68327693.0, |
| "step": 5140 |
| }, |
| { |
| "entropy": 2.4972184780985116, |
| "epoch": 4.582206643298281, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 3.3725, |
| "mean_token_accuracy": 0.4355587847530842, |
| "num_tokens": 68458149.0, |
| "step": 5150 |
| }, |
| { |
| "entropy": 2.5136324256658553, |
| "epoch": 4.591108885550548, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 1.0757777777777778e-05, |
| "loss": 3.4436, |
| "mean_token_accuracy": 0.42923225853592156, |
| "num_tokens": 68587415.0, |
| "step": 5160 |
| }, |
| { |
| "entropy": 2.4325045540928842, |
| "epoch": 4.600011127802816, |
| "grad_norm": 0.01043701171875, |
| "learning_rate": 1.0735555555555557e-05, |
| "loss": 3.343, |
| "mean_token_accuracy": 0.4392604473978281, |
| "num_tokens": 68731081.0, |
| "step": 5170 |
| }, |
| { |
| "entropy": 2.3738025426864624, |
| "epoch": 4.608913370055083, |
| "grad_norm": 0.0123291015625, |
| "learning_rate": 1.0713333333333333e-05, |
| "loss": 3.3045, |
| "mean_token_accuracy": 0.4423565460368991, |
| "num_tokens": 68864837.0, |
| "step": 5180 |
| }, |
| { |
| "entropy": 2.5630047604441644, |
| "epoch": 4.61781561230735, |
| "grad_norm": 0.00982666015625, |
| "learning_rate": 1.0691111111111112e-05, |
| "loss": 3.4978, |
| "mean_token_accuracy": 0.42465658094733955, |
| "num_tokens": 68995140.0, |
| "step": 5190 |
| }, |
| { |
| "entropy": 2.5928178526461125, |
| "epoch": 4.626717854559617, |
| "grad_norm": 0.007537841796875, |
| "learning_rate": 1.0668888888888892e-05, |
| "loss": 3.5253, |
| "mean_token_accuracy": 0.4215786622837186, |
| "num_tokens": 69120367.0, |
| "step": 5200 |
| }, |
| { |
| "epoch": 4.626717854559617, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 69120367.0, |
| "eval_coding_runtime": 96.1427, |
| "eval_coding_samples_per_second": 5.201, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 5200 |
| }, |
| { |
| "epoch": 4.626717854559617, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 69120367.0, |
| "eval_chemistry_runtime": 53.7985, |
| "eval_chemistry_samples_per_second": 9.294, |
| "eval_chemistry_steps_per_second": 4.647, |
| "step": 5200 |
| }, |
| { |
| "entropy": 2.5130502417683602, |
| "epoch": 4.635620096811884, |
| "grad_norm": 0.01104736328125, |
| "learning_rate": 1.0646666666666668e-05, |
| "loss": 3.4487, |
| "mean_token_accuracy": 0.4305033221840858, |
| "num_tokens": 69249033.0, |
| "step": 5210 |
| }, |
| { |
| "entropy": 2.5364614691585303, |
| "epoch": 4.644522339064152, |
| "grad_norm": 0.01214599609375, |
| "learning_rate": 1.0624444444444447e-05, |
| "loss": 3.4751, |
| "mean_token_accuracy": 0.428230413608253, |
| "num_tokens": 69385814.0, |
| "step": 5220 |
| }, |
| { |
| "entropy": 2.608835220336914, |
| "epoch": 4.653424581316419, |
| "grad_norm": 0.007720947265625, |
| "learning_rate": 1.0602222222222223e-05, |
| "loss": 3.5612, |
| "mean_token_accuracy": 0.41845532692968845, |
| "num_tokens": 69512425.0, |
| "step": 5230 |
| }, |
| { |
| "entropy": 2.4466094307601454, |
| "epoch": 4.662326823568686, |
| "grad_norm": 0.00750732421875, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 3.3642, |
| "mean_token_accuracy": 0.4332047704607248, |
| "num_tokens": 69648905.0, |
| "step": 5240 |
| }, |
| { |
| "entropy": 2.4500359155237676, |
| "epoch": 4.671229065820953, |
| "grad_norm": 0.01904296875, |
| "learning_rate": 1.0557777777777778e-05, |
| "loss": 3.3605, |
| "mean_token_accuracy": 0.4391206895932555, |
| "num_tokens": 69782478.0, |
| "step": 5250 |
| }, |
| { |
| "entropy": 2.5554568499326704, |
| "epoch": 4.680131308073221, |
| "grad_norm": 0.0108642578125, |
| "learning_rate": 1.0535555555555557e-05, |
| "loss": 3.4708, |
| "mean_token_accuracy": 0.4264969429001212, |
| "num_tokens": 69910674.0, |
| "step": 5260 |
| }, |
| { |
| "entropy": 2.5495406486094, |
| "epoch": 4.6890335503254885, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 1.0513333333333333e-05, |
| "loss": 3.4846, |
| "mean_token_accuracy": 0.4262933300808072, |
| "num_tokens": 70041943.0, |
| "step": 5270 |
| }, |
| { |
| "entropy": 2.5517571955919265, |
| "epoch": 4.697935792577756, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 1.0491111111111112e-05, |
| "loss": 3.4791, |
| "mean_token_accuracy": 0.4226047061383724, |
| "num_tokens": 70171207.0, |
| "step": 5280 |
| }, |
| { |
| "entropy": 2.4664143405854704, |
| "epoch": 4.706838034830023, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 1.046888888888889e-05, |
| "loss": 3.3864, |
| "mean_token_accuracy": 0.4341844154521823, |
| "num_tokens": 70304062.0, |
| "step": 5290 |
| }, |
| { |
| "entropy": 2.5869156286120414, |
| "epoch": 4.71574027708229, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 1.0446666666666668e-05, |
| "loss": 3.5102, |
| "mean_token_accuracy": 0.4167873265221715, |
| "num_tokens": 70428466.0, |
| "step": 5300 |
| }, |
| { |
| "epoch": 4.71574027708229, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 70428466.0, |
| "eval_coding_runtime": 96.0544, |
| "eval_coding_samples_per_second": 5.205, |
| "eval_coding_steps_per_second": 2.603, |
| "step": 5300 |
| }, |
| { |
| "epoch": 4.71574027708229, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 70428466.0, |
| "eval_chemistry_runtime": 53.8363, |
| "eval_chemistry_samples_per_second": 9.287, |
| "eval_chemistry_steps_per_second": 4.644, |
| "step": 5300 |
| }, |
| { |
| "entropy": 2.501867598295212, |
| "epoch": 4.724642519334557, |
| "grad_norm": 0.00665283203125, |
| "learning_rate": 1.0424444444444445e-05, |
| "loss": 3.3629, |
| "mean_token_accuracy": 0.43458408284932376, |
| "num_tokens": 70562587.0, |
| "step": 5310 |
| }, |
| { |
| "entropy": 2.4744629479944704, |
| "epoch": 4.733544761586825, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.0402222222222223e-05, |
| "loss": 3.3689, |
| "mean_token_accuracy": 0.43399940356612204, |
| "num_tokens": 70695641.0, |
| "step": 5320 |
| }, |
| { |
| "entropy": 2.4937205776572227, |
| "epoch": 4.742447003839092, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 1.038e-05, |
| "loss": 3.4293, |
| "mean_token_accuracy": 0.4300399720668793, |
| "num_tokens": 70827392.0, |
| "step": 5330 |
| }, |
| { |
| "entropy": 2.5173247322440147, |
| "epoch": 4.751349246091359, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.0357777777777778e-05, |
| "loss": 3.4068, |
| "mean_token_accuracy": 0.4295641141012311, |
| "num_tokens": 70960848.0, |
| "step": 5340 |
| }, |
| { |
| "entropy": 2.5341571643948555, |
| "epoch": 4.760251488343626, |
| "grad_norm": 0.0096435546875, |
| "learning_rate": 1.0335555555555556e-05, |
| "loss": 3.4986, |
| "mean_token_accuracy": 0.42371637765318154, |
| "num_tokens": 71093331.0, |
| "step": 5350 |
| }, |
| { |
| "entropy": 2.513373174518347, |
| "epoch": 4.769153730595894, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 1.0313333333333335e-05, |
| "loss": 3.4572, |
| "mean_token_accuracy": 0.4241326037794352, |
| "num_tokens": 71226088.0, |
| "step": 5360 |
| }, |
| { |
| "entropy": 2.5231514535844326, |
| "epoch": 4.778055972848161, |
| "grad_norm": 0.0174560546875, |
| "learning_rate": 1.0291111111111111e-05, |
| "loss": 3.4606, |
| "mean_token_accuracy": 0.4277513949200511, |
| "num_tokens": 71357161.0, |
| "step": 5370 |
| }, |
| { |
| "entropy": 2.44051893055439, |
| "epoch": 4.786958215100428, |
| "grad_norm": 0.0164794921875, |
| "learning_rate": 1.026888888888889e-05, |
| "loss": 3.3843, |
| "mean_token_accuracy": 0.4341155607253313, |
| "num_tokens": 71487103.0, |
| "step": 5380 |
| }, |
| { |
| "entropy": 2.4529117070138455, |
| "epoch": 4.7958604573526955, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.0246666666666666e-05, |
| "loss": 3.3863, |
| "mean_token_accuracy": 0.4364540001377463, |
| "num_tokens": 71623116.0, |
| "step": 5390 |
| }, |
| { |
| "entropy": 2.50192252472043, |
| "epoch": 4.804762699604963, |
| "grad_norm": 0.00830078125, |
| "learning_rate": 1.0224444444444446e-05, |
| "loss": 3.4233, |
| "mean_token_accuracy": 0.4286077202297747, |
| "num_tokens": 71755520.0, |
| "step": 5400 |
| }, |
| { |
| "epoch": 4.804762699604963, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 71755520.0, |
| "eval_coding_runtime": 96.0895, |
| "eval_coding_samples_per_second": 5.203, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 5400 |
| }, |
| { |
| "epoch": 4.804762699604963, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 71755520.0, |
| "eval_chemistry_runtime": 54.004, |
| "eval_chemistry_samples_per_second": 9.259, |
| "eval_chemistry_steps_per_second": 4.629, |
| "step": 5400 |
| }, |
| { |
| "entropy": 2.427228733897209, |
| "epoch": 4.813664941857231, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 1.0202222222222221e-05, |
| "loss": 3.3441, |
| "mean_token_accuracy": 0.43763389717787504, |
| "num_tokens": 71887313.0, |
| "step": 5410 |
| }, |
| { |
| "entropy": 2.4905957020819187, |
| "epoch": 4.822567184109498, |
| "grad_norm": 0.0072021484375, |
| "learning_rate": 1.018e-05, |
| "loss": 3.4096, |
| "mean_token_accuracy": 0.4305798047222197, |
| "num_tokens": 72021619.0, |
| "step": 5420 |
| }, |
| { |
| "entropy": 2.555949030816555, |
| "epoch": 4.831469426361765, |
| "grad_norm": 0.00982666015625, |
| "learning_rate": 1.0157777777777777e-05, |
| "loss": 3.4573, |
| "mean_token_accuracy": 0.42647946225479244, |
| "num_tokens": 72152969.0, |
| "step": 5430 |
| }, |
| { |
| "entropy": 2.481585122272372, |
| "epoch": 4.840371668614032, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 1.0135555555555556e-05, |
| "loss": 3.3853, |
| "mean_token_accuracy": 0.4319996147416532, |
| "num_tokens": 72283800.0, |
| "step": 5440 |
| }, |
| { |
| "entropy": 2.4411033555865287, |
| "epoch": 4.8492739108663, |
| "grad_norm": 0.013671875, |
| "learning_rate": 1.0113333333333335e-05, |
| "loss": 3.3347, |
| "mean_token_accuracy": 0.43567943051457403, |
| "num_tokens": 72416362.0, |
| "step": 5450 |
| }, |
| { |
| "entropy": 2.4901481226086615, |
| "epoch": 4.858176153118567, |
| "grad_norm": 0.008544921875, |
| "learning_rate": 1.0091111111111111e-05, |
| "loss": 3.3892, |
| "mean_token_accuracy": 0.4326670720241964, |
| "num_tokens": 72555896.0, |
| "step": 5460 |
| }, |
| { |
| "entropy": 2.47064997702837, |
| "epoch": 4.867078395370834, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 1.006888888888889e-05, |
| "loss": 3.4166, |
| "mean_token_accuracy": 0.43486586278304457, |
| "num_tokens": 72688351.0, |
| "step": 5470 |
| }, |
| { |
| "entropy": 2.453525426983833, |
| "epoch": 4.875980637623101, |
| "grad_norm": 0.01611328125, |
| "learning_rate": 1.0046666666666666e-05, |
| "loss": 3.2979, |
| "mean_token_accuracy": 0.44417268354445694, |
| "num_tokens": 72831506.0, |
| "step": 5480 |
| }, |
| { |
| "entropy": 2.516221982985735, |
| "epoch": 4.884882879875368, |
| "grad_norm": 0.00872802734375, |
| "learning_rate": 1.0024444444444446e-05, |
| "loss": 3.3998, |
| "mean_token_accuracy": 0.42740293592214584, |
| "num_tokens": 72964432.0, |
| "step": 5490 |
| }, |
| { |
| "entropy": 2.478584685921669, |
| "epoch": 4.893785122127636, |
| "grad_norm": 0.007415771484375, |
| "learning_rate": 1.0002222222222222e-05, |
| "loss": 3.3975, |
| "mean_token_accuracy": 0.4316392602398992, |
| "num_tokens": 73096691.0, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.893785122127636, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 73096691.0, |
| "eval_coding_runtime": 95.8589, |
| "eval_coding_samples_per_second": 5.216, |
| "eval_coding_steps_per_second": 2.608, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.893785122127636, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 73096691.0, |
