| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 6.3904, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 3.238465867936611, |
| "epoch": 0.032, |
| "grad_norm": 209.0, |
| "learning_rate": 1.8e-07, |
| "loss": 7.9136, |
| "mean_token_accuracy": 0.17908816603012384, |
| "num_tokens": 99789.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 3.293700896203518, |
| "epoch": 0.064, |
| "grad_norm": 198.0, |
| "learning_rate": 3.8e-07, |
| "loss": 7.9121, |
| "mean_token_accuracy": 0.17522176005877554, |
| "num_tokens": 190891.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 3.2724268689751623, |
| "epoch": 0.096, |
| "grad_norm": 195.0, |
| "learning_rate": 5.800000000000001e-07, |
| "loss": 7.7884, |
| "mean_token_accuracy": 0.18085262291133403, |
| "num_tokens": 286198.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 3.4351663440465927, |
| "epoch": 0.128, |
| "grad_norm": 172.0, |
| "learning_rate": 7.8e-07, |
| "loss": 7.5047, |
| "mean_token_accuracy": 0.18496066313236953, |
| "num_tokens": 385320.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 3.6431967571377752, |
| "epoch": 0.16, |
| "grad_norm": 155.0, |
| "learning_rate": 9.800000000000001e-07, |
| "loss": 7.2582, |
| "mean_token_accuracy": 0.18825604091398418, |
| "num_tokens": 486218.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 4.013543170690537, |
| "epoch": 0.192, |
| "grad_norm": 121.0, |
| "learning_rate": 1.1800000000000001e-06, |
| "loss": 6.9069, |
| "mean_token_accuracy": 0.18985986979678274, |
| "num_tokens": 581717.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 4.40448921173811, |
| "epoch": 0.224, |
| "grad_norm": 94.0, |
| "learning_rate": 1.3800000000000001e-06, |
| "loss": 6.3322, |
| "mean_token_accuracy": 0.20462538227438926, |
| "num_tokens": 676500.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 4.721661370992661, |
| "epoch": 0.256, |
| "grad_norm": 64.5, |
| "learning_rate": 1.5800000000000001e-06, |
| "loss": 5.8578, |
| "mean_token_accuracy": 0.2137928050942719, |
| "num_tokens": 775542.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 4.859676861763001, |
| "epoch": 0.288, |
| "grad_norm": 47.25, |
| "learning_rate": 1.7800000000000001e-06, |
| "loss": 5.4304, |
| "mean_token_accuracy": 0.22805662611499428, |
| "num_tokens": 872932.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 4.871997155249119, |
| "epoch": 0.32, |
| "grad_norm": 29.375, |
| "learning_rate": 1.98e-06, |
| "loss": 5.1109, |
| "mean_token_accuracy": 0.25303157502785323, |
| "num_tokens": 968340.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_coding_entropy": 1.926878339767456, |
| "eval_coding_loss": 2.3084304332733154, |
| "eval_coding_mean_token_accuracy": 0.48411355090141295, |
| "eval_coding_num_tokens": 968340.0, |
| "eval_coding_runtime": 80.1556, |
| "eval_coding_samples_per_second": 6.238, |
| "eval_coding_steps_per_second": 3.119, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_biology_entropy": 4.7239161949157715, |
| "eval_biology_loss": 4.968222618103027, |
| "eval_biology_mean_token_accuracy": 0.26778400188684465, |
| "eval_biology_num_tokens": 968340.0, |
| "eval_biology_runtime": 34.6244, |
| "eval_biology_samples_per_second": 14.441, |
| "eval_biology_steps_per_second": 7.22, |
| "step": 100 |
| }, |
| { |
| "entropy": 4.715248446166515, |
| "epoch": 0.352, |
| "grad_norm": 22.375, |
| "learning_rate": 2.1800000000000003e-06, |
| "loss": 4.8208, |
| "mean_token_accuracy": 0.27833458138629796, |
| "num_tokens": 1062214.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 4.553062991797924, |
| "epoch": 0.384, |
| "grad_norm": 24.0, |
| "learning_rate": 2.38e-06, |
| "loss": 4.6277, |
| "mean_token_accuracy": 0.29350204383954404, |
| "num_tokens": 1161800.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 4.52610100209713, |
| "epoch": 0.416, |
| "grad_norm": 17.625, |
