| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.8, |
| "eval_steps": 100, |
| "global_step": 6000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.9010286323726178, |
| "epoch": 0.008, |
| "grad_norm": 346.0, |
| "learning_rate": 1.8e-07, |
| "loss": 6.1315, |
| "mean_token_accuracy": 0.25839854273945095, |
| "num_tokens": 139100.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.966721160709858, |
| "epoch": 0.016, |
| "grad_norm": 332.0, |
| "learning_rate": 3.8e-07, |
| "loss": 6.1388, |
| "mean_token_accuracy": 0.24885686552152037, |
| "num_tokens": 282849.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 2.009664490073919, |
| "epoch": 0.024, |
| "grad_norm": 258.0, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 5.6986, |
| "mean_token_accuracy": 0.2627488659694791, |
| "num_tokens": 421949.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 2.1884495191276074, |
| "epoch": 0.032, |
| "grad_norm": 227.0, |
| "learning_rate": 7.8e-07, |
| "loss": 5.1409, |
| "mean_token_accuracy": 0.2678525767289102, |
| "num_tokens": 561726.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 2.3591181501746177, |
| "epoch": 0.04, |
| "grad_norm": 129.0, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 4.4527, |
| "mean_token_accuracy": 0.28993141520768406, |
| "num_tokens": 702070.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 2.718354082107544, |
| "epoch": 0.048, |
| "grad_norm": 65.5, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 3.7075, |
| "mean_token_accuracy": 0.31775791877880694, |
| "num_tokens": 842092.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 3.0432544946670532, |
| "epoch": 0.056, |
| "grad_norm": 41.25, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 3.2731, |
| "mean_token_accuracy": 0.3481094970367849, |
| "num_tokens": 987970.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 2.9290334582328796, |
| "epoch": 0.064, |
| "grad_norm": 29.0, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 2.9613, |
| "mean_token_accuracy": 0.37991682840511204, |
| "num_tokens": 1133349.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 2.647129815071821, |
| "epoch": 0.072, |
| "grad_norm": 25.625, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 2.6778, |
| "mean_token_accuracy": 0.41725210128352047, |
| "num_tokens": 1281864.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 2.506124369800091, |
| "epoch": 0.08, |
| "grad_norm": 30.0, |
| "learning_rate": 1.98e-06, |
| "loss": 2.564, |
| "mean_token_accuracy": 0.44079732317477466, |
| "num_tokens": 1422340.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_coding_entropy": 2.221115604877472, |
| "eval_coding_loss": 2.1600260734558105, |
| "eval_coding_mean_token_accuracy": 0.5053749171495437, |
| "eval_coding_num_tokens": 1422340.0, |
| "eval_coding_runtime": 80.7557, |
| "eval_coding_samples_per_second": 6.192, |
| "eval_coding_steps_per_second": 3.096, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_math_entropy": 2.4396081223487855, |
| "eval_math_loss": 2.4733667373657227, |
| "eval_math_mean_token_accuracy": 0.4529748194217682, |
| "eval_math_num_tokens": 1422340.0, |
| "eval_math_runtime": 45.3598, |
| "eval_math_samples_per_second": 11.023, |
| "eval_math_steps_per_second": 5.511, |
| "step": 100 |
| }, |
| { |
| "entropy": 2.3657003082334995, |
| "epoch": 0.088, |
| "grad_norm": 32.5, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 2.4185, |
| "mean_token_accuracy": 0.4646760683506727, |
| "num_tokens": 1569066.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 2.322612264752388, |
| "epoch": 0.096, |
| "grad_norm": 23.375, |
| "learning_rate": 2.38e-06, |
| "loss": 2.3229, |
| "mean_token_accuracy": 0.4779530819505453, |
| "num_tokens": 1711821.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 2.1566879101097585, |
| "epoch": 0.104, |
| "grad_norm": 27.25, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 2.19, |
| "mean_token_accuracy": 0.49810839612036945, |
| "num_tokens": 1851719.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 2.159694240242243, |
| "epoch": 0.112, |
| "grad_norm": 21.625, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 2.171, |
| "mean_token_accuracy": 0.5024791698902845, |
| "num_tokens": 1991855.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 2.1190536238253115, |
| "epoch": 0.12, |
| "grad_norm": 22.0, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 2.1148, |
| "mean_token_accuracy": 0.5197057416662574, |
| "num_tokens": 2139364.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 2.0801223501563073, |
| "epoch": 0.128, |
| "grad_norm": 33.0, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 2.0751, |
| "mean_token_accuracy": 0.5247604927048087, |
| "num_tokens": 2288720.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 1.9671796321868897, |
| "epoch": 0.136, |
| "grad_norm": 25.125, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 1.9964, |
| "mean_token_accuracy": 0.5339346826076508, |
| "num_tokens": 2427707.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 2.015957037359476, |
| "epoch": 0.144, |
| "grad_norm": 18.25, |
| "learning_rate": 3.58e-06, |
| "loss": 1.9768, |
| "mean_token_accuracy": 0.5422469139099121, |
| "num_tokens": 2564352.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 1.8770011462271214, |
| "epoch": 0.152, |
| "grad_norm": 20.75, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 1.899, |
| "mean_token_accuracy": 0.5532767685130239, |
| "num_tokens": 2703068.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 1.9259245485067367, |
| "epoch": 0.16, |
| "grad_norm": 24.125, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 1.9061, |
| "mean_token_accuracy": 0.5531875314190984, |
| "num_tokens": 2841876.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_coding_entropy": 2.074140534877777, |
| "eval_coding_loss": 2.048957586288452, |
| "eval_coding_mean_token_accuracy": 0.5282607427835464, |
| "eval_coding_num_tokens": 2841876.0, |
| "eval_coding_runtime": 80.2175, |
| "eval_coding_samples_per_second": 6.233, |
| "eval_coding_steps_per_second": 3.117, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_math_entropy": 1.883848337650299, |
| "eval_math_loss": 1.914016842842102, |
| "eval_math_mean_token_accuracy": 0.5553335778713226, |
| "eval_math_num_tokens": 2841876.0, |
| "eval_math_runtime": 45.3451, |
| "eval_math_samples_per_second": 11.027, |
| "eval_math_steps_per_second": 5.513, |
| "step": 200 |
| }, |
| { |
| "entropy": 1.8962259449064731, |
| "epoch": 0.168, |
| "grad_norm": 22.0, |
| "learning_rate": 4.18e-06, |
| "loss": 1.9095, |
| "mean_token_accuracy": 0.55602634716779, |
| "num_tokens": 2984096.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 1.8549927972257136, |
| "epoch": 0.176, |
| "grad_norm": 18.375, |
| "learning_rate": 4.38e-06, |
| "loss": 1.8083, |
| "mean_token_accuracy": 0.5690627742558718, |
| "num_tokens": 3129237.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 1.8978061877191066, |
| "epoch": 0.184, |
| "grad_norm": 21.0, |
| "learning_rate": 4.58e-06, |
| "loss": 1.8838, |
| "mean_token_accuracy": 0.5588616924360394, |
| "num_tokens": 3270850.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 1.8063146702945232, |
| "epoch": 0.192, |
| "grad_norm": 23.375, |
| "learning_rate": 4.78e-06, |
| "loss": 1.8173, |
| "mean_token_accuracy": 0.5768360491842032, |
| "num_tokens": 3414875.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 1.7750558860599994, |
| "epoch": 0.2, |
| "grad_norm": 21.625, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 1.7699, |
| "mean_token_accuracy": 0.5815087039023638, |
| "num_tokens": 3554899.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 1.800792108476162, |
| "epoch": 0.208, |
| "grad_norm": 29.75, |
| "learning_rate": 5.18e-06, |
| "loss": 1.7691, |
| "mean_token_accuracy": 0.5826054384931922, |
| "num_tokens": 3695623.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 1.7630318112671375, |
| "epoch": 0.216, |
| "grad_norm": 27.375, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 1.7276, |
| "mean_token_accuracy": 0.584668792411685, |
| "num_tokens": 3840400.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 1.756739466637373, |
| "epoch": 0.224, |
| "grad_norm": 16.375, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 1.7495, |
| "mean_token_accuracy": 0.5884332289919257, |
| "num_tokens": 3977666.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 1.6858243882656097, |
| "epoch": 0.232, |
| "grad_norm": 20.75, |
| "learning_rate": 5.78e-06, |
| "loss": 1.672, |
| "mean_token_accuracy": 0.5939588043838739, |
| "num_tokens": 4123665.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 1.6902806080877781, |
| "epoch": 0.24, |
| "grad_norm": 20.125, |
| "learning_rate": 5.98e-06, |
| "loss": 1.6792, |
| "mean_token_accuracy": 0.5978236820548772, |
| "num_tokens": 4269287.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_coding_entropy": 2.038021125793457, |
| "eval_coding_loss": 2.009580373764038, |
| "eval_coding_mean_token_accuracy": 0.5356820751428604, |
| "eval_coding_num_tokens": 4269287.0, |
| "eval_coding_runtime": 80.2243, |
| "eval_coding_samples_per_second": 6.233, |
| "eval_coding_steps_per_second": 3.116, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_math_entropy": 1.705006594657898, |
| "eval_math_loss": 1.684179663658142, |
| "eval_math_mean_token_accuracy": 0.5993804135322571, |
| "eval_math_num_tokens": 4269287.0, |
| "eval_math_runtime": 45.3771, |
| "eval_math_samples_per_second": 11.019, |
| "eval_math_steps_per_second": 5.509, |
| "step": 300 |
| }, |
| { |
| "entropy": 1.6501292690634728, |
| "epoch": 0.248, |
| "grad_norm": 27.5, |
| "learning_rate": 6.18e-06, |
| "loss": 1.622, |
| "mean_token_accuracy": 0.603010005876422, |
| "num_tokens": 4413984.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 1.6583089262247086, |
| "epoch": 0.256, |
| "grad_norm": 22.0, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 1.6323, |
| "mean_token_accuracy": 0.6021805582568049, |
| "num_tokens": 4558632.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 1.6304452382028103, |
| "epoch": 0.264, |
| "grad_norm": 22.0, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 1.6136, |
| "mean_token_accuracy": 0.6088442349806428, |
| "num_tokens": 4697507.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 1.6436940476298332, |
| "epoch": 0.272, |
| "grad_norm": 17.75, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 1.6032, |
| "mean_token_accuracy": 0.6129889598116278, |
| "num_tokens": 4839270.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 1.5658629022538662, |
| "epoch": 0.28, |
| "grad_norm": 19.625, |
| "learning_rate": 6.98e-06, |
| "loss": 1.5637, |
| "mean_token_accuracy": 0.6190130662173032, |
| "num_tokens": 4984993.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 1.590892932936549, |
| "epoch": 0.288, |
| "grad_norm": 17.125, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 1.575, |
| "mean_token_accuracy": 0.6215819546952843, |
| "num_tokens": 5125447.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 1.5730001308023929, |
| "epoch": 0.296, |
| "grad_norm": 17.5, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 1.5567, |
| "mean_token_accuracy": 0.6201104786247015, |
| "num_tokens": 5272211.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 1.5572829633951186, |
| "epoch": 0.304, |
| "grad_norm": 19.5, |
| "learning_rate": 7.58e-06, |
| "loss": 1.5193, |
| "mean_token_accuracy": 0.6289611663669348, |
| "num_tokens": 5415925.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 1.564134069532156, |
| "epoch": 0.312, |
| "grad_norm": 18.75, |
| "learning_rate": 7.78e-06, |
| "loss": 1.5464, |
| "mean_token_accuracy": 0.6214223591610789, |
| "num_tokens": 5563727.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 1.5101773023605347, |
| "epoch": 0.32, |
| "grad_norm": 18.75, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 1.4881, |
| "mean_token_accuracy": 0.635349939763546, |
| "num_tokens": 5710178.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_coding_entropy": 1.957896315574646, |
| "eval_coding_loss": 2.0061168670654297, |
| "eval_coding_mean_token_accuracy": 0.5387240793704987, |
| "eval_coding_num_tokens": 5710178.0, |
| "eval_coding_runtime": 80.2564, |
| "eval_coding_samples_per_second": 6.23, |
| "eval_coding_steps_per_second": 3.115, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_math_entropy": 1.492733929157257, |
| "eval_math_loss": 1.532265305519104, |
| "eval_math_mean_token_accuracy": 0.6304122837781906, |
| "eval_math_num_tokens": 5710178.0, |
| "eval_math_runtime": 45.4084, |
| "eval_math_samples_per_second": 11.011, |
| "eval_math_steps_per_second": 5.506, |
| "step": 400 |
| }, |
| { |
| "entropy": 1.5742981527000666, |
| "epoch": 0.328, |
| "grad_norm": 23.375, |
| "learning_rate": 8.18e-06, |
| "loss": 1.5481, |
| "mean_token_accuracy": 0.6223406890407205, |
| "num_tokens": 5851095.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 1.529792605340481, |
| "epoch": 0.336, |
| "grad_norm": 22.0, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 1.5155, |
| "mean_token_accuracy": 0.6262415384873747, |
| "num_tokens": 5993843.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 1.5310558043420315, |
| "epoch": 0.344, |
| "grad_norm": 22.625, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 1.496, |
| "mean_token_accuracy": 0.634973325394094, |
| "num_tokens": 6138231.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 1.4365405213087796, |
| "epoch": 0.352, |
| "grad_norm": 18.375, |
| "learning_rate": 8.78e-06, |
| "loss": 1.4261, |
| "mean_token_accuracy": 0.6425879798829556, |
| "num_tokens": 6279137.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 1.491362702846527, |
| "epoch": 0.36, |
| "grad_norm": 19.375, |
| "learning_rate": 8.98e-06, |
| "loss": 1.4506, |
| "mean_token_accuracy": 0.6380296373739839, |
| "num_tokens": 6415141.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 1.4470719736069442, |
| "epoch": 0.368, |
| "grad_norm": 19.5, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 1.4295, |
| "mean_token_accuracy": 0.6446485066786408, |
| "num_tokens": 6562387.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 1.467770417779684, |
| "epoch": 0.376, |
| "grad_norm": 15.4375, |
| "learning_rate": 9.38e-06, |
| "loss": 1.4541, |
| "mean_token_accuracy": 0.6424220692366361, |
| "num_tokens": 6708941.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 1.4349520716816186, |
| "epoch": 0.384, |
| "grad_norm": 16.125, |
| "learning_rate": 9.58e-06, |
| "loss": 1.4215, |
| "mean_token_accuracy": 0.6473838938400149, |
| "num_tokens": 6851520.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 1.4698528438806533, |
| "epoch": 0.392, |
| "grad_norm": 18.25, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 1.4189, |
| "mean_token_accuracy": 0.6451966688036919, |
| "num_tokens": 6995283.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 1.4960964936763048, |
| "epoch": 0.4, |
| "grad_norm": 21.75, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 1.4938, |
| "mean_token_accuracy": 0.6361222166568041, |
| "num_tokens": 7138932.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_coding_entropy": 2.0384827246665953, |
| "eval_coding_loss": 2.0063071250915527, |
| "eval_coding_mean_token_accuracy": 0.5387867102622986, |
| "eval_coding_num_tokens": 7138932.0, |
| "eval_coding_runtime": 80.2013, |
| "eval_coding_samples_per_second": 6.234, |
| "eval_coding_steps_per_second": 3.117, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_math_entropy": 1.4683585102558137, |
| "eval_math_loss": 1.4378676414489746, |
| "eval_math_mean_token_accuracy": 0.6471751933097839, |
| "eval_math_num_tokens": 7138932.0, |
| "eval_math_runtime": 45.2776, |
| "eval_math_samples_per_second": 11.043, |
| "eval_math_steps_per_second": 5.521, |
| "step": 500 |
| }, |
| { |
| "entropy": 1.4108689952641726, |
| "epoch": 0.408, |
| "grad_norm": 21.125, |
| "learning_rate": 1.018e-05, |
| "loss": 1.3867, |
| "mean_token_accuracy": 0.6560342008247971, |
| "num_tokens": 7281934.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 1.4268736977130174, |
| "epoch": 0.416, |
| "grad_norm": 23.625, |
| "learning_rate": 1.038e-05, |
| "loss": 1.4084, |
| "mean_token_accuracy": 0.6534364787861705, |
| "num_tokens": 7420720.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 1.400152337178588, |
| "epoch": 0.424, |
| "grad_norm": 14.125, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 1.3646, |
| "mean_token_accuracy": 0.6582854194566607, |
| "num_tokens": 7557196.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 1.3399375207722186, |
| "epoch": 0.432, |
| "grad_norm": 20.625, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 1.3223, |
| "mean_token_accuracy": 0.664572624117136, |
| "num_tokens": 7695518.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 1.4023700952529907, |
| "epoch": 0.44, |
| "grad_norm": 12.6875, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 1.364, |
| "mean_token_accuracy": 0.6588193815201521, |
| "num_tokens": 7839775.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 1.3175839744508266, |
| "epoch": 0.448, |
| "grad_norm": 12.9375, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 1.2952, |
| "mean_token_accuracy": 0.670157365873456, |
| "num_tokens": 7983543.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 1.3483419816941022, |
| "epoch": 0.456, |
| "grad_norm": 16.125, |
| "learning_rate": 1.138e-05, |
| "loss": 1.3367, |
| "mean_token_accuracy": 0.6675028461962939, |
| "num_tokens": 8128785.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 1.3787965573370458, |
| "epoch": 0.464, |
| "grad_norm": 16.5, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 1.3559, |
| "mean_token_accuracy": 0.662639387883246, |
| "num_tokens": 8271787.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 1.385534007847309, |
| "epoch": 0.472, |
| "grad_norm": 14.5, |
| "learning_rate": 1.178e-05, |
| "loss": 1.3516, |
| "mean_token_accuracy": 0.6594217313453555, |
| "num_tokens": 8416787.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 1.3437833908945322, |
| "epoch": 0.48, |
| "grad_norm": 14.4375, |
| "learning_rate": 1.198e-05, |
| "loss": 1.3461, |
| "mean_token_accuracy": 0.665634086355567, |
| "num_tokens": 8556170.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_coding_entropy": 2.031046444416046, |
| "eval_coding_loss": 2.020702362060547, |
| "eval_coding_mean_token_accuracy": 0.5379636323451996, |
| "eval_coding_num_tokens": 8556170.0, |
| "eval_coding_runtime": 80.1969, |
| "eval_coding_samples_per_second": 6.235, |
| "eval_coding_steps_per_second": 3.117, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_math_entropy": 1.3622242574691772, |
| "eval_math_loss": 1.3451071977615356, |
| "eval_math_mean_token_accuracy": 0.6687758553028107, |
| "eval_math_num_tokens": 8556170.0, |
| "eval_math_runtime": 45.3457, |
| "eval_math_samples_per_second": 11.026, |
| "eval_math_steps_per_second": 5.513, |
| "step": 600 |
| }, |
| { |
| "entropy": 1.3342523202300072, |
| "epoch": 0.488, |
| "grad_norm": 15.125, |
| "learning_rate": 1.218e-05, |
| "loss": 1.3023, |
| "mean_token_accuracy": 0.6706280615180731, |
| "num_tokens": 8697365.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 1.2956569973379373, |
| "epoch": 0.496, |
| "grad_norm": 19.375, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 1.2753, |
| "mean_token_accuracy": 0.6751342266798019, |
| "num_tokens": 8840590.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 1.3236910991370678, |
| "epoch": 0.504, |
| "grad_norm": 12.75, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 1.301, |
| "mean_token_accuracy": 0.676119314134121, |
| "num_tokens": 8987815.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 1.2810744147747755, |
| "epoch": 0.512, |
| "grad_norm": 13.875, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 1.2445, |
| "mean_token_accuracy": 0.6812406960874796, |
| "num_tokens": 9131903.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 1.2942697513848542, |
| "epoch": 0.52, |
| "grad_norm": 17.375, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 1.2705, |
| "mean_token_accuracy": 0.6787693897262216, |
| "num_tokens": 9275021.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 1.2697163943201304, |
| "epoch": 0.528, |
| "grad_norm": 13.75, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 1.245, |
| "mean_token_accuracy": 0.6814844202250242, |
| "num_tokens": 9429519.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 1.2942543551325798, |
| "epoch": 0.536, |
| "grad_norm": 14.375, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 1.2757, |
| "mean_token_accuracy": 0.6822770915925502, |
| "num_tokens": 9572443.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 1.2603163443505765, |
| "epoch": 0.544, |
| "grad_norm": 11.375, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 1.237, |
| "mean_token_accuracy": 0.6849679309874773, |
| "num_tokens": 9718247.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 1.3230578549206258, |
| "epoch": 0.552, |
| "grad_norm": 12.875, |
| "learning_rate": 1.378e-05, |
| "loss": 1.3118, |
| "mean_token_accuracy": 0.6753747440874577, |
| "num_tokens": 9851190.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 1.2490842442959547, |
| "epoch": 0.56, |
| "grad_norm": 12.4375, |
| "learning_rate": 1.398e-05, |
| "loss": 1.219, |
| "mean_token_accuracy": 0.6826988641172648, |
| "num_tokens": 9991100.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_coding_entropy": 1.95146404504776, |
| "eval_coding_loss": 2.0403237342834473, |
| "eval_coding_mean_token_accuracy": 0.5367712031602859, |
| "eval_coding_num_tokens": 9991100.0, |
| "eval_coding_runtime": 80.176, |
| "eval_coding_samples_per_second": 6.236, |
| "eval_coding_steps_per_second": 3.118, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.56, |
| "eval_math_entropy": 1.250319209575653, |
| "eval_math_loss": 1.2708617448806763, |
| "eval_math_mean_token_accuracy": 0.6818764185905457, |
| "eval_math_num_tokens": 9991100.0, |
| "eval_math_runtime": 45.3268, |
| "eval_math_samples_per_second": 11.031, |
| "eval_math_steps_per_second": 5.516, |
| "step": 700 |
| }, |
| { |
| "entropy": 1.234171437099576, |
| "epoch": 0.568, |
| "grad_norm": 16.25, |
| "learning_rate": 1.418e-05, |
| "loss": 1.2138, |
| "mean_token_accuracy": 0.689500842988491, |
| "num_tokens": 10131843.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 1.262464852258563, |
| "epoch": 0.576, |
| "grad_norm": 13.625, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 1.2519, |
| "mean_token_accuracy": 0.6837401404976845, |
| "num_tokens": 10276090.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 1.246568062528968, |
| "epoch": 0.584, |
| "grad_norm": 12.5625, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 1.224, |
| "mean_token_accuracy": 0.6878476399928332, |
| "num_tokens": 10416356.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 1.2306802656501532, |
| "epoch": 0.592, |
| "grad_norm": 11.1875, |
| "learning_rate": 1.478e-05, |
| "loss": 1.2034, |
| "mean_token_accuracy": 0.6904714822769165, |
| "num_tokens": 10566762.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 1.2370665207505227, |
| "epoch": 0.6, |
| "grad_norm": 17.25, |
| "learning_rate": 1.498e-05, |
| "loss": 1.2279, |
| "mean_token_accuracy": 0.688570697978139, |
| "num_tokens": 10707778.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 1.2397097527980805, |
| "epoch": 0.608, |
| "grad_norm": 12.125, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 1.2209, |
| "mean_token_accuracy": 0.6899642575532198, |
| "num_tokens": 10857002.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 1.19944361820817, |
| "epoch": 0.616, |
| "grad_norm": 12.0625, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 1.2, |
| "mean_token_accuracy": 0.6944263610988856, |
| "num_tokens": 11006038.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 1.211577532812953, |
| "epoch": 0.624, |
| "grad_norm": 14.25, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 1.1892, |