| "eval_chemistry_runtime": 53.6612, |
| "eval_chemistry_samples_per_second": 9.318, |
| "eval_chemistry_steps_per_second": 4.659, |
| "step": 5500 |
| }, |
| { |
| "entropy": 2.5298805542290213, |
| "epoch": 4.902687364379903, |
| "grad_norm": 0.01080322265625, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 3.4774, |
| "mean_token_accuracy": 0.42651538643985987, |
| "num_tokens": 73228254.0, |
| "step": 5510 |
| }, |
| { |
| "entropy": 2.382762984186411, |
| "epoch": 4.9115896066321705, |
| "grad_norm": 0.0164794921875, |
| "learning_rate": 9.957777777777779e-06, |
| "loss": 3.264, |
| "mean_token_accuracy": 0.44377621114253996, |
| "num_tokens": 73362943.0, |
| "step": 5520 |
| }, |
| { |
| "entropy": 2.472093004733324, |
| "epoch": 4.920491848884438, |
| "grad_norm": 0.007415771484375, |
| "learning_rate": 9.935555555555556e-06, |
| "loss": 3.4049, |
| "mean_token_accuracy": 0.4317916492931545, |
| "num_tokens": 73492566.0, |
| "step": 5530 |
| }, |
| { |
| "entropy": 2.562423176318407, |
| "epoch": 4.929394091136705, |
| "grad_norm": 0.00762939453125, |
| "learning_rate": 9.913333333333334e-06, |
| "loss": 3.4896, |
| "mean_token_accuracy": 0.4213337803259492, |
| "num_tokens": 73622919.0, |
| "step": 5540 |
| }, |
| { |
| "entropy": 2.5246185824275016, |
| "epoch": 4.938296333388973, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 9.891111111111113e-06, |
| "loss": 3.4677, |
| "mean_token_accuracy": 0.4274196891114116, |
| "num_tokens": 73753437.0, |
| "step": 5550 |
| }, |
| { |
| "entropy": 2.438191378861666, |
| "epoch": 4.94719857564124, |
| "grad_norm": 0.0091552734375, |
| "learning_rate": 9.86888888888889e-06, |
| "loss": 3.346, |
| "mean_token_accuracy": 0.4397026361897588, |
| "num_tokens": 73889876.0, |
| "step": 5560 |
| }, |
| { |
| "entropy": 2.517542506754398, |
| "epoch": 4.956100817893507, |
| "grad_norm": 0.01080322265625, |
| "learning_rate": 9.846666666666668e-06, |
| "loss": 3.4205, |
| "mean_token_accuracy": 0.432649065926671, |
| "num_tokens": 74023888.0, |
| "step": 5570 |
| }, |
| { |
| "entropy": 2.5115950651466847, |
| "epoch": 4.965003060145774, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 9.824444444444446e-06, |
| "loss": 3.4187, |
| "mean_token_accuracy": 0.43286951901391146, |
| "num_tokens": 74155331.0, |
| "step": 5580 |
| }, |
| { |
| "entropy": 2.499591101706028, |
| "epoch": 4.973905302398041, |
| "grad_norm": 0.00750732421875, |
| "learning_rate": 9.802222222222224e-06, |
| "loss": 3.4144, |
| "mean_token_accuracy": 0.4308878821320832, |
| "num_tokens": 74285971.0, |
| "step": 5590 |
| }, |
| { |
| "entropy": 2.601413767784834, |
| "epoch": 4.982807544650309, |
| "grad_norm": 0.01300048828125, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 3.5654, |
| "mean_token_accuracy": 0.4171551912091672, |
| "num_tokens": 74414112.0, |
| "step": 5600 |
| }, |
| { |
| "epoch": 4.982807544650309, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 74414112.0, |
| "eval_coding_runtime": 96.0729, |
| "eval_coding_samples_per_second": 5.204, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 5600 |
| }, |
| { |
| "epoch": 4.982807544650309, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 74414112.0, |
| "eval_chemistry_runtime": 53.8957, |
| "eval_chemistry_samples_per_second": 9.277, |
| "eval_chemistry_steps_per_second": 4.639, |
| "step": 5600 |
| }, |
| { |
| "entropy": 2.5589482344686987, |
| "epoch": 4.991709786902576, |
| "grad_norm": 0.01300048828125, |
| "learning_rate": 9.757777777777779e-06, |
| "loss": 3.5263, |
| "mean_token_accuracy": 0.4180930620059371, |
| "num_tokens": 74542230.0, |
| "step": 5610 |
| }, |
| { |
| "entropy": 2.4767904705649255, |
| "epoch": 5.0, |
| "grad_norm": 0.017822265625, |
| "learning_rate": 9.735555555555556e-06, |
| "loss": 3.3742, |
| "mean_token_accuracy": 0.433748621468576, |
| "num_tokens": 74664450.0, |
| "step": 5620 |
| }, |
| { |
| "entropy": 2.51421831920743, |
| "epoch": 5.008902242252267, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 9.713333333333334e-06, |
| "loss": 3.481, |
| "mean_token_accuracy": 0.42637166297063234, |
| "num_tokens": 74789045.0, |
| "step": 5630 |
| }, |
| { |
| "entropy": 2.5585424259305, |
| "epoch": 5.017804484504534, |
| "grad_norm": 0.011962890625, |
| "learning_rate": 9.691111111111112e-06, |
| "loss": 3.536, |
| "mean_token_accuracy": 0.4221954869106412, |
| "num_tokens": 74920219.0, |
| "step": 5640 |
| }, |
| { |
| "entropy": 2.4705275617539884, |
| "epoch": 5.026706726756802, |
| "grad_norm": 0.00830078125, |
| "learning_rate": 9.66888888888889e-06, |
| "loss": 3.3775, |
| "mean_token_accuracy": 0.43359337486326693, |
| "num_tokens": 75055353.0, |
| "step": 5650 |
| }, |
| { |
| "entropy": 2.4017206080257893, |
| "epoch": 5.035608969009069, |
| "grad_norm": 0.0181884765625, |
| "learning_rate": 9.646666666666667e-06, |
| "loss": 3.341, |
| "mean_token_accuracy": 0.43982536615803836, |
| "num_tokens": 75192549.0, |
| "step": 5660 |
| }, |
| { |
| "entropy": 2.5172601625323296, |
| "epoch": 5.044511211261336, |
| "grad_norm": 0.0181884765625, |
| "learning_rate": 9.624444444444445e-06, |
| "loss": 3.4378, |
| "mean_token_accuracy": 0.42869628984481095, |
| "num_tokens": 75332784.0, |
| "step": 5670 |
| }, |
| { |
| "entropy": 2.440136446803808, |
| "epoch": 5.0534134535136035, |
| "grad_norm": 0.01519775390625, |
| "learning_rate": 9.602222222222222e-06, |
| "loss": 3.3567, |
| "mean_token_accuracy": 0.44069963935762646, |
| "num_tokens": 75468065.0, |
| "step": 5680 |
| }, |
| { |
| "entropy": 2.5362621888518335, |
| "epoch": 5.062315695765871, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 9.58e-06, |
| "loss": 3.4758, |
| "mean_token_accuracy": 0.4249008771032095, |
| "num_tokens": 75601558.0, |
| "step": 5690 |
| }, |
| { |
| "entropy": 2.508413332700729, |
| "epoch": 5.071217938018139, |
| "grad_norm": 0.0147705078125, |
| "learning_rate": 9.557777777777777e-06, |
| "loss": 3.3666, |
| "mean_token_accuracy": 0.43379055354744195, |
| "num_tokens": 75739068.0, |
| "step": 5700 |
| }, |
| { |
| "epoch": 5.071217938018139, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 75739068.0, |
| "eval_coding_runtime": 96.4314, |
| "eval_coding_samples_per_second": 5.185, |
| "eval_coding_steps_per_second": 2.593, |
| "step": 5700 |
| }, |
| { |
| "epoch": 5.071217938018139, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 75739068.0, |
| "eval_chemistry_runtime": 54.0913, |
| "eval_chemistry_samples_per_second": 9.244, |
| "eval_chemistry_steps_per_second": 4.622, |
| "step": 5700 |
| }, |
| { |
| "entropy": 2.4802520789206026, |
| "epoch": 5.080120180270406, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 9.535555555555557e-06, |
| "loss": 3.3795, |
| "mean_token_accuracy": 0.4303091328591108, |
| "num_tokens": 75873195.0, |
| "step": 5710 |
| }, |
| { |
| "entropy": 2.4554261356592177, |
| "epoch": 5.089022422522673, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 9.513333333333334e-06, |
| "loss": 3.3848, |
| "mean_token_accuracy": 0.43654916174709796, |
| "num_tokens": 76004415.0, |
| "step": 5720 |
| }, |
| { |
| "entropy": 2.5274557061493397, |
| "epoch": 5.09792466477494, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 9.491111111111112e-06, |
| "loss": 3.4028, |
| "mean_token_accuracy": 0.42681868355721236, |
| "num_tokens": 76134930.0, |
| "step": 5730 |
| }, |
| { |
| "entropy": 2.603742365539074, |
| "epoch": 5.106826907027208, |
| "grad_norm": 0.011474609375, |
| "learning_rate": 9.46888888888889e-06, |
| "loss": 3.4915, |
| "mean_token_accuracy": 0.4225413169711828, |
| "num_tokens": 76268461.0, |
| "step": 5740 |
| }, |
| { |
| "entropy": 2.4755712531507017, |
| "epoch": 5.115729149279475, |
| "grad_norm": 0.01165771484375, |
| "learning_rate": 9.446666666666667e-06, |
| "loss": 3.3945, |
| "mean_token_accuracy": 0.4350418847054243, |
| "num_tokens": 76405447.0, |
| "step": 5750 |
| }, |
| { |
| "entropy": 2.4808173961937428, |
| "epoch": 5.124631391531742, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 9.424444444444445e-06, |
| "loss": 3.4039, |
| "mean_token_accuracy": 0.43662428483366966, |
| "num_tokens": 76542099.0, |
| "step": 5760 |
| }, |
| { |
| "entropy": 2.4800227262079715, |
| "epoch": 5.133533633784009, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 9.402222222222222e-06, |
| "loss": 3.3727, |
| "mean_token_accuracy": 0.4319386541843414, |
| "num_tokens": 76673172.0, |
| "step": 5770 |
| }, |
| { |
| "entropy": 2.4812305107712747, |
| "epoch": 5.142435876036276, |
| "grad_norm": 0.0238037109375, |
| "learning_rate": 9.38e-06, |
| "loss": 3.396, |
| "mean_token_accuracy": 0.4331910008564591, |
| "num_tokens": 76806284.0, |
| "step": 5780 |
| }, |
| { |
| "entropy": 2.4800585940480233, |
| "epoch": 5.151338118288544, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 9.35777777777778e-06, |
| "loss": 3.3909, |
| "mean_token_accuracy": 0.4347142428159714, |
| "num_tokens": 76931944.0, |
| "step": 5790 |
| }, |
| { |
| "entropy": 2.4100775457918644, |
| "epoch": 5.160240360540811, |
| "grad_norm": 0.01263427734375, |
| "learning_rate": 9.335555555555557e-06, |
| "loss": 3.3139, |
| "mean_token_accuracy": 0.4378079754300416, |
| "num_tokens": 77067082.0, |
| "step": 5800 |
| }, |
| { |
| "epoch": 5.160240360540811, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 77067082.0, |
| "eval_coding_runtime": 96.1166, |
| "eval_coding_samples_per_second": 5.202, |
| "eval_coding_steps_per_second": 2.601, |
| "step": 5800 |
| }, |
| { |
| "epoch": 5.160240360540811, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 77067082.0, |
| "eval_chemistry_runtime": 53.8626, |
| "eval_chemistry_samples_per_second": 9.283, |
| "eval_chemistry_steps_per_second": 4.641, |
| "step": 5800 |
| }, |
| { |
| "entropy": 2.5165654100477695, |
| "epoch": 5.1691426027930785, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 9.313333333333335e-06, |
| "loss": 3.4045, |
| "mean_token_accuracy": 0.42999832993373277, |
| "num_tokens": 77199220.0, |
| "step": 5810 |
| }, |
| { |
| "entropy": 2.4463331483304502, |
| "epoch": 5.178044845045346, |
| "grad_norm": 0.007476806640625, |
| "learning_rate": 9.291111111111112e-06, |
| "loss": 3.375, |
| "mean_token_accuracy": 0.4334684856235981, |
| "num_tokens": 77331038.0, |
| "step": 5820 |
| }, |
| { |
| "entropy": 2.519996690005064, |
| "epoch": 5.186947087297613, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 9.26888888888889e-06, |
| "loss": 3.4274, |
| "mean_token_accuracy": 0.427511371858418, |
| "num_tokens": 77468211.0, |
| "step": 5830 |
| }, |
| { |
| "entropy": 2.4938265405595303, |
| "epoch": 5.195849329549881, |
| "grad_norm": 0.0103759765625, |
| "learning_rate": 9.246666666666667e-06, |
| "loss": 3.4678, |
| "mean_token_accuracy": 0.42385053662583233, |
| "num_tokens": 77598400.0, |
| "step": 5840 |
| }, |
| { |
| "entropy": 2.4477928623557093, |
| "epoch": 5.204751571802148, |
| "grad_norm": 0.007781982421875, |
| "learning_rate": 9.224444444444445e-06, |
| "loss": 3.3877, |
| "mean_token_accuracy": 0.43123094700276854, |
| "num_tokens": 77735458.0, |
| "step": 5850 |
| }, |
| { |
| "entropy": 2.5270394928753377, |
| "epoch": 5.213653814054415, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 9.202222222222224e-06, |
| "loss": 3.4925, |
| "mean_token_accuracy": 0.4220319422893226, |
| "num_tokens": 77860470.0, |
| "step": 5860 |
| }, |
| { |
| "entropy": 2.5107815079391003, |
| "epoch": 5.222556056306682, |