| "learning_rate": 2.5800000000000003e-06, |
| "loss": 4.5457, |
| "mean_token_accuracy": 0.3010273904539645, |
| "num_tokens": 1259894.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 4.3714508935809135, |
| "epoch": 0.448, |
| "grad_norm": 24.5, |
| "learning_rate": 2.7800000000000005e-06, |
| "loss": 4.3595, |
| "mean_token_accuracy": 0.3121769287623465, |
| "num_tokens": 1358545.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 4.206141373515129, |
| "epoch": 0.48, |
| "grad_norm": 20.0, |
| "learning_rate": 2.9800000000000003e-06, |
| "loss": 4.2206, |
| "mean_token_accuracy": 0.32826427686959503, |
| "num_tokens": 1462493.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 4.1862502753734585, |
| "epoch": 0.512, |
| "grad_norm": 15.6875, |
| "learning_rate": 3.1800000000000005e-06, |
| "loss": 4.2016, |
| "mean_token_accuracy": 0.33233328983187677, |
| "num_tokens": 1557047.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 4.118708249926567, |
| "epoch": 0.544, |
| "grad_norm": 16.375, |
| "learning_rate": 3.3800000000000007e-06, |
| "loss": 4.0964, |
| "mean_token_accuracy": 0.3453528699465096, |
| "num_tokens": 1658711.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 4.135542331635952, |
| "epoch": 0.576, |
| "grad_norm": 17.25, |
| "learning_rate": 3.58e-06, |
| "loss": 4.1201, |
| "mean_token_accuracy": 0.34580441238358617, |
| "num_tokens": 1756413.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 4.003959603607655, |
| "epoch": 0.608, |
| "grad_norm": 18.375, |
| "learning_rate": 3.7800000000000002e-06, |
| "loss": 4.0091, |
| "mean_token_accuracy": 0.3496086118742824, |
| "num_tokens": 1853054.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 4.024305436015129, |
| "epoch": 0.64, |
| "grad_norm": 16.5, |
| "learning_rate": 3.980000000000001e-06, |
| "loss": 3.9643, |
| "mean_token_accuracy": 0.35361671894788743, |
| "num_tokens": 1953273.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_coding_entropy": 1.8568662638664246, |
| "eval_coding_loss": 2.1912217140197754, |
| "eval_coding_mean_token_accuracy": 0.5063816961050034, |
| "eval_coding_num_tokens": 1953273.0, |
| "eval_coding_runtime": 79.6784, |
| "eval_coding_samples_per_second": 6.275, |
| "eval_coding_steps_per_second": 3.138, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.64, |
| "eval_biology_entropy": 3.9995334424972535, |
| "eval_biology_loss": 3.9905741214752197, |
| "eval_biology_mean_token_accuracy": 0.36172509467601777, |
| "eval_biology_num_tokens": 1953273.0, |
| "eval_biology_runtime": 35.4091, |
| "eval_biology_samples_per_second": 14.121, |
| "eval_biology_steps_per_second": 7.06, |
| "step": 200 |
| }, |
| { |
| "entropy": 3.998256941139698, |
| "epoch": 0.672, |
| "grad_norm": 22.25, |
| "learning_rate": 4.18e-06, |
| "loss": 3.9371, |
| "mean_token_accuracy": 0.3586596045643091, |
| "num_tokens": 2050255.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 3.8787728905677796, |
| "epoch": 0.704, |
| "grad_norm": 18.5, |
| "learning_rate": 4.38e-06, |
| "loss": 3.8922, |
| "mean_token_accuracy": 0.363262693118304, |
| "num_tokens": 2150396.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 3.881282126903534, |
| "epoch": 0.736, |
| "grad_norm": 15.375, |
| "learning_rate": 4.58e-06, |
| "loss": 3.8711, |
| "mean_token_accuracy": 0.36838123146444557, |
| "num_tokens": 2250726.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 3.8907541409134865, |
| "epoch": 0.768, |
| "grad_norm": 19.0, |
| "learning_rate": 4.78e-06, |
| "loss": 3.8523, |
| "mean_token_accuracy": 0.36881062425673006, |
| "num_tokens": 2346168.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 3.8764460012316704, |
| "epoch": 0.8, |
| "grad_norm": 16.75, |
| "learning_rate": 4.980000000000001e-06, |
| "loss": 3.833, |
| "mean_token_accuracy": 0.37063299771398306, |