| "mean_token_accuracy": 0.6928001435473561, |
| "num_tokens": 11147326.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 1.1774040956050158, |
| "epoch": 0.632, |
| "grad_norm": 9.3125, |
| "learning_rate": 1.578e-05, |
| "loss": 1.1714, |
| "mean_token_accuracy": 0.6989362832158804, |
| "num_tokens": 11292873.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 1.2005819905549289, |
| "epoch": 0.64, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 1.1825, |
| "mean_token_accuracy": 0.6982070229947567, |
| "num_tokens": 11437710.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_coding_entropy": 1.9956619744300843, |
| "eval_coding_loss": 2.0632827281951904, |
| "eval_coding_mean_token_accuracy": 0.5326479307413101, |
| "eval_coding_num_tokens": 11437710.0, |
| "eval_coding_runtime": 80.1692, |
| "eval_coding_samples_per_second": 6.237, |
| "eval_coding_steps_per_second": 3.118, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_math_entropy": 1.2291223180294037, |
| "eval_math_loss": 1.207571268081665, |
| "eval_math_mean_token_accuracy": 0.6958818504810333, |
| "eval_math_num_tokens": 11437710.0, |
| "eval_math_runtime": 45.3259, |
| "eval_math_samples_per_second": 11.031, |
| "eval_math_steps_per_second": 5.516, |
| "step": 800 |
| }, |
| { |
| "entropy": 1.1634460400789977, |
| "epoch": 0.648, |
| "grad_norm": 11.3125, |
| "learning_rate": 1.618e-05, |
| "loss": 1.1484, |
| "mean_token_accuracy": 0.7010497888550162, |
| "num_tokens": 11583861.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 1.1786630261689424, |
| "epoch": 0.656, |
| "grad_norm": 8.375, |
| "learning_rate": 1.638e-05, |
| "loss": 1.1623, |
| "mean_token_accuracy": 0.7017704050987958, |
| "num_tokens": 11727268.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 1.1895627338439225, |
| "epoch": 0.664, |
| "grad_norm": 10.3125, |
| "learning_rate": 1.658e-05, |
| "loss": 1.1755, |
| "mean_token_accuracy": 0.6996823458001018, |
| "num_tokens": 11869608.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 1.1601045448333025, |
| "epoch": 0.672, |
| "grad_norm": 11.0625, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 1.1607, |
| "mean_token_accuracy": 0.7032990284264088, |
| "num_tokens": 12020478.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 1.1435815680772066, |
| "epoch": 0.68, |
| "grad_norm": 11.1875, |
| "learning_rate": 1.698e-05, |
| "loss": 1.1403, |
| "mean_token_accuracy": 0.704500000551343, |
| "num_tokens": 12162133.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 1.1609381169080735, |
| "epoch": 0.688, |
| "grad_norm": 9.5625, |
| "learning_rate": 1.718e-05, |
| "loss": 1.1385, |
| "mean_token_accuracy": 0.7051008731126786, |
| "num_tokens": 12300380.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 1.181287159770727, |
| "epoch": 0.696, |
| "grad_norm": 12.75, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 1.1488, |
| "mean_token_accuracy": 0.7007136529311537, |
| "num_tokens": 12445277.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 1.1309138476848601, |
| "epoch": 0.704, |
| "grad_norm": 8.25, |
| "learning_rate": 1.758e-05, |
| "loss": 1.1013, |
| "mean_token_accuracy": 0.7131508827209473, |
| "num_tokens": 12591366.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 1.1794448290020227, |
| "epoch": 0.712, |
| "grad_norm": 8.75, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 1.1501, |
| "mean_token_accuracy": 0.7011325571686029, |
| "num_tokens": 12732100.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 1.1434503257274629, |
| "epoch": 0.72, |
| "grad_norm": 11.625, |
| "learning_rate": 1.798e-05, |
| "loss": 1.1327, |
| "mean_token_accuracy": 0.7088143475353718, |
| "num_tokens": 12869618.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_coding_entropy": 1.9485954036712647, |
| "eval_coding_loss": 2.090169668197632, |
| "eval_coding_mean_token_accuracy": 0.5306968300342559, |
| "eval_coding_num_tokens": 12869618.0, |
| "eval_coding_runtime": 80.4241, |
| "eval_coding_samples_per_second": 6.217, |
| "eval_coding_steps_per_second": 3.109, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.72, |
| "eval_math_entropy": 1.1418632171154022, |
| "eval_math_loss": 1.1755626201629639, |
| "eval_math_mean_token_accuracy": 0.7029146280288696, |
| "eval_math_num_tokens": 12869618.0, |
| "eval_math_runtime": 45.3564, |
| "eval_math_samples_per_second": 11.024, |
| "eval_math_steps_per_second": 5.512, |
| "step": 900 |
| }, |
| { |
| "entropy": 1.1645388562232255, |
| "epoch": 0.728, |
| "grad_norm": 11.25, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 1.1468, |
| "mean_token_accuracy": 0.7067818889394403, |
| "num_tokens": 13009904.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 1.124797797203064, |
| "epoch": 0.736, |
| "grad_norm": 9.9375, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 1.1083, |
| "mean_token_accuracy": 0.7103868834674358, |
| "num_tokens": 13156311.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 1.1238487865775824, |
| "epoch": 0.744, |
| "grad_norm": 7.90625, |
| "learning_rate": 1.858e-05, |
| "loss": 1.1068, |
| "mean_token_accuracy": 0.7117075510323048, |
| "num_tokens": 13297031.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 1.0840136419981719, |
| "epoch": 0.752, |
| "grad_norm": 9.3125, |
| "learning_rate": 1.878e-05, |
| "loss": 1.0765, |
| "mean_token_accuracy": 0.7179766517132521, |
| "num_tokens": 13438911.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 1.0820568643510342, |
| "epoch": 0.76, |
| "grad_norm": 7.3125, |
| "learning_rate": 1.898e-05, |
| "loss": 1.0811, |
| "mean_token_accuracy": 0.7211926929652691, |
| "num_tokens": 13591931.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 1.0984113983809949, |
| "epoch": 0.768, |
| "grad_norm": 10.875, |
| "learning_rate": 1.918e-05, |
| "loss": 1.0869, |
| "mean_token_accuracy": 0.7214216101914644, |
| "num_tokens": 13735397.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 1.1237307298928498, |
| "epoch": 0.776, |
| "grad_norm": 6.9375, |
| "learning_rate": 1.938e-05, |
| "loss": 1.117, |
| "mean_token_accuracy": 0.7158737700432539, |
| "num_tokens": 13877049.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 1.1685453273355961, |
| "epoch": 0.784, |
| "grad_norm": 9.75, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 1.1514, |
| "mean_token_accuracy": 0.7037497881799937, |
| "num_tokens": 14024497.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 1.1096330046653748, |
| "epoch": 0.792, |
| "grad_norm": 6.78125, |
| "learning_rate": 1.978e-05, |
| "loss": 1.1127, |
| "mean_token_accuracy": 0.7135461080819369, |
| "num_tokens": 14170760.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 1.1440908499062061, |
| "epoch": 0.8, |
| "grad_norm": 10.1875, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 1.1286, |
| "mean_token_accuracy": 0.7088436063379049, |
| "num_tokens": 14316427.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_coding_entropy": 2.042564573287964, |
| "eval_coding_loss": 2.1235098838806152, |
| "eval_coding_mean_token_accuracy": 0.5269861690998078, |
| "eval_coding_num_tokens": 14316427.0, |
| "eval_coding_runtime": 80.2281, |
| "eval_coding_samples_per_second": 6.232, |
| "eval_coding_steps_per_second": 3.116, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.8, |
| "eval_math_entropy": 1.1564667975902558, |
| "eval_math_loss": 1.1225173473358154, |
| "eval_math_mean_token_accuracy": 0.7147891137599945, |
| "eval_math_num_tokens": 14316427.0, |
| "eval_math_runtime": 45.2271, |
| "eval_math_samples_per_second": 11.055, |
| "eval_math_steps_per_second": 5.528, |
| "step": 1000 |
| }, |
| { |
| "entropy": 1.1050428252667188, |
| "epoch": 0.808, |
| "grad_norm": 7.71875, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 1.0746, |
| "mean_token_accuracy": 0.7183676421642303, |
| "num_tokens": 14460896.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 1.1038101874291897, |
| "epoch": 0.816, |
| "grad_norm": 7.09375, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 1.0895, |
| "mean_token_accuracy": 0.7137791357934475, |
| "num_tokens": 14607136.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 1.1085384231060744, |
| "epoch": 0.824, |
| "grad_norm": 8.375, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 1.0908, |
| "mean_token_accuracy": 0.7163524894043803, |
| "num_tokens": 14752291.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 1.0919262439012527, |
| "epoch": 0.832, |
| "grad_norm": 8.1875, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 1.0797, |
| "mean_token_accuracy": 0.7174479674547911, |
| "num_tokens": 14894084.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 1.1038660164922476, |
| "epoch": 0.84, |
| "grad_norm": 7.59375, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 1.0847, |
| "mean_token_accuracy": 0.7196218792349101, |
| "num_tokens": 15034548.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 1.1151995915919541, |
| "epoch": 0.848, |
| "grad_norm": 5.59375, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 1.1088, |
| "mean_token_accuracy": 0.7122140154242516, |
| "num_tokens": 15177039.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 1.1070417396724224, |
| "epoch": 0.856, |
| "grad_norm": 6.59375, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 1.0954, |
| "mean_token_accuracy": 0.7178661905229091, |
| "num_tokens": 15320396.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 1.071781432814896, |
| "epoch": 0.864, |
| "grad_norm": 7.0625, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 1.0514, |
| "mean_token_accuracy": 0.7243800666183233, |
| "num_tokens": 15463393.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 1.0959462771192192, |
| "epoch": 0.872, |
| "grad_norm": 6.59375, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 1.0756, |
| "mean_token_accuracy": 0.7194638095796109, |
| "num_tokens": 15606730.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 1.1222398560494184, |
| "epoch": 0.88, |
| "grad_norm": 8.8125, |
| "learning_rate": 1.978e-05, |
| "loss": 1.0913, |
| "mean_token_accuracy": 0.7151848122477531, |
| "num_tokens": 15747943.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_coding_entropy": 1.8464843645095825, |
| "eval_coding_loss": 2.152169704437256, |
| "eval_coding_mean_token_accuracy": 0.5262901128530503, |
| "eval_coding_num_tokens": 15747943.0, |
| "eval_coding_runtime": 80.2552, |
| "eval_coding_samples_per_second": 6.23, |
| "eval_coding_steps_per_second": 3.115, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_math_entropy": 1.0326574013233185, |
| "eval_math_loss": 1.0848861932754517, |
| "eval_math_mean_token_accuracy": 0.722681262254715, |
| "eval_math_num_tokens": 15747943.0, |
| "eval_math_runtime": 45.3881, |
| "eval_math_samples_per_second": 11.016, |
| "eval_math_steps_per_second": 5.508, |
| "step": 1100 |
| }, |
| { |
| "entropy": 1.064740102738142, |
| "epoch": 0.888, |
| "grad_norm": 7.71875, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 1.0544, |
| "mean_token_accuracy": 0.7254730749875307, |
| "num_tokens": 15885971.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 1.1040192453190685, |
| "epoch": 0.896, |
| "grad_norm": 6.09375, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 1.0916, |
| "mean_token_accuracy": 0.7160311482846737, |
| "num_tokens": 16027931.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 1.034827933087945, |
| "epoch": 0.904, |
| "grad_norm": 6.5625, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 1.0138, |
| "mean_token_accuracy": 0.732599625363946, |
| "num_tokens": 16170943.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 1.0492325933650135, |
| "epoch": 0.912, |
| "grad_norm": 6.53125, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 1.0339, |
| "mean_token_accuracy": 0.7263875484466553, |
| "num_tokens": 16317127.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 1.0222572326660155, |
| "epoch": 0.92, |
| "grad_norm": 10.5, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 1.0146, |
| "mean_token_accuracy": 0.7330152779817581, |
| "num_tokens": 16457337.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 1.0255462951958179, |
| "epoch": 0.928, |
| "grad_norm": 7.6875, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 1.0095, |
| "mean_token_accuracy": 0.7334649574011565, |
| "num_tokens": 16598757.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 1.0513048900291324, |
| "epoch": 0.936, |
| "grad_norm": 6.53125, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 1.0319, |
| "mean_token_accuracy": 0.7290561534464359, |
| "num_tokens": 16740107.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 1.0610596431419252, |
| "epoch": 0.944, |
| "grad_norm": 7.375, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 1.0552, |
| "mean_token_accuracy": 0.7230983316898346, |
| "num_tokens": 16882236.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 1.0357142012566327, |
| "epoch": 0.952, |
| "grad_norm": 6.78125, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 1.0286, |
| "mean_token_accuracy": 0.7313166920095682, |
| "num_tokens": 17023937.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 1.0776148246601225, |
| "epoch": 0.96, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 1.0456, |
| "mean_token_accuracy": 0.7257028739899397, |
| "num_tokens": 17169495.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_coding_entropy": 1.8717318296432495, |
| "eval_coding_loss": 2.1234161853790283, |
| "eval_coding_mean_token_accuracy": 0.5292520480155944, |
| "eval_coding_num_tokens": 17169495.0, |
| "eval_coding_runtime": 80.2106, |
| "eval_coding_samples_per_second": 6.234, |
| "eval_coding_steps_per_second": 3.117, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_math_entropy": 1.0023325213193894, |
| "eval_math_loss": 1.0324506759643555, |
| "eval_math_mean_token_accuracy": 0.7347698690891266, |
| "eval_math_num_tokens": 17169495.0, |
| "eval_math_runtime": 45.5774, |
| "eval_math_samples_per_second": 10.97, |
| "eval_math_steps_per_second": 5.485, |
| "step": 1200 |
| }, |
| { |
| "entropy": 1.0140543539077043, |
| "epoch": 0.968, |
| "grad_norm": 6.09375, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 0.9982, |
| "mean_token_accuracy": 0.7358098488301039, |
| "num_tokens": 17313887.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 1.035257386416197, |
| "epoch": 0.976, |
| "grad_norm": 6.4375, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 1.017, |
| "mean_token_accuracy": 0.7321158088743687, |
| "num_tokens": 17454147.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 1.0174118569120765, |
| "epoch": 0.984, |
| "grad_norm": 6.5625, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 0.9963, |
| "mean_token_accuracy": 0.7360445655882358, |
| "num_tokens": 17596920.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 1.0450362712144852, |
| "epoch": 0.992, |
| "grad_norm": 7.25, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 1.0117, |
| "mean_token_accuracy": 0.7314894199371338, |
| "num_tokens": 17742596.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 1.0050787944346666, |
| "epoch": 1.0, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 1.0115, |
| "mean_token_accuracy": 0.7343492582440376, |
| "num_tokens": 17886424.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 0.9094233620911837, |
| "epoch": 1.008, |
| "grad_norm": 4.875, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 0.898, |
| "mean_token_accuracy": 0.754911670833826, |
| "num_tokens": 18033540.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 0.9372691879048943, |
| "epoch": 1.016, |
| "grad_norm": 6.53125, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 0.9139, |
| "mean_token_accuracy": 0.7527765788137912, |
| "num_tokens": 18178632.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 0.9098639283329248, |
| "epoch": 1.024, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.938e-05, |
| "loss": 0.901, |
| "mean_token_accuracy": 0.7548334702849389, |
| "num_tokens": 18316346.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 0.9618702568113804, |
| "epoch": 1.032, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 0.9393, |
| "mean_token_accuracy": 0.747647900134325, |
| "num_tokens": 18458494.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 0.8774365060031414, |
| "epoch": 1.04, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 0.8709, |
| "mean_token_accuracy": 0.7649227738380432, |
| "num_tokens": 18595230.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_coding_entropy": 1.741557788848877, |
| "eval_coding_loss": 2.1706783771514893, |
| "eval_coding_mean_token_accuracy": 0.5256800745725632, |
| "eval_coding_num_tokens": 18595230.0, |
| "eval_coding_runtime": 80.4097, |
| "eval_coding_samples_per_second": 6.218, |
| "eval_coding_steps_per_second": 3.109, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.04, |
| "eval_math_entropy": 0.9422389795780182, |
| "eval_math_loss": 1.012914776802063, |
| "eval_math_mean_token_accuracy": 0.7398272805213928, |
| "eval_math_num_tokens": 18595230.0, |
| "eval_math_runtime": 45.4927, |
| "eval_math_samples_per_second": 10.991, |
| "eval_math_steps_per_second": 5.495, |
| "step": 1300 |
| }, |
| { |
| "entropy": 0.9533776409924031, |
| "epoch": 1.048, |
| "grad_norm": 5.46875, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 0.9395, |
| "mean_token_accuracy": 0.7498730301856995, |
| "num_tokens": 18741221.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 0.9226375747472048, |
| "epoch": 1.056, |
| "grad_norm": 5.5, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 0.9111, |
| "mean_token_accuracy": 0.7551251873373985, |
| "num_tokens": 18885547.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 0.9371993999928236, |
| "epoch": 1.064, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 0.9403, |
| "mean_token_accuracy": 0.7490979980677366, |
| "num_tokens": 19031374.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 0.9045422596856951, |
| "epoch": 1.072, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 0.8924, |
| "mean_token_accuracy": 0.7566539086401463, |
| "num_tokens": 19169083.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 0.9355957299470902, |
| "epoch": 1.08, |
| "grad_norm": 4.75, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 0.9096, |
| "mean_token_accuracy": 0.7521375376731158, |
| "num_tokens": 19316856.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 0.9026603888720274, |
| "epoch": 1.088, |
| "grad_norm": 5.5, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 0.8909, |
| "mean_token_accuracy": 0.7602829448878765, |
| "num_tokens": 19461465.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 0.9450714871287346, |
| "epoch": 1.096, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.918e-05, |
| "loss": 0.9053, |
| "mean_token_accuracy": 0.753120943531394, |
| "num_tokens": 19610931.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 0.9266160767525434, |
| "epoch": 1.104, |
| "grad_norm": 5.59375, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 0.9018, |
| "mean_token_accuracy": 0.754140418022871, |
| "num_tokens": 19755027.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 0.8790432747453452, |
| "epoch": 1.112, |
| "grad_norm": 5.90625, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 0.884, |
| "mean_token_accuracy": 0.7613258976489305, |
| "num_tokens": 19895772.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 0.8988591339439154, |
| "epoch": 1.12, |
| "grad_norm": 5.9375, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 0.8849, |
| "mean_token_accuracy": 0.7603809408843517, |
| "num_tokens": 20043293.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_coding_entropy": 1.6949054861068726, |
| "eval_coding_loss": 2.156839370727539, |
| "eval_coding_mean_token_accuracy": 0.5291718513965606, |
| "eval_coding_num_tokens": 20043293.0, |
| "eval_coding_runtime": 80.2326, |
| "eval_coding_samples_per_second": 6.232, |
| "eval_coding_steps_per_second": 3.116, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.12, |
| "eval_math_entropy": 0.9113483186960221, |
| "eval_math_loss": 0.9880151152610779, |
| "eval_math_mean_token_accuracy": 0.7434683785438537, |
| "eval_math_num_tokens": 20043293.0, |
| "eval_math_runtime": 45.3699, |
| "eval_math_samples_per_second": 11.021, |
| "eval_math_steps_per_second": 5.51, |
| "step": 1400 |
| }, |
| { |
| "entropy": 0.8970432108268142, |
| "epoch": 1.1280000000000001, |
| "grad_norm": 6.25, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 0.8938, |
| "mean_token_accuracy": 0.7592019069939852, |
| "num_tokens": 20181910.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 0.8997785529121757, |
| "epoch": 1.1360000000000001, |
| "grad_norm": 5.71875, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 0.8919, |
| "mean_token_accuracy": 0.7560406625270844, |
| "num_tokens": 20323010.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 0.9096224224194884, |
| "epoch": 1.144, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 0.893, |
| "mean_token_accuracy": 0.7575655911117792, |
| "num_tokens": 20466029.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 0.923757266253233, |
| "epoch": 1.152, |
| "grad_norm": 6.59375, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 0.9181, |
| "mean_token_accuracy": 0.7567033864557743, |
| "num_tokens": 20613266.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 0.8967335695400834, |
| "epoch": 1.16, |
| "grad_norm": 6.65625, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 0.8905, |
| "mean_token_accuracy": 0.7599862135946751, |
| "num_tokens": 20765543.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 0.8982609620317816, |
| "epoch": 1.168, |
| "grad_norm": 5.75, |
| "learning_rate": 1.898e-05, |
| "loss": 0.8869, |
| "mean_token_accuracy": 0.7594386037439108, |
| "num_tokens": 20910404.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 0.8946591690182686, |
| "epoch": 1.176, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 0.8758, |
| "mean_token_accuracy": 0.7606226697564125, |
| "num_tokens": 21054261.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 0.8773745242506266, |
| "epoch": 1.184, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 0.8687, |
| "mean_token_accuracy": 0.7651932623237372, |
| "num_tokens": 21200197.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 0.8704428110271692, |
| "epoch": 1.192, |
| "grad_norm": 5.46875, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 0.875, |
| "mean_token_accuracy": 0.7662943832576274, |
| "num_tokens": 21348853.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 0.8863117309287191, |
| "epoch": 1.2, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 0.875, |
| "mean_token_accuracy": 0.7623905159533024, |
| "num_tokens": 21496441.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_coding_entropy": 1.790007698059082, |
| "eval_coding_loss": 2.1545250415802, |
| "eval_coding_mean_token_accuracy": 0.5271972445249558, |
| "eval_coding_num_tokens": 21496441.0, |
| "eval_coding_runtime": 80.2484, |
| "eval_coding_samples_per_second": 6.231, |
| "eval_coding_steps_per_second": 3.115, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.2, |
| "eval_math_entropy": 0.9137404448986054, |
| "eval_math_loss": 0.9646264314651489, |
| "eval_math_mean_token_accuracy": 0.7506821854114533, |
| "eval_math_num_tokens": 21496441.0, |
| "eval_math_runtime": 45.2544, |
| "eval_math_samples_per_second": 11.049, |
| "eval_math_steps_per_second": 5.524, |
| "step": 1500 |
| }, |
| { |
| "entropy": 0.9007376233115792, |
| "epoch": 1.208, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 0.8869, |
| "mean_token_accuracy": 0.7582313328981399, |
| "num_tokens": 21639067.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 0.8966150769963861, |
| "epoch": 1.216, |
| "grad_norm": 4.75, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 0.8658, |
| "mean_token_accuracy": 0.7638150323182344, |
| "num_tokens": 21778030.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 0.890181316435337, |
| "epoch": 1.224, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 0.8708, |
| "mean_token_accuracy": 0.7650891322642565, |