| "grad_norm": 0.0111083984375, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 3.455, |
| "mean_token_accuracy": 0.4282518906518817, |
| "num_tokens": 77990141.0, |
| "step": 5870 |
| }, |
| { |
| "entropy": 2.491736387461424, |
| "epoch": 5.23145829855895, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 9.15777777777778e-06, |
| "loss": 3.3697, |
| "mean_token_accuracy": 0.4359064460732043, |
| "num_tokens": 78127641.0, |
| "step": 5880 |
| }, |
| { |
| "entropy": 2.5040958702564238, |
| "epoch": 5.240360540811217, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 9.135555555555557e-06, |
| "loss": 3.4701, |
| "mean_token_accuracy": 0.43055264884606004, |
| "num_tokens": 78256808.0, |
| "step": 5890 |
| }, |
| { |
| "entropy": 2.4481254391372205, |
| "epoch": 5.249262783063484, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 9.113333333333335e-06, |
| "loss": 3.3284, |
| "mean_token_accuracy": 0.43865708420053123, |
| "num_tokens": 78392541.0, |
| "step": 5900 |
| }, |
| { |
| "epoch": 5.249262783063484, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 78392541.0, |
| "eval_coding_runtime": 96.0682, |
| "eval_coding_samples_per_second": 5.205, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 5900 |
| }, |
| { |
| "epoch": 5.249262783063484, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 78392541.0, |
| "eval_chemistry_runtime": 53.8408, |
| "eval_chemistry_samples_per_second": 9.287, |
| "eval_chemistry_steps_per_second": 4.643, |
| "step": 5900 |
| }, |
| { |
| "entropy": 2.4499885328114033, |
| "epoch": 5.258165025315751, |
| "grad_norm": 0.0225830078125, |
| "learning_rate": 9.091111111111112e-06, |
| "loss": 3.3971, |
| "mean_token_accuracy": 0.4344657910056412, |
| "num_tokens": 78525339.0, |
| "step": 5910 |
| }, |
| { |
| "entropy": 2.4558506585657596, |
| "epoch": 5.267067267568018, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 9.06888888888889e-06, |
| "loss": 3.3396, |
| "mean_token_accuracy": 0.43714132644236087, |
| "num_tokens": 78655681.0, |
| "step": 5920 |
| }, |
| { |
| "entropy": 2.535300438851118, |
| "epoch": 5.275969509820286, |
| "grad_norm": 0.018798828125, |
| "learning_rate": 9.046666666666668e-06, |
| "loss": 3.4972, |
| "mean_token_accuracy": 0.425995328463614, |
| "num_tokens": 78791770.0, |
| "step": 5930 |
| }, |
| { |
| "entropy": 2.517005608975887, |
| "epoch": 5.284871752072553, |
| "grad_norm": 0.00732421875, |
| "learning_rate": 9.024444444444445e-06, |
| "loss": 3.3915, |
| "mean_token_accuracy": 0.4262471618130803, |
| "num_tokens": 78927986.0, |
| "step": 5940 |
| }, |
| { |
| "entropy": 2.554993372410536, |
| "epoch": 5.2937739943248205, |
| "grad_norm": 0.01324462890625, |
| "learning_rate": 9.002222222222223e-06, |
| "loss": 3.4539, |
| "mean_token_accuracy": 0.42558628180995584, |
| "num_tokens": 79062131.0, |
| "step": 5950 |
| }, |
| { |
| "entropy": 2.48376282453537, |
| "epoch": 5.302676236577088, |
| "grad_norm": 0.007476806640625, |
| "learning_rate": 8.98e-06, |
| "loss": 3.3774, |
| "mean_token_accuracy": 0.42769679743796585, |
| "num_tokens": 79198429.0, |
| "step": 5960 |
| }, |
| { |
| "entropy": 2.433156369626522, |
| "epoch": 5.311578478829355, |
| "grad_norm": 0.0203857421875, |
| "learning_rate": 8.957777777777778e-06, |
| "loss": 3.324, |
| "mean_token_accuracy": 0.43958458956331015, |
| "num_tokens": 79327926.0, |
| "step": 5970 |
| }, |
| { |
| "entropy": 2.4852182611823084, |
| "epoch": 5.320480721081623, |
| "grad_norm": 0.007476806640625, |
| "learning_rate": 8.935555555555556e-06, |
| "loss": 3.3994, |
| "mean_token_accuracy": 0.43158692624419925, |
| "num_tokens": 79458407.0, |
| "step": 5980 |
| }, |
| { |
| "entropy": 2.4302474945783614, |
| "epoch": 5.32938296333389, |
| "grad_norm": 0.0142822265625, |
| "learning_rate": 8.913333333333333e-06, |
| "loss": 3.3853, |
| "mean_token_accuracy": 0.4374588614329696, |
| "num_tokens": 79590887.0, |
| "step": 5990 |
| }, |
| { |
| "entropy": 2.4680772449821236, |
| "epoch": 5.338285205586157, |
| "grad_norm": 0.00726318359375, |
| "learning_rate": 8.891111111111111e-06, |
| "loss": 3.418, |
| "mean_token_accuracy": 0.43037008214741945, |
| "num_tokens": 79724575.0, |
| "step": 6000 |
| }, |
| { |
| "epoch": 5.338285205586157, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 79724575.0, |
| "eval_coding_runtime": 96.0662, |
| "eval_coding_samples_per_second": 5.205, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 6000 |
| }, |
| { |
| "epoch": 5.338285205586157, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 79724575.0, |
| "eval_chemistry_runtime": 53.8447, |
| "eval_chemistry_samples_per_second": 9.286, |
| "eval_chemistry_steps_per_second": 4.643, |
| "step": 6000 |
| }, |
| { |
| "entropy": 2.501242730766535, |
| "epoch": 5.347187447838424, |
| "grad_norm": 0.009521484375, |
| "learning_rate": 8.868888888888888e-06, |
| "loss": 3.4371, |
| "mean_token_accuracy": 0.4284817291423678, |
| "num_tokens": 79850804.0, |
| "step": 6010 |
| }, |
| { |
| "entropy": 2.497699758410454, |
| "epoch": 5.356089690090691, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 8.846666666666668e-06, |
| "loss": 3.423, |
| "mean_token_accuracy": 0.42744764816015957, |
| "num_tokens": 79982285.0, |
| "step": 6020 |
| }, |
| { |
| "entropy": 2.4926821291446686, |
| "epoch": 5.364991932342959, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 8.824444444444445e-06, |
| "loss": 3.4281, |
| "mean_token_accuracy": 0.4338597496971488, |
| "num_tokens": 80113888.0, |
| "step": 6030 |
| }, |
| { |
| "entropy": 2.540310859680176, |
| "epoch": 5.373894174595226, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 8.802222222222223e-06, |
| "loss": 3.4406, |
| "mean_token_accuracy": 0.4294831304810941, |
| "num_tokens": 80247460.0, |
| "step": 6040 |
| }, |
| { |
| "entropy": 2.441784296184778, |
| "epoch": 5.382796416847493, |
| "grad_norm": 0.0096435546875, |
| "learning_rate": 8.78e-06, |
| "loss": 3.3584, |
| "mean_token_accuracy": 0.43861055970191953, |
| "num_tokens": 80386391.0, |
| "step": 6050 |
| }, |
| { |
| "entropy": 2.434786938875914, |
| "epoch": 5.39169865909976, |
| "grad_norm": 0.007720947265625, |
| "learning_rate": 8.757777777777778e-06, |
| "loss": 3.3533, |
| "mean_token_accuracy": 0.4356680458411574, |
| "num_tokens": 80515154.0, |
| "step": 6060 |
| }, |
| { |
| "entropy": 2.5079306304454803, |
| "epoch": 5.400600901352028, |
| "grad_norm": 0.011474609375, |
| "learning_rate": 8.735555555555556e-06, |
| "loss": 3.4349, |
| "mean_token_accuracy": 0.42870584819465873, |
| "num_tokens": 80647333.0, |
| "step": 6070 |
| }, |
| { |
| "entropy": 2.520564942806959, |
| "epoch": 5.4095031436042955, |
| "grad_norm": 0.0118408203125, |
| "learning_rate": 8.713333333333333e-06, |
| "loss": 3.4201, |
| "mean_token_accuracy": 0.42633711621165277, |
| "num_tokens": 80785851.0, |
| "step": 6080 |
| }, |
| { |
| "entropy": 2.5415661334991455, |
| "epoch": 5.418405385856563, |
| "grad_norm": 0.01019287109375, |
| "learning_rate": 8.691111111111111e-06, |
| "loss": 3.4499, |
| "mean_token_accuracy": 0.42608464881777763, |
| "num_tokens": 80916483.0, |
| "step": 6090 |
| }, |
| { |
| "entropy": 2.513637960702181, |
| "epoch": 5.42730762810883, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 8.66888888888889e-06, |
| "loss": 3.4316, |
| "mean_token_accuracy": 0.4285950168967247, |
| "num_tokens": 81052407.0, |
| "step": 6100 |
| }, |
| { |
| "epoch": 5.42730762810883, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 81052407.0, |
| "eval_coding_runtime": 96.3445, |
| "eval_coding_samples_per_second": 5.19, |
| "eval_coding_steps_per_second": 2.595, |
| "step": 6100 |
| }, |
| { |
| "epoch": 5.42730762810883, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 81052407.0, |
| "eval_chemistry_runtime": 53.8786, |
| "eval_chemistry_samples_per_second": 9.28, |
| "eval_chemistry_steps_per_second": 4.64, |
| "step": 6100 |
| }, |
| { |
| "entropy": 2.5025472305715084, |
| "epoch": 5.436209870361097, |
| "grad_norm": 0.009765625, |
| "learning_rate": 8.646666666666668e-06, |
| "loss": 3.438, |
| "mean_token_accuracy": 0.4295802541077137, |
| "num_tokens": 81179853.0, |
| "step": 6110 |
| }, |
| { |
| "entropy": 2.4906201288104057, |
| "epoch": 5.445112112613365, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 8.624444444444446e-06, |
| "loss": 3.3953, |
| "mean_token_accuracy": 0.43400709666311743, |
| "num_tokens": 81313243.0, |
| "step": 6120 |
| }, |
| { |
| "entropy": 2.611784942448139, |
| "epoch": 5.454014354865632, |
| "grad_norm": 0.01116943359375, |
| "learning_rate": 8.602222222222223e-06, |
| "loss": 3.488, |
| "mean_token_accuracy": 0.42277059275656936, |
| "num_tokens": 81442204.0, |
| "step": 6130 |
| }, |
| { |
| "entropy": 2.4712997011840345, |
| "epoch": 5.462916597117899, |
| "grad_norm": 0.01544189453125, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 3.3982, |
| "mean_token_accuracy": 0.43449760554358363, |
| "num_tokens": 81577943.0, |
| "step": 6140 |
| }, |
| { |
| "entropy": 2.5050548873841763, |
| "epoch": 5.471818839370166, |
| "grad_norm": 0.016845703125, |
| "learning_rate": 8.557777777777778e-06, |
| "loss": 3.3912, |
| "mean_token_accuracy": 0.43150261491537095, |
| "num_tokens": 81712547.0, |
| "step": 6150 |
| }, |
| { |
| "entropy": 2.5267510414123535, |
| "epoch": 5.480721081622434, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 8.535555555555556e-06, |
| "loss": 3.4195, |
| "mean_token_accuracy": 0.42423103014007213, |
| "num_tokens": 81851089.0, |
| "step": 6160 |
| }, |
| { |
| "entropy": 2.4870824836194516, |
| "epoch": 5.489623323874701, |
| "grad_norm": 0.0108642578125, |
| "learning_rate": 8.513333333333335e-06, |
| "loss": 3.333, |
| "mean_token_accuracy": 0.43592707309871914, |
| "num_tokens": 81986911.0, |
| "step": 6170 |
| }, |
| { |
| "entropy": 2.527715814113617, |
| "epoch": 5.498525566126968, |
| "grad_norm": 0.0091552734375, |
| "learning_rate": 8.491111111111113e-06, |
| "loss": 3.4532, |
| "mean_token_accuracy": 0.4267021115869284, |
| "num_tokens": 82119911.0, |
| "step": 6180 |
| }, |
| { |
| "entropy": 2.520997066050768, |
| "epoch": 5.507427808379235, |
| "grad_norm": 0.0103759765625, |
| "learning_rate": 8.46888888888889e-06, |
| "loss": 3.4247, |
| "mean_token_accuracy": 0.424376964289695, |
| "num_tokens": 82250498.0, |
| "step": 6190 |
| }, |
| { |
| "entropy": 2.525718642026186, |
| "epoch": 5.5163300506315025, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 8.446666666666668e-06, |
| "loss": 3.5029, |
| "mean_token_accuracy": 0.4253832670859993, |
| "num_tokens": 82380463.0, |
| "step": 6200 |
| }, |
| { |
| "epoch": 5.5163300506315025, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 82380463.0, |
| "eval_coding_runtime": 96.1359, |
| "eval_coding_samples_per_second": 5.201, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 6200 |
| }, |
| { |
| "epoch": 5.5163300506315025, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 82380463.0, |
| "eval_chemistry_runtime": 53.8019, |
| "eval_chemistry_samples_per_second": 9.293, |
| "eval_chemistry_steps_per_second": 4.647, |
| "step": 6200 |
| }, |
| { |
| "entropy": 2.5128908447921274, |
| "epoch": 5.52523229288377, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 8.424444444444446e-06, |
| "loss": 3.4245, |
| "mean_token_accuracy": 0.4285768648609519, |
| "num_tokens": 82516241.0, |
| "step": 6210 |
| }, |
| { |
| "entropy": 2.439585692435503, |
| "epoch": 5.534134535136038, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 8.402222222222223e-06, |
| "loss": 3.4019, |