| "num_tokens": 2446303.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 3.8517253383994103, |
| "epoch": 0.832, |
| "grad_norm": 16.5, |
| "learning_rate": 5.18e-06, |
| "loss": 3.8114, |
| "mean_token_accuracy": 0.3745407974347472, |
| "num_tokens": 2540899.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 3.778804798424244, |
| "epoch": 0.864, |
| "grad_norm": 15.9375, |
| "learning_rate": 5.380000000000001e-06, |
| "loss": 3.73, |
| "mean_token_accuracy": 0.37728464547544716, |
| "num_tokens": 2638613.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 3.7176755860447885, |
| "epoch": 0.896, |
| "grad_norm": 19.875, |
| "learning_rate": 5.580000000000001e-06, |
| "loss": 3.6871, |
| "mean_token_accuracy": 0.3835112711414695, |
| "num_tokens": 2743826.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 3.7258670911192895, |
| "epoch": 0.928, |
| "grad_norm": 21.75, |
| "learning_rate": 5.78e-06, |
| "loss": 3.6856, |
| "mean_token_accuracy": 0.38429632876068354, |
| "num_tokens": 2847721.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 3.712230232357979, |
| "epoch": 0.96, |
| "grad_norm": 15.6875, |
| "learning_rate": 5.98e-06, |
| "loss": 3.6164, |
| "mean_token_accuracy": 0.3888826366513968, |
| "num_tokens": 2948735.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_coding_entropy": 1.794304684638977, |
| "eval_coding_loss": 2.1738245487213135, |
| "eval_coding_mean_token_accuracy": 0.5117835735082626, |
| "eval_coding_num_tokens": 2948735.0, |
| "eval_coding_runtime": 78.9169, |
| "eval_coding_samples_per_second": 6.336, |
| "eval_coding_steps_per_second": 3.168, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.96, |
| "eval_biology_entropy": 3.593994511604309, |
| "eval_biology_loss": 3.689100503921509, |
| "eval_biology_mean_token_accuracy": 0.39202056431770327, |
| "eval_biology_num_tokens": 2948735.0, |
| "eval_biology_runtime": 34.5492, |
| "eval_biology_samples_per_second": 14.472, |
| "eval_biology_steps_per_second": 7.236, |
| "step": 300 |
| }, |
| { |
| "entropy": 3.6798228099942207, |
| "epoch": 0.992, |
| "grad_norm": 18.375, |
| "learning_rate": 6.18e-06, |
| "loss": 3.6273, |
| "mean_token_accuracy": 0.3867473939433694, |
| "num_tokens": 3049436.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 3.6835175109537026, |
| "epoch": 1.0224, |
| "grad_norm": 15.875, |
| "learning_rate": 6.380000000000001e-06, |
| "loss": 3.626, |
| "mean_token_accuracy": 0.3879838499583696, |
| "num_tokens": 3140209.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 3.61849829852581, |
| "epoch": 1.0544, |
| "grad_norm": 15.1875, |
| "learning_rate": 6.5800000000000005e-06, |
| "loss": 3.5842, |
| "mean_token_accuracy": 0.3950715743005276, |
| "num_tokens": 3237743.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 3.6082335516810415, |
| "epoch": 1.0864, |
| "grad_norm": 15.125, |
| "learning_rate": 6.780000000000001e-06, |
| "loss": 3.556, |
| "mean_token_accuracy": 0.3969528621062636, |
| "num_tokens": 3336343.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 3.549923361837864, |
| "epoch": 1.1184, |
| "grad_norm": 18.75, |
| "learning_rate": 6.98e-06, |
| "loss": 3.4731, |
| "mean_token_accuracy": 0.4048408783972263, |
| "num_tokens": 3432312.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 3.5383049845695496, |
| "epoch": 1.1504, |
| "grad_norm": 17.125, |
| "learning_rate": 7.180000000000001e-06, |
| "loss": 3.485, |
| "mean_token_accuracy": 0.4019928781315684, |
| "num_tokens": 3529743.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 3.550269864499569, |
| "epoch": 1.1824, |
| "grad_norm": 18.625, |
| "learning_rate": 7.3800000000000005e-06, |
| "loss": 3.4992, |
| "mean_token_accuracy": 0.40656839609146117, |
| "num_tokens": 3628421.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 3.549488727748394, |
| "epoch": 1.2144, |
| "grad_norm": 15.6875, |
| "learning_rate": 7.58e-06, |
| "loss": 3.5189, |