| "num_tokens": 21920512.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 0.8583106055855751, |
| "epoch": 1.232, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 0.8536, |
| "mean_token_accuracy": 0.7647566251456738, |
| "num_tokens": 22060851.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 0.8502283520996571, |
| "epoch": 1.24, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.878e-05, |
| "loss": 0.8405, |
| "mean_token_accuracy": 0.7698996499180794, |
| "num_tokens": 22206786.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 0.8690088042989373, |
| "epoch": 1.248, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 0.8465, |
| "mean_token_accuracy": 0.7664700176566839, |
| "num_tokens": 22348738.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 0.8944761479273439, |
| "epoch": 1.256, |
| "grad_norm": 5.78125, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 0.8814, |
| "mean_token_accuracy": 0.7623841091990471, |
| "num_tokens": 22492991.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 0.9210197998210787, |
| "epoch": 1.264, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 0.8894, |
| "mean_token_accuracy": 0.7565367586910725, |
| "num_tokens": 22635039.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 0.8753552945330739, |
| "epoch": 1.272, |
| "grad_norm": 5.59375, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 0.8587, |
| "mean_token_accuracy": 0.7652978543192148, |
| "num_tokens": 22769285.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 0.8754111222922802, |
| "epoch": 1.28, |
| "grad_norm": 4.5, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 0.8562, |
| "mean_token_accuracy": 0.7660590428858995, |
| "num_tokens": 22915190.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_coding_entropy": 1.8190780267715454, |
| "eval_coding_loss": 2.1373343467712402, |
| "eval_coding_mean_token_accuracy": 0.5272793205976486, |
| "eval_coding_num_tokens": 22915190.0, |
| "eval_coding_runtime": 80.5353, |
| "eval_coding_samples_per_second": 6.208, |
| "eval_coding_steps_per_second": 3.104, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.28, |
| "eval_math_entropy": 0.9109430243968963, |
| "eval_math_loss": 0.941487729549408, |
| "eval_math_mean_token_accuracy": 0.7543970160484313, |
| "eval_math_num_tokens": 22915190.0, |
| "eval_math_runtime": 45.4717, |
| "eval_math_samples_per_second": 10.996, |
| "eval_math_steps_per_second": 5.498, |
| "step": 1600 |
| }, |
| { |
| "entropy": 0.8607332542538643, |
| "epoch": 1.288, |
| "grad_norm": 6.875, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 0.849, |
| "mean_token_accuracy": 0.7669094074517488, |
| "num_tokens": 23055764.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 0.8501356575638056, |
| "epoch": 1.296, |
| "grad_norm": 11.625, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 0.8352, |
| "mean_token_accuracy": 0.7705895308405161, |
| "num_tokens": 23194947.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 0.8666134277358651, |
| "epoch": 1.304, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 0.8526, |
| "mean_token_accuracy": 0.7642605971544981, |
| "num_tokens": 23339508.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 0.87587650436908, |
| "epoch": 1.312, |
| "grad_norm": 6.0, |
| "learning_rate": 1.858e-05, |
| "loss": 0.8831, |
| "mean_token_accuracy": 0.7628425531089306, |
| "num_tokens": 23483496.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 0.8858171083033085, |
| "epoch": 1.32, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 0.862, |
| "mean_token_accuracy": 0.7631197202950716, |
| "num_tokens": 23624582.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 0.8816602308303118, |
| "epoch": 1.328, |
| "grad_norm": 5.875, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 0.8757, |
| "mean_token_accuracy": 0.7643136404454708, |
| "num_tokens": 23762589.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 0.8687862390652299, |
| "epoch": 1.336, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 0.8495, |
| "mean_token_accuracy": 0.7662194889038801, |
| "num_tokens": 23901568.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 0.8407161304727196, |
| "epoch": 1.3439999999999999, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 0.8401, |
| "mean_token_accuracy": 0.7718680150806904, |
| "num_tokens": 24041497.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 0.8547042405232788, |
| "epoch": 1.3519999999999999, |
| "grad_norm": 5.25, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 0.839, |
| "mean_token_accuracy": 0.770446752756834, |
| "num_tokens": 24184558.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 0.8688854144886136, |
| "epoch": 1.3599999999999999, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 0.8496, |
| "mean_token_accuracy": 0.7686848890036344, |
| "num_tokens": 24330591.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "eval_coding_entropy": 1.6994317145347595, |
| "eval_coding_loss": 2.1480090618133545, |
| "eval_coding_mean_token_accuracy": 0.5283526113033294, |
| "eval_coding_num_tokens": 24330591.0, |
| "eval_coding_runtime": 80.395, |
| "eval_coding_samples_per_second": 6.219, |
| "eval_coding_steps_per_second": 3.11, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.3599999999999999, |
| "eval_math_entropy": 0.8663188086748124, |
| "eval_math_loss": 0.9338840842247009, |
| "eval_math_mean_token_accuracy": 0.756821269750595, |
| "eval_math_num_tokens": 24330591.0, |
| "eval_math_runtime": 45.5195, |
| "eval_math_samples_per_second": 10.984, |
| "eval_math_steps_per_second": 5.492, |
| "step": 1700 |
| }, |
| { |
| "entropy": 0.878911854326725, |
| "epoch": 1.3679999999999999, |
| "grad_norm": 5.6875, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 0.876, |
| "mean_token_accuracy": 0.7649706695228815, |
| "num_tokens": 24470245.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 0.9300806663930417, |
| "epoch": 1.376, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 0.918, |
| "mean_token_accuracy": 0.753387338668108, |
| "num_tokens": 24611420.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 0.8447118682786823, |
| "epoch": 1.384, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 0.8272, |
| "mean_token_accuracy": 0.7724191237241029, |
| "num_tokens": 24752101.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 0.8587421044707299, |
| "epoch": 1.392, |
| "grad_norm": 4.875, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 0.8554, |
| "mean_token_accuracy": 0.7666859105229378, |
| "num_tokens": 24894694.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 0.8358689803630114, |
| "epoch": 1.4, |
| "grad_norm": 5.75, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 0.8156, |
| "mean_token_accuracy": 0.7759720867499709, |
| "num_tokens": 25041723.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 0.8464876215904951, |
| "epoch": 1.408, |
| "grad_norm": 5.46875, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 0.8324, |
| "mean_token_accuracy": 0.7729319699108601, |
| "num_tokens": 25184946.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 0.8552038704976439, |
| "epoch": 1.416, |
| "grad_norm": 5.25, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 0.8433, |
| "mean_token_accuracy": 0.7681127458810806, |
| "num_tokens": 25328147.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 0.8203344563022256, |
| "epoch": 1.424, |
| "grad_norm": 3.953125, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 0.8099, |
| "mean_token_accuracy": 0.7773803893476725, |
| "num_tokens": 25474746.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 0.8401063056662679, |
| "epoch": 1.432, |
| "grad_norm": 5.625, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 0.8302, |
| "mean_token_accuracy": 0.7718876168131829, |
| "num_tokens": 25617380.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 0.8981215547770262, |
| "epoch": 1.44, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 0.8847, |
| "mean_token_accuracy": 0.761121940985322, |
| "num_tokens": 25758550.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_coding_entropy": 1.7459067511558533, |
| "eval_coding_loss": 2.139580488204956, |
| "eval_coding_mean_token_accuracy": 0.5292461370229721, |
| "eval_coding_num_tokens": 25758550.0, |
| "eval_coding_runtime": 80.0857, |
| "eval_coding_samples_per_second": 6.243, |
| "eval_coding_steps_per_second": 3.122, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.44, |
| "eval_math_entropy": 0.8692929556369782, |
| "eval_math_loss": 0.9176215529441833, |
| "eval_math_mean_token_accuracy": 0.7594350986480712, |
| "eval_math_num_tokens": 25758550.0, |
| "eval_math_runtime": 45.2992, |
| "eval_math_samples_per_second": 11.038, |
| "eval_math_steps_per_second": 5.519, |
| "step": 1800 |
| }, |
| { |
| "entropy": 0.8652363793924451, |
| "epoch": 1.448, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 0.8406, |
| "mean_token_accuracy": 0.7703289985656738, |
| "num_tokens": 25898353.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 0.821346534229815, |
| "epoch": 1.456, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 0.8127, |
| "mean_token_accuracy": 0.7736376784741879, |
| "num_tokens": 26039862.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 0.8591237479820848, |
| "epoch": 1.464, |
| "grad_norm": 3.796875, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 0.8476, |
| "mean_token_accuracy": 0.7689028561115265, |
| "num_tokens": 26180423.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 0.8386457657441497, |
| "epoch": 1.472, |
| "grad_norm": 3.609375, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 0.8366, |
| "mean_token_accuracy": 0.7718514669686556, |
| "num_tokens": 26318666.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 0.86677349768579, |
| "epoch": 1.48, |
| "grad_norm": 4.75, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 0.8562, |
| "mean_token_accuracy": 0.7676644265651703, |
| "num_tokens": 26454916.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 0.8781768172979355, |
| "epoch": 1.488, |
| "grad_norm": 5.5, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 0.8703, |
| "mean_token_accuracy": 0.7649923708289862, |
| "num_tokens": 26594867.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 0.8159094477072358, |
| "epoch": 1.496, |
| "grad_norm": 4.375, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 0.8094, |
| "mean_token_accuracy": 0.7793155841529369, |
| "num_tokens": 26740426.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 0.828849502466619, |
| "epoch": 1.504, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 0.8178, |
| "mean_token_accuracy": 0.7788573730736971, |
| "num_tokens": 26887233.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 0.8783965388312935, |
| "epoch": 1.512, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 0.8674, |
| "mean_token_accuracy": 0.7655724518001079, |
| "num_tokens": 27028948.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 0.844552518799901, |
| "epoch": 1.52, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 0.8297, |
| "mean_token_accuracy": 0.7694923385977746, |
| "num_tokens": 27177448.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_coding_entropy": 1.683746066570282, |
| "eval_coding_loss": 2.1466054916381836, |
| "eval_coding_mean_token_accuracy": 0.5325500816106796, |
| "eval_coding_num_tokens": 27177448.0, |
| "eval_coding_runtime": 79.7243, |
| "eval_coding_samples_per_second": 6.272, |
| "eval_coding_steps_per_second": 3.136, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.52, |
| "eval_math_entropy": 0.827024330496788, |
| "eval_math_loss": 0.9100779891014099, |
| "eval_math_mean_token_accuracy": 0.7613636610507966, |
| "eval_math_num_tokens": 27177448.0, |
| "eval_math_runtime": 45.3081, |
| "eval_math_samples_per_second": 11.036, |
| "eval_math_steps_per_second": 5.518, |
| "step": 1900 |
| }, |
| { |
| "entropy": 0.8612215500324965, |
| "epoch": 1.528, |
| "grad_norm": 5.0, |
| "learning_rate": 1.798e-05, |
| "loss": 0.8509, |
| "mean_token_accuracy": 0.7680411688983441, |
| "num_tokens": 27319831.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 0.8220518652349711, |
| "epoch": 1.536, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 0.8195, |
| "mean_token_accuracy": 0.7773624457418918, |
| "num_tokens": 27467857.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 0.8079562712460756, |
| "epoch": 1.544, |
| "grad_norm": 4.625, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 0.798, |
| "mean_token_accuracy": 0.7781507484614849, |
| "num_tokens": 27608050.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 0.8312527898699045, |
| "epoch": 1.552, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 0.8278, |
| "mean_token_accuracy": 0.7739048127084971, |
| "num_tokens": 27755138.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 0.821598158031702, |
| "epoch": 1.56, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 0.7995, |
| "mean_token_accuracy": 0.7789873193949461, |
| "num_tokens": 27900865.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 0.8338596414774656, |
| "epoch": 1.568, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 0.8229, |
| "mean_token_accuracy": 0.7787407040596008, |
| "num_tokens": 28049363.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 0.8418713506311178, |
| "epoch": 1.576, |
| "grad_norm": 5.53125, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 0.8391, |
| "mean_token_accuracy": 0.7727061875164509, |
| "num_tokens": 28190993.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 0.8218053575605154, |
| "epoch": 1.584, |
| "grad_norm": 5.0, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 0.8086, |
| "mean_token_accuracy": 0.7779126781970263, |
| "num_tokens": 28337890.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 0.8246776606887579, |
| "epoch": 1.592, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 0.8105, |
| "mean_token_accuracy": 0.7750975411385298, |
| "num_tokens": 28478533.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 0.8497464153915644, |
| "epoch": 1.6, |
| "grad_norm": 3.921875, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 0.8426, |
| "mean_token_accuracy": 0.769883980229497, |
| "num_tokens": 28626068.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_coding_entropy": 1.7501705222129822, |
| "eval_coding_loss": 2.115278720855713, |
| "eval_coding_mean_token_accuracy": 0.5340379716157914, |
| "eval_coding_num_tokens": 28626068.0, |
| "eval_coding_runtime": 79.9509, |
| "eval_coding_samples_per_second": 6.254, |
| "eval_coding_steps_per_second": 3.127, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_math_entropy": 0.8623598643541336, |
| "eval_math_loss": 0.8901376724243164, |
| "eval_math_mean_token_accuracy": 0.7659974377155304, |
| "eval_math_num_tokens": 28626068.0, |
| "eval_math_runtime": 45.0463, |
| "eval_math_samples_per_second": 11.1, |
| "eval_math_steps_per_second": 5.55, |
| "step": 2000 |
| }, |
| { |
| "entropy": 0.837820990756154, |
| "epoch": 1.608, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.7757777777777777e-05, |
| "loss": 0.8229, |
| "mean_token_accuracy": 0.7745513282716274, |
| "num_tokens": 28766128.0, |
| "step": 2010 |
| }, |
| { |
| "entropy": 0.8420733030885458, |
| "epoch": 1.616, |
| "grad_norm": 3.53125, |
| "learning_rate": 1.7735555555555558e-05, |
| "loss": 0.8462, |
| "mean_token_accuracy": 0.7718999199569225, |
| "num_tokens": 28913208.0, |
| "step": 2020 |
| }, |
| { |
| "entropy": 0.8126023454591632, |
| "epoch": 1.624, |
| "grad_norm": 3.359375, |
| "learning_rate": 1.7713333333333335e-05, |
| "loss": 0.81, |
| "mean_token_accuracy": 0.7789839718490839, |
| "num_tokens": 29050814.0, |
| "step": 2030 |
| }, |
| { |
| "entropy": 0.837799658626318, |
| "epoch": 1.6320000000000001, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.7691111111111113e-05, |
| "loss": 0.8254, |
| "mean_token_accuracy": 0.772352834045887, |
| "num_tokens": 29188917.0, |
| "step": 2040 |
| }, |
| { |
| "entropy": 0.7966205691918731, |
| "epoch": 1.6400000000000001, |
| "grad_norm": 6.28125, |
| "learning_rate": 1.766888888888889e-05, |
| "loss": 0.7858, |
| "mean_token_accuracy": 0.7840570934116841, |
| "num_tokens": 29330204.0, |
| "step": 2050 |
| }, |
| { |
| "entropy": 0.827992920577526, |
| "epoch": 1.6480000000000001, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.7646666666666668e-05, |
| "loss": 0.8186, |
| "mean_token_accuracy": 0.7782142274081707, |
| "num_tokens": 29477292.0, |
| "step": 2060 |
| }, |
| { |
| "entropy": 0.8256994063034654, |
| "epoch": 1.6560000000000001, |
| "grad_norm": 4.625, |
| "learning_rate": 1.7624444444444446e-05, |
| "loss": 0.8087, |
| "mean_token_accuracy": 0.775632555782795, |
| "num_tokens": 29625992.0, |
| "step": 2070 |
| }, |
| { |
| "entropy": 0.8139445355162025, |
| "epoch": 1.6640000000000001, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.7602222222222223e-05, |
| "loss": 0.7977, |
| "mean_token_accuracy": 0.7794208023697138, |
| "num_tokens": 29772480.0, |
| "step": 2080 |
| }, |
| { |
| "entropy": 0.854400297254324, |
| "epoch": 1.6720000000000002, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.758e-05, |
| "loss": 0.8466, |
| "mean_token_accuracy": 0.7679815135896206, |
| "num_tokens": 29918880.0, |
| "step": 2090 |
| }, |
| { |
| "entropy": 0.8064053267240524, |
| "epoch": 1.6800000000000002, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.755777777777778e-05, |
| "loss": 0.7888, |
| "mean_token_accuracy": 0.784536050260067, |
| "num_tokens": 30061007.0, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "eval_coding_entropy": 1.7241254634857177, |
| "eval_coding_loss": 2.12294340133667, |
| "eval_coding_mean_token_accuracy": 0.5336118508577347, |
| "eval_coding_num_tokens": 30061007.0, |
| "eval_coding_runtime": 80.0809, |
| "eval_coding_samples_per_second": 6.244, |
| "eval_coding_steps_per_second": 3.122, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.6800000000000002, |
| "eval_math_entropy": 0.8144398384094238, |
| "eval_math_loss": 0.8816776871681213, |
| "eval_math_mean_token_accuracy": 0.7675907769203186, |
| "eval_math_num_tokens": 30061007.0, |
| "eval_math_runtime": 45.3264, |
| "eval_math_samples_per_second": 11.031, |
| "eval_math_steps_per_second": 5.516, |
| "step": 2100 |
| }, |
| { |
| "entropy": 0.8489811306819319, |
| "epoch": 1.688, |
| "grad_norm": 4.75, |
| "learning_rate": 1.7535555555555556e-05, |
| "loss": 0.8253, |
| "mean_token_accuracy": 0.7727938439697027, |
| "num_tokens": 30199529.0, |
| "step": 2110 |
| }, |
| { |
| "entropy": 0.8082182152196765, |
| "epoch": 1.696, |
| "grad_norm": 5.9375, |
| "learning_rate": 1.7513333333333334e-05, |
| "loss": 0.8025, |
| "mean_token_accuracy": 0.779643752425909, |
| "num_tokens": 30344415.0, |
| "step": 2120 |
| }, |
| { |
| "entropy": 0.8107613243162632, |
| "epoch": 1.704, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.749111111111111e-05, |
| "loss": 0.7969, |
| "mean_token_accuracy": 0.7823810841888189, |
| "num_tokens": 30489871.0, |
| "step": 2130 |
| }, |
| { |
| "entropy": 0.8439682397991419, |
| "epoch": 1.712, |
| "grad_norm": 4.25, |
| "learning_rate": 1.746888888888889e-05, |
| "loss": 0.8148, |
| "mean_token_accuracy": 0.7743747126311064, |
| "num_tokens": 30640891.0, |
| "step": 2140 |
| }, |
| { |
| "entropy": 0.7898154929280281, |
| "epoch": 1.72, |
| "grad_norm": 3.765625, |
| "learning_rate": 1.7446666666666667e-05, |
| "loss": 0.7833, |
| "mean_token_accuracy": 0.7805226560682058, |
| "num_tokens": 30780476.0, |
| "step": 2150 |
| }, |
| { |
| "entropy": 0.8036614634096623, |
| "epoch": 1.728, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.7424444444444444e-05, |
| "loss": 0.785, |
| "mean_token_accuracy": 0.7833148058503866, |
| "num_tokens": 30921295.0, |
| "step": 2160 |
| }, |
| { |
| "entropy": 0.8018153687939048, |
| "epoch": 1.736, |
| "grad_norm": 6.15625, |
| "learning_rate": 1.7402222222222222e-05, |
| "loss": 0.7885, |
| "mean_token_accuracy": 0.7808650441467762, |
| "num_tokens": 31062675.0, |
| "step": 2170 |
| }, |
| { |
| "entropy": 0.8291892172768712, |
| "epoch": 1.744, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 0.8308, |
| "mean_token_accuracy": 0.7759933091700078, |
| "num_tokens": 31206375.0, |
| "step": 2180 |
| }, |
| { |
| "entropy": 0.7900549739599227, |
| "epoch": 1.752, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.735777777777778e-05, |
| "loss": 0.7808, |
| "mean_token_accuracy": 0.7833611223846674, |
| "num_tokens": 31349892.0, |
| "step": 2190 |
| }, |
| { |
| "entropy": 0.8141259755939245, |
| "epoch": 1.76, |
| "grad_norm": 3.84375, |
| "learning_rate": 1.7335555555555558e-05, |
| "loss": 0.8063, |
| "mean_token_accuracy": 0.7753262374550104, |
| "num_tokens": 31491856.0, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_coding_entropy": 1.736436007976532, |
| "eval_coding_loss": 2.0947279930114746, |
| "eval_coding_mean_token_accuracy": 0.5359265700578689, |
| "eval_coding_num_tokens": 31491856.0, |
| "eval_coding_runtime": 80.0668, |
| "eval_coding_samples_per_second": 6.245, |
| "eval_coding_steps_per_second": 3.122, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.76, |
| "eval_math_entropy": 0.8346000514030456, |
| "eval_math_loss": 0.8691937327384949, |
| "eval_math_mean_token_accuracy": 0.7694536716938019, |
| "eval_math_num_tokens": 31491856.0, |
| "eval_math_runtime": 45.657, |
| "eval_math_samples_per_second": 10.951, |
| "eval_math_steps_per_second": 5.476, |
| "step": 2200 |
| }, |
| { |
| "entropy": 0.8055585173889994, |
| "epoch": 1.768, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.7313333333333336e-05, |
| "loss": 0.7946, |
| "mean_token_accuracy": 0.7799558952450752, |
| "num_tokens": 31639387.0, |
| "step": 2210 |
| }, |
| { |
| "entropy": 0.813026848807931, |
| "epoch": 1.776, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.7291111111111113e-05, |
| "loss": 0.7969, |
| "mean_token_accuracy": 0.779479306936264, |
| "num_tokens": 31782829.0, |
| "step": 2220 |
| }, |
| { |
| "entropy": 0.8508387736976146, |
| "epoch": 1.784, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.726888888888889e-05, |
| "loss": 0.8152, |
| "mean_token_accuracy": 0.7752707224339247, |
| "num_tokens": 31927682.0, |
| "step": 2230 |
| }, |
| { |
| "entropy": 0.8000576250255108, |
| "epoch": 1.792, |
| "grad_norm": 3.8125, |
| "learning_rate": 1.724666666666667e-05, |
| "loss": 0.7944, |
| "mean_token_accuracy": 0.7807411305606365, |
| "num_tokens": 32070187.0, |
| "step": 2240 |
| }, |
| { |
| "entropy": 0.8331243453547359, |
| "epoch": 1.8, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.7224444444444446e-05, |
| "loss": 0.8175, |
| "mean_token_accuracy": 0.7756631281226873, |
| "num_tokens": 32209783.0, |
| "step": 2250 |
| }, |
| { |
| "entropy": 0.8137514226138591, |
| "epoch": 1.808, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.7202222222222224e-05, |
| "loss": 0.7899, |
| "mean_token_accuracy": 0.7820551082491874, |
| "num_tokens": 32355072.0, |
| "step": 2260 |
| }, |
| { |
| "entropy": 0.7685122527182102, |
| "epoch": 1.8159999999999998, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.718e-05, |
| "loss": 0.7611, |
| "mean_token_accuracy": 0.7883212938904762, |
| "num_tokens": 32493575.0, |
| "step": 2270 |
| }, |
| { |
| "entropy": 0.7748165799304843, |
| "epoch": 1.8239999999999998, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.715777777777778e-05, |
| "loss": 0.7457, |
| "mean_token_accuracy": 0.7897661957889796, |
| "num_tokens": 32634013.0, |
| "step": 2280 |
| }, |
| { |
| "entropy": 0.8222803980112076, |
| "epoch": 1.8319999999999999, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.7135555555555557e-05, |
| "loss": 0.8206, |
| "mean_token_accuracy": 0.7788942668586969, |
| "num_tokens": 32779380.0, |
| "step": 2290 |
| }, |
| { |
| "entropy": 0.7715628230944276, |
| "epoch": 1.8399999999999999, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.7113333333333334e-05, |
| "loss": 0.7635, |
| "mean_token_accuracy": 0.78925103507936, |
| "num_tokens": 32917222.0, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "eval_coding_entropy": 1.736720157146454, |