| "mean_token_accuracy": 0.43478961642831565, |
| "num_tokens": 82650788.0, |
| "step": 6220 |
| }, |
| { |
| "entropy": 2.540559572726488, |
| "epoch": 5.543036777388305, |
| "grad_norm": 0.0107421875, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 3.4324, |
| "mean_token_accuracy": 0.4311952818185091, |
| "num_tokens": 82788379.0, |
| "step": 6230 |
| }, |
| { |
| "entropy": 2.4810707099735736, |
| "epoch": 5.551939019640572, |
| "grad_norm": 0.007720947265625, |
| "learning_rate": 8.357777777777779e-06, |
| "loss": 3.3777, |
| "mean_token_accuracy": 0.43144108168780804, |
| "num_tokens": 82930302.0, |
| "step": 6240 |
| }, |
| { |
| "entropy": 2.450982717424631, |
| "epoch": 5.560841261892839, |
| "grad_norm": 0.013671875, |
| "learning_rate": 8.335555555555556e-06, |
| "loss": 3.3857, |
| "mean_token_accuracy": 0.4328729661181569, |
| "num_tokens": 83063791.0, |
| "step": 6250 |
| }, |
| { |
| "entropy": 2.543005682528019, |
| "epoch": 5.569743504145107, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 8.313333333333334e-06, |
| "loss": 3.4802, |
| "mean_token_accuracy": 0.42440302260220053, |
| "num_tokens": 83197683.0, |
| "step": 6260 |
| }, |
| { |
| "entropy": 2.5175108201801777, |
| "epoch": 5.578645746397374, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 8.291111111111112e-06, |
| "loss": 3.4521, |
| "mean_token_accuracy": 0.4274182070046663, |
| "num_tokens": 83327451.0, |
| "step": 6270 |
| }, |
| { |
| "entropy": 2.5078105956315992, |
| "epoch": 5.587547988649641, |
| "grad_norm": 0.00958251953125, |
| "learning_rate": 8.268888888888889e-06, |
| "loss": 3.4708, |
| "mean_token_accuracy": 0.4265906841494143, |
| "num_tokens": 83455212.0, |
| "step": 6280 |
| }, |
| { |
| "entropy": 2.5451072968542574, |
| "epoch": 5.596450230901908, |
| "grad_norm": 0.00946044921875, |
| "learning_rate": 8.246666666666667e-06, |
| "loss": 3.4787, |
| "mean_token_accuracy": 0.4233635565266013, |
| "num_tokens": 83584894.0, |
| "step": 6290 |
| }, |
| { |
| "entropy": 2.6029953725636004, |
| "epoch": 5.605352473154175, |
| "grad_norm": 0.007720947265625, |
| "learning_rate": 8.224444444444444e-06, |
| "loss": 3.5227, |
| "mean_token_accuracy": 0.42024144623428583, |
| "num_tokens": 83714248.0, |
| "step": 6300 |
| }, |
| { |
| "epoch": 5.605352473154175, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 83714248.0, |
| "eval_coding_runtime": 96.091, |
| "eval_coding_samples_per_second": 5.203, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 6300 |
| }, |
| { |
| "epoch": 5.605352473154175, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 83714248.0, |
| "eval_chemistry_runtime": 54.1311, |
| "eval_chemistry_samples_per_second": 9.237, |
| "eval_chemistry_steps_per_second": 4.618, |
| "step": 6300 |
| }, |
| { |
| "entropy": 2.5744390614330768, |
| "epoch": 5.614254715406443, |
| "grad_norm": 0.0098876953125, |
| "learning_rate": 8.202222222222222e-06, |
| "loss": 3.5003, |
| "mean_token_accuracy": 0.41896801330149175, |
| "num_tokens": 83847363.0, |
| "step": 6310 |
| }, |
| { |
| "entropy": 2.524623380601406, |
| "epoch": 5.62315695765871, |
| "grad_norm": 0.01068115234375, |
| "learning_rate": 8.18e-06, |
| "loss": 3.4049, |
| "mean_token_accuracy": 0.43059933921322224, |
| "num_tokens": 83983112.0, |
| "step": 6320 |
| }, |
| { |
| "entropy": 2.5099828019738197, |
| "epoch": 5.6320591999109775, |
| "grad_norm": 0.015869140625, |
| "learning_rate": 8.157777777777779e-06, |
| "loss": 3.4666, |
| "mean_token_accuracy": 0.4304792143404484, |
| "num_tokens": 84113799.0, |
| "step": 6330 |
| }, |
| { |
| "entropy": 2.5512323528528214, |
| "epoch": 5.640961442163245, |
| "grad_norm": 0.01397705078125, |
| "learning_rate": 8.135555555555557e-06, |
| "loss": 3.4242, |
| "mean_token_accuracy": 0.4255604514852166, |
| "num_tokens": 84246512.0, |
| "step": 6340 |
| }, |
| { |
| "entropy": 2.4761642038822176, |
| "epoch": 5.649863684415513, |
| "grad_norm": 0.010009765625, |
| "learning_rate": 8.113333333333334e-06, |
| "loss": 3.3876, |
| "mean_token_accuracy": 0.43336022151634096, |
| "num_tokens": 84372011.0, |
| "step": 6350 |
| }, |
| { |
| "entropy": 2.4591390673071145, |
| "epoch": 5.65876592666778, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 8.091111111111112e-06, |
| "loss": 3.3653, |
| "mean_token_accuracy": 0.43248563539236784, |
| "num_tokens": 84503035.0, |
| "step": 6360 |
| }, |
| { |
| "entropy": 2.512166740745306, |
| "epoch": 5.667668168920047, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 8.06888888888889e-06, |
| "loss": 3.4415, |
| "mean_token_accuracy": 0.4289271808229387, |
| "num_tokens": 84641054.0, |
| "step": 6370 |
| }, |
| { |
| "entropy": 2.4857188440859317, |
| "epoch": 5.676570411172314, |
| "grad_norm": 0.012939453125, |
| "learning_rate": 8.046666666666667e-06, |
| "loss": 3.4064, |
| "mean_token_accuracy": 0.4323981337249279, |
| "num_tokens": 84768004.0, |
| "step": 6380 |
| }, |
| { |
| "entropy": 2.506490983068943, |
| "epoch": 5.685472653424581, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 8.024444444444445e-06, |
| "loss": 3.3922, |
| "mean_token_accuracy": 0.4339711846783757, |
| "num_tokens": 84896277.0, |
| "step": 6390 |
| }, |
| { |
| "entropy": 2.498063915967941, |
| "epoch": 5.694374895676849, |
| "grad_norm": 0.0126953125, |
| "learning_rate": 8.002222222222222e-06, |
| "loss": 3.449, |
| "mean_token_accuracy": 0.42802249658852815, |
| "num_tokens": 85035009.0, |
| "step": 6400 |
| }, |
| { |
| "epoch": 5.694374895676849, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 85035009.0, |
| "eval_coding_runtime": 96.075, |
| "eval_coding_samples_per_second": 5.204, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 6400 |
| }, |
| { |
| "epoch": 5.694374895676849, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 85035009.0, |
| "eval_chemistry_runtime": 53.7985, |
| "eval_chemistry_samples_per_second": 9.294, |
| "eval_chemistry_steps_per_second": 4.647, |
| "step": 6400 |
| }, |
| { |
| "entropy": 2.4345531940460203, |
| "epoch": 5.703277137929116, |
| "grad_norm": 0.01025390625, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 3.3744, |
| "mean_token_accuracy": 0.43794034253805875, |
| "num_tokens": 85167902.0, |
| "step": 6410 |
| }, |
| { |
| "entropy": 2.5272544123232366, |
| "epoch": 5.712179380181383, |
| "grad_norm": 0.0137939453125, |
| "learning_rate": 7.957777777777779e-06, |
| "loss": 3.4629, |
| "mean_token_accuracy": 0.4239718407392502, |
| "num_tokens": 85302580.0, |
| "step": 6420 |
| }, |
| { |
| "entropy": 2.431706988066435, |
| "epoch": 5.72108162243365, |
| "grad_norm": 0.011474609375, |
| "learning_rate": 7.935555555555557e-06, |
| "loss": 3.3264, |
| "mean_token_accuracy": 0.44290522038936614, |
| "num_tokens": 85437544.0, |
| "step": 6430 |
| }, |
| { |
| "entropy": 2.4396374851465223, |
| "epoch": 5.729983864685918, |
| "grad_norm": 0.007476806640625, |
| "learning_rate": 7.913333333333334e-06, |
| "loss": 3.3369, |
| "mean_token_accuracy": 0.4370517207309604, |
| "num_tokens": 85573308.0, |
| "step": 6440 |
| }, |
| { |
| "entropy": 2.4476514376699923, |
| "epoch": 5.738886106938185, |
| "grad_norm": 0.010009765625, |
| "learning_rate": 7.891111111111112e-06, |
| "loss": 3.2999, |
| "mean_token_accuracy": 0.43856146819889547, |
| "num_tokens": 85710949.0, |
| "step": 6450 |
| }, |
| { |
| "entropy": 2.4396982535719873, |
| "epoch": 5.7477883491904525, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 7.86888888888889e-06, |
| "loss": 3.36, |
| "mean_token_accuracy": 0.4344617579132318, |
| "num_tokens": 85842459.0, |
| "step": 6460 |
| }, |
| { |
| "entropy": 2.5628412067890167, |
| "epoch": 5.7566905914427196, |
| "grad_norm": 0.014404296875, |
| "learning_rate": 7.846666666666667e-06, |
| "loss": 3.4595, |
| "mean_token_accuracy": 0.4248708474449813, |
| "num_tokens": 85973499.0, |
| "step": 6470 |
| }, |
| { |
| "entropy": 2.52545180991292, |
| "epoch": 5.765592833694987, |
| "grad_norm": 0.015869140625, |
| "learning_rate": 7.824444444444445e-06, |
| "loss": 3.4347, |
| "mean_token_accuracy": 0.4277951089665294, |
| "num_tokens": 86104434.0, |
| "step": 6480 |
| }, |
| { |
| "entropy": 2.504948926717043, |
| "epoch": 5.774495075947254, |
| "grad_norm": 0.00897216796875, |
| "learning_rate": 7.802222222222222e-06, |
| "loss": 3.4467, |
| "mean_token_accuracy": 0.4269657224416733, |
| "num_tokens": 86236930.0, |
| "step": 6490 |
| }, |
| { |
| "entropy": 2.4826805599033834, |
| "epoch": 5.783397318199522, |
| "grad_norm": 0.012451171875, |
| "learning_rate": 7.78e-06, |
| "loss": 3.4297, |
| "mean_token_accuracy": 0.43146911114454267, |
| "num_tokens": 86369082.0, |
| "step": 6500 |
| }, |
| { |
| "epoch": 5.783397318199522, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 86369082.0, |
| "eval_coding_runtime": 96.0066, |
| "eval_coding_samples_per_second": 5.208, |
| "eval_coding_steps_per_second": 2.604, |
| "step": 6500 |
| }, |
| { |
| "epoch": 5.783397318199522, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 86369082.0, |
| "eval_chemistry_runtime": 53.8514, |
| "eval_chemistry_samples_per_second": 9.285, |
| "eval_chemistry_steps_per_second": 4.642, |
| "step": 6500 |
| }, |
| { |
| "entropy": 2.4843976750969885, |
| "epoch": 5.792299560451789, |
| "grad_norm": 0.0089111328125, |
| "learning_rate": 7.75777777777778e-06, |
| "loss": 3.4182, |
| "mean_token_accuracy": 0.432208720035851, |
| "num_tokens": 86500590.0, |
| "step": 6510 |
| }, |
| { |
| "entropy": 2.4628002405166627, |
| "epoch": 5.801201802704056, |
| "grad_norm": 0.01025390625, |
| "learning_rate": 7.735555555555557e-06, |
| "loss": 3.3908, |
| "mean_token_accuracy": 0.4326698398217559, |
| "num_tokens": 86633561.0, |
| "step": 6520 |
| }, |
| { |
| "entropy": 2.4886969231069087, |
| "epoch": 5.810104044956323, |
| "grad_norm": 0.0101318359375, |
| "learning_rate": 7.713333333333335e-06, |
| "loss": 3.4051, |
| "mean_token_accuracy": 0.43189607802778485, |
| "num_tokens": 86766464.0, |
| "step": 6530 |
| }, |
| { |
| "entropy": 2.445012257993221, |
| "epoch": 5.819006287208591, |
| "grad_norm": 0.015625, |
| "learning_rate": 7.691111111111112e-06, |
| "loss": 3.3542, |
| "mean_token_accuracy": 0.4378402501344681, |
| "num_tokens": 86902165.0, |
| "step": 6540 |
| }, |
| { |
| "entropy": 2.5058255933225153, |
| "epoch": 5.827908529460858, |
| "grad_norm": 0.0194091796875, |
| "learning_rate": 7.66888888888889e-06, |
| "loss": 3.4675, |
| "mean_token_accuracy": 0.4307621283456683, |
| "num_tokens": 87034298.0, |
| "step": 6550 |
| }, |
| { |
| "entropy": 2.530172634869814, |
| "epoch": 5.836810771713125, |
| "grad_norm": 0.01043701171875, |
| "learning_rate": 7.646666666666667e-06, |
| "loss": 3.449, |
| "mean_token_accuracy": 0.42743550054728985, |
| "num_tokens": 87163315.0, |
| "step": 6560 |
| }, |
| { |
| "entropy": 2.5022324703633787, |
| "epoch": 5.845713013965392, |
| "grad_norm": 0.0096435546875, |
| "learning_rate": 7.624444444444445e-06, |
| "loss": 3.4292, |
| "mean_token_accuracy": 0.42965292911976577, |
| "num_tokens": 87293488.0, |
| "step": 6570 |
| }, |
| { |
| "entropy": 2.571312975138426, |
| "epoch": 5.854615256217659, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 7.602222222222223e-06, |
| "loss": 3.5246, |
| "mean_token_accuracy": 0.4220321110449731, |
| "num_tokens": 87424228.0, |
| "step": 6580 |
| }, |
| { |
| "entropy": 2.4508534274995326, |
| "epoch": 5.863517498469927, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 7.58e-06, |
| "loss": 3.3452, |
| "mean_token_accuracy": 0.438654142152518, |
| "num_tokens": 87559509.0, |