| "mean_token_accuracy": 0.40171824395656586, |
| "num_tokens": 3731754.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 3.5428646177053453, |
| "epoch": 1.2464, |
| "grad_norm": 16.875, |
| "learning_rate": 7.78e-06, |
| "loss": 3.4652, |
| "mean_token_accuracy": 0.40349574480205774, |
| "num_tokens": 3833669.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 3.412770676612854, |
| "epoch": 1.2784, |
| "grad_norm": 19.375, |
| "learning_rate": 7.980000000000002e-06, |
| "loss": 3.3885, |
| "mean_token_accuracy": 0.4114306215196848, |
| "num_tokens": 3933489.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.2784, |
| "eval_coding_entropy": 1.864494191646576, |
| "eval_coding_loss": 2.1695075035095215, |
| "eval_coding_mean_token_accuracy": 0.5126164103746415, |
| "eval_coding_num_tokens": 3933489.0, |
| "eval_coding_runtime": 79.6045, |
| "eval_coding_samples_per_second": 6.281, |
| "eval_coding_steps_per_second": 3.141, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.2784, |
| "eval_biology_entropy": 3.4962758731842043, |
| "eval_biology_loss": 3.484936475753784, |
| "eval_biology_mean_token_accuracy": 0.4131394236087799, |
| "eval_biology_num_tokens": 3933489.0, |
| "eval_biology_runtime": 34.5802, |
| "eval_biology_samples_per_second": 14.459, |
| "eval_biology_steps_per_second": 7.23, |
| "step": 400 |
| }, |
| { |
| "entropy": 3.447491915524006, |
| "epoch": 1.3104, |
| "grad_norm": 17.375, |
| "learning_rate": 8.18e-06, |
| "loss": 3.397, |
| "mean_token_accuracy": 0.4184817912057042, |
| "num_tokens": 4028501.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 3.491288626194, |
| "epoch": 1.3424, |
| "grad_norm": 22.0, |
| "learning_rate": 8.380000000000001e-06, |
| "loss": 3.4228, |
| "mean_token_accuracy": 0.4082869263365865, |
| "num_tokens": 4128412.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 3.4260040298104286, |
| "epoch": 1.3744, |
| "grad_norm": 18.25, |
| "learning_rate": 8.580000000000001e-06, |
| "loss": 3.3652, |
| "mean_token_accuracy": 0.417889959923923, |
| "num_tokens": 4228703.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 3.4077636525034904, |
| "epoch": 1.4064, |
| "grad_norm": 20.375, |
| "learning_rate": 8.78e-06, |
| "loss": 3.3644, |
| "mean_token_accuracy": 0.41263102628290654, |
| "num_tokens": 4329623.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 3.4000454649329184, |
| "epoch": 1.4384000000000001, |
| "grad_norm": 17.25, |
| "learning_rate": 8.98e-06, |
| "loss": 3.3629, |
| "mean_token_accuracy": 0.41560379061847924, |
| "num_tokens": 4424877.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 3.473749025166035, |
| "epoch": 1.4704, |
| "grad_norm": 18.0, |
| "learning_rate": 9.180000000000002e-06, |
| "loss": 3.3787, |
| "mean_token_accuracy": 0.4140065621584654, |
| "num_tokens": 4518810.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 3.418153963983059, |
| "epoch": 1.5024, |
| "grad_norm": 21.875, |
| "learning_rate": 9.38e-06, |
| "loss": 3.3539, |
| "mean_token_accuracy": 0.41980997528880837, |
| "num_tokens": 4617323.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 3.410584717988968, |
| "epoch": 1.5344, |
| "grad_norm": 20.875, |
| "learning_rate": 9.58e-06, |
| "loss": 3.3493, |
| "mean_token_accuracy": 0.4202044107019901, |
| "num_tokens": 4719316.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 3.3172555193305016, |
| "epoch": 1.5664, |
| "grad_norm": 23.75, |
| "learning_rate": 9.780000000000001e-06, |
| "loss": 3.2781, |
| "mean_token_accuracy": 0.42654727064073084, |
| "num_tokens": 4818127.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 3.3434231370687484, |
| "epoch": 1.5984, |
| "grad_norm": 17.75, |
| "learning_rate": 9.980000000000001e-06, |
| "loss": 3.2655, |
| "mean_token_accuracy": 0.42654075026512145, |
| "num_tokens": 4915672.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.5984, |
