| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.4, |
| "eval_steps": 100, |
| "global_step": 3000, |
| "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 |
| } |
| ], |
| "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": 2.718543355992791e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