| "step": 6590 |
| }, |
| { |
| "entropy": 2.442332284525037, |
| "epoch": 5.8724197407221945, |
| "grad_norm": 0.01434326171875, |
| "learning_rate": 7.557777777777779e-06, |
| "loss": 3.3969, |
| "mean_token_accuracy": 0.433063118904829, |
| "num_tokens": 87691978.0, |
| "step": 6600 |
| }, |
| { |
| "epoch": 5.8724197407221945, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 87691978.0, |
| "eval_coding_runtime": 96.0615, |
| "eval_coding_samples_per_second": 5.205, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 6600 |
| }, |
| { |
| "epoch": 5.8724197407221945, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 87691978.0, |
| "eval_chemistry_runtime": 53.8034, |
| "eval_chemistry_samples_per_second": 9.293, |
| "eval_chemistry_steps_per_second": 4.647, |
| "step": 6600 |
| }, |
| { |
| "entropy": 2.4951433286070825, |
| "epoch": 5.881321982974462, |
| "grad_norm": 0.009765625, |
| "learning_rate": 7.535555555555556e-06, |
| "loss": 3.4058, |
| "mean_token_accuracy": 0.43170636221766473, |
| "num_tokens": 87822328.0, |
| "step": 6610 |
| }, |
| { |
| "entropy": 2.5601726777851583, |
| "epoch": 5.890224225226729, |
| "grad_norm": 0.010009765625, |
| "learning_rate": 7.513333333333334e-06, |
| "loss": 3.4955, |
| "mean_token_accuracy": 0.4233979659155011, |
| "num_tokens": 87954735.0, |
| "step": 6620 |
| }, |
| { |
| "entropy": 2.4996988527476787, |
| "epoch": 5.899126467478997, |
| "grad_norm": 0.01080322265625, |
| "learning_rate": 7.4911111111111116e-06, |
| "loss": 3.4099, |
| "mean_token_accuracy": 0.43844735734164714, |
| "num_tokens": 88088134.0, |
| "step": 6630 |
| }, |
| { |
| "entropy": 2.550193040072918, |
| "epoch": 5.908028709731264, |
| "grad_norm": 0.01336669921875, |
| "learning_rate": 7.46888888888889e-06, |
| "loss": 3.4842, |
| "mean_token_accuracy": 0.42592169158160686, |
| "num_tokens": 88221722.0, |
| "step": 6640 |
| }, |
| { |
| "entropy": 2.5889481589198113, |
| "epoch": 5.916930951983531, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 7.446666666666668e-06, |
| "loss": 3.5807, |
| "mean_token_accuracy": 0.41430754475295545, |
| "num_tokens": 88350216.0, |
| "step": 6650 |
| }, |
| { |
| "entropy": 2.421783010661602, |
| "epoch": 5.925833194235798, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 7.424444444444445e-06, |
| "loss": 3.3262, |
| "mean_token_accuracy": 0.43862591590732336, |
| "num_tokens": 88485541.0, |
| "step": 6660 |
| }, |
| { |
| "entropy": 2.4927210204303263, |
| "epoch": 5.934735436488065, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 7.402222222222223e-06, |
| "loss": 3.3737, |
| "mean_token_accuracy": 0.43310569478198885, |
| "num_tokens": 88621410.0, |
| "step": 6670 |
| }, |
| { |
| "entropy": 2.379143285006285, |
| "epoch": 5.943637678740333, |
| "grad_norm": 0.01007080078125, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 3.2598, |
| "mean_token_accuracy": 0.44978996366262436, |
| "num_tokens": 88759175.0, |
| "step": 6680 |
| }, |
| { |
| "entropy": 2.555007755756378, |
| "epoch": 5.9525399209926, |
| "grad_norm": 0.01251220703125, |
| "learning_rate": 7.357777777777778e-06, |
| "loss": 3.5111, |
| "mean_token_accuracy": 0.4206316273659468, |
| "num_tokens": 88892113.0, |
| "step": 6690 |
| }, |
| { |
| "entropy": 2.4999193131923674, |
| "epoch": 5.961442163244867, |
| "grad_norm": 0.01263427734375, |
| "learning_rate": 7.335555555555556e-06, |
| "loss": 3.4427, |
| "mean_token_accuracy": 0.4279798803851008, |
| "num_tokens": 89022141.0, |
| "step": 6700 |
| }, |
| { |
| "epoch": 5.961442163244867, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 89022141.0, |
| "eval_coding_runtime": 96.0411, |
| "eval_coding_samples_per_second": 5.206, |
| "eval_coding_steps_per_second": 2.603, |
| "step": 6700 |
| }, |
| { |
| "epoch": 5.961442163244867, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 89022141.0, |
| "eval_chemistry_runtime": 53.8922, |
| "eval_chemistry_samples_per_second": 9.278, |
| "eval_chemistry_steps_per_second": 4.639, |
| "step": 6700 |
| }, |
| { |
| "entropy": 2.466172583401203, |
| "epoch": 5.970344405497134, |
| "grad_norm": 0.008056640625, |
| "learning_rate": 7.313333333333333e-06, |
| "loss": 3.3604, |
| "mean_token_accuracy": 0.43419276885688307, |
| "num_tokens": 89155157.0, |
| "step": 6710 |
| }, |
| { |
| "entropy": 2.4998693864792587, |
| "epoch": 5.9792466477494015, |
| "grad_norm": 0.01019287109375, |
| "learning_rate": 7.291111111111112e-06, |
| "loss": 3.4258, |
| "mean_token_accuracy": 0.4315438823774457, |
| "num_tokens": 89284373.0, |
| "step": 6720 |
| }, |
| { |
| "entropy": 2.5144196592271326, |
| "epoch": 5.9881488900016695, |
| "grad_norm": 0.0111083984375, |
| "learning_rate": 7.268888888888889e-06, |
| "loss": 3.4709, |
| "mean_token_accuracy": 0.4295299300923944, |
| "num_tokens": 89417047.0, |
| "step": 6730 |
| }, |
| { |
| "entropy": 2.532013650983572, |
| "epoch": 5.997051132253937, |
| "grad_norm": 0.0076904296875, |
| "learning_rate": 7.246666666666667e-06, |
| "loss": 3.4134, |
| "mean_token_accuracy": 0.4276865027844906, |
| "num_tokens": 89549854.0, |
| "step": 6740 |
| }, |
| { |
| "entropy": 2.498489917524709, |
| "epoch": 6.00534134535136, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 7.224444444444445e-06, |
| "loss": 3.4195, |
| "mean_token_accuracy": 0.42868580994189986, |
| "num_tokens": 89671858.0, |
| "step": 6750 |
| }, |
| { |
| "entropy": 2.602482482790947, |
| "epoch": 6.0142435876036275, |
| "grad_norm": 0.0179443359375, |
| "learning_rate": 7.202222222222222e-06, |
| "loss": 3.5308, |
| "mean_token_accuracy": 0.4163219309411943, |
| "num_tokens": 89801780.0, |
| "step": 6760 |
| }, |
| { |
| "entropy": 2.4988586612045767, |
| "epoch": 6.023145829855895, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 3.435, |
| "mean_token_accuracy": 0.4281375538557768, |
| "num_tokens": 89936329.0, |
| "step": 6770 |
| }, |
| { |
| "entropy": 2.4863852575421332, |
| "epoch": 6.032048072108163, |
| "grad_norm": 0.0078125, |
| "learning_rate": 7.157777777777778e-06, |
| "loss": 3.411, |
| "mean_token_accuracy": 0.42743010092526673, |
| "num_tokens": 90074675.0, |
| "step": 6780 |
| }, |
| { |
| "entropy": 2.4680812403559687, |
| "epoch": 6.04095031436043, |
| "grad_norm": 0.0107421875, |
| "learning_rate": 7.135555555555557e-06, |
| "loss": 3.3645, |
| "mean_token_accuracy": 0.43903659004718065, |
| "num_tokens": 90206586.0, |
| "step": 6790 |
| }, |
| { |
| "entropy": 2.492842423170805, |
| "epoch": 6.049852556612697, |
| "grad_norm": 0.0157470703125, |
| "learning_rate": 7.113333333333334e-06, |
| "loss": 3.4434, |
| "mean_token_accuracy": 0.428391483053565, |
| "num_tokens": 90341380.0, |
| "step": 6800 |
| }, |
| { |
| "epoch": 6.049852556612697, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 90341380.0, |
| "eval_coding_runtime": 96.0362, |
| "eval_coding_samples_per_second": 5.206, |
| "eval_coding_steps_per_second": 2.603, |
| "step": 6800 |
| }, |
| { |
| "epoch": 6.049852556612697, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 90341380.0, |
| "eval_chemistry_runtime": 53.7841, |
| "eval_chemistry_samples_per_second": 9.296, |
| "eval_chemistry_steps_per_second": 4.648, |
| "step": 6800 |
| }, |
| { |
| "entropy": 2.464235895872116, |
| "epoch": 6.058754798864964, |
| "grad_norm": 0.0086669921875, |
| "learning_rate": 7.091111111111112e-06, |
| "loss": 3.3873, |
| "mean_token_accuracy": 0.43350021336227657, |
| "num_tokens": 90472190.0, |
| "step": 6810 |
| }, |
| { |
| "entropy": 2.4852027095854283, |
| "epoch": 6.067657041117231, |
| "grad_norm": 0.01348876953125, |
| "learning_rate": 7.06888888888889e-06, |
| "loss": 3.39, |
| "mean_token_accuracy": 0.43165316954255106, |
| "num_tokens": 90606763.0, |
| "step": 6820 |
| }, |
| { |
| "entropy": 2.4813057355582715, |
| "epoch": 6.076559283369499, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 7.046666666666667e-06, |
| "loss": 3.3502, |
| "mean_token_accuracy": 0.4323008943349123, |
| "num_tokens": 90739642.0, |
| "step": 6830 |
| }, |
| { |
| "entropy": 2.5161099396646023, |
| "epoch": 6.085461525621766, |
| "grad_norm": 0.01214599609375, |
| "learning_rate": 7.024444444444445e-06, |
| "loss": 3.4332, |
| "mean_token_accuracy": 0.4263143142685294, |
| "num_tokens": 90873649.0, |
| "step": 6840 |
| }, |
| { |
| "entropy": 2.464642623066902, |
| "epoch": 6.094363767874033, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 7.0022222222222225e-06, |
| "loss": 3.4035, |
| "mean_token_accuracy": 0.4362926648929715, |
| "num_tokens": 91004431.0, |
| "step": 6850 |
| }, |
| { |
| "entropy": 2.464748837053776, |
| "epoch": 6.1032660101263, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 6.98e-06, |
| "loss": 3.3865, |
| "mean_token_accuracy": 0.43288208059966565, |
| "num_tokens": 91135503.0, |
| "step": 6860 |
| }, |
| { |
| "entropy": 2.4363988876342773, |
| "epoch": 6.112168252378567, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 6.9577777777777785e-06, |
| "loss": 3.3665, |
| "mean_token_accuracy": 0.4368050819262862, |
| "num_tokens": 91273603.0, |
| "step": 6870 |
| }, |
| { |
| "entropy": 2.4948987126350404, |
| "epoch": 6.121070494630835, |
| "grad_norm": 0.0106201171875, |
| "learning_rate": 6.935555555555556e-06, |
| "loss": 3.3735, |
| "mean_token_accuracy": 0.43682076446712015, |
| "num_tokens": 91406138.0, |
| "step": 6880 |
| }, |
| { |
| "entropy": 2.546030812710524, |
| "epoch": 6.1299727368831025, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 6.913333333333334e-06, |
| "loss": 3.4874, |
| "mean_token_accuracy": 0.4242988735437393, |
| "num_tokens": 91534665.0, |
| "step": 6890 |
| }, |
| { |
| "entropy": 2.552320735156536, |
| "epoch": 6.13887497913537, |
| "grad_norm": 0.01025390625, |
| "learning_rate": 6.891111111111111e-06, |
| "loss": 3.4824, |
| "mean_token_accuracy": 0.4236021380871534, |
| "num_tokens": 91666764.0, |
| "step": 6900 |
| }, |
| { |
| "epoch": 6.13887497913537, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 91666764.0, |
| "eval_coding_runtime": 96.2862, |
| "eval_coding_samples_per_second": 5.193, |
| "eval_coding_steps_per_second": 2.596, |
| "step": 6900 |
| }, |
| { |
| "epoch": 6.13887497913537, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 91666764.0, |
| "eval_chemistry_runtime": 53.7618, |
| "eval_chemistry_samples_per_second": 9.3, |
| "eval_chemistry_steps_per_second": 4.65, |
| "step": 6900 |
| }, |
| { |
| "entropy": 2.3952490091323853, |
| "epoch": 6.147777221387637, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 6.868888888888889e-06, |
| "loss": 3.2876, |
| "mean_token_accuracy": 0.44642102662473915, |
| "num_tokens": 91802264.0, |
| "step": 6910 |
| }, |
| { |
| "entropy": 2.457097998261452, |
| "epoch": 6.156679463639905, |
| "grad_norm": 0.014892578125, |
| "learning_rate": 6.846666666666667e-06, |
| "loss": 3.3808, |
| "mean_token_accuracy": 0.4335661603137851, |
| "num_tokens": 91930098.0, |
| "step": 6920 |
| }, |
| { |
| "entropy": 2.4704342782497406, |
| "epoch": 6.165581705892172, |
| "grad_norm": 0.0103759765625, |
| "learning_rate": 6.824444444444444e-06, |
| "loss": 3.3542, |
| "mean_token_accuracy": 0.43675631619989874, |
| "num_tokens": 92062681.0, |
| "step": 6930 |
| }, |
| { |
| "entropy": 2.455590482801199, |
| "epoch": 6.174483948144439, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 6.802222222222223e-06, |
| "loss": 3.392, |
| "mean_token_accuracy": 0.437709685228765, |
| "num_tokens": 92203160.0, |
| "step": 6940 |
| }, |
| { |
| "entropy": 2.484208692610264, |
| "epoch": 6.183386190396706, |