| "eval_coding_entropy": 1.9044507808685303, |
| "eval_coding_loss": 2.163560152053833, |
| "eval_coding_mean_token_accuracy": 0.5137092983722686, |
| "eval_coding_num_tokens": 4915672.0, |
| "eval_coding_runtime": 78.8714, |
| "eval_coding_samples_per_second": 6.339, |
| "eval_coding_steps_per_second": 3.17, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.5984, |
| "eval_biology_entropy": 3.2849253273010253, |
| "eval_biology_loss": 3.3220765590667725, |
| "eval_biology_mean_token_accuracy": 0.4304432601928711, |
| "eval_biology_num_tokens": 4915672.0, |
| "eval_biology_runtime": 34.5563, |
| "eval_biology_samples_per_second": 14.469, |
| "eval_biology_steps_per_second": 7.235, |
| "step": 500 |
| }, |
| { |
| "entropy": 3.276486141234636, |
| "epoch": 1.6303999999999998, |
| "grad_norm": 19.5, |
| "learning_rate": 1.018e-05, |
| "loss": 3.2382, |
| "mean_token_accuracy": 0.42907340787351134, |
| "num_tokens": 5015381.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 3.3447541266679766, |
| "epoch": 1.6623999999999999, |
| "grad_norm": 23.25, |
| "learning_rate": 1.038e-05, |
| "loss": 3.2677, |
| "mean_token_accuracy": 0.42632252164185047, |
| "num_tokens": 5117126.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 3.3336540818214417, |
| "epoch": 1.6944, |
| "grad_norm": 20.75, |
| "learning_rate": 1.0580000000000002e-05, |
| "loss": 3.299, |
| "mean_token_accuracy": 0.4259482711553574, |
| "num_tokens": 5213073.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 3.319507985562086, |
| "epoch": 1.7264, |
| "grad_norm": 20.0, |
| "learning_rate": 1.0780000000000002e-05, |
| "loss": 3.2425, |
| "mean_token_accuracy": 0.4222921848297119, |
| "num_tokens": 5308786.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 3.249939414858818, |
| "epoch": 1.7584, |
| "grad_norm": 23.625, |
| "learning_rate": 1.0980000000000002e-05, |
| "loss": 3.2109, |
| "mean_token_accuracy": 0.4321354482322931, |
| "num_tokens": 5405141.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 3.2699292853474615, |
| "epoch": 1.7904, |
| "grad_norm": 24.25, |
| "learning_rate": 1.1180000000000001e-05, |
| "loss": 3.2126, |
| "mean_token_accuracy": 0.4329648604616523, |
| "num_tokens": 5505162.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 3.2213672384619714, |
| "epoch": 1.8224, |
| "grad_norm": 19.375, |
| "learning_rate": 1.138e-05, |
| "loss": 3.1498, |
| "mean_token_accuracy": 0.4372351700440049, |
| "num_tokens": 5604409.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 3.158064991235733, |
| "epoch": 1.8544, |
| "grad_norm": 25.625, |
| "learning_rate": 1.1580000000000001e-05, |
| "loss": 3.0863, |
| "mean_token_accuracy": 0.44516869317740204, |
| "num_tokens": 5705344.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 3.2133444771170616, |
| "epoch": 1.8864, |
| "grad_norm": 21.125, |
| "learning_rate": 1.178e-05, |
| "loss": 3.1518, |
| "mean_token_accuracy": 0.4377427864819765, |
| "num_tokens": 5794262.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 3.2123986557126045, |
| "epoch": 1.9184, |
| "grad_norm": 21.0, |
| "learning_rate": 1.198e-05, |
| "loss": 3.1537, |
| "mean_token_accuracy": 0.439489153586328, |
| "num_tokens": 5895496.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.9184, |
| "eval_coding_entropy": 1.892683578968048, |
| "eval_coding_loss": 2.1593310832977295, |
| "eval_coding_mean_token_accuracy": 0.514176591873169, |
| "eval_coding_num_tokens": 5895496.0, |
| "eval_coding_runtime": 78.9615, |
| "eval_coding_samples_per_second": 6.332, |
| "eval_coding_steps_per_second": 3.166, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.9184, |
| "eval_biology_entropy": 3.14973046875, |
| "eval_biology_loss": 3.1956348419189453, |
| "eval_biology_mean_token_accuracy": 0.4422305516004562, |
| "eval_biology_num_tokens": 5895496.0, |
| "eval_biology_runtime": 34.5624, |