| "eval_coding_loss": 2.0968034267425537, |
| "eval_coding_mean_token_accuracy": 0.5346400545835495, |
| "eval_coding_num_tokens": 32917222.0, |
| "eval_coding_runtime": 80.2838, |
| "eval_coding_samples_per_second": 6.228, |
| "eval_coding_steps_per_second": 3.114, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.8399999999999999, |
| "eval_math_entropy": 0.8437877731323242, |
| "eval_math_loss": 0.8606641292572021, |
| "eval_math_mean_token_accuracy": 0.7722736403942109, |
| "eval_math_num_tokens": 32917222.0, |
| "eval_math_runtime": 45.4568, |
| "eval_math_samples_per_second": 10.999, |
| "eval_math_steps_per_second": 5.5, |
| "step": 2300 |
| }, |
| { |
| "entropy": 0.8428482884541154, |
| "epoch": 1.8479999999999999, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.7091111111111112e-05, |
| "loss": 0.8228, |
| "mean_token_accuracy": 0.774522565305233, |
| "num_tokens": 33058167.0, |
| "step": 2310 |
| }, |
| { |
| "entropy": 0.795220804400742, |
| "epoch": 1.8559999999999999, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.706888888888889e-05, |
| "loss": 0.7785, |
| "mean_token_accuracy": 0.7837839517742395, |
| "num_tokens": 33207147.0, |
| "step": 2320 |
| }, |
| { |
| "entropy": 0.809249134361744, |
| "epoch": 1.8639999999999999, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.704666666666667e-05, |
| "loss": 0.801, |
| "mean_token_accuracy": 0.7804740186780691, |
| "num_tokens": 33351298.0, |
| "step": 2330 |
| }, |
| { |
| "entropy": 0.796000718139112, |
| "epoch": 1.8719999999999999, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.7024444444444445e-05, |
| "loss": 0.7813, |
| "mean_token_accuracy": 0.7847625344991684, |
| "num_tokens": 33496572.0, |
| "step": 2340 |
| }, |
| { |
| "entropy": 0.7839631149545312, |
| "epoch": 1.88, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.7002222222222226e-05, |
| "loss": 0.7809, |
| "mean_token_accuracy": 0.786628844961524, |
| "num_tokens": 33637848.0, |
| "step": 2350 |
| }, |
| { |
| "entropy": 0.8307819724082947, |
| "epoch": 1.888, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.698e-05, |
| "loss": 0.8106, |
| "mean_token_accuracy": 0.7778272360563279, |
| "num_tokens": 33775788.0, |
| "step": 2360 |
| }, |
| { |
| "entropy": 0.8217816924676299, |
| "epoch": 1.896, |
| "grad_norm": 3.921875, |
| "learning_rate": 1.695777777777778e-05, |
| "loss": 0.8116, |
| "mean_token_accuracy": 0.7768773242831231, |
| "num_tokens": 33913914.0, |
| "step": 2370 |
| }, |
| { |
| "entropy": 0.7895107766613364, |
| "epoch": 1.904, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.6935555555555555e-05, |
| "loss": 0.7809, |
| "mean_token_accuracy": 0.7821679543703794, |
| "num_tokens": 34052322.0, |
| "step": 2380 |
| }, |
| { |
| "entropy": 0.7899815816432237, |
| "epoch": 1.912, |
| "grad_norm": 4.0, |
| "learning_rate": 1.6913333333333336e-05, |
| "loss": 0.7681, |
| "mean_token_accuracy": 0.7880932979285717, |
| "num_tokens": 34193156.0, |
| "step": 2390 |
| }, |
| { |
| "entropy": 0.7880993392318487, |
| "epoch": 1.92, |
| "grad_norm": 6.09375, |
| "learning_rate": 1.689111111111111e-05, |
| "loss": 0.7853, |
| "mean_token_accuracy": 0.781376288831234, |
| "num_tokens": 34331468.0, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_coding_entropy": 1.7088720936775208, |
| "eval_coding_loss": 2.094611406326294, |
| "eval_coding_mean_token_accuracy": 0.5360869452953338, |
| "eval_coding_num_tokens": 34331468.0, |
| "eval_coding_runtime": 80.2278, |
| "eval_coding_samples_per_second": 6.232, |
| "eval_coding_steps_per_second": 3.116, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.92, |
| "eval_math_entropy": 0.83945922768116, |
| "eval_math_loss": 0.8572995662689209, |
| "eval_math_mean_token_accuracy": 0.7724003307819366, |
| "eval_math_num_tokens": 34331468.0, |
| "eval_math_runtime": 45.3778, |
| "eval_math_samples_per_second": 11.019, |
| "eval_math_steps_per_second": 5.509, |
| "step": 2400 |
| }, |
| { |
| "entropy": 0.7887925436720252, |
| "epoch": 1.928, |
| "grad_norm": 3.875, |
| "learning_rate": 1.686888888888889e-05, |
| "loss": 0.7671, |
| "mean_token_accuracy": 0.7853432096540928, |
| "num_tokens": 34473829.0, |
| "step": 2410 |
| }, |
| { |
| "entropy": 0.8162721384316682, |
| "epoch": 1.936, |
| "grad_norm": 3.671875, |
| "learning_rate": 1.684666666666667e-05, |
| "loss": 0.795, |
| "mean_token_accuracy": 0.7794598631560803, |
| "num_tokens": 34612743.0, |
| "step": 2420 |
| }, |
| { |
| "entropy": 0.8256091730669141, |
| "epoch": 1.944, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.6824444444444447e-05, |
| "loss": 0.8107, |
| "mean_token_accuracy": 0.7800196815282107, |
| "num_tokens": 34751346.0, |
| "step": 2430 |
| }, |
| { |
| "entropy": 0.7719887597486377, |
| "epoch": 1.952, |
| "grad_norm": 3.953125, |
| "learning_rate": 1.6802222222222224e-05, |
| "loss": 0.761, |
| "mean_token_accuracy": 0.7870867639780045, |
| "num_tokens": 34891115.0, |
| "step": 2440 |
| }, |
| { |
| "entropy": 0.7895837098360061, |
| "epoch": 1.96, |
| "grad_norm": 3.78125, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 0.7795, |
| "mean_token_accuracy": 0.7846163548529148, |
| "num_tokens": 35035012.0, |
| "step": 2450 |
| }, |
| { |
| "entropy": 0.8195503711700439, |
| "epoch": 1.968, |
| "grad_norm": 6.28125, |
| "learning_rate": 1.675777777777778e-05, |
| "loss": 0.8125, |
| "mean_token_accuracy": 0.7801840104162693, |
| "num_tokens": 35179505.0, |
| "step": 2460 |
| }, |
| { |
| "entropy": 0.7725927256047725, |
| "epoch": 1.976, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.6735555555555557e-05, |
| "loss": 0.7705, |
| "mean_token_accuracy": 0.7864962343126536, |
| "num_tokens": 35327065.0, |
| "step": 2470 |
| }, |
| { |
| "entropy": 0.7860851565375924, |
| "epoch": 1.984, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.6713333333333335e-05, |
| "loss": 0.774, |
| "mean_token_accuracy": 0.7879247676581145, |
| "num_tokens": 35472596.0, |
| "step": 2480 |
| }, |
| { |
| "entropy": 0.7687792839482427, |
| "epoch": 1.992, |
| "grad_norm": 4.25, |
| "learning_rate": 1.6691111111111112e-05, |
| "loss": 0.7587, |
| "mean_token_accuracy": 0.7887185402214527, |
| "num_tokens": 35629001.0, |
| "step": 2490 |
| }, |
| { |
| "entropy": 0.7837625231593848, |
| "epoch": 2.0, |
| "grad_norm": 3.625, |
| "learning_rate": 1.666888888888889e-05, |
| "loss": 0.7845, |
| "mean_token_accuracy": 0.7862223986536264, |
| "num_tokens": 35772848.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_coding_entropy": 1.7319149060249328, |
| "eval_coding_loss": 2.0931284427642822, |
| "eval_coding_mean_token_accuracy": 0.5360672180652618, |
| "eval_coding_num_tokens": 35772848.0, |
| "eval_coding_runtime": 80.2404, |
| "eval_coding_samples_per_second": 6.231, |
| "eval_coding_steps_per_second": 3.116, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_math_entropy": 0.828873330950737, |
| "eval_math_loss": 0.8462316989898682, |
| "eval_math_mean_token_accuracy": 0.7752388260364532, |
| "eval_math_num_tokens": 35772848.0, |
| "eval_math_runtime": 45.0166, |
| "eval_math_samples_per_second": 11.107, |
| "eval_math_steps_per_second": 5.554, |
| "step": 2500 |
| }, |
| { |
| "entropy": 0.6735414253547788, |
| "epoch": 2.008, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.6646666666666668e-05, |
| "loss": 0.6274, |
| "mean_token_accuracy": 0.8130374934524298, |
| "num_tokens": 35916850.0, |
| "step": 2510 |
| }, |
| { |
| "entropy": 0.6529837347567081, |
| "epoch": 2.016, |
| "grad_norm": 3.265625, |
| "learning_rate": 1.6624444444444445e-05, |
| "loss": 0.6102, |
| "mean_token_accuracy": 0.8186526905745268, |
| "num_tokens": 36060330.0, |
| "step": 2520 |
| }, |
| { |
| "entropy": 0.6179780930280685, |
| "epoch": 2.024, |
| "grad_norm": 3.75, |
| "learning_rate": 1.6602222222222223e-05, |
| "loss": 0.6025, |
| "mean_token_accuracy": 0.819010778516531, |
| "num_tokens": 36196242.0, |
| "step": 2530 |
| }, |
| { |
| "entropy": 0.6179583409801126, |
| "epoch": 2.032, |
| "grad_norm": 3.671875, |
| "learning_rate": 1.658e-05, |
| "loss": 0.6081, |
| "mean_token_accuracy": 0.8189381070435047, |
| "num_tokens": 36343894.0, |
| "step": 2540 |
| }, |
| { |
| "entropy": 0.6131366770714521, |
| "epoch": 2.04, |
| "grad_norm": 3.875, |
| "learning_rate": 1.6557777777777778e-05, |
| "loss": 0.6036, |
| "mean_token_accuracy": 0.8218087896704673, |
| "num_tokens": 36491138.0, |
| "step": 2550 |
| }, |
| { |
| "entropy": 0.600029761530459, |
| "epoch": 2.048, |
| "grad_norm": 3.6875, |
| "learning_rate": 1.6535555555555556e-05, |
| "loss": 0.5895, |
| "mean_token_accuracy": 0.8269942883402109, |
| "num_tokens": 36636410.0, |
| "step": 2560 |
| }, |
| { |
| "entropy": 0.5947625052183867, |
| "epoch": 2.056, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.6513333333333333e-05, |
| "loss": 0.5855, |
| "mean_token_accuracy": 0.8234040837734937, |
| "num_tokens": 36779876.0, |
| "step": 2570 |
| }, |
| { |
| "entropy": 0.617282728664577, |
| "epoch": 2.064, |
| "grad_norm": 3.453125, |
| "learning_rate": 1.6491111111111114e-05, |
| "loss": 0.5946, |
| "mean_token_accuracy": 0.820942048728466, |
| "num_tokens": 36925382.0, |
| "step": 2580 |
| }, |
| { |
| "entropy": 0.6160542763769626, |
| "epoch": 2.072, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.646888888888889e-05, |
| "loss": 0.5958, |
| "mean_token_accuracy": 0.8195015273988246, |
| "num_tokens": 37072484.0, |
| "step": 2590 |
| }, |
| { |
| "entropy": 0.5932559167966247, |
| "epoch": 2.08, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.644666666666667e-05, |
| "loss": 0.5788, |
| "mean_token_accuracy": 0.8278630424290896, |
| "num_tokens": 37208672.0, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_coding_entropy": 1.3430861020088196, |
| "eval_coding_loss": 2.309582471847534, |
| "eval_coding_mean_token_accuracy": 0.528821860074997, |
| "eval_coding_num_tokens": 37208672.0, |
| "eval_coding_runtime": 80.2684, |
| "eval_coding_samples_per_second": 6.229, |
| "eval_coding_steps_per_second": 3.115, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.08, |
| "eval_math_entropy": 0.6452965697646141, |
| "eval_math_loss": 0.8981713056564331, |
| "eval_math_mean_token_accuracy": 0.7727058820724487, |
| "eval_math_num_tokens": 37208672.0, |
| "eval_math_runtime": 45.3659, |
| "eval_math_samples_per_second": 11.021, |
| "eval_math_steps_per_second": 5.511, |
| "step": 2600 |
| }, |
| { |
| "entropy": 0.5977367775514721, |
| "epoch": 2.088, |
| "grad_norm": 4.25, |
| "learning_rate": 1.6424444444444444e-05, |
| "loss": 0.5868, |
| "mean_token_accuracy": 0.824899784475565, |
| "num_tokens": 37351172.0, |
| "step": 2610 |
| }, |
| { |
| "entropy": 0.6267562232911587, |
| "epoch": 2.096, |
| "grad_norm": 3.515625, |
| "learning_rate": 1.6402222222222225e-05, |
| "loss": 0.6152, |
| "mean_token_accuracy": 0.8204413119703531, |
| "num_tokens": 37491954.0, |
| "step": 2620 |
| }, |
| { |
| "entropy": 0.5948652492836117, |
| "epoch": 2.104, |
| "grad_norm": 3.40625, |
| "learning_rate": 1.638e-05, |
| "loss": 0.5701, |
| "mean_token_accuracy": 0.82628378495574, |
| "num_tokens": 37640186.0, |
| "step": 2630 |
| }, |
| { |
| "entropy": 0.6065684221684933, |
| "epoch": 2.112, |
| "grad_norm": 3.625, |
| "learning_rate": 1.635777777777778e-05, |
| "loss": 0.5944, |
| "mean_token_accuracy": 0.8194048006087542, |
| "num_tokens": 37787033.0, |
| "step": 2640 |
| }, |
| { |
| "entropy": 0.6144680310040712, |
| "epoch": 2.12, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.6335555555555558e-05, |
| "loss": 0.6046, |
| "mean_token_accuracy": 0.822132907435298, |
| "num_tokens": 37924697.0, |
| "step": 2650 |
| }, |
| { |
| "entropy": 0.5829701878130435, |
| "epoch": 2.128, |
| "grad_norm": 3.71875, |
| "learning_rate": 1.6313333333333335e-05, |
| "loss": 0.5627, |
| "mean_token_accuracy": 0.830169964209199, |
| "num_tokens": 38068955.0, |
| "step": 2660 |
| }, |
| { |
| "entropy": 0.5947947616688907, |
| "epoch": 2.136, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.6291111111111113e-05, |
| "loss": 0.5906, |
| "mean_token_accuracy": 0.8241069782525301, |
| "num_tokens": 38213816.0, |
| "step": 2670 |
| }, |
| { |
| "entropy": 0.6134304277598858, |
| "epoch": 2.144, |
| "grad_norm": 3.671875, |
| "learning_rate": 1.626888888888889e-05, |
| "loss": 0.5949, |
| "mean_token_accuracy": 0.8213455483317376, |
| "num_tokens": 38359914.0, |
| "step": 2680 |
| }, |
| { |
| "entropy": 0.6099502108991146, |
| "epoch": 2.152, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.6246666666666668e-05, |
| "loss": 0.6023, |
| "mean_token_accuracy": 0.8206498011946678, |
| "num_tokens": 38497639.0, |
| "step": 2690 |
| }, |
| { |
| "entropy": 0.6001534506678581, |
| "epoch": 2.16, |
| "grad_norm": 3.640625, |
| "learning_rate": 1.6224444444444446e-05, |
| "loss": 0.5808, |
| "mean_token_accuracy": 0.8278015557676554, |
| "num_tokens": 38649684.0, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.16, |
| "eval_coding_entropy": 1.358636585712433, |
| "eval_coding_loss": 2.3387935161590576, |
| "eval_coding_mean_token_accuracy": 0.5271971690654754, |
| "eval_coding_num_tokens": 38649684.0, |
| "eval_coding_runtime": 80.2442, |
| "eval_coding_samples_per_second": 6.231, |
| "eval_coding_steps_per_second": 3.115, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.16, |
| "eval_math_entropy": 0.6467408466339112, |
| "eval_math_loss": 0.8951907753944397, |
| "eval_math_mean_token_accuracy": 0.7731802122592926, |
| "eval_math_num_tokens": 38649684.0, |
| "eval_math_runtime": 45.347, |
| "eval_math_samples_per_second": 11.026, |
| "eval_math_steps_per_second": 5.513, |
| "step": 2700 |
| }, |
| { |
| "entropy": 0.584844472259283, |
| "epoch": 2.168, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.6202222222222223e-05, |
| "loss": 0.575, |
| "mean_token_accuracy": 0.8267203096300364, |
| "num_tokens": 38790127.0, |
| "step": 2710 |
| }, |
| { |
| "entropy": 0.5814568039029837, |
| "epoch": 2.176, |
| "grad_norm": 5.875, |
| "learning_rate": 1.618e-05, |
| "loss": 0.5622, |
| "mean_token_accuracy": 0.8280694082379341, |
| "num_tokens": 38938155.0, |
| "step": 2720 |
| }, |
| { |
| "entropy": 0.6045277355238795, |
| "epoch": 2.184, |
| "grad_norm": 3.71875, |
| "learning_rate": 1.615777777777778e-05, |
| "loss": 0.5899, |
| "mean_token_accuracy": 0.8225381962954998, |
| "num_tokens": 39083921.0, |
| "step": 2730 |
| }, |
| { |
| "entropy": 0.6171120654791593, |
| "epoch": 2.192, |
| "grad_norm": 5.25, |
| "learning_rate": 1.6135555555555556e-05, |
| "loss": 0.6059, |
| "mean_token_accuracy": 0.8195779286324978, |
| "num_tokens": 39228800.0, |
| "step": 2740 |
| }, |
| { |
| "entropy": 0.6280247885733843, |
| "epoch": 2.2, |
| "grad_norm": 4.0, |
| "learning_rate": 1.6113333333333334e-05, |
| "loss": 0.6179, |
| "mean_token_accuracy": 0.8163844373077154, |
| "num_tokens": 39367604.0, |
| "step": 2750 |
| }, |
| { |
| "entropy": 0.5978868333622813, |
| "epoch": 2.208, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.609111111111111e-05, |
| "loss": 0.5865, |
| "mean_token_accuracy": 0.8261822827160359, |
| "num_tokens": 39513118.0, |
| "step": 2760 |
| }, |
| { |
| "entropy": 0.6030507305637002, |
| "epoch": 2.216, |
| "grad_norm": 3.4375, |
| "learning_rate": 1.606888888888889e-05, |
| "loss": 0.5908, |
| "mean_token_accuracy": 0.8233633246272802, |
| "num_tokens": 39657164.0, |
| "step": 2770 |
| }, |
| { |
| "entropy": 0.5979916060343384, |
| "epoch": 2.224, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.6046666666666667e-05, |
| "loss": 0.5844, |
| "mean_token_accuracy": 0.8251450814306736, |
| "num_tokens": 39804212.0, |
| "step": 2780 |
| }, |
| { |
| "entropy": 0.6049735843203962, |
| "epoch": 2.232, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.6024444444444444e-05, |
| "loss": 0.59, |
| "mean_token_accuracy": 0.8252436757087708, |
| "num_tokens": 39948164.0, |
| "step": 2790 |
| }, |
| { |
| "entropy": 0.6112302895635366, |
| "epoch": 2.24, |
| "grad_norm": 3.6875, |
| "learning_rate": 1.6002222222222222e-05, |
| "loss": 0.6102, |
| "mean_token_accuracy": 0.8215605676174164, |
| "num_tokens": 40092012.0, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_coding_entropy": 1.367234740257263, |
| "eval_coding_loss": 2.310321092605591, |
| "eval_coding_mean_token_accuracy": 0.5290858243703842, |
| "eval_coding_num_tokens": 40092012.0, |
| "eval_coding_runtime": 80.4951, |
| "eval_coding_samples_per_second": 6.212, |
| "eval_coding_steps_per_second": 3.106, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.24, |
| "eval_math_entropy": 0.6641956909894943, |
| "eval_math_loss": 0.8882858753204346, |
| "eval_math_mean_token_accuracy": 0.7742883183956146, |
| "eval_math_num_tokens": 40092012.0, |
| "eval_math_runtime": 45.3581, |
| "eval_math_samples_per_second": 11.023, |
| "eval_math_steps_per_second": 5.512, |
| "step": 2800 |
| }, |
| { |
| "entropy": 0.6122245856560766, |
| "epoch": 2.248, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 0.593, |
| "mean_token_accuracy": 0.8206648476421833, |
| "num_tokens": 40239280.0, |
| "step": 2810 |
| }, |
| { |
| "entropy": 0.5890989113599062, |
| "epoch": 2.2560000000000002, |
| "grad_norm": 3.84375, |
| "learning_rate": 1.5957777777777777e-05, |
| "loss": 0.5799, |
| "mean_token_accuracy": 0.8253789249807596, |
| "num_tokens": 40380069.0, |
| "step": 2820 |
| }, |
| { |
| "entropy": 0.6324972776696086, |
| "epoch": 2.2640000000000002, |
| "grad_norm": 3.796875, |
| "learning_rate": 1.5935555555555558e-05, |
| "loss": 0.6229, |
| "mean_token_accuracy": 0.8179504320025444, |
| "num_tokens": 40518358.0, |
| "step": 2830 |
| }, |
| { |
| "entropy": 0.6080603180453181, |
| "epoch": 2.2720000000000002, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.5913333333333332e-05, |
| "loss": 0.5876, |
| "mean_token_accuracy": 0.8210589576512575, |
| "num_tokens": 40655903.0, |
| "step": 2840 |
| }, |
| { |
| "entropy": 0.5939296236261725, |
| "epoch": 2.2800000000000002, |
| "grad_norm": 3.703125, |
| "learning_rate": 1.5891111111111113e-05, |
| "loss": 0.5817, |
| "mean_token_accuracy": 0.8257537618279457, |
| "num_tokens": 40805856.0, |
| "step": 2850 |
| }, |
| { |
| "entropy": 0.626422270014882, |
| "epoch": 2.288, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.5868888888888888e-05, |
| "loss": 0.6103, |
| "mean_token_accuracy": 0.8156515300273895, |
| "num_tokens": 40944791.0, |
| "step": 2860 |
| }, |
| { |
| "entropy": 0.602643084153533, |
| "epoch": 2.296, |
| "grad_norm": 5.125, |
| "learning_rate": 1.584666666666667e-05, |
| "loss": 0.5793, |
| "mean_token_accuracy": 0.8242363192141056, |
| "num_tokens": 41092054.0, |
| "step": 2870 |
| }, |
| { |
| "entropy": 0.6078485570847988, |
| "epoch": 2.304, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.5824444444444446e-05, |
| "loss": 0.59, |
| "mean_token_accuracy": 0.8246722422540188, |
| "num_tokens": 41227146.0, |
| "step": 2880 |
| }, |
| { |
| "entropy": 0.6273407065309584, |
| "epoch": 2.312, |
| "grad_norm": 3.90625, |
| "learning_rate": 1.5802222222222224e-05, |
| "loss": 0.6203, |
| "mean_token_accuracy": 0.8199824463576079, |
| "num_tokens": 41367070.0, |
| "step": 2890 |
| }, |
| { |
| "entropy": 0.6072839391417801, |
| "epoch": 2.32, |
| "grad_norm": 3.890625, |
| "learning_rate": 1.578e-05, |
| "loss": 0.5969, |
| "mean_token_accuracy": 0.8211672656238079, |
| "num_tokens": 41509322.0, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.32, |
| "eval_coding_entropy": 1.3690478925704956, |
| "eval_coding_loss": 2.325049877166748, |
| "eval_coding_mean_token_accuracy": 0.5277329924106597, |
| "eval_coding_num_tokens": 41509322.0, |
| "eval_coding_runtime": 80.3505, |
| "eval_coding_samples_per_second": 6.223, |
| "eval_coding_steps_per_second": 3.111, |
| "step": 2900 |
| }, |
| { |
| "epoch": 2.32, |
| "eval_math_entropy": 0.6644403736591339, |
| "eval_math_loss": 0.8825833797454834, |
| "eval_math_mean_token_accuracy": 0.7753834664821625, |
| "eval_math_num_tokens": 41509322.0, |
| "eval_math_runtime": 45.3945, |
| "eval_math_samples_per_second": 11.015, |
| "eval_math_steps_per_second": 5.507, |
| "step": 2900 |
| }, |
| { |
| "entropy": 0.6207097603008151, |
| "epoch": 2.328, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.575777777777778e-05, |
| "loss": 0.6098, |
| "mean_token_accuracy": 0.8228004731237888, |
| "num_tokens": 41647972.0, |
| "step": 2910 |
| }, |
| { |
| "entropy": 0.5956490144133568, |
| "epoch": 2.336, |
| "grad_norm": 3.859375, |
| "learning_rate": 1.5735555555555557e-05, |
| "loss": 0.5802, |
| "mean_token_accuracy": 0.8249830607324838, |
| "num_tokens": 41792799.0, |
| "step": 2920 |
| }, |
| { |
| "entropy": 0.6102036783471704, |
| "epoch": 2.344, |
| "grad_norm": 3.53125, |
| "learning_rate": 1.5713333333333334e-05, |
| "loss": 0.5957, |
| "mean_token_accuracy": 0.8222156394273042, |
| "num_tokens": 41937917.0, |
| "step": 2930 |
| }, |
| { |
| "entropy": 0.6381065575405955, |
| "epoch": 2.352, |
| "grad_norm": 4.125, |
| "learning_rate": 1.5691111111111112e-05, |
| "loss": 0.623, |
| "mean_token_accuracy": 0.8146627087146043, |
| "num_tokens": 42083080.0, |
| "step": 2940 |
| }, |
| { |
| "entropy": 0.5995131282135844, |
| "epoch": 2.36, |
| "grad_norm": 3.765625, |
| "learning_rate": 1.5668888888888893e-05, |
| "loss": 0.5903, |
| "mean_token_accuracy": 0.8256115213036537, |
| "num_tokens": 42225109.0, |
| "step": 2950 |
| }, |
| { |
| "entropy": 0.6044453907757997, |
| "epoch": 2.368, |
| "grad_norm": 4.1875, |
| "learning_rate": 1.5646666666666667e-05, |
| "loss": 0.5917, |
| "mean_token_accuracy": 0.8233913883566857, |
| "num_tokens": 42368442.0, |
| "step": 2960 |
| }, |
| { |
| "entropy": 0.612680570781231, |
| "epoch": 2.376, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.5624444444444448e-05, |
| "loss": 0.6082, |
| "mean_token_accuracy": 0.8218625281006098, |
| "num_tokens": 42519610.0, |
| "step": 2970 |
| }, |
| { |
| "entropy": 0.6331062354147434, |
| "epoch": 2.384, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.5602222222222222e-05, |
| "loss": 0.6226, |
| "mean_token_accuracy": 0.8189941883087158, |
| "num_tokens": 42663926.0, |
| "step": 2980 |
| }, |
| { |
| "entropy": 0.602106521371752, |
| "epoch": 2.392, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 0.5926, |
| "mean_token_accuracy": 0.8226938959211111, |
| "num_tokens": 42806515.0, |
| "step": 2990 |
| }, |
| { |
| "entropy": 0.6206376163288951, |
| "epoch": 2.4, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.5557777777777778e-05, |
| "loss": 0.6051, |
| "mean_token_accuracy": 0.8227528646588326, |
| "num_tokens": 42946351.0, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_coding_entropy": 1.3359435107707978, |
| "eval_coding_loss": 2.3117940425872803, |
| "eval_coding_mean_token_accuracy": 0.5303116244077682, |
| "eval_coding_num_tokens": 42946351.0, |
| "eval_coding_runtime": 80.2007, |
| "eval_coding_samples_per_second": 6.234, |
| "eval_coding_steps_per_second": 3.117, |
| "step": 3000 |
| }, |
| { |
| "epoch": 2.4, |
| "eval_math_entropy": 0.6451345187425613, |
| "eval_math_loss": 0.8811075091362, |
| "eval_math_mean_token_accuracy": 0.7760629127025604, |
| "eval_math_num_tokens": 42946351.0, |
| "eval_math_runtime": 45.2237, |
| "eval_math_samples_per_second": 11.056, |
| "eval_math_steps_per_second": 5.528, |
| "step": 3000 |
| }, |
| { |
| "entropy": 0.6195723777636886, |
| "epoch": 2.408, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.553555555555556e-05, |
| "loss": 0.6106, |
| "mean_token_accuracy": 0.8195767261087894, |
| "num_tokens": 43086271.0, |
| "step": 3010 |
| }, |
| { |
| "entropy": 0.6303073976188898, |
| "epoch": 2.416, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.5513333333333333e-05, |
| "loss": 0.614, |
| "mean_token_accuracy": 0.8191704362630844, |
| "num_tokens": 43227782.0, |
| "step": 3020 |
| }, |
| { |
| "entropy": 0.62420126888901, |
| "epoch": 2.424, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.5491111111111114e-05, |
| "loss": 0.6211, |
| "mean_token_accuracy": 0.818870022520423, |
| "num_tokens": 43368240.0, |
| "step": 3030 |
| }, |
| { |