| "grad_norm": 0.00762939453125, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 3.4201, |
| "mean_token_accuracy": 0.433217691257596, |
| "num_tokens": 92331881.0, |
| "step": 6950 |
| }, |
| { |
| "entropy": 2.4681537471711636, |
| "epoch": 6.192288432648973, |
| "grad_norm": 0.007781982421875, |
| "learning_rate": 6.757777777777779e-06, |
| "loss": 3.3539, |
| "mean_token_accuracy": 0.43540666438639164, |
| "num_tokens": 92462984.0, |
| "step": 6960 |
| }, |
| { |
| "entropy": 2.5262075044214725, |
| "epoch": 6.201190674901241, |
| "grad_norm": 0.00860595703125, |
| "learning_rate": 6.735555555555556e-06, |
| "loss": 3.4468, |
| "mean_token_accuracy": 0.4259802566841245, |
| "num_tokens": 92592863.0, |
| "step": 6970 |
| }, |
| { |
| "entropy": 2.5219677701592444, |
| "epoch": 6.210092917153508, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 6.713333333333334e-06, |
| "loss": 3.4365, |
| "mean_token_accuracy": 0.4248078852891922, |
| "num_tokens": 92723682.0, |
| "step": 6980 |
| }, |
| { |
| "entropy": 2.4787024706602097, |
| "epoch": 6.218995159405775, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 6.691111111111112e-06, |
| "loss": 3.3779, |
| "mean_token_accuracy": 0.43126047775149345, |
| "num_tokens": 92860099.0, |
| "step": 6990 |
| }, |
| { |
| "entropy": 2.4372297123074533, |
| "epoch": 6.227897401658042, |
| "grad_norm": 0.00909423828125, |
| "learning_rate": 6.668888888888889e-06, |
| "loss": 3.3133, |
| "mean_token_accuracy": 0.43854769729077814, |
| "num_tokens": 92990269.0, |
| "step": 7000 |
| }, |
| { |
| "epoch": 6.227897401658042, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 92990269.0, |
| "eval_coding_runtime": 96.0299, |
| "eval_coding_samples_per_second": 5.207, |
| "eval_coding_steps_per_second": 2.603, |
| "step": 7000 |
| }, |
| { |
| "epoch": 6.227897401658042, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 92990269.0, |
| "eval_chemistry_runtime": 53.6629, |
| "eval_chemistry_samples_per_second": 9.317, |
| "eval_chemistry_steps_per_second": 4.659, |
| "step": 7000 |
| }, |
| { |
| "entropy": 2.483369851857424, |
| "epoch": 6.2367996439103095, |
| "grad_norm": 0.01953125, |
| "learning_rate": 6.646666666666667e-06, |
| "loss": 3.4276, |
| "mean_token_accuracy": 0.4321822114288807, |
| "num_tokens": 93121907.0, |
| "step": 7010 |
| }, |
| { |
| "entropy": 2.491847258806229, |
| "epoch": 6.2457018861625775, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 6.6244444444444445e-06, |
| "loss": 3.3971, |
| "mean_token_accuracy": 0.42967430241405963, |
| "num_tokens": 93255976.0, |
| "step": 7020 |
| }, |
| { |
| "entropy": 2.518848204612732, |
| "epoch": 6.254604128414845, |
| "grad_norm": 0.01416015625, |
| "learning_rate": 6.602222222222223e-06, |
| "loss": 3.4225, |
| "mean_token_accuracy": 0.43131844196468594, |
| "num_tokens": 93392677.0, |
| "step": 7030 |
| }, |
| { |
| "entropy": 2.472347294539213, |
| "epoch": 6.263506370667112, |
| "grad_norm": 0.00921630859375, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 3.3946, |
| "mean_token_accuracy": 0.4363216387107968, |
| "num_tokens": 93525783.0, |
| "step": 7040 |
| }, |
| { |
| "entropy": 2.5548721224069597, |
| "epoch": 6.272408612919379, |
| "grad_norm": 0.0126953125, |
| "learning_rate": 6.557777777777778e-06, |
| "loss": 3.5132, |
| "mean_token_accuracy": 0.4248511105775833, |
| "num_tokens": 93657444.0, |
| "step": 7050 |
| }, |
| { |
| "entropy": 2.4779123783111574, |
| "epoch": 6.281310855171647, |
| "grad_norm": 0.016357421875, |
| "learning_rate": 6.535555555555556e-06, |
| "loss": 3.4119, |
| "mean_token_accuracy": 0.4322746017947793, |
| "num_tokens": 93785838.0, |
| "step": 7060 |
| }, |
| { |
| "entropy": 2.4434043914079666, |
| "epoch": 6.290213097423914, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 6.513333333333333e-06, |
| "loss": 3.3354, |
| "mean_token_accuracy": 0.43630710542201995, |
| "num_tokens": 93922422.0, |
| "step": 7070 |
| }, |
| { |
| "entropy": 2.537347627431154, |
| "epoch": 6.299115339676181, |
| "grad_norm": 0.007598876953125, |
| "learning_rate": 6.491111111111111e-06, |
| "loss": 3.4671, |
| "mean_token_accuracy": 0.4240851294249296, |
| "num_tokens": 94057453.0, |
| "step": 7080 |
| }, |
| { |
| "entropy": 2.5193082474172117, |
| "epoch": 6.308017581928448, |
| "grad_norm": 0.0087890625, |
| "learning_rate": 6.468888888888889e-06, |
| "loss": 3.4713, |
| "mean_token_accuracy": 0.4296780854463577, |
| "num_tokens": 94192949.0, |
| "step": 7090 |
| }, |
| { |
| "entropy": 2.4920457646250727, |
| "epoch": 6.316919824180715, |
| "grad_norm": 0.0115966796875, |
| "learning_rate": 6.446666666666668e-06, |
| "loss": 3.4091, |
| "mean_token_accuracy": 0.4310237796977162, |
| "num_tokens": 94324885.0, |
| "step": 7100 |
| }, |
| { |
| "epoch": 6.316919824180715, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 94324885.0, |
| "eval_coding_runtime": 96.1202, |
| "eval_coding_samples_per_second": 5.202, |
| "eval_coding_steps_per_second": 2.601, |
| "step": 7100 |
| }, |
| { |
| "epoch": 6.316919824180715, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 94324885.0, |
| "eval_chemistry_runtime": 53.9836, |
| "eval_chemistry_samples_per_second": 9.262, |
| "eval_chemistry_steps_per_second": 4.631, |
| "step": 7100 |
| }, |
| { |
| "entropy": 2.5310638435184956, |
| "epoch": 6.325822066432983, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 6.4244444444444455e-06, |
| "loss": 3.4547, |
| "mean_token_accuracy": 0.4299845080822706, |
| "num_tokens": 94450990.0, |
| "step": 7110 |
| }, |
| { |
| "entropy": 2.481987812370062, |
| "epoch": 6.33472430868525, |
| "grad_norm": 0.00726318359375, |
| "learning_rate": 6.402222222222223e-06, |
| "loss": 3.4458, |
| "mean_token_accuracy": 0.4303594870492816, |
| "num_tokens": 94584487.0, |
| "step": 7120 |
| }, |
| { |
| "entropy": 2.538077312707901, |
| "epoch": 6.343626550937517, |
| "grad_norm": 0.007171630859375, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 3.4017, |
| "mean_token_accuracy": 0.43017367608845236, |
| "num_tokens": 94720088.0, |
| "step": 7130 |
| }, |
| { |
| "entropy": 2.456275135278702, |
| "epoch": 6.3525287931897845, |
| "grad_norm": 0.0089111328125, |
| "learning_rate": 6.357777777777778e-06, |
| "loss": 3.4158, |
| "mean_token_accuracy": 0.4358561382628977, |
| "num_tokens": 94852872.0, |
| "step": 7140 |
| }, |
| { |
| "entropy": 2.4109351888298987, |
| "epoch": 6.361431035442052, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 6.335555555555556e-06, |
| "loss": 3.3178, |
| "mean_token_accuracy": 0.4392677407711744, |
| "num_tokens": 94982813.0, |
| "step": 7150 |
| }, |
| { |
| "entropy": 2.4356062449514866, |
| "epoch": 6.37033327769432, |
| "grad_norm": 0.017578125, |
| "learning_rate": 6.313333333333334e-06, |
| "loss": 3.3377, |
| "mean_token_accuracy": 0.4346992772072554, |
| "num_tokens": 95118997.0, |
| "step": 7160 |
| }, |
| { |
| "entropy": 2.5708038546144962, |
| "epoch": 6.379235519946587, |
| "grad_norm": 0.01446533203125, |
| "learning_rate": 6.291111111111111e-06, |
| "loss": 3.463, |
| "mean_token_accuracy": 0.421726209577173, |
| "num_tokens": 95257483.0, |
| "step": 7170 |
| }, |
| { |
| "entropy": 2.5586150221526625, |
| "epoch": 6.388137762198854, |
| "grad_norm": 0.0093994140625, |
| "learning_rate": 6.26888888888889e-06, |
| "loss": 3.4739, |
| "mean_token_accuracy": 0.4253166550770402, |
| "num_tokens": 95389414.0, |
| "step": 7180 |
| }, |
| { |
| "entropy": 2.5243496797978877, |
| "epoch": 6.397040004451121, |
| "grad_norm": 0.01019287109375, |
| "learning_rate": 6.246666666666667e-06, |
| "loss": 3.5097, |
| "mean_token_accuracy": 0.421814835537225, |
| "num_tokens": 95519502.0, |
| "step": 7190 |
| }, |
| { |
| "entropy": 2.464592805504799, |
| "epoch": 6.405942246703388, |
| "grad_norm": 0.0081787109375, |
| "learning_rate": 6.224444444444445e-06, |
| "loss": 3.338, |
| "mean_token_accuracy": 0.4362541016191244, |
| "num_tokens": 95652357.0, |
| "step": 7200 |
| }, |
| { |
| "epoch": 6.405942246703388, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 95652357.0, |
| "eval_coding_runtime": 96.1388, |
| "eval_coding_samples_per_second": 5.201, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 7200 |
| }, |
| { |
| "epoch": 6.405942246703388, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 95652357.0, |
| "eval_chemistry_runtime": 53.8512, |
| "eval_chemistry_samples_per_second": 9.285, |
| "eval_chemistry_steps_per_second": 4.642, |
| "step": 7200 |
| }, |
| { |
| "entropy": 2.532952569797635, |
| "epoch": 6.414844488955656, |
| "grad_norm": 0.009033203125, |
| "learning_rate": 6.2022222222222225e-06, |
| "loss": 3.4453, |
| "mean_token_accuracy": 0.42813959503546356, |
| "num_tokens": 95782859.0, |
| "step": 7210 |
| }, |
| { |
| "entropy": 2.50785236954689, |
| "epoch": 6.423746731207923, |
| "grad_norm": 0.0076904296875, |
| "learning_rate": 6.18e-06, |
| "loss": 3.4139, |
| "mean_token_accuracy": 0.42927719466388226, |
| "num_tokens": 95918079.0, |
| "step": 7220 |
| }, |
| { |
| "entropy": 2.5400012150406837, |
| "epoch": 6.43264897346019, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 6.157777777777778e-06, |
| "loss": 3.4913, |
| "mean_token_accuracy": 0.4187105377204716, |
| "num_tokens": 96040641.0, |
| "step": 7230 |
| }, |
| { |
| "entropy": 2.5495525658130647, |
| "epoch": 6.441551215712457, |
| "grad_norm": 0.01165771484375, |
| "learning_rate": 6.135555555555555e-06, |
| "loss": 3.505, |
| "mean_token_accuracy": 0.4231853854842484, |
| "num_tokens": 96171926.0, |
| "step": 7240 |
| }, |
| { |
| "entropy": 2.5429927080869676, |
| "epoch": 6.450453457964725, |
| "grad_norm": 0.006622314453125, |
| "learning_rate": 6.113333333333333e-06, |
| "loss": 3.4363, |
| "mean_token_accuracy": 0.4289200007915497, |
| "num_tokens": 96302056.0, |
| "step": 7250 |
| }, |
| { |
| "entropy": 2.4564895257353783, |
| "epoch": 6.459355700216992, |
| "grad_norm": 0.0084228515625, |
| "learning_rate": 6.091111111111112e-06, |
| "loss": 3.3177, |
| "mean_token_accuracy": 0.4375461803749204, |
| "num_tokens": 96440327.0, |
| "step": 7260 |
| }, |
| { |
| "entropy": 2.472339541465044, |
| "epoch": 6.4682579424692594, |
| "grad_norm": 0.0103759765625, |
| "learning_rate": 6.06888888888889e-06, |
| "loss": 3.3745, |
| "mean_token_accuracy": 0.43724719900637865, |
| "num_tokens": 96578594.0, |
| "step": 7270 |
| }, |
| { |
| "entropy": 2.528899458050728, |
| "epoch": 6.4771601847215265, |
| "grad_norm": 0.0079345703125, |
| "learning_rate": 6.0466666666666675e-06, |
| "loss": 3.4088, |
| "mean_token_accuracy": 0.42949474006891253, |
| "num_tokens": 96714394.0, |
| "step": 7280 |
| }, |
| { |
| "entropy": 2.5160689525306226, |
| "epoch": 6.486062426973794, |
| "grad_norm": 0.009765625, |
| "learning_rate": 6.024444444444445e-06, |
| "loss": 3.4235, |
| "mean_token_accuracy": 0.43323270697146654, |
| "num_tokens": 96851878.0, |
| "step": 7290 |
| }, |
| { |
| "entropy": 2.50460671633482, |
| "epoch": 6.494964669226062, |
| "grad_norm": 0.0089111328125, |
| "learning_rate": 6.002222222222223e-06, |
| "loss": 3.4287, |
| "mean_token_accuracy": 0.43280349038541316, |
| "num_tokens": 96984616.0, |
| "step": 7300 |
| }, |
| { |
| "epoch": 6.494964669226062, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 96984616.0, |
| "eval_coding_runtime": 96.1397, |
| "eval_coding_samples_per_second": 5.201, |
| "eval_coding_steps_per_second": 2.6, |
| "step": 7300 |
| }, |
| { |
| "epoch": 6.494964669226062, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 96984616.0, |
| "eval_chemistry_runtime": 53.8052, |