| "eval_biology_samples_per_second": 14.467, |
| "eval_biology_steps_per_second": 7.233, |
| "step": 600 |
| }, |
| { |
| "entropy": 3.2023044779896734, |
| "epoch": 1.9504000000000001, |
| "grad_norm": 20.375, |
| "learning_rate": 1.218e-05, |
| "loss": 3.145, |
| "mean_token_accuracy": 0.4402220517396927, |
| "num_tokens": 5992443.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 3.173665773868561, |
| "epoch": 1.9824000000000002, |
| "grad_norm": 19.375, |
| "learning_rate": 1.2380000000000002e-05, |
| "loss": 3.1235, |
| "mean_token_accuracy": 0.43916483521461486, |
| "num_tokens": 6092718.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 3.130789705012974, |
| "epoch": 2.0128, |
| "grad_norm": 21.5, |
| "learning_rate": 1.2580000000000002e-05, |
| "loss": 3.0485, |
| "mean_token_accuracy": 0.4457206469225256, |
| "num_tokens": 6184982.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 3.024783915281296, |
| "epoch": 2.0448, |
| "grad_norm": 19.875, |
| "learning_rate": 1.2780000000000001e-05, |
| "loss": 2.992, |
| "mean_token_accuracy": 0.45380019545555117, |
| "num_tokens": 6286901.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 3.1106227308511736, |
| "epoch": 2.0768, |
| "grad_norm": 20.25, |
| "learning_rate": 1.2980000000000001e-05, |
| "loss": 3.0474, |
| "mean_token_accuracy": 0.4507150810211897, |
| "num_tokens": 6381706.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 3.070780225098133, |
| "epoch": 2.1088, |
| "grad_norm": 22.125, |
| "learning_rate": 1.3180000000000001e-05, |
| "loss": 3.014, |
| "mean_token_accuracy": 0.45487339831888673, |
| "num_tokens": 6478891.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 3.1098087787628175, |
| "epoch": 2.1408, |
| "grad_norm": 25.75, |
| "learning_rate": 1.3380000000000002e-05, |
| "loss": 3.0755, |
| "mean_token_accuracy": 0.44592468421906234, |
| "num_tokens": 6576566.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 3.086433382332325, |
| "epoch": 2.1728, |
| "grad_norm": 18.5, |
| "learning_rate": 1.3580000000000002e-05, |
| "loss": 3.0116, |
| "mean_token_accuracy": 0.4536932479590178, |
| "num_tokens": 6678216.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 3.0678569465875625, |
| "epoch": 2.2048, |
| "grad_norm": 19.5, |
| "learning_rate": 1.378e-05, |
| "loss": 2.9845, |
| "mean_token_accuracy": 0.4514392616227269, |
| "num_tokens": 6771313.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 3.0470827482640743, |
| "epoch": 2.2368, |
| "grad_norm": 23.0, |
| "learning_rate": 1.398e-05, |
| "loss": 2.9636, |
| "mean_token_accuracy": 0.4556872248649597, |
| "num_tokens": 6872028.0, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.2368, |
| "eval_coding_entropy": 1.8572090373039245, |
| "eval_coding_loss": 2.177124500274658, |
| "eval_coding_mean_token_accuracy": 0.513177493929863, |
| "eval_coding_num_tokens": 6872028.0, |
| "eval_coding_runtime": 78.9661, |
| "eval_coding_samples_per_second": 6.332, |
| "eval_coding_steps_per_second": 3.166, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.2368, |
| "eval_biology_entropy": 2.983607707500458, |
| "eval_biology_loss": 3.0935444831848145, |
| "eval_biology_mean_token_accuracy": 0.4546413152217865, |
| "eval_biology_num_tokens": 6872028.0, |
| "eval_biology_runtime": 34.9245, |
| "eval_biology_samples_per_second": 14.317, |
| "eval_biology_steps_per_second": 7.158, |
| "step": 700 |
| }, |
| { |
| "entropy": 3.0512571170926095, |
| "epoch": 2.2688, |
| "grad_norm": 18.75, |
| "learning_rate": 1.418e-05, |
| "loss": 2.9764, |
| "mean_token_accuracy": 0.45544591508805754, |
| "num_tokens": 6971802.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 3.079524612426758, |
| "epoch": 2.3008, |
| "grad_norm": 22.0, |
| "learning_rate": 1.4380000000000001e-05, |
| "loss": 3.0142, |
| "mean_token_accuracy": 0.45264575611799956, |