| "entropy": 0.5791218370199204, |
| "epoch": 2.432, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.546888888888889e-05, |
| "loss": 0.5759, |
| "mean_token_accuracy": 0.8293201316148042, |
| "num_tokens": 43511939.0, |
| "step": 3040 |
| }, |
| { |
| "entropy": 0.5735427383333445, |
| "epoch": 2.44, |
| "grad_norm": 4.125, |
| "learning_rate": 1.544666666666667e-05, |
| "loss": 0.5512, |
| "mean_token_accuracy": 0.8342619281262159, |
| "num_tokens": 43663309.0, |
| "step": 3050 |
| }, |
| { |
| "entropy": 0.6128185350447893, |
| "epoch": 2.448, |
| "grad_norm": 4.25, |
| "learning_rate": 1.5424444444444447e-05, |
| "loss": 0.6083, |
| "mean_token_accuracy": 0.8209333509206772, |
| "num_tokens": 43811005.0, |
| "step": 3060 |
| }, |
| { |
| "entropy": 0.6115518560633063, |
| "epoch": 2.456, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.5402222222222224e-05, |
| "loss": 0.6022, |
| "mean_token_accuracy": 0.8200164671987296, |
| "num_tokens": 43958354.0, |
| "step": 3070 |
| }, |
| { |
| "entropy": 0.616653636842966, |
| "epoch": 2.464, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 0.6011, |
| "mean_token_accuracy": 0.8216471027582883, |
| "num_tokens": 44106223.0, |
| "step": 3080 |
| }, |
| { |
| "entropy": 0.6570189191028476, |
| "epoch": 2.472, |
| "grad_norm": 3.78125, |
| "learning_rate": 1.535777777777778e-05, |
| "loss": 0.6351, |
| "mean_token_accuracy": 0.8137945652008056, |
| "num_tokens": 44243030.0, |
| "step": 3090 |
| }, |
| { |
| "entropy": 0.6116200280375779, |
| "epoch": 2.48, |
| "grad_norm": 4.125, |
| "learning_rate": 1.5335555555555557e-05, |
| "loss": 0.5943, |
| "mean_token_accuracy": 0.8211002852767706, |
| "num_tokens": 44379095.0, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.48, |
| "eval_coding_entropy": 1.3962756509780885, |
| "eval_coding_loss": 2.285658359527588, |
| "eval_coding_mean_token_accuracy": 0.5313360285758972, |
| "eval_coding_num_tokens": 44379095.0, |
| "eval_coding_runtime": 80.5498, |
| "eval_coding_samples_per_second": 6.207, |
| "eval_coding_steps_per_second": 3.104, |
| "step": 3100 |
| }, |
| { |
| "epoch": 2.48, |
| "eval_math_entropy": 0.6653561153411865, |
| "eval_math_loss": 0.8721400499343872, |
| "eval_math_mean_token_accuracy": 0.7776806814670563, |
| "eval_math_num_tokens": 44379095.0, |
| "eval_math_runtime": 45.5684, |
| "eval_math_samples_per_second": 10.973, |
| "eval_math_steps_per_second": 5.486, |
| "step": 3100 |
| }, |
| { |
| "entropy": 0.6091651175171137, |
| "epoch": 2.488, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.5313333333333335e-05, |
| "loss": 0.5956, |
| "mean_token_accuracy": 0.8218154896050691, |
| "num_tokens": 44523801.0, |
| "step": 3110 |
| }, |
| { |
| "entropy": 0.6286436004564167, |
| "epoch": 2.496, |
| "grad_norm": 4.125, |
| "learning_rate": 1.5291111111111112e-05, |
| "loss": 0.6202, |
| "mean_token_accuracy": 0.8169801935553551, |
| "num_tokens": 44670316.0, |
| "step": 3120 |
| }, |
| { |
| "entropy": 0.626076702773571, |
| "epoch": 2.504, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.526888888888889e-05, |
| "loss": 0.6112, |
| "mean_token_accuracy": 0.8196502018719911, |
| "num_tokens": 44812412.0, |
| "step": 3130 |
| }, |
| { |
| "entropy": 0.5890089662745595, |
| "epoch": 2.512, |
| "grad_norm": 4.25, |
| "learning_rate": 1.5246666666666668e-05, |
| "loss": 0.5845, |
| "mean_token_accuracy": 0.8265151169151068, |
| "num_tokens": 44952877.0, |
| "step": 3140 |
| }, |
| { |
| "entropy": 0.5783503481186927, |
| "epoch": 2.52, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.5224444444444447e-05, |
| "loss": 0.5587, |
| "mean_token_accuracy": 0.8326530661433935, |
| "num_tokens": 45093032.0, |
| "step": 3150 |
| }, |
| { |
| "entropy": 0.6343388651497662, |
| "epoch": 2.528, |
| "grad_norm": 3.75, |
| "learning_rate": 1.5202222222222223e-05, |
| "loss": 0.6286, |
| "mean_token_accuracy": 0.8168471794575453, |
| "num_tokens": 45237653.0, |
| "step": 3160 |
| }, |
| { |
| "entropy": 0.6116237677633762, |
| "epoch": 2.536, |
| "grad_norm": 3.734375, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 0.606, |
| "mean_token_accuracy": 0.8205171436071396, |
| "num_tokens": 45382845.0, |
| "step": 3170 |
| }, |
| { |
| "entropy": 0.5925072895362973, |
| "epoch": 2.544, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.5157777777777778e-05, |
| "loss": 0.5789, |
| "mean_token_accuracy": 0.8276698347181082, |
| "num_tokens": 45524866.0, |
| "step": 3180 |
| }, |
| { |
| "entropy": 0.6094906609505415, |
| "epoch": 2.552, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.5135555555555557e-05, |
| "loss": 0.6068, |
| "mean_token_accuracy": 0.820575163513422, |
| "num_tokens": 45658825.0, |
| "step": 3190 |
| }, |
| { |
| "entropy": 0.5766677531413734, |
| "epoch": 2.56, |
| "grad_norm": 3.75, |
| "learning_rate": 1.5113333333333335e-05, |
| "loss": 0.5561, |
| "mean_token_accuracy": 0.8320007264614105, |
| "num_tokens": 45803783.0, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_coding_entropy": 1.3296539869308472, |
| "eval_coding_loss": 2.3300273418426514, |
| "eval_coding_mean_token_accuracy": 0.5285363863706589, |
| "eval_coding_num_tokens": 45803783.0, |
| "eval_coding_runtime": 80.4282, |
| "eval_coding_samples_per_second": 6.217, |
| "eval_coding_steps_per_second": 3.108, |
| "step": 3200 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_math_entropy": 0.6420052548646927, |
| "eval_math_loss": 0.8757193684577942, |
| "eval_math_mean_token_accuracy": 0.7778271386623382, |
| "eval_math_num_tokens": 45803783.0, |
| "eval_math_runtime": 45.4668, |
| "eval_math_samples_per_second": 10.997, |
| "eval_math_steps_per_second": 5.499, |
| "step": 3200 |
| }, |
| { |
| "entropy": 0.5817652337253094, |
| "epoch": 2.568, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.5091111111111113e-05, |
| "loss": 0.5828, |
| "mean_token_accuracy": 0.8272556405514478, |
| "num_tokens": 45950221.0, |
| "step": 3210 |
| }, |
| { |
| "entropy": 0.6090063130483031, |
| "epoch": 2.576, |
| "grad_norm": 3.765625, |
| "learning_rate": 1.506888888888889e-05, |
| "loss": 0.5998, |
| "mean_token_accuracy": 0.8237855039536953, |
| "num_tokens": 46090713.0, |
| "step": 3220 |
| }, |
| { |
| "entropy": 0.6048751763999463, |
| "epoch": 2.584, |
| "grad_norm": 3.53125, |
| "learning_rate": 1.5046666666666668e-05, |
| "loss": 0.5924, |
| "mean_token_accuracy": 0.8245016437023878, |
| "num_tokens": 46237183.0, |
| "step": 3230 |
| }, |
| { |
| "entropy": 0.5832300594076514, |
| "epoch": 2.592, |
| "grad_norm": 3.390625, |
| "learning_rate": 1.5024444444444445e-05, |
| "loss": 0.5773, |
| "mean_token_accuracy": 0.8293544840067625, |
| "num_tokens": 46384084.0, |
| "step": 3240 |
| }, |
| { |
| "entropy": 0.5941741293296218, |
| "epoch": 2.6, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.5002222222222223e-05, |
| "loss": 0.572, |
| "mean_token_accuracy": 0.8266665641218424, |
| "num_tokens": 46527852.0, |
| "step": 3250 |
| }, |
| { |
| "entropy": 0.6402363840490579, |
| "epoch": 2.608, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.498e-05, |
| "loss": 0.6342, |
| "mean_token_accuracy": 0.814816839247942, |
| "num_tokens": 46667552.0, |
| "step": 3260 |
| }, |
| { |
| "entropy": 0.5911107852123678, |
| "epoch": 2.616, |
| "grad_norm": 3.53125, |
| "learning_rate": 1.495777777777778e-05, |
| "loss": 0.5698, |
| "mean_token_accuracy": 0.8299769192934037, |
| "num_tokens": 46817714.0, |
| "step": 3270 |
| }, |
| { |
| "entropy": 0.5668670304119587, |
| "epoch": 2.624, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.4935555555555556e-05, |
| "loss": 0.552, |
| "mean_token_accuracy": 0.8342004626989364, |
| "num_tokens": 46961124.0, |
| "step": 3280 |
| }, |
| { |
| "entropy": 0.6150367921218276, |
| "epoch": 2.632, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.4913333333333335e-05, |
| "loss": 0.604, |
| "mean_token_accuracy": 0.8216619595885277, |
| "num_tokens": 47100906.0, |
| "step": 3290 |
| }, |
| { |
| "entropy": 0.6032324797473848, |
| "epoch": 2.64, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.4891111111111111e-05, |
| "loss": 0.5914, |
| "mean_token_accuracy": 0.8257958121597767, |
| "num_tokens": 47245326.0, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.64, |
| "eval_coding_entropy": 1.3401714746952056, |
| "eval_coding_loss": 2.317732810974121, |
| "eval_coding_mean_token_accuracy": 0.5300027307271957, |
| "eval_coding_num_tokens": 47245326.0, |
| "eval_coding_runtime": 80.4688, |
| "eval_coding_samples_per_second": 6.214, |
| "eval_coding_steps_per_second": 3.107, |
| "step": 3300 |
| }, |
| { |
| "epoch": 2.64, |
| "eval_math_entropy": 0.6371326298713684, |
| "eval_math_loss": 0.874556303024292, |
| "eval_math_mean_token_accuracy": 0.7783330268859864, |
| "eval_math_num_tokens": 47245326.0, |
| "eval_math_runtime": 45.4681, |
| "eval_math_samples_per_second": 10.997, |
| "eval_math_steps_per_second": 5.498, |
| "step": 3300 |
| }, |
| { |
| "entropy": 0.5668387338519096, |
| "epoch": 2.648, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.486888888888889e-05, |
| "loss": 0.5596, |
| "mean_token_accuracy": 0.8290297947824001, |
| "num_tokens": 47383295.0, |
| "step": 3310 |
| }, |
| { |
| "entropy": 0.5910252512432634, |
| "epoch": 2.656, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.4846666666666666e-05, |
| "loss": 0.5758, |
| "mean_token_accuracy": 0.8262805126607418, |
| "num_tokens": 47527347.0, |
| "step": 3320 |
| }, |
| { |
| "entropy": 0.6149861505255103, |
| "epoch": 2.664, |
| "grad_norm": 3.8125, |
| "learning_rate": 1.4824444444444446e-05, |
| "loss": 0.5953, |
| "mean_token_accuracy": 0.8206445593386889, |
| "num_tokens": 47672245.0, |
| "step": 3330 |
| }, |
| { |
| "entropy": 0.6189125653356313, |
| "epoch": 2.672, |
| "grad_norm": 5.46875, |
| "learning_rate": 1.4802222222222225e-05, |
| "loss": 0.607, |
| "mean_token_accuracy": 0.8170826382935047, |
| "num_tokens": 47808167.0, |
| "step": 3340 |
| }, |
| { |
| "entropy": 0.6008067956194282, |
| "epoch": 2.68, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.478e-05, |
| "loss": 0.5837, |
| "mean_token_accuracy": 0.8247155662626028, |
| "num_tokens": 47948898.0, |
| "step": 3350 |
| }, |
| { |
| "entropy": 0.6006117646582425, |
| "epoch": 2.6879999999999997, |
| "grad_norm": 3.71875, |
| "learning_rate": 1.475777777777778e-05, |
| "loss": 0.5913, |
| "mean_token_accuracy": 0.8237164456397295, |
| "num_tokens": 48092035.0, |
| "step": 3360 |
| }, |
| { |
| "entropy": 0.5730712346732616, |
| "epoch": 2.6959999999999997, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.4735555555555556e-05, |
| "loss": 0.5527, |
| "mean_token_accuracy": 0.8319510795176029, |
| "num_tokens": 48235772.0, |
| "step": 3370 |
| }, |
| { |
| "entropy": 0.6454704931005836, |
| "epoch": 2.7039999999999997, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.4713333333333335e-05, |
| "loss": 0.6369, |
| "mean_token_accuracy": 0.8137039370834828, |
| "num_tokens": 48386231.0, |
| "step": 3380 |
| }, |
| { |
| "entropy": 0.6155837027356028, |
| "epoch": 2.7119999999999997, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.4691111111111111e-05, |
| "loss": 0.6111, |
| "mean_token_accuracy": 0.8185550339519978, |
| "num_tokens": 48525612.0, |
| "step": 3390 |
| }, |
| { |
| "entropy": 0.569570798240602, |
| "epoch": 2.7199999999999998, |
| "grad_norm": 3.953125, |
| "learning_rate": 1.466888888888889e-05, |
| "loss": 0.5519, |
| "mean_token_accuracy": 0.8328184049576521, |
| "num_tokens": 48663262.0, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_coding_entropy": 1.3565053853988647, |
| "eval_coding_loss": 2.325847864151001, |
| "eval_coding_mean_token_accuracy": 0.5291582072973251, |
| "eval_coding_num_tokens": 48663262.0, |
| "eval_coding_runtime": 80.4263, |
| "eval_coding_samples_per_second": 6.217, |
| "eval_coding_steps_per_second": 3.108, |
| "step": 3400 |
| }, |
| { |
| "epoch": 2.7199999999999998, |
| "eval_math_entropy": 0.6437968505620957, |
| "eval_math_loss": 0.8697450160980225, |
| "eval_math_mean_token_accuracy": 0.77901797914505, |
| "eval_math_num_tokens": 48663262.0, |
| "eval_math_runtime": 45.431, |
| "eval_math_samples_per_second": 11.006, |
| "eval_math_steps_per_second": 5.503, |
| "step": 3400 |
| }, |
| { |
| "entropy": 0.5994617763906718, |
| "epoch": 2.7279999999999998, |
| "grad_norm": 4.5, |
| "learning_rate": 1.4646666666666666e-05, |
| "loss": 0.5845, |
| "mean_token_accuracy": 0.8254001647233963, |
| "num_tokens": 48806650.0, |
| "step": 3410 |
| }, |
| { |
| "entropy": 0.5916474238969386, |
| "epoch": 2.7359999999999998, |
| "grad_norm": 4.125, |
| "learning_rate": 1.4624444444444446e-05, |
| "loss": 0.585, |
| "mean_token_accuracy": 0.8274773966521025, |
| "num_tokens": 48949456.0, |
| "step": 3420 |
| }, |
| { |
| "entropy": 0.5937440561130642, |
| "epoch": 2.7439999999999998, |
| "grad_norm": 3.6875, |
| "learning_rate": 1.4602222222222225e-05, |
| "loss": 0.5769, |
| "mean_token_accuracy": 0.8288517195731402, |
| "num_tokens": 49095798.0, |
| "step": 3430 |
| }, |
| { |
| "entropy": 0.5918507298454643, |
| "epoch": 2.752, |
| "grad_norm": 3.953125, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 0.5929, |
| "mean_token_accuracy": 0.8225170582532882, |
| "num_tokens": 49235751.0, |
| "step": 3440 |
| }, |
| { |
| "entropy": 0.6349924458190799, |
| "epoch": 2.76, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.455777777777778e-05, |
| "loss": 0.6132, |
| "mean_token_accuracy": 0.8191227190196514, |
| "num_tokens": 49378604.0, |
| "step": 3450 |
| }, |
| { |
| "entropy": 0.5800085199065507, |
| "epoch": 2.768, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.4535555555555556e-05, |
| "loss": 0.5769, |
| "mean_token_accuracy": 0.8278793919831514, |
| "num_tokens": 49524628.0, |
| "step": 3460 |
| }, |
| { |
| "entropy": 0.6092205945402384, |
| "epoch": 2.776, |
| "grad_norm": 4.0, |
| "learning_rate": 1.4513333333333336e-05, |
| "loss": 0.6009, |
| "mean_token_accuracy": 0.8238059796392918, |
| "num_tokens": 49659115.0, |
| "step": 3470 |
| }, |
| { |
| "entropy": 0.6272034887224436, |
| "epoch": 2.784, |
| "grad_norm": 4.375, |
| "learning_rate": 1.4491111111111111e-05, |
| "loss": 0.6103, |
| "mean_token_accuracy": 0.8182276464998722, |
| "num_tokens": 49799604.0, |
| "step": 3480 |
| }, |
| { |
| "entropy": 0.5972219604998827, |
| "epoch": 2.792, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.446888888888889e-05, |
| "loss": 0.5931, |
| "mean_token_accuracy": 0.826616058126092, |
| "num_tokens": 49937944.0, |
| "step": 3490 |
| }, |
| { |
| "entropy": 0.592647529952228, |
| "epoch": 2.8, |
| "grad_norm": 3.25, |
| "learning_rate": 1.4446666666666668e-05, |
| "loss": 0.5807, |
| "mean_token_accuracy": 0.8266494810581207, |
| "num_tokens": 50079307.0, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_coding_entropy": 1.3411680674552917, |
| "eval_coding_loss": 2.3138482570648193, |
| "eval_coding_mean_token_accuracy": 0.5302999067306519, |
| "eval_coding_num_tokens": 50079307.0, |
| "eval_coding_runtime": 85.508, |
| "eval_coding_samples_per_second": 5.847, |
| "eval_coding_steps_per_second": 2.924, |
| "step": 3500 |
| }, |
| { |
| "epoch": 2.8, |
| "eval_math_entropy": 0.6460445786714554, |
| "eval_math_loss": 0.8624674081802368, |
| "eval_math_mean_token_accuracy": 0.7803667221069336, |
| "eval_math_num_tokens": 50079307.0, |
| "eval_math_runtime": 45.1518, |
| "eval_math_samples_per_second": 11.074, |
| "eval_math_steps_per_second": 5.537, |
| "step": 3500 |
| }, |
| { |
| "entropy": 0.6025649834424257, |
| "epoch": 2.808, |
| "grad_norm": 3.9375, |
| "learning_rate": 1.4424444444444446e-05, |
| "loss": 0.592, |
| "mean_token_accuracy": 0.8237518947571516, |
| "num_tokens": 50226527.0, |
| "step": 3510 |
| }, |
| { |
| "entropy": 0.6065294581465424, |
| "epoch": 2.816, |
| "grad_norm": 4.09375, |
| "learning_rate": 1.4402222222222224e-05, |
| "loss": 0.5761, |
| "mean_token_accuracy": 0.8253755778074264, |
| "num_tokens": 50372216.0, |
| "step": 3520 |
| }, |
| { |
| "entropy": 0.570460251159966, |
| "epoch": 2.824, |
| "grad_norm": 3.96875, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 0.5598, |
| "mean_token_accuracy": 0.8309069123119116, |
| "num_tokens": 50513700.0, |
| "step": 3530 |
| }, |
| { |
| "entropy": 0.5963032564148307, |
| "epoch": 2.832, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.4357777777777779e-05, |
| "loss": 0.5892, |
| "mean_token_accuracy": 0.8245172429829836, |
| "num_tokens": 50657211.0, |
| "step": 3540 |
| }, |
| { |
| "entropy": 0.5929852182045579, |
| "epoch": 2.84, |
| "grad_norm": 3.71875, |
| "learning_rate": 1.4335555555555556e-05, |
| "loss": 0.5884, |
| "mean_token_accuracy": 0.8264828316867352, |
| "num_tokens": 50800221.0, |
| "step": 3550 |
| }, |
| { |
| "entropy": 0.6365296164527535, |
| "epoch": 2.848, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.4313333333333334e-05, |
| "loss": 0.6198, |
| "mean_token_accuracy": 0.8176385808736086, |
| "num_tokens": 50946503.0, |
| "step": 3560 |
| }, |
| { |
| "entropy": 0.6030083609744906, |
| "epoch": 2.856, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.4291111111111113e-05, |
| "loss": 0.594, |
| "mean_token_accuracy": 0.8221963416785002, |
| "num_tokens": 51087706.0, |
| "step": 3570 |
| }, |
| { |
| "entropy": 0.5978926762007177, |
| "epoch": 2.864, |
| "grad_norm": 3.8125, |
| "learning_rate": 1.426888888888889e-05, |
| "loss": 0.5832, |
| "mean_token_accuracy": 0.8252336475998163, |
| "num_tokens": 51229298.0, |
| "step": 3580 |
| }, |
| { |
| "entropy": 0.5961005238816142, |
| "epoch": 2.872, |
| "grad_norm": 4.5, |
| "learning_rate": 1.4246666666666669e-05, |
| "loss": 0.5785, |
| "mean_token_accuracy": 0.8258848875761032, |
| "num_tokens": 51369931.0, |
| "step": 3590 |
| }, |
| { |
| "entropy": 0.6119486084207892, |
| "epoch": 2.88, |
| "grad_norm": 3.953125, |
| "learning_rate": 1.4224444444444445e-05, |
| "loss": 0.5909, |
| "mean_token_accuracy": 0.823506697267294, |
| "num_tokens": 51509622.0, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_coding_entropy": 1.3437452661991118, |
| "eval_coding_loss": 2.3273701667785645, |
| "eval_coding_mean_token_accuracy": 0.5298208997249604, |
| "eval_coding_num_tokens": 51509622.0, |
| "eval_coding_runtime": 80.4059, |
| "eval_coding_samples_per_second": 6.218, |
| "eval_coding_steps_per_second": 3.109, |
| "step": 3600 |
| }, |
| { |
| "epoch": 2.88, |
| "eval_math_entropy": 0.6390358245372773, |
| "eval_math_loss": 0.8618610501289368, |
| "eval_math_mean_token_accuracy": 0.781146463394165, |
| "eval_math_num_tokens": 51509622.0, |
| "eval_math_runtime": 45.2644, |
| "eval_math_samples_per_second": 11.046, |
| "eval_math_steps_per_second": 5.523, |
| "step": 3600 |
| }, |
| { |
| "entropy": 0.5799249410629272, |
| "epoch": 2.888, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.4202222222222224e-05, |
| "loss": 0.5793, |
| "mean_token_accuracy": 0.8289754271507264, |
| "num_tokens": 51647367.0, |
| "step": 3610 |
| }, |
| { |
| "entropy": 0.596883624047041, |
| "epoch": 2.896, |
| "grad_norm": 3.8125, |
| "learning_rate": 1.418e-05, |
| "loss": 0.5732, |
| "mean_token_accuracy": 0.8276680398732423, |
| "num_tokens": 51790856.0, |
| "step": 3620 |
| }, |
| { |
| "entropy": 0.587921068072319, |
| "epoch": 2.904, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.4157777777777779e-05, |
| "loss": 0.5936, |
| "mean_token_accuracy": 0.8243591144680977, |
| "num_tokens": 51934568.0, |
| "step": 3630 |
| }, |
| { |
| "entropy": 0.6033103983849287, |
| "epoch": 2.912, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.4135555555555555e-05, |
| "loss": 0.5859, |
| "mean_token_accuracy": 0.8248248245567084, |
| "num_tokens": 52074393.0, |
| "step": 3640 |
| }, |
| { |
| "entropy": 0.5971073280088603, |
| "epoch": 2.92, |
| "grad_norm": 4.25, |
| "learning_rate": 1.4113333333333334e-05, |
| "loss": 0.5862, |
| "mean_token_accuracy": 0.8287141636013985, |
| "num_tokens": 52218473.0, |
| "step": 3650 |
| }, |
| { |
| "entropy": 0.6164516042917967, |
| "epoch": 2.928, |
| "grad_norm": 3.875, |
| "learning_rate": 1.4091111111111114e-05, |
| "loss": 0.5983, |
| "mean_token_accuracy": 0.822106447070837, |
| "num_tokens": 52366467.0, |
| "step": 3660 |
| }, |
| { |
| "entropy": 0.5970261240378022, |
| "epoch": 2.936, |
| "grad_norm": 4.03125, |
| "learning_rate": 1.406888888888889e-05, |
| "loss": 0.5939, |
| "mean_token_accuracy": 0.8235851936042309, |
| "num_tokens": 52505962.0, |
| "step": 3670 |
| }, |
| { |
| "entropy": 0.5658912505954504, |
| "epoch": 2.944, |
| "grad_norm": 4.0, |
| "learning_rate": 1.4046666666666669e-05, |
| "loss": 0.5623, |
| "mean_token_accuracy": 0.8313923060894013, |
| "num_tokens": 52657863.0, |
| "step": 3680 |
| }, |
| { |
| "entropy": 0.6137504337355495, |
| "epoch": 2.952, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.4024444444444445e-05, |
| "loss": 0.5989, |
| "mean_token_accuracy": 0.8215026266872882, |
| "num_tokens": 52802222.0, |
| "step": 3690 |
| }, |
| { |
| "entropy": 0.5801763826981187, |
| "epoch": 2.96, |
| "grad_norm": 4.75, |
| "learning_rate": 1.4002222222222224e-05, |
| "loss": 0.5797, |
| "mean_token_accuracy": 0.8291269116103649, |
| "num_tokens": 52947183.0, |
| "step": 3700 |
| }, |
| { |
| "epoch": 2.96, |
| "eval_coding_entropy": 1.3707081127166747, |
| "eval_coding_loss": 2.3238508701324463, |
| "eval_coding_mean_token_accuracy": 0.5291933624744415, |
| "eval_coding_num_tokens": 52947183.0, |
| "eval_coding_runtime": 80.2723, |
| "eval_coding_samples_per_second": 6.229, |
| "eval_coding_steps_per_second": 3.114, |
| "step": 3700 |
| }, |
| { |
| "epoch": 2.96, |
| "eval_math_entropy": 0.6480549938678741, |
| "eval_math_loss": 0.858568012714386, |
| "eval_math_mean_token_accuracy": 0.7815098395347595, |
| "eval_math_num_tokens": 52947183.0, |
| "eval_math_runtime": 45.3883, |
| "eval_math_samples_per_second": 11.016, |
| "eval_math_steps_per_second": 5.508, |
| "step": 3700 |
| }, |
| { |
| "entropy": 0.6031183777377009, |
| "epoch": 2.968, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.398e-05, |
| "loss": 0.5811, |
| "mean_token_accuracy": 0.8253906678408385, |
| "num_tokens": 53079727.0, |
| "step": 3710 |
| }, |
| { |
| "entropy": 0.5833488658070565, |
| "epoch": 2.976, |
| "grad_norm": 3.984375, |
| "learning_rate": 1.395777777777778e-05, |
| "loss": 0.5744, |
| "mean_token_accuracy": 0.8283881366252899, |
| "num_tokens": 53226984.0, |
| "step": 3720 |
| }, |
| { |
| "entropy": 0.5871400842443109, |
| "epoch": 2.984, |
| "grad_norm": 4.375, |
| "learning_rate": 1.3935555555555557e-05, |
| "loss": 0.5827, |
| "mean_token_accuracy": 0.8275302708148956, |
| "num_tokens": 53370611.0, |
| "step": 3730 |
| }, |
| { |
| "entropy": 0.586006616614759, |
| "epoch": 2.992, |
| "grad_norm": 4.0, |
| "learning_rate": 1.3913333333333335e-05, |
| "loss": 0.5695, |
| "mean_token_accuracy": 0.8304219745099545, |
| "num_tokens": 53515204.0, |
| "step": 3740 |
| }, |
| { |
| "entropy": 0.589054942689836, |
| "epoch": 3.0, |
| "grad_norm": 3.890625, |
| "learning_rate": 1.3891111111111114e-05, |
| "loss": 0.5859, |
| "mean_token_accuracy": 0.825661375373602, |
| "num_tokens": 53659272.0, |
| "step": 3750 |
| }, |
| { |
| "entropy": 0.444517913274467, |
| "epoch": 3.008, |
| "grad_norm": 5.875, |
| "learning_rate": 1.386888888888889e-05, |
| "loss": 0.3976, |
| "mean_token_accuracy": 0.8780367832630873, |
| "num_tokens": 53800592.0, |
| "step": 3760 |
| }, |
| { |
| "entropy": 0.39136042818427086, |
| "epoch": 3.016, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.3846666666666669e-05, |
| "loss": 0.3735, |
| "mean_token_accuracy": 0.8823954571038485, |
| "num_tokens": 53944783.0, |
| "step": 3770 |
| }, |
| { |
| "entropy": 0.40102586671710017, |
| "epoch": 3.024, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.3824444444444445e-05, |
| "loss": 0.3803, |
| "mean_token_accuracy": 0.8799105253070593, |
| "num_tokens": 54092017.0, |
| "step": 3780 |
| }, |
| { |
| "entropy": 0.37874664990231394, |
| "epoch": 3.032, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.3802222222222224e-05, |
| "loss": 0.3507, |