| "eval_chemistry_samples_per_second": 9.293, |
| "eval_chemistry_steps_per_second": 4.646, |
| "step": 7300 |
| }, |
| { |
| "entropy": 2.4880875654518606, |
| "epoch": 6.503866911478329, |
| "grad_norm": 0.012939453125, |
| "learning_rate": 5.98e-06, |
| "loss": 3.4382, |
| "mean_token_accuracy": 0.4333686189725995, |
| "num_tokens": 97116492.0, |
| "step": 7310 |
| }, |
| { |
| "entropy": 2.5349057361483576, |
| "epoch": 6.512769153730596, |
| "grad_norm": 0.013427734375, |
| "learning_rate": 5.957777777777778e-06, |
| "loss": 3.4585, |
| "mean_token_accuracy": 0.425345850456506, |
| "num_tokens": 97252268.0, |
| "step": 7320 |
| }, |
| { |
| "entropy": 2.4609145797789096, |
| "epoch": 6.521671395982863, |
| "grad_norm": 0.007232666015625, |
| "learning_rate": 5.9355555555555556e-06, |
| "loss": 3.3552, |
| "mean_token_accuracy": 0.4344397310167551, |
| "num_tokens": 97383736.0, |
| "step": 7330 |
| }, |
| { |
| "entropy": 2.463261793553829, |
| "epoch": 6.530573638235131, |
| "grad_norm": 0.0076904296875, |
| "learning_rate": 5.913333333333334e-06, |
| "loss": 3.4206, |
| "mean_token_accuracy": 0.4310577527619898, |
| "num_tokens": 97518049.0, |
| "step": 7340 |
| }, |
| { |
| "entropy": 2.411273017525673, |
| "epoch": 6.539475880487398, |
| "grad_norm": 0.0086669921875, |
| "learning_rate": 5.891111111111112e-06, |
| "loss": 3.3492, |
| "mean_token_accuracy": 0.43940184358507395, |
| "num_tokens": 97651794.0, |
| "step": 7350 |
| }, |
| { |
| "entropy": 2.4915858305990697, |
| "epoch": 6.548378122739665, |
| "grad_norm": 0.01434326171875, |
| "learning_rate": 5.868888888888889e-06, |
| "loss": 3.4, |
| "mean_token_accuracy": 0.43103193417191504, |
| "num_tokens": 97787713.0, |
| "step": 7360 |
| }, |
| { |
| "entropy": 2.5162147298455237, |
| "epoch": 6.557280364991932, |
| "grad_norm": 0.0172119140625, |
| "learning_rate": 5.846666666666667e-06, |
| "loss": 3.429, |
| "mean_token_accuracy": 0.4287140166386962, |
| "num_tokens": 97917928.0, |
| "step": 7370 |
| }, |
| { |
| "entropy": 2.4405299685895443, |
| "epoch": 6.566182607244199, |
| "grad_norm": 0.008544921875, |
| "learning_rate": 5.8244444444444445e-06, |
| "loss": 3.3506, |
| "mean_token_accuracy": 0.4359043904580176, |
| "num_tokens": 98049388.0, |
| "step": 7380 |
| }, |
| { |
| "entropy": 2.499612495303154, |
| "epoch": 6.575084849496466, |
| "grad_norm": 0.0133056640625, |
| "learning_rate": 5.802222222222222e-06, |
| "loss": 3.4628, |
| "mean_token_accuracy": 0.4284491116181016, |
| "num_tokens": 98176301.0, |
| "step": 7390 |
| }, |
| { |
| "entropy": 2.4712293349206447, |
| "epoch": 6.583987091748734, |
| "grad_norm": 0.007720947265625, |
| "learning_rate": 5.78e-06, |
| "loss": 3.3879, |
| "mean_token_accuracy": 0.4354086763225496, |
| "num_tokens": 98312063.0, |
| "step": 7400 |
| }, |
| { |
| "epoch": 6.583987091748734, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 98312063.0, |
| "eval_coding_runtime": 96.0453, |
| "eval_coding_samples_per_second": 5.206, |
| "eval_coding_steps_per_second": 2.603, |
| "step": 7400 |
| }, |
| { |
| "epoch": 6.583987091748734, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 98312063.0, |
| "eval_chemistry_runtime": 53.8225, |
| "eval_chemistry_samples_per_second": 9.29, |
| "eval_chemistry_steps_per_second": 4.645, |
| "step": 7400 |
| }, |
| { |
| "entropy": 2.5102811083197594, |
| "epoch": 6.5928893340010015, |
| "grad_norm": 0.0091552734375, |
| "learning_rate": 5.757777777777779e-06, |
| "loss": 3.4789, |
| "mean_token_accuracy": 0.4250242728739977, |
| "num_tokens": 98439856.0, |
| "step": 7410 |
| }, |
| { |
| "entropy": 2.4445991814136505, |
| "epoch": 6.601791576253269, |
| "grad_norm": 0.0068359375, |
| "learning_rate": 5.735555555555557e-06, |
| "loss": 3.39, |
| "mean_token_accuracy": 0.43826481867581607, |
| "num_tokens": 98579402.0, |
| "step": 7420 |
| }, |
| { |
| "entropy": 2.4795279905200003, |
| "epoch": 6.610693818505536, |
| "grad_norm": 0.00933837890625, |
| "learning_rate": 5.713333333333334e-06, |
| "loss": 3.3835, |
| "mean_token_accuracy": 0.43464959431439637, |
| "num_tokens": 98712611.0, |
| "step": 7430 |
| }, |
| { |
| "entropy": 2.5811775140464306, |
| "epoch": 6.619596060757804, |
| "grad_norm": 0.0076904296875, |
| "learning_rate": 5.691111111111112e-06, |
| "loss": 3.5088, |
| "mean_token_accuracy": 0.42061771992594005, |
| "num_tokens": 98850798.0, |
| "step": 7440 |
| }, |
| { |
| "entropy": 2.4562979079782963, |
| "epoch": 6.628498303010071, |
| "grad_norm": 0.00927734375, |
| "learning_rate": 5.6688888888888895e-06, |
| "loss": 3.368, |
| "mean_token_accuracy": 0.43350073117762805, |
| "num_tokens": 98984028.0, |
| "step": 7450 |
| }, |
| { |
| "entropy": 2.5175287291407584, |
| "epoch": 6.637400545262338, |
| "grad_norm": 0.01190185546875, |
| "learning_rate": 5.646666666666667e-06, |
| "loss": 3.4521, |
| "mean_token_accuracy": 0.4270141161978245, |
| "num_tokens": 99111643.0, |
| "step": 7460 |
| }, |
| { |
| "entropy": 2.5121711291372777, |
| "epoch": 6.646302787514605, |
| "grad_norm": 0.00946044921875, |
| "learning_rate": 5.624444444444445e-06, |
| "loss": 3.42, |
| "mean_token_accuracy": 0.4303737297654152, |
| "num_tokens": 99242289.0, |
| "step": 7470 |
| }, |
| { |
| "entropy": 2.450058939307928, |
| "epoch": 6.655205029766872, |
| "grad_norm": 0.007568359375, |
| "learning_rate": 5.602222222222222e-06, |
| "loss": 3.386, |
| "mean_token_accuracy": 0.43809804543852804, |
| "num_tokens": 99378974.0, |
| "step": 7480 |
| }, |
| { |
| "entropy": 2.4624539002776147, |
| "epoch": 6.66410727201914, |
| "grad_norm": 0.00762939453125, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 3.3172, |
| "mean_token_accuracy": 0.43682467341423037, |
| "num_tokens": 99517066.0, |
| "step": 7490 |
| }, |
| { |
| "entropy": 2.521768597513437, |
| "epoch": 6.673009514271407, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 5.557777777777778e-06, |
| "loss": 3.4754, |
| "mean_token_accuracy": 0.4253229197114706, |
| "num_tokens": 99648981.0, |
| "step": 7500 |
| }, |
| { |
| "epoch": 6.673009514271407, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 99648981.0, |
| "eval_coding_runtime": 96.1992, |
| "eval_coding_samples_per_second": 5.198, |
| "eval_coding_steps_per_second": 2.599, |
| "step": 7500 |
| }, |
| { |
| "epoch": 6.673009514271407, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 99648981.0, |
| "eval_chemistry_runtime": 53.7841, |
| "eval_chemistry_samples_per_second": 9.296, |
| "eval_chemistry_steps_per_second": 4.648, |
| "step": 7500 |
| }, |
| { |
| "entropy": 2.522602032124996, |
| "epoch": 6.681911756523674, |
| "grad_norm": 0.01611328125, |
| "learning_rate": 5.535555555555556e-06, |
| "loss": 3.4502, |
| "mean_token_accuracy": 0.42392852511256934, |
| "num_tokens": 99778644.0, |
| "step": 7510 |
| }, |
| { |
| "entropy": 2.5007221080362796, |
| "epoch": 6.690813998775941, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 5.513333333333334e-06, |
| "loss": 3.4298, |
| "mean_token_accuracy": 0.42853542100638153, |
| "num_tokens": 99911404.0, |
| "step": 7520 |
| }, |
| { |
| "entropy": 2.447383212298155, |
| "epoch": 6.699716241028209, |
| "grad_norm": 0.00872802734375, |
| "learning_rate": 5.491111111111111e-06, |
| "loss": 3.3504, |
| "mean_token_accuracy": 0.43351489007472993, |
| "num_tokens": 100043200.0, |
| "step": 7530 |
| }, |
| { |
| "entropy": 2.4391328021883965, |
| "epoch": 6.7086184832804765, |
| "grad_norm": 0.00994873046875, |
| "learning_rate": 5.468888888888889e-06, |
| "loss": 3.3782, |
| "mean_token_accuracy": 0.4376953698694706, |
| "num_tokens": 100179539.0, |
| "step": 7540 |
| }, |
| { |
| "entropy": 2.4792590856552126, |
| "epoch": 6.717520725532744, |
| "grad_norm": 0.0078125, |
| "learning_rate": 5.4466666666666665e-06, |
| "loss": 3.4166, |
| "mean_token_accuracy": 0.43456687349826095, |
| "num_tokens": 100312799.0, |
| "step": 7550 |
| }, |
| { |
| "entropy": 2.5195920787751676, |
| "epoch": 6.726422967785011, |
| "grad_norm": 0.010009765625, |
| "learning_rate": 5.424444444444444e-06, |
| "loss": 3.4455, |
| "mean_token_accuracy": 0.42921275170519946, |
| "num_tokens": 100449079.0, |
| "step": 7560 |
| }, |
| { |
| "entropy": 2.588169426470995, |
| "epoch": 6.735325210037278, |
| "grad_norm": 0.01226806640625, |
| "learning_rate": 5.402222222222223e-06, |
| "loss": 3.5392, |
| "mean_token_accuracy": 0.41795885330066085, |
| "num_tokens": 100581705.0, |
| "step": 7570 |
| }, |
| { |
| "entropy": 2.5045555576682093, |
| "epoch": 6.744227452289546, |
| "grad_norm": 0.01220703125, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 3.4323, |
| "mean_token_accuracy": 0.4288202838040888, |
| "num_tokens": 100712718.0, |
| "step": 7580 |
| }, |
| { |
| "entropy": 2.520971208438277, |
| "epoch": 6.753129694541813, |
| "grad_norm": 0.0089111328125, |
| "learning_rate": 5.357777777777779e-06, |
| "loss": 3.4269, |
| "mean_token_accuracy": 0.4290512563660741, |
| "num_tokens": 100845295.0, |
| "step": 7590 |
| }, |
| { |
| "entropy": 2.5159785486757755, |
| "epoch": 6.76203193679408, |
| "grad_norm": 0.0108642578125, |
| "learning_rate": 5.335555555555556e-06, |
| "loss": 3.4727, |
| "mean_token_accuracy": 0.42586620785295964, |
| "num_tokens": 100982382.0, |
| "step": 7600 |
| }, |
| { |
| "epoch": 6.76203193679408, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 100982382.0, |
| "eval_coding_runtime": 96.0903, |
| "eval_coding_samples_per_second": 5.203, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 7600 |
| }, |
| { |
| "epoch": 6.76203193679408, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 100982382.0, |
| "eval_chemistry_runtime": 54.1028, |
| "eval_chemistry_samples_per_second": 9.242, |
| "eval_chemistry_steps_per_second": 4.621, |
| "step": 7600 |
| }, |
| { |
| "entropy": 2.5083722554147245, |
| "epoch": 6.770934179046347, |
| "grad_norm": 0.01458740234375, |
| "learning_rate": 5.313333333333334e-06, |
| "loss": 3.4472, |
| "mean_token_accuracy": 0.42705939766019585, |
| "num_tokens": 101109412.0, |
| "step": 7610 |
| }, |
| { |
| "entropy": 2.5104228369891644, |
| "epoch": 6.779836421298615, |
| "grad_norm": 0.0152587890625, |
| "learning_rate": 5.2911111111111115e-06, |
| "loss": 3.3607, |
| "mean_token_accuracy": 0.4343923058360815, |
| "num_tokens": 101243042.0, |
| "step": 7620 |
| }, |
| { |
| "entropy": 2.529687734693289, |
| "epoch": 6.788738663550882, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 5.268888888888889e-06, |
| "loss": 3.4752, |
| "mean_token_accuracy": 0.42584536513313653, |
| "num_tokens": 101375004.0, |
| "step": 7630 |
| }, |
| { |
| "entropy": 2.4341720774769784, |
| "epoch": 6.797640905803149, |
| "grad_norm": 0.00885009765625, |
| "learning_rate": 5.246666666666667e-06, |
| "loss": 3.3952, |
| "mean_token_accuracy": 0.4374100027605891, |
| "num_tokens": 101510927.0, |
| "step": 7640 |
| }, |
| { |
| "entropy": 2.481144145131111, |
| "epoch": 6.806543148055416, |
| "grad_norm": 0.00994873046875, |
| "learning_rate": 5.224444444444445e-06, |
| "loss": 3.3911, |
| "mean_token_accuracy": 0.4383924352005124, |
| "num_tokens": 101641311.0, |
| "step": 7650 |
| }, |
| { |
| "entropy": 2.5052821427583694, |
| "epoch": 6.8154453903076835, |
| "grad_norm": 0.01153564453125, |
| "learning_rate": 5.202222222222223e-06, |
| "loss": 3.4406, |
| "mean_token_accuracy": 0.4294231103733182, |
| "num_tokens": 101769432.0, |
| "step": 7660 |
| }, |
| { |
| "entropy": 2.490334639698267, |
| "epoch": 6.824347632559951, |
| "grad_norm": 0.007659912109375, |
| "learning_rate": 5.18e-06, |
| "loss": 3.3928, |
