| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.890224225226729, |
| "eval_steps": 100, |
| "global_step": 1000, |
| "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 |
| } |
| ], |
| "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": 1.1048995053730529e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