| "mean_token_accuracy": 0.8864472284913063, |
| "num_tokens": 54233983.0, |
| "step": 3790 |
| }, |
| { |
| "entropy": 0.38700895812362435, |
| "epoch": 3.04, |
| "grad_norm": 5.0, |
| "learning_rate": 1.378e-05, |
| "loss": 0.366, |
| "mean_token_accuracy": 0.8837040297687053, |
| "num_tokens": 54380158.0, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_coding_entropy": 0.9976171374320983, |
| "eval_coding_loss": 2.8449416160583496, |
| "eval_coding_mean_token_accuracy": 0.5104463682174682, |
| "eval_coding_num_tokens": 54380158.0, |
| "eval_coding_runtime": 80.2677, |
| "eval_coding_samples_per_second": 6.229, |
| "eval_coding_steps_per_second": 3.115, |
| "step": 3800 |
| }, |
| { |
| "epoch": 3.04, |
| "eval_math_entropy": 0.45957636153697967, |
| "eval_math_loss": 1.0391985177993774, |
| "eval_math_mean_token_accuracy": 0.7721266651153564, |
| "eval_math_num_tokens": 54380158.0, |
| "eval_math_runtime": 45.3349, |
| "eval_math_samples_per_second": 11.029, |
| "eval_math_steps_per_second": 5.515, |
| "step": 3800 |
| }, |
| { |
| "entropy": 0.38679480627179147, |
| "epoch": 3.048, |
| "grad_norm": 4.5, |
| "learning_rate": 1.375777777777778e-05, |
| "loss": 0.3692, |
| "mean_token_accuracy": 0.8850649714469909, |
| "num_tokens": 54517614.0, |
| "step": 3810 |
| }, |
| { |
| "entropy": 0.363014649040997, |
| "epoch": 3.056, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.3735555555555557e-05, |
| "loss": 0.3429, |
| "mean_token_accuracy": 0.8884872987866401, |
| "num_tokens": 54658093.0, |
| "step": 3820 |
| }, |
| { |
| "entropy": 0.38303085109218954, |
| "epoch": 3.064, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.3713333333333335e-05, |
| "loss": 0.3639, |
| "mean_token_accuracy": 0.883158453553915, |
| "num_tokens": 54804961.0, |
| "step": 3830 |
| }, |
| { |
| "entropy": 0.3767666126601398, |
| "epoch": 3.072, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.3691111111111112e-05, |
| "loss": 0.3541, |
| "mean_token_accuracy": 0.8856425810605287, |
| "num_tokens": 54949938.0, |
| "step": 3840 |
| }, |
| { |
| "entropy": 0.3665018714033067, |
| "epoch": 3.08, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.366888888888889e-05, |
| "loss": 0.3442, |
| "mean_token_accuracy": 0.8890479180961848, |
| "num_tokens": 55087269.0, |
| "step": 3850 |
| }, |
| { |
| "entropy": 0.35775307659059763, |
| "epoch": 3.088, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.3646666666666668e-05, |
| "loss": 0.3409, |
| "mean_token_accuracy": 0.890467518568039, |
| "num_tokens": 55234643.0, |
| "step": 3860 |
| }, |
| { |
| "entropy": 0.3798696202225983, |
| "epoch": 3.096, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.3624444444444445e-05, |
| "loss": 0.3619, |
| "mean_token_accuracy": 0.8841665253043175, |
| "num_tokens": 55376099.0, |
| "step": 3870 |
| }, |
| { |
| "entropy": 0.3838302061893046, |
| "epoch": 3.104, |
| "grad_norm": 4.28125, |
| "learning_rate": 1.3602222222222223e-05, |
| "loss": 0.3652, |
| "mean_token_accuracy": 0.8823526252061129, |
| "num_tokens": 55519679.0, |
| "step": 3880 |
| }, |
| { |
| "entropy": 0.3730103265494108, |
| "epoch": 3.112, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 0.3558, |
| "mean_token_accuracy": 0.8874844864010811, |
| "num_tokens": 55665189.0, |
| "step": 3890 |
| }, |
| { |
| "entropy": 0.3636066407896578, |
| "epoch": 3.12, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.3557777777777778e-05, |
| "loss": 0.35, |
| "mean_token_accuracy": 0.8876182068139314, |
| "num_tokens": 55812397.0, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.12, |
| "eval_coding_entropy": 0.9763415265083313, |
| "eval_coding_loss": 2.881798505783081, |
| "eval_coding_mean_token_accuracy": 0.5110503755807877, |
| "eval_coding_num_tokens": 55812397.0, |
| "eval_coding_runtime": 80.214, |
| "eval_coding_samples_per_second": 6.233, |
| "eval_coding_steps_per_second": 3.117, |
| "step": 3900 |
| }, |
| { |
| "epoch": 3.12, |
| "eval_math_entropy": 0.4509195798635483, |
| "eval_math_loss": 1.0534956455230713, |
| "eval_math_mean_token_accuracy": 0.7725037877559662, |
| "eval_math_num_tokens": 55812397.0, |
| "eval_math_runtime": 45.3698, |
| "eval_math_samples_per_second": 11.021, |
| "eval_math_steps_per_second": 5.51, |
| "step": 3900 |
| }, |
| { |
| "entropy": 0.37728115255013106, |
| "epoch": 3.128, |
| "grad_norm": 4.375, |
| "learning_rate": 1.3535555555555557e-05, |
| "loss": 0.3553, |
| "mean_token_accuracy": 0.8857639830559492, |
| "num_tokens": 55952921.0, |
| "step": 3910 |
| }, |
| { |
| "entropy": 0.3796572177670896, |
| "epoch": 3.136, |
| "grad_norm": 4.46875, |
| "learning_rate": 1.3513333333333333e-05, |
| "loss": 0.3617, |
| "mean_token_accuracy": 0.8842459708452225, |
| "num_tokens": 56101550.0, |
| "step": 3920 |
| }, |
| { |
| "entropy": 0.36837306870147585, |
| "epoch": 3.144, |
| "grad_norm": 4.375, |
| "learning_rate": 1.3491111111111113e-05, |
| "loss": 0.3494, |
| "mean_token_accuracy": 0.8870921719819307, |
| "num_tokens": 56240370.0, |
| "step": 3930 |
| }, |
| { |
| "entropy": 0.3695572538301349, |
| "epoch": 3.152, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.3468888888888888e-05, |
| "loss": 0.3544, |
| "mean_token_accuracy": 0.8864484548568725, |
| "num_tokens": 56387662.0, |
| "step": 3940 |
| }, |
| { |
| "entropy": 0.3955209769308567, |
| "epoch": 3.16, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.3446666666666668e-05, |
| "loss": 0.3775, |
| "mean_token_accuracy": 0.8805460192263126, |
| "num_tokens": 56532459.0, |
| "step": 3950 |
| }, |
| { |
| "entropy": 0.3788654290139675, |
| "epoch": 3.168, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.3424444444444447e-05, |
| "loss": 0.3681, |
| "mean_token_accuracy": 0.8828589104115963, |
| "num_tokens": 56673426.0, |
| "step": 3960 |
| }, |
| { |
| "entropy": 0.3717520281672478, |
| "epoch": 3.176, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.3402222222222223e-05, |
| "loss": 0.347, |
| "mean_token_accuracy": 0.8868261199444533, |
| "num_tokens": 56815667.0, |
| "step": 3970 |
| }, |
| { |
| "entropy": 0.37956275697797537, |
| "epoch": 3.184, |
| "grad_norm": 5.0, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 0.3641, |
| "mean_token_accuracy": 0.8828948926180601, |
| "num_tokens": 56957044.0, |
| "step": 3980 |
| }, |
| { |
| "entropy": 0.3702836920507252, |
| "epoch": 3.192, |
| "grad_norm": 4.5, |
| "learning_rate": 1.3357777777777778e-05, |
| "loss": 0.3483, |
| "mean_token_accuracy": 0.8869308207184077, |
| "num_tokens": 57103391.0, |
| "step": 3990 |
| }, |
| { |
| "entropy": 0.3733728963881731, |
| "epoch": 3.2, |
| "grad_norm": 4.875, |
| "learning_rate": 1.3335555555555558e-05, |
| "loss": 0.3622, |
| "mean_token_accuracy": 0.8857541091740131, |
| "num_tokens": 57250097.0, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_coding_entropy": 0.9959763128757477, |
| "eval_coding_loss": 2.889766216278076, |
| "eval_coding_mean_token_accuracy": 0.5086574196815491, |
| "eval_coding_num_tokens": 57250097.0, |
| "eval_coding_runtime": 80.0247, |
| "eval_coding_samples_per_second": 6.248, |
| "eval_coding_steps_per_second": 3.124, |
| "step": 4000 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_math_entropy": 0.45689878618717195, |
| "eval_math_loss": 1.0550038814544678, |
| "eval_math_mean_token_accuracy": 0.7724210150241851, |
| "eval_math_num_tokens": 57250097.0, |
| "eval_math_runtime": 45.0826, |
| "eval_math_samples_per_second": 11.091, |
| "eval_math_steps_per_second": 5.545, |
| "step": 4000 |
| }, |
| { |
| "entropy": 0.3935958819463849, |
| "epoch": 3.208, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.3313333333333333e-05, |
| "loss": 0.3766, |
| "mean_token_accuracy": 0.8812199637293816, |
| "num_tokens": 57391222.0, |
| "step": 4010 |
| }, |
| { |
| "entropy": 0.3712491972371936, |
| "epoch": 3.216, |
| "grad_norm": 5.25, |
| "learning_rate": 1.3291111111111113e-05, |
| "loss": 0.3558, |
| "mean_token_accuracy": 0.8870142463594675, |
| "num_tokens": 57532026.0, |
| "step": 4020 |
| }, |
| { |
| "entropy": 0.36472889482975007, |
| "epoch": 3.224, |
| "grad_norm": 4.15625, |
| "learning_rate": 1.3268888888888889e-05, |
| "loss": 0.3498, |
| "mean_token_accuracy": 0.8890798524022102, |
| "num_tokens": 57678101.0, |
| "step": 4030 |
| }, |
| { |
| "entropy": 0.3772866981104016, |
| "epoch": 3.232, |
| "grad_norm": 5.375, |
| "learning_rate": 1.3246666666666668e-05, |
| "loss": 0.3674, |
| "mean_token_accuracy": 0.8839915424585343, |
| "num_tokens": 57814853.0, |
| "step": 4040 |
| }, |
| { |
| "entropy": 0.3516525051556528, |
| "epoch": 3.24, |
| "grad_norm": 4.25, |
| "learning_rate": 1.3224444444444446e-05, |
| "loss": 0.3363, |
| "mean_token_accuracy": 0.8922900687903166, |
| "num_tokens": 57955700.0, |
| "step": 4050 |
| }, |
| { |
| "entropy": 0.3661925207823515, |
| "epoch": 3.248, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.3202222222222223e-05, |
| "loss": 0.3398, |
| "mean_token_accuracy": 0.8890674687922001, |
| "num_tokens": 58094190.0, |
| "step": 4060 |
| }, |
| { |
| "entropy": 0.3697372685186565, |
| "epoch": 3.2560000000000002, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 0.3499, |
| "mean_token_accuracy": 0.8869231604039669, |
| "num_tokens": 58234084.0, |
| "step": 4070 |
| }, |
| { |
| "entropy": 0.3708636582829058, |
| "epoch": 3.2640000000000002, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.3157777777777778e-05, |
| "loss": 0.3508, |
| "mean_token_accuracy": 0.8864367742091417, |
| "num_tokens": 58374378.0, |
| "step": 4080 |
| }, |
| { |
| "entropy": 0.36966984616592524, |
| "epoch": 3.2720000000000002, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.3135555555555558e-05, |
| "loss": 0.3601, |
| "mean_token_accuracy": 0.8859989456832409, |
| "num_tokens": 58516575.0, |
| "step": 4090 |
| }, |
| { |
| "entropy": 0.37621888155117633, |
| "epoch": 3.2800000000000002, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.3113333333333334e-05, |
| "loss": 0.3558, |
| "mean_token_accuracy": 0.8852376375347376, |
| "num_tokens": 58652789.0, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.2800000000000002, |
| "eval_coding_entropy": 0.9915046076774597, |
| "eval_coding_loss": 2.889876365661621, |
| "eval_coding_mean_token_accuracy": 0.5089448739290238, |
| "eval_coding_num_tokens": 58652789.0, |
| "eval_coding_runtime": 80.4227, |
| "eval_coding_samples_per_second": 6.217, |
| "eval_coding_steps_per_second": 3.109, |
| "step": 4100 |
| }, |
| { |
| "epoch": 3.2800000000000002, |
| "eval_math_entropy": 0.46175317239761354, |
| "eval_math_loss": 1.0507140159606934, |
| "eval_math_mean_token_accuracy": 0.7714938235282898, |
| "eval_math_num_tokens": 58652789.0, |
| "eval_math_runtime": 45.458, |
| "eval_math_samples_per_second": 10.999, |
| "eval_math_steps_per_second": 5.5, |
| "step": 4100 |
| }, |
| { |
| "entropy": 0.3793962474912405, |
| "epoch": 3.288, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.3091111111111113e-05, |
| "loss": 0.3718, |
| "mean_token_accuracy": 0.8843492668122053, |
| "num_tokens": 58787621.0, |
| "step": 4110 |
| }, |
| { |
| "entropy": 0.3836384235881269, |
| "epoch": 3.296, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.306888888888889e-05, |
| "loss": 0.3625, |
| "mean_token_accuracy": 0.8842442166060209, |
| "num_tokens": 58930014.0, |
| "step": 4120 |
| }, |
| { |
| "entropy": 0.36728960955515505, |
| "epoch": 3.304, |
| "grad_norm": 5.375, |
| "learning_rate": 1.3046666666666668e-05, |
| "loss": 0.3585, |
| "mean_token_accuracy": 0.8854542460292578, |
| "num_tokens": 59069679.0, |
| "step": 4130 |
| }, |
| { |
| "entropy": 0.38970273407176137, |
| "epoch": 3.312, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.3024444444444446e-05, |
| "loss": 0.3697, |
| "mean_token_accuracy": 0.8815676849335432, |
| "num_tokens": 59210270.0, |
| "step": 4140 |
| }, |
| { |
| "entropy": 0.36535215443000196, |
| "epoch": 3.32, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.3002222222222223e-05, |
| "loss": 0.3477, |
| "mean_token_accuracy": 0.8877509526908398, |
| "num_tokens": 59352767.0, |
| "step": 4150 |
| }, |
| { |
| "entropy": 0.374912076164037, |
| "epoch": 3.328, |
| "grad_norm": 4.875, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 0.3559, |
| "mean_token_accuracy": 0.8860851589590311, |
| "num_tokens": 59496408.0, |
| "step": 4160 |
| }, |
| { |
| "entropy": 0.3589112013578415, |
| "epoch": 3.336, |
| "grad_norm": 5.84375, |
| "learning_rate": 1.2957777777777779e-05, |
| "loss": 0.3505, |
| "mean_token_accuracy": 0.8878632180392743, |
| "num_tokens": 59635459.0, |
| "step": 4170 |
| }, |
| { |
| "entropy": 0.37663340773433446, |
| "epoch": 3.344, |
| "grad_norm": 4.75, |
| "learning_rate": 1.2935555555555556e-05, |
| "loss": 0.3528, |
| "mean_token_accuracy": 0.8863626789301634, |
| "num_tokens": 59779722.0, |
| "step": 4180 |
| }, |
| { |
| "entropy": 0.3957128500565886, |
| "epoch": 3.352, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.2913333333333336e-05, |
| "loss": 0.376, |
| "mean_token_accuracy": 0.8786687582731247, |
| "num_tokens": 59924246.0, |
| "step": 4190 |
| }, |
| { |
| "entropy": 0.39835441708564756, |
| "epoch": 3.36, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.2891111111111112e-05, |
| "loss": 0.3748, |
| "mean_token_accuracy": 0.879026398807764, |
| "num_tokens": 60061562.0, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_coding_entropy": 1.0046057679653169, |
| "eval_coding_loss": 2.8688178062438965, |
| "eval_coding_mean_token_accuracy": 0.5100064791440964, |
| "eval_coding_num_tokens": 60061562.0, |
| "eval_coding_runtime": 80.5815, |
| "eval_coding_samples_per_second": 6.205, |
| "eval_coding_steps_per_second": 3.102, |
| "step": 4200 |
| }, |
| { |
| "epoch": 3.36, |
| "eval_math_entropy": 0.4630183935165405, |
| "eval_math_loss": 1.049013614654541, |
| "eval_math_mean_token_accuracy": 0.7723689336776733, |
| "eval_math_num_tokens": 60061562.0, |
| "eval_math_runtime": 45.4533, |
| "eval_math_samples_per_second": 11.0, |
| "eval_math_steps_per_second": 5.5, |
| "step": 4200 |
| }, |
| { |
| "entropy": 0.3695203845389187, |
| "epoch": 3.368, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.2868888888888891e-05, |
| "loss": 0.3532, |
| "mean_token_accuracy": 0.8871517650783062, |
| "num_tokens": 60203128.0, |
| "step": 4210 |
| }, |
| { |
| "entropy": 0.39565675128251315, |
| "epoch": 3.376, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.2846666666666667e-05, |
| "loss": 0.3822, |
| "mean_token_accuracy": 0.8784398142248392, |
| "num_tokens": 60346574.0, |
| "step": 4220 |
| }, |
| { |
| "entropy": 0.37085114419460297, |
| "epoch": 3.384, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.2824444444444446e-05, |
| "loss": 0.3547, |
| "mean_token_accuracy": 0.8863418217748403, |
| "num_tokens": 60492492.0, |
| "step": 4230 |
| }, |
| { |
| "entropy": 0.35558655243366954, |
| "epoch": 3.392, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.2802222222222222e-05, |
| "loss": 0.3374, |
| "mean_token_accuracy": 0.8919463641941547, |
| "num_tokens": 60636820.0, |
| "step": 4240 |
| }, |
| { |
| "entropy": 0.3644601003266871, |
| "epoch": 3.4, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 0.3463, |
| "mean_token_accuracy": 0.8881781507283449, |
| "num_tokens": 60780515.0, |
| "step": 4250 |
| }, |
| { |
| "entropy": 0.371591713745147, |
| "epoch": 3.408, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.2757777777777777e-05, |
| "loss": 0.3566, |
| "mean_token_accuracy": 0.8859987486153841, |
| "num_tokens": 60922033.0, |
| "step": 4260 |
| }, |
| { |
| "entropy": 0.38073564767837526, |
| "epoch": 3.416, |
| "grad_norm": 4.875, |
| "learning_rate": 1.2735555555555557e-05, |
| "loss": 0.3646, |
| "mean_token_accuracy": 0.8825856417417526, |
| "num_tokens": 61066753.0, |
| "step": 4270 |
| }, |
| { |
| "entropy": 0.3755529592745006, |
| "epoch": 3.424, |
| "grad_norm": 5.5, |
| "learning_rate": 1.2713333333333336e-05, |
| "loss": 0.3581, |
| "mean_token_accuracy": 0.8837402831763029, |
| "num_tokens": 61209156.0, |
| "step": 4280 |
| }, |
| { |
| "entropy": 0.376344402693212, |
| "epoch": 3.432, |
| "grad_norm": 5.34375, |
| "learning_rate": 1.2691111111111112e-05, |
| "loss": 0.3584, |
| "mean_token_accuracy": 0.8857293087989092, |
| "num_tokens": 61350188.0, |
| "step": 4290 |
| }, |
| { |
| "entropy": 0.3725393834523857, |
| "epoch": 3.44, |
| "grad_norm": 4.625, |
| "learning_rate": 1.2668888888888891e-05, |
| "loss": 0.3569, |
| "mean_token_accuracy": 0.8862212751060724, |
| "num_tokens": 61491228.0, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.44, |
| "eval_coding_entropy": 0.9676942622661591, |
| "eval_coding_loss": 2.9207324981689453, |
| "eval_coding_mean_token_accuracy": 0.5097229942083359, |
| "eval_coding_num_tokens": 61491228.0, |
| "eval_coding_runtime": 80.4, |
| "eval_coding_samples_per_second": 6.219, |
| "eval_coding_steps_per_second": 3.109, |
| "step": 4300 |
| }, |
| { |
| "epoch": 3.44, |
| "eval_math_entropy": 0.4544974520206451, |
| "eval_math_loss": 1.0598396062850952, |
| "eval_math_mean_token_accuracy": 0.7718669784069061, |
| "eval_math_num_tokens": 61491228.0, |
| "eval_math_runtime": 45.4527, |
| "eval_math_samples_per_second": 11.0, |
| "eval_math_steps_per_second": 5.5, |
| "step": 4300 |
| }, |
| { |
| "entropy": 0.3861199578270316, |
| "epoch": 3.448, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.2646666666666667e-05, |
| "loss": 0.3696, |
| "mean_token_accuracy": 0.8823696870356799, |
| "num_tokens": 61637706.0, |
| "step": 4310 |
| }, |
| { |
| "entropy": 0.37499388279393314, |
| "epoch": 3.456, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.2624444444444446e-05, |
| "loss": 0.3632, |
| "mean_token_accuracy": 0.8855489175766706, |
| "num_tokens": 61773033.0, |
| "step": 4320 |
| }, |
| { |
| "entropy": 0.38619182985275985, |
| "epoch": 3.464, |
| "grad_norm": 4.75, |
| "learning_rate": 1.2602222222222222e-05, |
| "loss": 0.3757, |
| "mean_token_accuracy": 0.8817618012428283, |
| "num_tokens": 61919678.0, |
| "step": 4330 |
| }, |
| { |
| "entropy": 0.38958541546016934, |
| "epoch": 3.472, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 0.3691, |
| "mean_token_accuracy": 0.8797871727496386, |
| "num_tokens": 62061802.0, |
| "step": 4340 |
| }, |
| { |
| "entropy": 0.3728965263813734, |
| "epoch": 3.48, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.2557777777777779e-05, |
| "loss": 0.3609, |
| "mean_token_accuracy": 0.8867140613496304, |
| "num_tokens": 62206556.0, |
| "step": 4350 |
| }, |
| { |
| "entropy": 0.37456957083195447, |
| "epoch": 3.488, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.2535555555555557e-05, |
| "loss": 0.3534, |
| "mean_token_accuracy": 0.8862990573048591, |
| "num_tokens": 62347490.0, |
| "step": 4360 |
| }, |
| { |
| "entropy": 0.38473188849166035, |
| "epoch": 3.496, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.2513333333333334e-05, |
| "loss": 0.378, |
| "mean_token_accuracy": 0.8817499719560147, |
| "num_tokens": 62489233.0, |
| "step": 4370 |
| }, |
| { |
| "entropy": 0.36225079353898765, |
| "epoch": 3.504, |
| "grad_norm": 4.75, |
| "learning_rate": 1.2491111111111112e-05, |
| "loss": 0.3375, |
| "mean_token_accuracy": 0.8891223490238189, |
| "num_tokens": 62636811.0, |
| "step": 4380 |
| }, |
| { |
| "entropy": 0.37174071511253715, |
| "epoch": 3.512, |
| "grad_norm": 5.71875, |
| "learning_rate": 1.246888888888889e-05, |
| "loss": 0.3588, |
| "mean_token_accuracy": 0.8847623728215694, |
| "num_tokens": 62779188.0, |
| "step": 4390 |
| }, |
| { |
| "entropy": 0.3783165253698826, |
| "epoch": 3.52, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.2446666666666667e-05, |
| "loss": 0.3664, |
| "mean_token_accuracy": 0.8849261485040187, |
| "num_tokens": 62928536.0, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_coding_entropy": 1.0008562405109405, |
| "eval_coding_loss": 2.873396873474121, |
| "eval_coding_mean_token_accuracy": 0.509664377450943, |
| "eval_coding_num_tokens": 62928536.0, |
| "eval_coding_runtime": 80.405, |
| "eval_coding_samples_per_second": 6.219, |
| "eval_coding_steps_per_second": 3.109, |
| "step": 4400 |
| }, |
| { |
| "epoch": 3.52, |
| "eval_math_entropy": 0.46665811783075334, |
| "eval_math_loss": 1.0437169075012207, |
| "eval_math_mean_token_accuracy": 0.7728277809619903, |
| "eval_math_num_tokens": 62928536.0, |
| "eval_math_runtime": 45.5432, |
| "eval_math_samples_per_second": 10.979, |
| "eval_math_steps_per_second": 5.489, |
| "step": 4400 |
| }, |
| { |
| "entropy": 0.3748255460523069, |
| "epoch": 3.528, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.2424444444444445e-05, |
| "loss": 0.3605, |
| "mean_token_accuracy": 0.8849743034690618, |
| "num_tokens": 63069075.0, |
| "step": 4410 |
| }, |
| { |
| "entropy": 0.37385046752169726, |
| "epoch": 3.536, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.2402222222222222e-05, |
| "loss": 0.3614, |
| "mean_token_accuracy": 0.8848837114870548, |
| "num_tokens": 63212496.0, |
| "step": 4420 |
| }, |
| { |
| "entropy": 0.3783167893998325, |
| "epoch": 3.544, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 0.3554, |
| "mean_token_accuracy": 0.8844420034438372, |
| "num_tokens": 63362927.0, |
| "step": 4430 |
| }, |
| { |
| "entropy": 0.36881880927830935, |
| "epoch": 3.552, |
| "grad_norm": 4.4375, |
| "learning_rate": 1.235777777777778e-05, |
| "loss": 0.3579, |
| "mean_token_accuracy": 0.8859864506870508, |
| "num_tokens": 63506921.0, |
| "step": 4440 |
| }, |
| { |
| "entropy": 0.3825777556747198, |
| "epoch": 3.56, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.2335555555555557e-05, |
| "loss": 0.3577, |
| "mean_token_accuracy": 0.8835489846765995, |
| "num_tokens": 63656188.0, |
| "step": 4450 |
| }, |
| { |
| "entropy": 0.3539299752563238, |
| "epoch": 3.568, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.2313333333333335e-05, |
| "loss": 0.3414, |
| "mean_token_accuracy": 0.8887290176004171, |
| "num_tokens": 63803208.0, |
| "step": 4460 |
| }, |
| { |
| "entropy": 0.38577949684113266, |
| "epoch": 3.576, |
| "grad_norm": 4.9375, |
| "learning_rate": 1.2291111111111112e-05, |
| "loss": 0.3785, |
| "mean_token_accuracy": 0.882218312844634, |
| "num_tokens": 63943451.0, |
| "step": 4470 |
| }, |
| { |
| "entropy": 0.3646447826176882, |
| "epoch": 3.584, |
| "grad_norm": 5.375, |
| "learning_rate": 1.226888888888889e-05, |
| "loss": 0.3424, |
| "mean_token_accuracy": 0.890447198599577, |
| "num_tokens": 64084867.0, |
| "step": 4480 |
| }, |
| { |
| "entropy": 0.36499870689585806, |
| "epoch": 3.592, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.2246666666666667e-05, |
| "loss": 0.3551, |
| "mean_token_accuracy": 0.8863568991422653, |
| "num_tokens": 64233342.0, |
| "step": 4490 |
| }, |
| { |
| "entropy": 0.3507792026735842, |
| "epoch": 3.6, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.2224444444444445e-05, |
| "loss": 0.3382, |
| "mean_token_accuracy": 0.8924963969737292, |
| "num_tokens": 64377046.0, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.6, |
| "eval_coding_entropy": 0.9826929938793182, |
| "eval_coding_loss": 2.9040191173553467, |
| "eval_coding_mean_token_accuracy": 0.509175629734993, |
| "eval_coding_num_tokens": 64377046.0, |
| "eval_coding_runtime": 80.1588, |
| "eval_coding_samples_per_second": 6.238, |
| "eval_coding_steps_per_second": 3.119, |
| "step": 4500 |
| }, |
| { |
| "epoch": 3.6, |
| "eval_math_entropy": 0.4600372558236122, |
| "eval_math_loss": 1.0517936944961548, |
| "eval_math_mean_token_accuracy": 0.7727461197376251, |
| "eval_math_num_tokens": 64377046.0, |
| "eval_math_runtime": 45.1814, |
| "eval_math_samples_per_second": 11.067, |
| "eval_math_steps_per_second": 5.533, |
| "step": 4500 |
| }, |
| { |
| "entropy": 0.382382083311677, |
| "epoch": 3.608, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.2202222222222224e-05, |
| "loss": 0.3631, |
| "mean_token_accuracy": 0.8832461632788181, |
| "num_tokens": 64519519.0, |
| "step": 4510 |
| }, |
| { |
| "entropy": 0.38277862733229995, |
| "epoch": 3.616, |
| "grad_norm": 4.625, |
| "learning_rate": 1.218e-05, |
| "loss": 0.3704, |
| "mean_token_accuracy": 0.8824177011847496, |