| "mean_token_accuracy": 0.43187982104718686, |
| "num_tokens": 101898196.0, |
| "step": 7670 |
| }, |
| { |
| "entropy": 2.504181128740311, |
| "epoch": 6.833249874812219, |
| "grad_norm": 0.014892578125, |
| "learning_rate": 5.157777777777778e-06, |
| "loss": 3.4545, |
| "mean_token_accuracy": 0.43120015701279046, |
| "num_tokens": 102031323.0, |
| "step": 7680 |
| }, |
| { |
| "entropy": 2.471246572583914, |
| "epoch": 6.842152117064486, |
| "grad_norm": 0.00787353515625, |
| "learning_rate": 5.135555555555556e-06, |
| "loss": 3.3435, |
| "mean_token_accuracy": 0.43862386075779797, |
| "num_tokens": 102167828.0, |
| "step": 7690 |
| }, |
| { |
| "entropy": 2.5251905120909215, |
| "epoch": 6.851054359316753, |
| "grad_norm": 0.0091552734375, |
| "learning_rate": 5.113333333333333e-06, |
| "loss": 3.4755, |
| "mean_token_accuracy": 0.426297792699188, |
| "num_tokens": 102296351.0, |
| "step": 7700 |
| }, |
| { |
| "epoch": 6.851054359316753, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 102296351.0, |
| "eval_coding_runtime": 96.0362, |
| "eval_coding_samples_per_second": 5.206, |
| "eval_coding_steps_per_second": 2.603, |
| "step": 7700 |
| }, |
| { |
| "epoch": 6.851054359316753, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 102296351.0, |
| "eval_chemistry_runtime": 53.7985, |
| "eval_chemistry_samples_per_second": 9.294, |
| "eval_chemistry_steps_per_second": 4.647, |
| "step": 7700 |
| }, |
| { |
| "entropy": 2.415869578719139, |
| "epoch": 6.85995660156902, |
| "grad_norm": 0.00897216796875, |
| "learning_rate": 5.091111111111111e-06, |
| "loss": 3.305, |
| "mean_token_accuracy": 0.4405446361750364, |
| "num_tokens": 102428516.0, |
| "step": 7710 |
| }, |
| { |
| "entropy": 2.4899660937488077, |
| "epoch": 6.868858843821288, |
| "grad_norm": 0.007598876953125, |
| "learning_rate": 5.06888888888889e-06, |
| "loss": 3.4196, |
| "mean_token_accuracy": 0.431217847764492, |
| "num_tokens": 102566033.0, |
| "step": 7720 |
| }, |
| { |
| "entropy": 2.4361541725695135, |
| "epoch": 6.877761086073555, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 5.046666666666668e-06, |
| "loss": 3.411, |
| "mean_token_accuracy": 0.4357116429135203, |
| "num_tokens": 102700229.0, |
| "step": 7730 |
| }, |
| { |
| "entropy": 2.4812551006674766, |
| "epoch": 6.886663328325822, |
| "grad_norm": 0.00823974609375, |
| "learning_rate": 5.024444444444445e-06, |
| "loss": 3.392, |
| "mean_token_accuracy": 0.43206119928508996, |
| "num_tokens": 102830839.0, |
| "step": 7740 |
| }, |
| { |
| "entropy": 2.4984117180109022, |
| "epoch": 6.895565570578089, |
| "grad_norm": 0.006561279296875, |
| "learning_rate": 5.002222222222223e-06, |
| "loss": 3.47, |
| "mean_token_accuracy": 0.431227030325681, |
| "num_tokens": 102960370.0, |
| "step": 7750 |
| }, |
| { |
| "entropy": 2.4913737319409845, |
| "epoch": 6.904467812830356, |
| "grad_norm": 0.00750732421875, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 3.4454, |
| "mean_token_accuracy": 0.4266663836315274, |
| "num_tokens": 103091130.0, |
| "step": 7760 |
| }, |
| { |
| "entropy": 2.5031268443912267, |
| "epoch": 6.913370055082624, |
| "grad_norm": 0.0135498046875, |
| "learning_rate": 4.957777777777778e-06, |
| "loss": 3.428, |
| "mean_token_accuracy": 0.4339109744876623, |
| "num_tokens": 103224686.0, |
| "step": 7770 |
| }, |
| { |
| "entropy": 2.490542209893465, |
| "epoch": 6.922272297334891, |
| "grad_norm": 0.0174560546875, |
| "learning_rate": 4.935555555555556e-06, |
| "loss": 3.4388, |
| "mean_token_accuracy": 0.42758956318721175, |
| "num_tokens": 103361015.0, |
| "step": 7780 |
| }, |
| { |
| "entropy": 2.577172574400902, |
| "epoch": 6.9311745395871585, |
| "grad_norm": 0.00872802734375, |
| "learning_rate": 4.9133333333333334e-06, |
| "loss": 3.5359, |
| "mean_token_accuracy": 0.4206360104493797, |
| "num_tokens": 103490753.0, |
| "step": 7790 |
| }, |
| { |
| "entropy": 2.560431795567274, |
| "epoch": 6.940076781839426, |
| "grad_norm": 0.00799560546875, |
| "learning_rate": 4.891111111111111e-06, |
| "loss": 3.4889, |
| "mean_token_accuracy": 0.4257276436313987, |
| "num_tokens": 103619762.0, |
| "step": 7800 |
| }, |
| { |
| "epoch": 6.940076781839426, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 103619762.0, |
| "eval_coding_runtime": 96.0993, |
| "eval_coding_samples_per_second": 5.203, |
| "eval_coding_steps_per_second": 2.601, |
| "step": 7800 |
| }, |
| { |
| "epoch": 6.940076781839426, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 103619762.0, |
| "eval_chemistry_runtime": 53.7993, |
| "eval_chemistry_samples_per_second": 9.294, |
| "eval_chemistry_steps_per_second": 4.647, |
| "step": 7800 |
| }, |
| { |
| "entropy": 2.4733123246580364, |
| "epoch": 6.9489790240916935, |
| "grad_norm": 0.01507568359375, |
| "learning_rate": 4.8688888888888895e-06, |
| "loss": 3.3776, |
| "mean_token_accuracy": 0.435265781916678, |
| "num_tokens": 103756199.0, |
| "step": 7810 |
| }, |
| { |
| "entropy": 2.533934009075165, |
| "epoch": 6.957881266343961, |
| "grad_norm": 0.01953125, |
| "learning_rate": 4.846666666666667e-06, |
| "loss": 3.4883, |
| "mean_token_accuracy": 0.41994120124727485, |
| "num_tokens": 103889266.0, |
| "step": 7820 |
| }, |
| { |
| "entropy": 2.561226685345173, |
| "epoch": 6.966783508596228, |
| "grad_norm": 0.0135498046875, |
| "learning_rate": 4.824444444444445e-06, |
| "loss": 3.5256, |
| "mean_token_accuracy": 0.4213694125413895, |
| "num_tokens": 104019503.0, |
| "step": 7830 |
| }, |
| { |
| "entropy": 2.517245050519705, |
| "epoch": 6.975685750848495, |
| "grad_norm": 0.01422119140625, |
| "learning_rate": 4.802222222222222e-06, |
| "loss": 3.4143, |
| "mean_token_accuracy": 0.42596928421407937, |
| "num_tokens": 104158232.0, |
| "step": 7840 |
| }, |
| { |
| "entropy": 2.5071246303617953, |
| "epoch": 6.984587993100762, |
| "grad_norm": 0.01031494140625, |
| "learning_rate": 4.78e-06, |
| "loss": 3.3799, |
| "mean_token_accuracy": 0.42881283489987254, |
| "num_tokens": 104289595.0, |
| "step": 7850 |
| }, |
| { |
| "entropy": 2.425448203086853, |
| "epoch": 6.993490235353029, |
| "grad_norm": 0.00830078125, |
| "learning_rate": 4.7577777777777784e-06, |
| "loss": 3.3817, |
| "mean_token_accuracy": 0.4351279839873314, |
| "num_tokens": 104426884.0, |
| "step": 7860 |
| }, |
| { |
| "entropy": 2.5756377293759543, |
| "epoch": 7.001780448450454, |
| "grad_norm": 0.0103759765625, |
| "learning_rate": 4.735555555555556e-06, |
| "loss": 3.4305, |
| "mean_token_accuracy": 0.42208054431732867, |
| "num_tokens": 104553776.0, |
| "step": 7870 |
| }, |
| { |
| "entropy": 2.5470064997673036, |
| "epoch": 7.010682690702721, |
| "grad_norm": 0.011474609375, |
| "learning_rate": 4.713333333333334e-06, |
| "loss": 3.4732, |
| "mean_token_accuracy": 0.4272332336753607, |
| "num_tokens": 104680255.0, |
| "step": 7880 |
| }, |
| { |
| "entropy": 2.4954129464924337, |
| "epoch": 7.019584932954988, |
| "grad_norm": 0.00811767578125, |
| "learning_rate": 4.691111111111111e-06, |
| "loss": 3.4048, |
| "mean_token_accuracy": 0.4318564634770155, |
| "num_tokens": 104818307.0, |
| "step": 7890 |
| }, |
| { |
| "entropy": 2.4294852819293737, |
| "epoch": 7.028487175207255, |
| "grad_norm": 0.00946044921875, |
| "learning_rate": 4.66888888888889e-06, |
| "loss": 3.3261, |
| "mean_token_accuracy": 0.4339137837290764, |
| "num_tokens": 104947980.0, |
| "step": 7900 |
| }, |
| { |
| "epoch": 7.028487175207255, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 104947980.0, |
| "eval_coding_runtime": 96.1195, |
| "eval_coding_samples_per_second": 5.202, |
| "eval_coding_steps_per_second": 2.601, |
| "step": 7900 |
| }, |
| { |
| "epoch": 7.028487175207255, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 104947980.0, |
| "eval_chemistry_runtime": 53.9014, |
| "eval_chemistry_samples_per_second": 9.276, |
| "eval_chemistry_steps_per_second": 4.638, |
| "step": 7900 |
| }, |
| { |
| "entropy": 2.540067868679762, |
| "epoch": 7.037389417459522, |
| "grad_norm": 0.009521484375, |
| "learning_rate": 4.646666666666667e-06, |
| "loss": 3.4727, |
| "mean_token_accuracy": 0.42681131660938265, |
| "num_tokens": 105075119.0, |
| "step": 7910 |
| }, |
| { |
| "entropy": 2.489720791578293, |
| "epoch": 7.04629165971179, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 4.624444444444445e-06, |
| "loss": 3.3746, |
| "mean_token_accuracy": 0.43236390585079787, |
| "num_tokens": 105207505.0, |
| "step": 7920 |
| }, |
| { |
| "entropy": 2.4379853047430515, |
| "epoch": 7.055193901964057, |
| "grad_norm": 0.007293701171875, |
| "learning_rate": 4.602222222222223e-06, |
| "loss": 3.4086, |
| "mean_token_accuracy": 0.43374104183167217, |
| "num_tokens": 105341914.0, |
| "step": 7930 |
| }, |
| { |
| "entropy": 2.523585247993469, |
| "epoch": 7.064096144216324, |
| "grad_norm": 0.0074462890625, |
| "learning_rate": 4.58e-06, |
| "loss": 3.4766, |
| "mean_token_accuracy": 0.42782812751829624, |
| "num_tokens": 105468312.0, |
| "step": 7940 |
| }, |
| { |
| "entropy": 2.5453101977705956, |
| "epoch": 7.0729983864685915, |
| "grad_norm": 0.007171630859375, |
| "learning_rate": 4.557777777777778e-06, |
| "loss": 3.4975, |
| "mean_token_accuracy": 0.4227051706984639, |
| "num_tokens": 105601514.0, |
| "step": 7950 |
| }, |
| { |
| "entropy": 2.460656537115574, |
| "epoch": 7.0819006287208595, |
| "grad_norm": 0.006988525390625, |
| "learning_rate": 4.535555555555555e-06, |
| "loss": 3.3504, |
| "mean_token_accuracy": 0.43799448730424045, |
| "num_tokens": 105733015.0, |
| "step": 7960 |
| }, |
| { |
| "entropy": 2.4749496310949324, |
| "epoch": 7.090802870973127, |
| "grad_norm": 0.00848388671875, |
| "learning_rate": 4.513333333333333e-06, |
| "loss": 3.4108, |
| "mean_token_accuracy": 0.4327717902138829, |
| "num_tokens": 105863056.0, |
| "step": 7970 |
| }, |
| { |
| "entropy": 2.4818050473928452, |
| "epoch": 7.099705113225394, |
| "grad_norm": 0.00970458984375, |
| "learning_rate": 4.4911111111111115e-06, |
| "loss": 3.448, |
| "mean_token_accuracy": 0.4283379612490535, |
| "num_tokens": 105990196.0, |
| "step": 7980 |
| }, |
| { |
| "entropy": 2.485025796294212, |
| "epoch": 7.108607355477661, |
| "grad_norm": 0.0146484375, |
| "learning_rate": 4.468888888888889e-06, |
| "loss": 3.4402, |
| "mean_token_accuracy": 0.43056870540604, |
| "num_tokens": 106124388.0, |
| "step": 7990 |
| }, |
| { |
| "entropy": 2.4852806739509106, |
| "epoch": 7.117509597729928, |
| "grad_norm": 0.00836181640625, |
| "learning_rate": 4.446666666666667e-06, |
| "loss": 3.4043, |
| "mean_token_accuracy": 0.4322856513783336, |
| "num_tokens": 106261923.0, |
| "step": 8000 |
| }, |
| { |
| "epoch": 7.117509597729928, |
| "eval_coding_entropy": 1.266241003036499, |
| "eval_coding_loss": 1.2625732421875, |
| "eval_coding_mean_token_accuracy": 0.6838934738636017, |
| "eval_coding_num_tokens": 106261923.0, |
| "eval_coding_runtime": 96.0886, |
| "eval_coding_samples_per_second": 5.204, |
| "eval_coding_steps_per_second": 2.602, |
| "step": 8000 |
| }, |
| { |
| "epoch": 7.117509597729928, |
| "eval_chemistry_entropy": 2.520227357387543, |
| "eval_chemistry_loss": 3.5777227878570557, |
| "eval_chemistry_mean_token_accuracy": 0.4262875291109085, |
| "eval_chemistry_num_tokens": 106261923.0, |
| "eval_chemistry_runtime": 53.6736, |
| "eval_chemistry_samples_per_second": 9.316, |
| "eval_chemistry_steps_per_second": 4.658, |
| "step": 8000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "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": 8.827471834320642e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