| "num_tokens": 64664301.0, |
| "step": 4520 |
| }, |
| { |
| "entropy": 0.3549914536997676, |
| "epoch": 3.624, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.215777777777778e-05, |
| "loss": 0.332, |
| "mean_token_accuracy": 0.8927140560001134, |
| "num_tokens": 64803219.0, |
| "step": 4530 |
| }, |
| { |
| "entropy": 0.37289916798472406, |
| "epoch": 3.632, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.2135555555555556e-05, |
| "loss": 0.3607, |
| "mean_token_accuracy": 0.8848063521087169, |
| "num_tokens": 64946172.0, |
| "step": 4540 |
| }, |
| { |
| "entropy": 0.35420147171244026, |
| "epoch": 3.64, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.2113333333333335e-05, |
| "loss": 0.3366, |
| "mean_token_accuracy": 0.8921176314353942, |
| "num_tokens": 65088973.0, |
| "step": 4550 |
| }, |
| { |
| "entropy": 0.36922263633459806, |
| "epoch": 3.648, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.209111111111111e-05, |
| "loss": 0.3529, |
| "mean_token_accuracy": 0.8881422884762287, |
| "num_tokens": 65230709.0, |
| "step": 4560 |
| }, |
| { |
| "entropy": 0.3910324684344232, |
| "epoch": 3.656, |
| "grad_norm": 5.71875, |
| "learning_rate": 1.206888888888889e-05, |
| "loss": 0.3814, |
| "mean_token_accuracy": 0.8797136139124632, |
| "num_tokens": 65372932.0, |
| "step": 4570 |
| }, |
| { |
| "entropy": 0.3851853286847472, |
| "epoch": 3.664, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.204666666666667e-05, |
| "loss": 0.3646, |
| "mean_token_accuracy": 0.8828969236463309, |
| "num_tokens": 65515738.0, |
| "step": 4580 |
| }, |
| { |
| "entropy": 0.3762326635420322, |
| "epoch": 3.672, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.2024444444444445e-05, |
| "loss": 0.3592, |
| "mean_token_accuracy": 0.8839790564030409, |
| "num_tokens": 65664163.0, |
| "step": 4590 |
| }, |
| { |
| "entropy": 0.36739573180675505, |
| "epoch": 3.68, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.2002222222222225e-05, |
| "loss": 0.3521, |
| "mean_token_accuracy": 0.8870802029967308, |
| "num_tokens": 65802680.0, |
| "step": 4600 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_coding_entropy": 0.990908629655838, |
| "eval_coding_loss": 2.897575855255127, |
| "eval_coding_mean_token_accuracy": 0.5093085201978683, |
| "eval_coding_num_tokens": 65802680.0, |
| "eval_coding_runtime": 80.4264, |
| "eval_coding_samples_per_second": 6.217, |
| "eval_coding_steps_per_second": 3.108, |
| "step": 4600 |
| }, |
| { |
| "epoch": 3.68, |
| "eval_math_entropy": 0.4573659638762474, |
| "eval_math_loss": 1.0522512197494507, |
| "eval_math_mean_token_accuracy": 0.7727820448875428, |
| "eval_math_num_tokens": 65802680.0, |
| "eval_math_runtime": 45.465, |
| "eval_math_samples_per_second": 10.997, |
| "eval_math_steps_per_second": 5.499, |
| "step": 4600 |
| }, |
| { |
| "entropy": 0.3736352989450097, |
| "epoch": 3.6879999999999997, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.198e-05, |
| "loss": 0.3576, |
| "mean_token_accuracy": 0.8849338546395302, |
| "num_tokens": 65944671.0, |
| "step": 4610 |
| }, |
| { |
| "entropy": 0.38910045521333814, |
| "epoch": 3.6959999999999997, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.195777777777778e-05, |
| "loss": 0.3734, |
| "mean_token_accuracy": 0.8828508943319321, |
| "num_tokens": 66089104.0, |
| "step": 4620 |
| }, |
| { |
| "entropy": 0.3689356703311205, |
| "epoch": 3.7039999999999997, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.1935555555555556e-05, |
| "loss": 0.3489, |
| "mean_token_accuracy": 0.8861795470118523, |
| "num_tokens": 66227510.0, |
| "step": 4630 |
| }, |
| { |
| "entropy": 0.3816040868870914, |
| "epoch": 3.7119999999999997, |
| "grad_norm": 5.59375, |
| "learning_rate": 1.1913333333333335e-05, |
| "loss": 0.3626, |
| "mean_token_accuracy": 0.8842772148549557, |
| "num_tokens": 66369895.0, |
| "step": 4640 |
| }, |
| { |
| "entropy": 0.3834704033099115, |
| "epoch": 3.7199999999999998, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.1891111111111111e-05, |
| "loss": 0.3676, |
| "mean_token_accuracy": 0.8825402859598398, |
| "num_tokens": 66509263.0, |
| "step": 4650 |
| }, |
| { |
| "entropy": 0.3672866901382804, |
| "epoch": 3.7279999999999998, |
| "grad_norm": 5.5625, |
| "learning_rate": 1.186888888888889e-05, |
| "loss": 0.3517, |
| "mean_token_accuracy": 0.8866676952689886, |
| "num_tokens": 66653165.0, |
| "step": 4660 |
| }, |
| { |
| "entropy": 0.3687026120722294, |
| "epoch": 3.7359999999999998, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.1846666666666668e-05, |
| "loss": 0.3488, |
| "mean_token_accuracy": 0.8869987074285746, |
| "num_tokens": 66794795.0, |
| "step": 4670 |
| }, |
| { |
| "entropy": 0.3771993276663125, |
| "epoch": 3.7439999999999998, |
| "grad_norm": 5.9375, |
| "learning_rate": 1.1824444444444445e-05, |
| "loss": 0.3633, |
| "mean_token_accuracy": 0.8843857437372208, |
| "num_tokens": 66941476.0, |
| "step": 4680 |
| }, |
| { |
| "entropy": 0.37563126338645814, |
| "epoch": 3.752, |
| "grad_norm": 6.125, |
| "learning_rate": 1.1802222222222223e-05, |
| "loss": 0.359, |
| "mean_token_accuracy": 0.8859431281685829, |
| "num_tokens": 67080833.0, |
| "step": 4690 |
| }, |
| { |
| "entropy": 0.37513453857973217, |
| "epoch": 3.76, |
| "grad_norm": 5.40625, |
| "learning_rate": 1.178e-05, |
| "loss": 0.3689, |
| "mean_token_accuracy": 0.8839717302471399, |
| "num_tokens": 67225380.0, |
| "step": 4700 |
| }, |
| { |
| "epoch": 3.76, |
| "eval_coding_entropy": 0.9823836164474488, |
| "eval_coding_loss": 2.8980584144592285, |
| "eval_coding_mean_token_accuracy": 0.5091208604574203, |
| "eval_coding_num_tokens": 67225380.0, |
| "eval_coding_runtime": 80.051, |
| "eval_coding_samples_per_second": 6.246, |
| "eval_coding_steps_per_second": 3.123, |
| "step": 4700 |
| }, |
| { |
| "epoch": 3.76, |
| "eval_math_entropy": 0.4617558936476707, |
| "eval_math_loss": 1.0457470417022705, |
| "eval_math_mean_token_accuracy": 0.7731688814163208, |
| "eval_math_num_tokens": 67225380.0, |
| "eval_math_runtime": 45.4184, |
| "eval_math_samples_per_second": 11.009, |
| "eval_math_steps_per_second": 5.504, |
| "step": 4700 |
| }, |
| { |
| "entropy": 0.382127748336643, |
| "epoch": 3.768, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.1757777777777778e-05, |
| "loss": 0.3654, |
| "mean_token_accuracy": 0.8842219535261393, |
| "num_tokens": 67372078.0, |
| "step": 4710 |
| }, |
| { |
| "entropy": 0.40150315770879386, |
| "epoch": 3.776, |
| "grad_norm": 4.875, |
| "learning_rate": 1.1735555555555556e-05, |
| "loss": 0.3891, |
| "mean_token_accuracy": 0.8769311863929034, |
| "num_tokens": 67516641.0, |
| "step": 4720 |
| }, |
| { |
| "entropy": 0.37217814614996314, |
| "epoch": 3.784, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.1713333333333334e-05, |
| "loss": 0.3545, |
| "mean_token_accuracy": 0.8866195354610682, |
| "num_tokens": 67663323.0, |
| "step": 4730 |
| }, |
| { |
| "entropy": 0.3578353689983487, |
| "epoch": 3.792, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.1691111111111113e-05, |
| "loss": 0.3483, |
| "mean_token_accuracy": 0.8901370160281659, |
| "num_tokens": 67808009.0, |
| "step": 4740 |
| }, |
| { |
| "entropy": 0.3813229700550437, |
| "epoch": 3.8, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.1668888888888889e-05, |
| "loss": 0.3595, |
| "mean_token_accuracy": 0.8861154112964869, |
| "num_tokens": 67951284.0, |
| "step": 4750 |
| }, |
| { |
| "entropy": 0.38187315631657837, |
| "epoch": 3.808, |
| "grad_norm": 4.96875, |
| "learning_rate": 1.1646666666666668e-05, |
| "loss": 0.3639, |
| "mean_token_accuracy": 0.8838838778436184, |
| "num_tokens": 68092996.0, |
| "step": 4760 |
| }, |
| { |
| "entropy": 0.3607623036019504, |
| "epoch": 3.816, |
| "grad_norm": 5.3125, |
| "learning_rate": 1.1624444444444446e-05, |
| "loss": 0.3496, |
| "mean_token_accuracy": 0.8890579253435135, |
| "num_tokens": 68232710.0, |
| "step": 4770 |
| }, |
| { |
| "entropy": 0.36887320773676036, |
| "epoch": 3.824, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.1602222222222223e-05, |
| "loss": 0.3494, |
| "mean_token_accuracy": 0.887510247156024, |
| "num_tokens": 68380631.0, |
| "step": 4780 |
| }, |
| { |
| "entropy": 0.37964276894927024, |
| "epoch": 3.832, |
| "grad_norm": 5.0, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 0.3609, |
| "mean_token_accuracy": 0.8827996332198381, |
| "num_tokens": 68524110.0, |
| "step": 4790 |
| }, |
| { |
| "entropy": 0.38416239712387323, |
| "epoch": 3.84, |
| "grad_norm": 5.125, |
| "learning_rate": 1.1557777777777779e-05, |
| "loss": 0.3658, |
| "mean_token_accuracy": 0.8818104423582553, |
| "num_tokens": 68666505.0, |
| "step": 4800 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_coding_entropy": 0.9897603187561035, |
| "eval_coding_loss": 2.8927347660064697, |
| "eval_coding_mean_token_accuracy": 0.5102743128538132, |
| "eval_coding_num_tokens": 68666505.0, |
| "eval_coding_runtime": 80.4077, |
| "eval_coding_samples_per_second": 6.218, |
| "eval_coding_steps_per_second": 3.109, |
| "step": 4800 |
| }, |
| { |
| "epoch": 3.84, |
| "eval_math_entropy": 0.4583911989331245, |
| "eval_math_loss": 1.047003984451294, |
| "eval_math_mean_token_accuracy": 0.7733059370517731, |
| "eval_math_num_tokens": 68666505.0, |
| "eval_math_runtime": 45.4985, |
| "eval_math_samples_per_second": 10.989, |
| "eval_math_steps_per_second": 5.495, |
| "step": 4800 |
| }, |
| { |
| "entropy": 0.3677236555144191, |
| "epoch": 3.848, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.1535555555555556e-05, |
| "loss": 0.3512, |
| "mean_token_accuracy": 0.8877127178013324, |
| "num_tokens": 68806562.0, |
| "step": 4810 |
| }, |
| { |
| "entropy": 0.3843054579570889, |
| "epoch": 3.856, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.1513333333333334e-05, |
| "loss": 0.3739, |
| "mean_token_accuracy": 0.881762932240963, |
| "num_tokens": 68951258.0, |
| "step": 4820 |
| }, |
| { |
| "entropy": 0.3714002256281674, |
| "epoch": 3.864, |
| "grad_norm": 4.625, |
| "learning_rate": 1.1491111111111113e-05, |
| "loss": 0.3572, |
| "mean_token_accuracy": 0.8855661526322365, |
| "num_tokens": 69099386.0, |
| "step": 4830 |
| }, |
| { |
| "entropy": 0.3957472760230303, |
| "epoch": 3.872, |
| "grad_norm": 5.375, |
| "learning_rate": 1.1468888888888889e-05, |
| "loss": 0.38, |
| "mean_token_accuracy": 0.8787306550890207, |
| "num_tokens": 69243705.0, |
| "step": 4840 |
| }, |
| { |
| "entropy": 0.3830735825933516, |
| "epoch": 3.88, |
| "grad_norm": 5.15625, |
| "learning_rate": 1.1446666666666668e-05, |
| "loss": 0.3593, |
| "mean_token_accuracy": 0.8831100359559059, |
| "num_tokens": 69385841.0, |
| "step": 4850 |
| }, |
| { |
| "entropy": 0.36585811246186495, |
| "epoch": 3.888, |
| "grad_norm": 4.875, |
| "learning_rate": 1.1424444444444444e-05, |
| "loss": 0.3602, |
| "mean_token_accuracy": 0.8849144950509071, |
| "num_tokens": 69529393.0, |
| "step": 4860 |
| }, |
| { |
| "entropy": 0.3830698226578534, |
| "epoch": 3.896, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.1402222222222224e-05, |
| "loss": 0.3597, |
| "mean_token_accuracy": 0.8860345799475908, |
| "num_tokens": 69675884.0, |
| "step": 4870 |
| }, |
| { |
| "entropy": 0.36876266533508895, |
| "epoch": 3.904, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.138e-05, |
| "loss": 0.3541, |
| "mean_token_accuracy": 0.8872621208429337, |
| "num_tokens": 69820950.0, |
| "step": 4880 |
| }, |
| { |
| "entropy": 0.3679444779641926, |
| "epoch": 3.912, |
| "grad_norm": 5.4375, |
| "learning_rate": 1.1357777777777779e-05, |
| "loss": 0.3511, |
| "mean_token_accuracy": 0.8861892752349376, |
| "num_tokens": 69954594.0, |
| "step": 4890 |
| }, |
| { |
| "entropy": 0.360114162042737, |
| "epoch": 3.92, |
| "grad_norm": 6.125, |
| "learning_rate": 1.1335555555555558e-05, |
| "loss": 0.3455, |
| "mean_token_accuracy": 0.8892067279666662, |
| "num_tokens": 70097736.0, |
| "step": 4900 |
| }, |
| { |
| "epoch": 3.92, |
| "eval_coding_entropy": 0.9788743360042572, |
| "eval_coding_loss": 2.917754888534546, |
| "eval_coding_mean_token_accuracy": 0.5092361553907394, |
| "eval_coding_num_tokens": 70097736.0, |
| "eval_coding_runtime": 80.794, |
| "eval_coding_samples_per_second": 6.189, |
| "eval_coding_steps_per_second": 3.094, |
| "step": 4900 |
| }, |
| { |
| "epoch": 3.92, |
| "eval_math_entropy": 0.4574748896360397, |
| "eval_math_loss": 1.0498664379119873, |
| "eval_math_mean_token_accuracy": 0.773719715833664, |
| "eval_math_num_tokens": 70097736.0, |
| "eval_math_runtime": 45.7175, |
| "eval_math_samples_per_second": 10.937, |
| "eval_math_steps_per_second": 5.468, |
| "step": 4900 |
| }, |
| { |
| "entropy": 0.3809856834821403, |
| "epoch": 3.928, |
| "grad_norm": 5.09375, |
| "learning_rate": 1.1313333333333334e-05, |
| "loss": 0.3642, |
| "mean_token_accuracy": 0.884857527166605, |
| "num_tokens": 70242064.0, |
| "step": 4910 |
| }, |
| { |
| "entropy": 0.3923514380119741, |
| "epoch": 3.936, |
| "grad_norm": 4.6875, |
| "learning_rate": 1.1291111111111113e-05, |
| "loss": 0.3782, |
| "mean_token_accuracy": 0.8797802284359932, |
| "num_tokens": 70395123.0, |
| "step": 4920 |
| }, |
| { |
| "entropy": 0.3779861328192055, |
| "epoch": 3.944, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.126888888888889e-05, |
| "loss": 0.3613, |
| "mean_token_accuracy": 0.886091311275959, |
| "num_tokens": 70538241.0, |
| "step": 4930 |
| }, |
| { |
| "entropy": 0.37287365915253756, |
| "epoch": 3.952, |
| "grad_norm": 4.875, |
| "learning_rate": 1.1246666666666669e-05, |
| "loss": 0.362, |
| "mean_token_accuracy": 0.8859498616307974, |
| "num_tokens": 70678153.0, |
| "step": 4940 |
| }, |
| { |
| "entropy": 0.375614822935313, |
| "epoch": 3.96, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.1224444444444444e-05, |
| "loss": 0.3602, |
| "mean_token_accuracy": 0.8855674635618925, |
| "num_tokens": 70821953.0, |
| "step": 4950 |
| }, |
| { |
| "entropy": 0.3887378931045532, |
| "epoch": 3.968, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.1202222222222224e-05, |
| "loss": 0.3668, |
| "mean_token_accuracy": 0.8815871883183718, |
| "num_tokens": 70963030.0, |
| "step": 4960 |
| }, |
| { |
| "entropy": 0.3692332307808101, |
| "epoch": 3.976, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 0.3542, |
| "mean_token_accuracy": 0.8868688520044088, |
| "num_tokens": 71114194.0, |
| "step": 4970 |
| }, |
| { |
| "entropy": 0.3624194352887571, |
| "epoch": 3.984, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.1157777777777779e-05, |
| "loss": 0.348, |
| "mean_token_accuracy": 0.8885762255638838, |
| "num_tokens": 71264525.0, |
| "step": 4980 |
| }, |
| { |
| "entropy": 0.3633726418949664, |
| "epoch": 3.992, |
| "grad_norm": 4.375, |
| "learning_rate": 1.1135555555555557e-05, |
| "loss": 0.3556, |
| "mean_token_accuracy": 0.8881875392049551, |
| "num_tokens": 71403783.0, |
| "step": 4990 |
| }, |
| { |
| "entropy": 0.38885025242343546, |
| "epoch": 4.0, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.1113333333333334e-05, |
| "loss": 0.3741, |
| "mean_token_accuracy": 0.8807745091617107, |
| "num_tokens": 71545696.0, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_coding_entropy": 0.9978849303722381, |
| "eval_coding_loss": 2.8970143795013428, |
| "eval_coding_mean_token_accuracy": 0.5093906437158584, |
| "eval_coding_num_tokens": 71545696.0, |
| "eval_coding_runtime": 80.727, |
| "eval_coding_samples_per_second": 6.194, |
| "eval_coding_steps_per_second": 3.097, |
| "step": 5000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_math_entropy": 0.4634436811208725, |
| "eval_math_loss": 1.0399599075317383, |
| "eval_math_mean_token_accuracy": 0.7740748710632325, |
| "eval_math_num_tokens": 71545696.0, |
| "eval_math_runtime": 45.4914, |
| "eval_math_samples_per_second": 10.991, |
| "eval_math_steps_per_second": 5.496, |
| "step": 5000 |
| }, |
| { |
| "entropy": 0.28919308301992713, |
| "epoch": 4.008, |
| "grad_norm": 5.65625, |
| "learning_rate": 1.1091111111111112e-05, |
| "loss": 0.2162, |
| "mean_token_accuracy": 0.9341955050826073, |
| "num_tokens": 71689540.0, |
| "step": 5010 |
| }, |
| { |
| "entropy": 0.2366358136292547, |
| "epoch": 4.016, |
| "grad_norm": 3.921875, |
| "learning_rate": 1.106888888888889e-05, |
| "loss": 0.1996, |
| "mean_token_accuracy": 0.9395219430327415, |
| "num_tokens": 71839909.0, |
| "step": 5020 |
| }, |
| { |
| "entropy": 0.25541594810783863, |
| "epoch": 4.024, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.1046666666666667e-05, |
| "loss": 0.2023, |
| "mean_token_accuracy": 0.9369995202869177, |
| "num_tokens": 71984981.0, |
| "step": 5030 |
| }, |
| { |
| "entropy": 0.2474956126883626, |
| "epoch": 4.032, |
| "grad_norm": 4.75, |
| "learning_rate": 1.1024444444444445e-05, |
| "loss": 0.2038, |
| "mean_token_accuracy": 0.9379708308726549, |
| "num_tokens": 72128860.0, |
| "step": 5040 |
| }, |
| { |
| "entropy": 0.24668635604903102, |
| "epoch": 4.04, |
| "grad_norm": 4.40625, |
| "learning_rate": 1.1002222222222222e-05, |
| "loss": 0.1986, |
| "mean_token_accuracy": 0.939756016805768, |
| "num_tokens": 72271134.0, |
| "step": 5050 |
| }, |
| { |
| "entropy": 0.23974487385712565, |
| "epoch": 4.048, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 0.1913, |
| "mean_token_accuracy": 0.9405794236809015, |
| "num_tokens": 72415886.0, |
| "step": 5060 |
| }, |
| { |
| "entropy": 0.2342471931129694, |
| "epoch": 4.056, |
| "grad_norm": 4.5, |
| "learning_rate": 1.0957777777777778e-05, |
| "loss": 0.1919, |
| "mean_token_accuracy": 0.9399378102272749, |
| "num_tokens": 72556364.0, |
| "step": 5070 |
| }, |
| { |
| "entropy": 0.254677475290373, |
| "epoch": 4.064, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.0935555555555557e-05, |
| "loss": 0.2069, |
| "mean_token_accuracy": 0.9358757227659226, |
| "num_tokens": 72695806.0, |
| "step": 5080 |
| }, |
| { |
| "entropy": 0.2395065507851541, |
| "epoch": 4.072, |
| "grad_norm": 4.0625, |
| "learning_rate": 1.0913333333333333e-05, |
| "loss": 0.1884, |
| "mean_token_accuracy": 0.9409325756132603, |
| "num_tokens": 72845497.0, |
| "step": 5090 |
| }, |
| { |
| "entropy": 0.23455773801542817, |
| "epoch": 4.08, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.0891111111111112e-05, |
| "loss": 0.1854, |
| "mean_token_accuracy": 0.9424024146050215, |
| "num_tokens": 72986978.0, |
| "step": 5100 |
| }, |
| { |
| "epoch": 4.08, |
| "eval_coding_entropy": 0.8721909387111664, |
| "eval_coding_loss": 3.243525505065918, |
| "eval_coding_mean_token_accuracy": 0.49728499686717986, |
| "eval_coding_num_tokens": 72986978.0, |
| "eval_coding_runtime": 80.649, |
| "eval_coding_samples_per_second": 6.2, |
| "eval_coding_steps_per_second": 3.1, |
| "step": 5100 |
| }, |
| { |
| "epoch": 4.08, |
| "eval_math_entropy": 0.37121138352155686, |
| "eval_math_loss": 1.21889328956604, |
| "eval_math_mean_token_accuracy": 0.7666704759597779, |
| "eval_math_num_tokens": 72986978.0, |
| "eval_math_runtime": 45.3763, |
| "eval_math_samples_per_second": 11.019, |
| "eval_math_steps_per_second": 5.509, |
| "step": 5100 |
| }, |
| { |
| "entropy": 0.24260664987377822, |
| "epoch": 4.088, |
| "grad_norm": 5.21875, |
| "learning_rate": 1.0868888888888888e-05, |
| "loss": 0.1941, |
| "mean_token_accuracy": 0.9385300673544407, |
| "num_tokens": 73127844.0, |
| "step": 5110 |
| }, |
| { |
| "entropy": 0.2419006848242134, |
| "epoch": 4.096, |
| "grad_norm": 5.28125, |
| "learning_rate": 1.0846666666666667e-05, |
| "loss": 0.1974, |
| "mean_token_accuracy": 0.9387755412608385, |
| "num_tokens": 73271299.0, |
| "step": 5120 |
| }, |
| { |
| "entropy": 0.23894086172804235, |
| "epoch": 4.104, |
| "grad_norm": 4.625, |
| "learning_rate": 1.0824444444444447e-05, |
| "loss": 0.1931, |
| "mean_token_accuracy": 0.9397589214146137, |
| "num_tokens": 73415299.0, |
| "step": 5130 |
| }, |
| { |
| "entropy": 0.2361506049055606, |
| "epoch": 4.112, |
| "grad_norm": 4.875, |
| "learning_rate": 1.0802222222222223e-05, |
| "loss": 0.1931, |
| "mean_token_accuracy": 0.9405748970806599, |
| "num_tokens": 73562479.0, |
| "step": 5140 |
| }, |
| { |
| "entropy": 0.2326064475812018, |
| "epoch": 4.12, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 0.1854, |
| "mean_token_accuracy": 0.9420107550919056, |
| "num_tokens": 73706744.0, |
| "step": 5150 |
| }, |
| { |
| "entropy": 0.24269476146437227, |
| "epoch": 4.128, |
| "grad_norm": 4.75, |
| "learning_rate": 1.0757777777777778e-05, |
| "loss": 0.2024, |
| "mean_token_accuracy": 0.93746776022017, |
| "num_tokens": 73845445.0, |
| "step": 5160 |
| }, |
| { |
| "entropy": 0.2444027409888804, |
| "epoch": 4.136, |
| "grad_norm": 4.53125, |
| "learning_rate": 1.0735555555555557e-05, |
| "loss": 0.1993, |
| "mean_token_accuracy": 0.9361980214715004, |
| "num_tokens": 73988841.0, |
| "step": 5170 |
| }, |
| { |
| "entropy": 0.235722887981683, |
| "epoch": 4.144, |
| "grad_norm": 3.796875, |
| "learning_rate": 1.0713333333333333e-05, |
| "loss": 0.1913, |
| "mean_token_accuracy": 0.9398902922868728, |
| "num_tokens": 74135425.0, |
| "step": 5180 |
| }, |
| { |
| "entropy": 0.24436253043822945, |
| "epoch": 4.152, |
| "grad_norm": 4.21875, |
| "learning_rate": 1.0691111111111112e-05, |
| "loss": 0.2008, |
| "mean_token_accuracy": 0.9367274407297372, |
| "num_tokens": 74288949.0, |
| "step": 5190 |
| }, |
| { |
| "entropy": 0.24575479007326068, |
| "epoch": 4.16, |
| "grad_norm": 4.5, |
| "learning_rate": 1.0668888888888892e-05, |
| "loss": 0.2012, |
| "mean_token_accuracy": 0.9368447102606297, |
| "num_tokens": 74431943.0, |
| "step": 5200 |
| }, |
| { |
| "epoch": 4.16, |
| "eval_coding_entropy": 0.8665661373138428, |
| "eval_coding_loss": 3.2714428901672363, |
| "eval_coding_mean_token_accuracy": 0.4959770722389221, |
| "eval_coding_num_tokens": 74431943.0, |
| "eval_coding_runtime": 80.2919, |
| "eval_coding_samples_per_second": 6.227, |
| "eval_coding_steps_per_second": 3.114, |
| "step": 5200 |
| }, |
| { |
| "epoch": 4.16, |
| "eval_math_entropy": 0.36944735699892045, |
| "eval_math_loss": 1.2288126945495605, |
| "eval_math_mean_token_accuracy": 0.7664795405864716, |
| "eval_math_num_tokens": 74431943.0, |
| "eval_math_runtime": 45.3911, |
| "eval_math_samples_per_second": 11.015, |
| "eval_math_steps_per_second": 5.508, |
| "step": 5200 |
| }, |
| { |
| "entropy": 0.24547731992788613, |
| "epoch": 4.168, |
| "grad_norm": 5.25, |
| "learning_rate": 1.0646666666666668e-05, |
| "loss": 0.2009, |
| "mean_token_accuracy": 0.9369833692908287, |
| "num_tokens": 74574799.0, |
| "step": 5210 |
| }, |
| { |
| "entropy": 0.2451748125255108, |
| "epoch": 4.176, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.0624444444444447e-05, |
| "loss": 0.2067, |
| "mean_token_accuracy": 0.9366228148341179, |
| "num_tokens": 74713404.0, |
| "step": 5220 |
| }, |
| { |
| "entropy": 0.23505266499705613, |
| "epoch": 4.184, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.0602222222222223e-05, |
| "loss": 0.1882, |
| "mean_token_accuracy": 0.9409615311771631, |
| "num_tokens": 74852126.0, |
| "step": 5230 |
| }, |
| { |
| "entropy": 0.24006395200267433, |
| "epoch": 4.192, |
| "grad_norm": 4.84375, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 0.1956, |
| "mean_token_accuracy": 0.9393050987273455, |
| "num_tokens": 74997493.0, |
| "step": 5240 |
| }, |
| { |
| "entropy": 0.23621616866439582, |
| "epoch": 4.2, |
| "grad_norm": 5.0, |
| "learning_rate": 1.0557777777777778e-05, |
| "loss": 0.1909, |
| "mean_token_accuracy": 0.9400795757770538, |
| "num_tokens": 75140086.0, |
| "step": 5250 |
| }, |
| { |
| "entropy": 0.2386924964375794, |
| "epoch": 4.208, |
| "grad_norm": 4.25, |
| "learning_rate": 1.0535555555555557e-05, |
| "loss": 0.1924, |
| "mean_token_accuracy": 0.9397358737885952, |
| "num_tokens": 75275375.0, |
| "step": 5260 |
| }, |
| { |
| "entropy": 0.23743885876610876, |
| "epoch": 4.216, |
| "grad_norm": 4.5625, |
| "learning_rate": 1.0513333333333333e-05, |
| "loss": 0.1905, |
| "mean_token_accuracy": 0.94017285220325, |
| "num_tokens": 75416848.0, |
| "step": 5270 |
| }, |
| { |
| "entropy": 0.2380893887486309, |
| "epoch": 4.224, |
| "grad_norm": 4.8125, |
| "learning_rate": 1.0491111111111112e-05, |
| "loss": 0.1957, |
| "mean_token_accuracy": 0.9389154739677906, |
| "num_tokens": 75563461.0, |
| "step": 5280 |
| }, |
| { |
| "entropy": 0.24170237150974572, |
| "epoch": 4.232, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.046888888888889e-05, |
| "loss": 0.199, |
| "mean_token_accuracy": 0.9385070417076349, |
| "num_tokens": 75711005.0, |
| "step": 5290 |
| }, |
| { |
| "entropy": 0.24652994736097752, |
| "epoch": 4.24, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.0446666666666668e-05, |
| "loss": 0.203, |
| "mean_token_accuracy": 0.936685860902071, |
| "num_tokens": 75854808.0, |
| "step": 5300 |
| }, |
| { |
| "epoch": 4.24, |
| "eval_coding_entropy": 0.867091385126114, |
| "eval_coding_loss": 3.2733445167541504, |
| "eval_coding_mean_token_accuracy": 0.4962546360492706, |
| "eval_coding_num_tokens": 75854808.0, |
| "eval_coding_runtime": 80.3009, |
| "eval_coding_samples_per_second": 6.227, |
| "eval_coding_steps_per_second": 3.113, |
| "step": 5300 |
| }, |
| { |
| "epoch": 4.24, |
| "eval_math_entropy": 0.369137930393219, |
| "eval_math_loss": 1.2312028408050537, |
| "eval_math_mean_token_accuracy": 0.766460196018219, |
| "eval_math_num_tokens": 75854808.0, |
| "eval_math_runtime": 45.4172, |
| "eval_math_samples_per_second": 11.009, |
| "eval_math_steps_per_second": 5.505, |
| "step": 5300 |
| }, |
| { |
| "entropy": 0.2501950033940375, |
| "epoch": 4.248, |
| "grad_norm": 5.03125, |
| "learning_rate": 1.0424444444444445e-05, |
| "loss": 0.2077, |
| "mean_token_accuracy": 0.935857379809022, |
| "num_tokens": 76003692.0, |
| "step": 5310 |
| }, |
| { |
| "entropy": 0.2273095187265426, |
| "epoch": 4.256, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.0402222222222223e-05, |
| "loss": 0.187, |
| "mean_token_accuracy": 0.9406390994787216, |
| "num_tokens": 76148449.0, |
| "step": 5320 |
| }, |
| { |
| "entropy": 0.23291684789583086, |
| "epoch": 4.264, |
| "grad_norm": 3.859375, |
| "learning_rate": 1.038e-05, |
| "loss": 0.1863, |
| "mean_token_accuracy": 0.9404033485800027, |
| "num_tokens": 76291734.0, |
| "step": 5330 |
| }, |
| { |
| "entropy": 0.2373313173186034, |
| "epoch": 4.272, |
| "grad_norm": 4.875, |
| "learning_rate": 1.0357777777777778e-05, |
| "loss": 0.1928, |
| "mean_token_accuracy": 0.9394815161824226, |
| "num_tokens": 76435793.0, |
| "step": 5340 |
| }, |
| { |
| "entropy": 0.23483251431025565, |
| "epoch": 4.28, |
| "grad_norm": 4.75, |
| "learning_rate": 1.0335555555555556e-05, |
| "loss": 0.1934, |
| "mean_token_accuracy": 0.9396228928118944, |
| "num_tokens": 76585530.0, |
| "step": 5350 |
| }, |
| { |
| "entropy": 0.2379099162761122, |
| "epoch": 4.288, |
| "grad_norm": 4.65625, |
| "learning_rate": 1.0313333333333335e-05, |
| "loss": 0.1958, |
| "mean_token_accuracy": 0.9397361688315868, |
| "num_tokens": 76726162.0, |
| "step": 5360 |
| }, |
| { |
| "entropy": 0.2431611659936607, |
| "epoch": 4.296, |
| "grad_norm": 4.3125, |
| "learning_rate": 1.0291111111111111e-05, |
| "loss": 0.1992, |
| "mean_token_accuracy": 0.9383690871298314, |
| "num_tokens": 76868425.0, |
| "step": 5370 |
| }, |
| { |
| "entropy": 0.23810519305989147, |
| "epoch": 4.304, |
| "grad_norm": 4.90625, |
| "learning_rate": 1.026888888888889e-05, |
| "loss": 0.1996, |
| "mean_token_accuracy": 0.9376580834388732, |
| "num_tokens": 77016858.0, |
| "step": 5380 |
| }, |
| { |
| "entropy": 0.2412513096816838, |
| "epoch": 4.312, |
| "grad_norm": 4.78125, |
| "learning_rate": 1.0246666666666666e-05, |
| "loss": 0.2013, |
| "mean_token_accuracy": 0.9362608548253775, |
| "num_tokens": 77163432.0, |
| "step": 5390 |
| }, |
| { |
| "entropy": 0.2439442121423781, |
| "epoch": 4.32, |
| "grad_norm": 4.875, |
| "learning_rate": 1.0224444444444446e-05, |
| "loss": 0.2021, |
| "mean_token_accuracy": 0.937019569799304, |
| "num_tokens": 77311424.0, |
| "step": 5400 |
| }, |
| { |
| "epoch": 4.32, |
| "eval_coding_entropy": 0.8642041590213776, |
| "eval_coding_loss": 3.2831966876983643, |
| "eval_coding_mean_token_accuracy": 0.4949408472776413, |
| "eval_coding_num_tokens": 77311424.0, |
| "eval_coding_runtime": 80.2691, |
| "eval_coding_samples_per_second": 6.229, |
| "eval_coding_steps_per_second": 3.115, |
| "step": 5400 |
| }, |
| { |
| "epoch": 4.32, |
| "eval_math_entropy": 0.37020832109451296, |
| "eval_math_loss": 1.2300353050231934, |
| "eval_math_mean_token_accuracy": 0.7663494193553925, |
| "eval_math_num_tokens": 77311424.0, |
| "eval_math_runtime": 45.396, |
| "eval_math_samples_per_second": 11.014, |
| "eval_math_steps_per_second": 5.507, |
| "step": 5400 |
| }, |
| { |
| "entropy": 0.22979448451660572, |
| "epoch": 4.328, |
| "grad_norm": 4.375, |
| "learning_rate": 1.0202222222222221e-05, |
| "loss": 0.1913, |
| "mean_token_accuracy": 0.9401264976710081, |
| "num_tokens": 77451500.0, |
| "step": 5410 |
| }, |
| { |
| "entropy": 0.24582726685330272, |
| "epoch": 4.336, |
| "grad_norm": 5.25, |
| "learning_rate": 1.018e-05, |
| "loss": 0.2017, |
| "mean_token_accuracy": 0.9370923280715943, |
| "num_tokens": 77596829.0, |
| "step": 5420 |
| }, |
| { |
| "entropy": 0.23917227215133607, |
| "epoch": 4.344, |
| "grad_norm": 4.71875, |
| "learning_rate": 1.0157777777777777e-05, |
| "loss": 0.1984, |
| "mean_token_accuracy": 0.9387729406356812, |
| "num_tokens": 77742947.0, |
| "step": 5430 |
| }, |
| { |
| "entropy": 0.24740321515128016, |
| "epoch": 4.352, |
| "grad_norm": 5.0625, |
| "learning_rate": 1.0135555555555556e-05, |
| "loss": 0.2018, |
| "mean_token_accuracy": 0.9359302509576082, |
| "num_tokens": 77889107.0, |
| "step": 5440 |
| }, |
| { |
| "entropy": 0.23968188627623022, |
| "epoch": 4.36, |
| "grad_norm": 4.25, |
| "learning_rate": 1.0113333333333335e-05, |
| "loss": 0.2014, |
| "mean_token_accuracy": 0.9373920358717441, |
| "num_tokens": 78033570.0, |
| "step": 5450 |
| }, |
| { |
| "entropy": 0.24595484188757838, |
| "epoch": 4.368, |
| "grad_norm": 5.1875, |
| "learning_rate": 1.0091111111111111e-05, |
| "loss": 0.2016, |
| "mean_token_accuracy": 0.9371922299265861, |
| "num_tokens": 78176273.0, |
| "step": 5460 |
| }, |
| { |
| "entropy": 0.23858330412767828, |
| "epoch": 4.376, |
| "grad_norm": 4.34375, |
| "learning_rate": 1.006888888888889e-05, |
| "loss": 0.1983, |
| "mean_token_accuracy": 0.9376890752464533, |
| "num_tokens": 78319367.0, |
| "step": 5470 |
| }, |
| { |
| "entropy": 0.2517506274394691, |
| "epoch": 4.384, |
| "grad_norm": 4.875, |
| "learning_rate": 1.0046666666666666e-05, |
| "loss": 0.2066, |
| "mean_token_accuracy": 0.9337670724838972, |
| "num_tokens": 78461335.0, |
| "step": 5480 |
| }, |
| { |
| "entropy": 0.25959395272657276, |
| "epoch": 4.392, |
| "grad_norm": 4.75, |
| "learning_rate": 1.0024444444444446e-05, |
| "loss": 0.2176, |
| "mean_token_accuracy": 0.9327810432761907, |
| "num_tokens": 78602893.0, |
| "step": 5490 |
| }, |
| { |
| "entropy": 0.23389470535330475, |
| "epoch": 4.4, |
| "grad_norm": 4.59375, |
| "learning_rate": 1.0002222222222222e-05, |
| "loss": 0.193, |
| "mean_token_accuracy": 0.9383365754038095, |
| "num_tokens": 78748521.0, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.4, |
| "eval_coding_entropy": 0.8639817719459534, |
| "eval_coding_loss": 3.284721851348877, |
| "eval_coding_mean_token_accuracy": 0.49526145589351656, |
| "eval_coding_num_tokens": 78748521.0, |
| "eval_coding_runtime": 80.2508, |
| "eval_coding_samples_per_second": 6.23, |
| "eval_coding_steps_per_second": 3.115, |
| "step": 5500 |
| }, |
| { |
| "epoch": 4.4, |
| "eval_math_entropy": 0.3674210168719292, |
| "eval_math_loss": 1.2383450269699097, |
| "eval_math_mean_token_accuracy": 0.7659579167366027, |
| "eval_math_num_tokens": 78748521.0, |
| "eval_math_runtime": 45.3216, |
| "eval_math_samples_per_second": 11.032, |
| "eval_math_steps_per_second": 5.516, |
| "step": 5500 |
| }, |
| { |
| "entropy": 0.2523001337889582, |
| "epoch": 4.408, |
| "grad_norm": 5.03125, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 0.2102, |
| "mean_token_accuracy": 0.9342300169169903, |
| "num_tokens": 78888559.0, |
| "step": 5510 |
| }, |
| { |
| "entropy": 0.23446680726483465, |
| "epoch": 4.416, |
| "grad_norm": 4.875, |
| "learning_rate": 9.957777777777779e-06, |
| "loss": 0.1929, |
| "mean_token_accuracy": 0.9398462988436223, |
| "num_tokens": 79028488.0, |
| "step": 5520 |
| }, |
| { |
| "entropy": 0.2429468885064125, |
| "epoch": 4.424, |
| "grad_norm": 5.34375, |
| "learning_rate": 9.935555555555556e-06, |
| "loss": 0.1999, |
| "mean_token_accuracy": 0.9371183041483164, |
| "num_tokens": 79176307.0, |
| "step": 5530 |
| }, |
| { |
| "entropy": 0.2349563892930746, |
| "epoch": 4.432, |
| "grad_norm": 4.75, |
| "learning_rate": 9.913333333333334e-06, |
| "loss": 0.1921, |
| "mean_token_accuracy": 0.9398663032799959, |
| "num_tokens": 79315788.0, |
| "step": 5540 |
| }, |
| { |
| "entropy": 0.2346967566292733, |
| "epoch": 4.44, |
| "grad_norm": 5.1875, |
| "learning_rate": 9.891111111111113e-06, |
| "loss": 0.1956, |
| "mean_token_accuracy": 0.9401642885059118, |
| "num_tokens": 79455782.0, |
| "step": 5550 |
| }, |
| { |
| "entropy": 0.2509564989712089, |
| "epoch": 4.448, |
| "grad_norm": 5.28125, |
| "learning_rate": 9.86888888888889e-06, |
| "loss": 0.2097, |
| "mean_token_accuracy": 0.9353599067777395, |
| "num_tokens": 79595473.0, |
| "step": 5560 |
| }, |
| { |
| "entropy": 0.24543018965050578, |
| "epoch": 4.456, |
| "grad_norm": 4.6875, |
| "learning_rate": 9.846666666666668e-06, |
| "loss": 0.2033, |
| "mean_token_accuracy": 0.9361506931483745, |
| "num_tokens": 79742205.0, |
| "step": 5570 |
| }, |
| { |
| "entropy": 0.24763995162211358, |
| "epoch": 4.464, |
| "grad_norm": 4.5625, |
| "learning_rate": 9.824444444444446e-06, |
| "loss": 0.2087, |
| "mean_token_accuracy": 0.9350599583238364, |
| "num_tokens": 79884376.0, |
| "step": 5580 |
| }, |
| { |
| "entropy": 0.24856340689584613, |
| "epoch": 4.4719999999999995, |
| "grad_norm": 4.46875, |
| "learning_rate": 9.802222222222224e-06, |
| "loss": 0.2061, |
| "mean_token_accuracy": 0.9357188276946544, |
| "num_tokens": 80021902.0, |
| "step": 5590 |
| }, |
| { |
| "entropy": 0.2427858689799905, |
| "epoch": 4.48, |
| "grad_norm": 4.84375, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 0.2034, |
| "mean_token_accuracy": 0.9366299249231815, |
| "num_tokens": 80163433.0, |
| "step": 5600 |
| }, |
| { |
| "epoch": 4.48, |
| "eval_coding_entropy": 0.8580341808795929, |
| "eval_coding_loss": 3.291886806488037, |
| "eval_coding_mean_token_accuracy": 0.4952204238176346, |
| "eval_coding_num_tokens": 80163433.0, |
| "eval_coding_runtime": 80.3982, |
| "eval_coding_samples_per_second": 6.219, |
| "eval_coding_steps_per_second": 3.11, |
| "step": 5600 |
| }, |
| { |
| "epoch": 4.48, |
| "eval_math_entropy": 0.36658355516195296, |
| "eval_math_loss": 1.2375688552856445, |
| "eval_math_mean_token_accuracy": 0.7660913934707642, |
| "eval_math_num_tokens": 80163433.0, |
| "eval_math_runtime": 45.4302, |
| "eval_math_samples_per_second": 11.006, |
| "eval_math_steps_per_second": 5.503, |
| "step": 5600 |
| }, |
| { |
| "entropy": 0.24371355529874564, |
| "epoch": 4.4879999999999995, |
| "grad_norm": 4.875, |
| "learning_rate": 9.757777777777779e-06, |
| "loss": 0.2043, |
| "mean_token_accuracy": 0.9371550094336272, |
| "num_tokens": 80300407.0, |
| "step": 5610 |
| }, |
| { |
| "entropy": 0.25914601678960025, |
| "epoch": 4.496, |
| "grad_norm": 5.21875, |
| "learning_rate": 9.735555555555556e-06, |
| "loss": 0.2176, |
| "mean_token_accuracy": 0.9322570439428091, |
| "num_tokens": 80443676.0, |
| "step": 5620 |
| }, |
| { |
| "entropy": 0.23960631545633077, |
| "epoch": 4.504, |
| "grad_norm": 5.6875, |
| "learning_rate": 9.713333333333334e-06, |
| "loss": 0.1983, |
| "mean_token_accuracy": 0.9372116841375828, |
| "num_tokens": 80582575.0, |
| "step": 5630 |
| }, |
| { |
| "entropy": 0.2339357567951083, |
| "epoch": 4.5120000000000005, |
| "grad_norm": 4.75, |
| "learning_rate": 9.691111111111112e-06, |
| "loss": 0.193, |
| "mean_token_accuracy": 0.9385768733918667, |
| "num_tokens": 80731924.0, |
| "step": 5640 |
| }, |
| { |
| "entropy": 0.24636179031804203, |
| "epoch": 4.52, |
| "grad_norm": 4.84375, |
| "learning_rate": 9.66888888888889e-06, |
| "loss": 0.2063, |
| "mean_token_accuracy": 0.935525069385767, |
| "num_tokens": 80877372.0, |
| "step": 5650 |
| }, |
| { |
| "entropy": 0.23936477806419135, |
| "epoch": 4.5280000000000005, |
| "grad_norm": 5.46875, |
| "learning_rate": 9.646666666666667e-06, |
| "loss": 0.1987, |
| "mean_token_accuracy": 0.9378582388162613, |
| "num_tokens": 81022957.0, |
| "step": 5660 |
| }, |
| { |
| "entropy": 0.24199356408789754, |
| "epoch": 4.536, |
| "grad_norm": 4.875, |
| "learning_rate": 9.624444444444445e-06, |
| "loss": 0.1996, |
| "mean_token_accuracy": 0.9372540459036827, |
| "num_tokens": 81163034.0, |
| "step": 5670 |
| }, |
| { |
| "entropy": 0.24176772963255644, |
| "epoch": 4.5440000000000005, |
| "grad_norm": 4.75, |
| "learning_rate": 9.602222222222222e-06, |
| "loss": 0.2001, |
| "mean_token_accuracy": 0.9374632082879544, |
| "num_tokens": 81309913.0, |
| "step": 5680 |
| }, |
| { |
| "entropy": 0.24366185194812714, |
| "epoch": 4.552, |
| "grad_norm": 5.21875, |
| "learning_rate": 9.58e-06, |
| "loss": 0.2058, |
| "mean_token_accuracy": 0.936968456208706, |
| "num_tokens": 81449547.0, |
| "step": 5690 |
| }, |
| { |
| "entropy": 0.23524297084659337, |
| "epoch": 4.5600000000000005, |
| "grad_norm": 5.34375, |
| "learning_rate": 9.557777777777777e-06, |
| "loss": 0.1942, |
| "mean_token_accuracy": 0.9392556425184011, |
| "num_tokens": 81589202.0, |
| "step": 5700 |
| }, |
| { |
| "epoch": 4.5600000000000005, |
| "eval_coding_entropy": 0.8490579767227173, |
| "eval_coding_loss": 3.2955329418182373, |
| "eval_coding_mean_token_accuracy": 0.49646771132946016, |
| "eval_coding_num_tokens": 81589202.0, |
| "eval_coding_runtime": 80.352, |
| "eval_coding_samples_per_second": 6.223, |
| "eval_coding_steps_per_second": 3.111, |
| "step": 5700 |
| }, |
| { |
| "epoch": 4.5600000000000005, |
| "eval_math_entropy": 0.3640580996274948, |
| "eval_math_loss": 1.2399064302444458, |
| "eval_math_mean_token_accuracy": 0.7662024302482605, |
| "eval_math_num_tokens": 81589202.0, |
| "eval_math_runtime": 45.415, |
| "eval_math_samples_per_second": 11.01, |
| "eval_math_steps_per_second": 5.505, |
| "step": 5700 |
| }, |
| { |
| "entropy": 0.24147932096384467, |
| "epoch": 4.568, |
| "grad_norm": 5.1875, |
| "learning_rate": 9.535555555555557e-06, |
| "loss": 0.2069, |
| "mean_token_accuracy": 0.9362010084092617, |
| "num_tokens": 81729019.0, |
| "step": 5710 |
| }, |
| { |
| "entropy": 0.23784119579941035, |
| "epoch": 4.576, |
| "grad_norm": 5.15625, |
| "learning_rate": 9.513333333333334e-06, |
| "loss": 0.1978, |
| "mean_token_accuracy": 0.9374346259981394, |
| "num_tokens": 81869804.0, |
| "step": 5720 |
| }, |
| { |
| "entropy": 0.24561254819855094, |
| "epoch": 4.584, |
| "grad_norm": 4.625, |
| "learning_rate": 9.491111111111112e-06, |
| "loss": 0.2039, |
| "mean_token_accuracy": 0.9363627798855305, |
| "num_tokens": 82004791.0, |
| "step": 5730 |
| }, |
| { |
| "entropy": 0.24887093310244382, |
| "epoch": 4.592, |
| "grad_norm": 4.59375, |
| "learning_rate": 9.46888888888889e-06, |
| "loss": 0.2067, |
| "mean_token_accuracy": 0.935890756547451, |
| "num_tokens": 82148464.0, |
| "step": 5740 |
| }, |
| { |
| "entropy": 0.23292487277649343, |
| "epoch": 4.6, |
| "grad_norm": 5.1875, |
| "learning_rate": 9.446666666666667e-06, |
| "loss": 0.1948, |
| "mean_token_accuracy": 0.9389873996376992, |
| "num_tokens": 82293216.0, |
| "step": 5750 |
| }, |
| { |
| "entropy": 0.2514194253366441, |
| "epoch": 4.608, |
| "grad_norm": 5.5625, |
| "learning_rate": 9.424444444444445e-06, |
| "loss": 0.208, |
| "mean_token_accuracy": 0.9355383906513453, |
| "num_tokens": 82427716.0, |
| "step": 5760 |
| }, |
| { |
| "entropy": 0.2385628822259605, |
| "epoch": 4.616, |
| "grad_norm": 5.4375, |
| "learning_rate": 9.402222222222222e-06, |
| "loss": 0.1938, |
| "mean_token_accuracy": 0.9385287798941135, |
| "num_tokens": 82567756.0, |
| "step": 5770 |
| }, |
| { |
| "entropy": 0.246723764995113, |
| "epoch": 4.624, |
| "grad_norm": 5.28125, |
| "learning_rate": 9.38e-06, |
| "loss": 0.2081, |
| "mean_token_accuracy": 0.9364254608750343, |
| "num_tokens": 82710604.0, |
| "step": 5780 |
| }, |
| { |
| "entropy": 0.24006328606046737, |
| "epoch": 4.632, |
| "grad_norm": 5.28125, |
| "learning_rate": 9.35777777777778e-06, |
| "loss": 0.2001, |
| "mean_token_accuracy": 0.9366243660449982, |
| "num_tokens": 82845295.0, |
| "step": 5790 |
| }, |
| { |
| "entropy": 0.24273840263485907, |
| "epoch": 4.64, |
| "grad_norm": 5.21875, |
| "learning_rate": 9.335555555555557e-06, |
| "loss": 0.2018, |
| "mean_token_accuracy": 0.9366823032498359, |
| "num_tokens": 82990401.0, |
| "step": 5800 |
| }, |
| { |
| "epoch": 4.64, |
| "eval_coding_entropy": 0.8574855062961578, |
| "eval_coding_loss": 3.284010171890259, |
| "eval_coding_mean_token_accuracy": 0.4962291866540909, |
| "eval_coding_num_tokens": 82990401.0, |
| "eval_coding_runtime": 80.3464, |
| "eval_coding_samples_per_second": 6.223, |
| "eval_coding_steps_per_second": 3.112, |
| "step": 5800 |
| }, |
| { |
| "epoch": 4.64, |
| "eval_math_entropy": 0.3680396780371666, |
| "eval_math_loss": 1.235351324081421, |
| "eval_math_mean_token_accuracy": 0.7662593610286713, |
| "eval_math_num_tokens": 82990401.0, |
| "eval_math_runtime": 45.4496, |
| "eval_math_samples_per_second": 11.001, |
| "eval_math_steps_per_second": 5.501, |
| "step": 5800 |
| }, |
| { |
| "entropy": 0.24379037078469992, |
| "epoch": 4.648, |
| "grad_norm": 5.78125, |
| "learning_rate": 9.313333333333335e-06, |
| "loss": 0.1945, |
| "mean_token_accuracy": 0.9394961971789598, |
| "num_tokens": 83127708.0, |
| "step": 5810 |
| }, |
| { |
| "entropy": 0.24079365273937584, |
| "epoch": 4.656, |
| "grad_norm": 5.4375, |
| "learning_rate": 9.291111111111112e-06, |
| "loss": 0.1978, |
| "mean_token_accuracy": 0.9378721848130226, |
| "num_tokens": 83269847.0, |
| "step": 5820 |
| }, |
| { |
| "entropy": 0.23888413924723864, |
| "epoch": 4.664, |
| "grad_norm": 4.8125, |
| "learning_rate": 9.26888888888889e-06, |
| "loss": 0.1968, |
| "mean_token_accuracy": 0.9375009935349226, |
| "num_tokens": 83409952.0, |
| "step": 5830 |
| }, |
| { |
| "entropy": 0.2453721878118813, |
| "epoch": 4.672, |
| "grad_norm": 4.96875, |
| "learning_rate": 9.246666666666667e-06, |
| "loss": 0.2064, |
| "mean_token_accuracy": 0.9357371307909489, |
| "num_tokens": 83550833.0, |
| "step": 5840 |
| }, |
| { |
| "entropy": 0.24028267255052924, |
| "epoch": 4.68, |
| "grad_norm": 4.875, |
| "learning_rate": 9.224444444444445e-06, |
| "loss": 0.204, |
| "mean_token_accuracy": 0.9366933092474937, |
| "num_tokens": 83695948.0, |
| "step": 5850 |
| }, |
| { |
| "entropy": 0.24287205799482764, |
| "epoch": 4.688, |
| "grad_norm": 5.0625, |
| "learning_rate": 9.202222222222224e-06, |
| "loss": 0.2046, |
| "mean_token_accuracy": 0.9369335785508156, |
| "num_tokens": 83840860.0, |
| "step": 5860 |
| }, |
| { |
| "entropy": 0.2388632460962981, |
| "epoch": 4.696, |
| "grad_norm": 5.25, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 0.197, |
| "mean_token_accuracy": 0.9387118477374315, |
| "num_tokens": 83984234.0, |
| "step": 5870 |
| }, |
| { |
| "entropy": 0.24010645528323948, |
| "epoch": 4.704, |
| "grad_norm": 4.875, |
| "learning_rate": 9.15777777777778e-06, |
| "loss": 0.2022, |
| "mean_token_accuracy": 0.9373754136264324, |
| "num_tokens": 84128625.0, |
| "step": 5880 |
| }, |
| { |
| "entropy": 0.2570457765366882, |
| "epoch": 4.712, |
| "grad_norm": 5.78125, |
| "learning_rate": 9.135555555555557e-06, |
| "loss": 0.2169, |
| "mean_token_accuracy": 0.9330484878271819, |
| "num_tokens": 84278278.0, |
| "step": 5890 |
| }, |
| { |
| "entropy": 0.2350900627207011, |
| "epoch": 4.72, |
| "grad_norm": 4.8125, |
| "learning_rate": 9.113333333333335e-06, |
| "loss": 0.1997, |
| "mean_token_accuracy": 0.9393527030944824, |
| "num_tokens": 84423857.0, |
| "step": 5900 |
| }, |
| { |
| "epoch": 4.72, |
| "eval_coding_entropy": 0.8491866540908813, |
| "eval_coding_loss": 3.288071393966675, |
| "eval_coding_mean_token_accuracy": 0.4968215843439102, |
| "eval_coding_num_tokens": 84423857.0, |
| "eval_coding_runtime": 80.3178, |
| "eval_coding_samples_per_second": 6.225, |
| "eval_coding_steps_per_second": 3.113, |
| "step": 5900 |
| }, |
| { |
| "epoch": 4.72, |
| "eval_math_entropy": 0.36505941808223724, |
| "eval_math_loss": 1.2386001348495483, |
| "eval_math_mean_token_accuracy": 0.7660931167602539, |
| "eval_math_num_tokens": 84423857.0, |
| "eval_math_runtime": 45.437, |
| "eval_math_samples_per_second": 11.004, |
| "eval_math_steps_per_second": 5.502, |
| "step": 5900 |
| }, |
| { |
| "entropy": 0.24601346775889396, |
| "epoch": 4.728, |
| "grad_norm": 4.78125, |
| "learning_rate": 9.091111111111112e-06, |
| "loss": 0.2043, |
| "mean_token_accuracy": 0.9365566540509462, |
| "num_tokens": 84563785.0, |
| "step": 5910 |
| }, |
| { |
| "entropy": 0.23536207475699483, |
| "epoch": 4.736, |
| "grad_norm": 5.125, |
| "learning_rate": 9.06888888888889e-06, |
| "loss": 0.1956, |
| "mean_token_accuracy": 0.9389339841902256, |
| "num_tokens": 84695402.0, |
| "step": 5920 |
| }, |
| { |
| "entropy": 0.23836666014976798, |
| "epoch": 4.744, |
| "grad_norm": 5.40625, |
| "learning_rate": 9.046666666666668e-06, |
| "loss": 0.1932, |
| "mean_token_accuracy": 0.938853096216917, |
| "num_tokens": 84838953.0, |
| "step": 5930 |
| }, |
| { |
| "entropy": 0.23160754293203353, |
| "epoch": 4.752, |
| "grad_norm": 5.03125, |
| "learning_rate": 9.024444444444445e-06, |
| "loss": 0.1924, |
| "mean_token_accuracy": 0.9394034445285797, |
| "num_tokens": 84980763.0, |
| "step": 5940 |
| }, |
| { |
| "entropy": 0.23382272492162884, |
| "epoch": 4.76, |
| "grad_norm": 4.5, |
| "learning_rate": 9.002222222222223e-06, |
| "loss": 0.1937, |
| "mean_token_accuracy": 0.9393350720405579, |
| "num_tokens": 85125207.0, |
| "step": 5950 |
| }, |
| { |
| "entropy": 0.25123192071914674, |
| "epoch": 4.768, |
| "grad_norm": 4.625, |
| "learning_rate": 8.98e-06, |
| "loss": 0.2114, |
| "mean_token_accuracy": 0.9343443937599659, |
| "num_tokens": 85272835.0, |
| "step": 5960 |
| }, |
| { |
| "entropy": 0.2455756193958223, |
| "epoch": 4.776, |
| "grad_norm": 4.90625, |
| "learning_rate": 8.957777777777778e-06, |
| "loss": 0.2034, |
| "mean_token_accuracy": 0.9348585840314627, |
| "num_tokens": 85420144.0, |
| "step": 5970 |
| }, |
| { |
| "entropy": 0.2482148847077042, |
| "epoch": 4.784, |
| "grad_norm": 5.375, |
| "learning_rate": 8.935555555555556e-06, |
| "loss": 0.2089, |
| "mean_token_accuracy": 0.934000676125288, |
| "num_tokens": 85568674.0, |
| "step": 5980 |
| }, |
| { |
| "entropy": 0.23058385020121933, |
| "epoch": 4.792, |
| "grad_norm": 5.125, |
| "learning_rate": 8.913333333333333e-06, |
| "loss": 0.1925, |
| "mean_token_accuracy": 0.9392436318099499, |
| "num_tokens": 85711653.0, |
| "step": 5990 |
| }, |
| { |
| "entropy": 0.24983597761020065, |
| "epoch": 4.8, |
| "grad_norm": 4.34375, |
| "learning_rate": 8.891111111111111e-06, |
| "loss": 0.2066, |
| "mean_token_accuracy": 0.9344624415040016, |
| "num_tokens": 85856157.0, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.8, |
| "eval_coding_entropy": 0.852398107290268, |
| "eval_coding_loss": 3.2907886505126953, |
| "eval_coding_mean_token_accuracy": 0.4957404453754425, |
| "eval_coding_num_tokens": 85856157.0, |
| "eval_coding_runtime": 80.3199, |
| "eval_coding_samples_per_second": 6.225, |
| "eval_coding_steps_per_second": 3.113, |
| "step": 6000 |
| }, |
| { |
| "epoch": 4.8, |
| "eval_math_entropy": 0.36472996628284454, |
| "eval_math_loss": 1.23907470703125, |
| "eval_math_mean_token_accuracy": 0.7664013624191284, |
| "eval_math_num_tokens": 85856157.0, |
| "eval_math_runtime": 45.2313, |
| "eval_math_samples_per_second": 11.054, |
| "eval_math_steps_per_second": 5.527, |
| "step": 6000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 8, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 5.430255106010929e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