| "num_tokens": 7067482.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 3.0822106420993807, |
| "epoch": 2.3327999999999998, |
| "grad_norm": 25.75, |
| "learning_rate": 1.4580000000000001e-05, |
| "loss": 3.0382, |
| "mean_token_accuracy": 0.45174078848212956, |
| "num_tokens": 7164629.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 3.01283405572176, |
| "epoch": 2.3648, |
| "grad_norm": 20.25, |
| "learning_rate": 1.478e-05, |
| "loss": 2.9356, |
| "mean_token_accuracy": 0.46082064919173715, |
| "num_tokens": 7260811.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 3.0809263423085214, |
| "epoch": 2.3968, |
| "grad_norm": 18.5, |
| "learning_rate": 1.498e-05, |
| "loss": 2.9877, |
| "mean_token_accuracy": 0.45245055090636016, |
| "num_tokens": 7353504.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 3.0187013909220695, |
| "epoch": 2.4288, |
| "grad_norm": 22.75, |
| "learning_rate": 1.5180000000000002e-05, |
| "loss": 2.9599, |
| "mean_token_accuracy": 0.4591849962249398, |
| "num_tokens": 7450239.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 2.969215099513531, |
| "epoch": 2.4608, |
| "grad_norm": 22.875, |
| "learning_rate": 1.5380000000000002e-05, |
| "loss": 2.8833, |
| "mean_token_accuracy": 0.4634178843349218, |
| "num_tokens": 7549458.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 2.9438143245875836, |
| "epoch": 2.4928, |
| "grad_norm": 18.5, |
| "learning_rate": 1.5580000000000003e-05, |
| "loss": 2.9044, |
| "mean_token_accuracy": 0.4644233400002122, |
| "num_tokens": 7646646.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 2.9995977893471717, |
| "epoch": 2.5248, |
| "grad_norm": 19.875, |
| "learning_rate": 1.578e-05, |
| "loss": 2.9524, |
| "mean_token_accuracy": 0.46077679973095653, |
| "num_tokens": 7743967.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 2.9542898386716843, |
| "epoch": 2.5568, |
| "grad_norm": 21.75, |
| "learning_rate": 1.5980000000000003e-05, |
| "loss": 2.8718, |
| "mean_token_accuracy": 0.4678089875727892, |
| "num_tokens": 7844679.0, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.5568, |
| "eval_coding_entropy": 1.8718931813240052, |
| "eval_coding_loss": 2.1724154949188232, |
| "eval_coding_mean_token_accuracy": 0.5149018170833588, |
| "eval_coding_num_tokens": 7844679.0, |
| "eval_coding_runtime": 78.9496, |
| "eval_coding_samples_per_second": 6.333, |
| "eval_coding_steps_per_second": 3.167, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.5568, |
| "eval_biology_entropy": 2.85398051404953, |
| "eval_biology_loss": 2.9827656745910645, |
| "eval_biology_mean_token_accuracy": 0.465399906039238, |
| "eval_biology_num_tokens": 7844679.0, |
| "eval_biology_runtime": 34.5946, |
| "eval_biology_samples_per_second": 14.453, |
| "eval_biology_steps_per_second": 7.227, |
| "step": 800 |
| }, |
| { |
| "entropy": 2.905091355741024, |
| "epoch": 2.5888, |
| "grad_norm": 20.0, |
| "learning_rate": 1.618e-05, |
| "loss": 2.8797, |
| "mean_token_accuracy": 0.47355432007461784, |
| "num_tokens": 7944653.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 2.941771225631237, |
| "epoch": 2.6208, |
| "grad_norm": 20.625, |
| "learning_rate": 1.638e-05, |
| "loss": 2.857, |
| "mean_token_accuracy": 0.4673197165131569, |
| "num_tokens": 8043783.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 2.992217946052551, |
| "epoch": 2.6528, |
| "grad_norm": 22.25, |
| "learning_rate": 1.658e-05, |
| "loss": 2.9048, |
| "mean_token_accuracy": 0.46489576790481807, |
| "num_tokens": 8142126.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 2.9804361656308176, |
| "epoch": 2.6848, |
| "grad_norm": 18.75, |
| "learning_rate": 1.6780000000000002e-05, |
| "loss": 2.9216, |
| "mean_token_accuracy": 0.462474500015378, |
| "num_tokens": 8235244.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 2.933942112326622, |
| "epoch": 2.7168, |
| "grad_norm": 18.125, |
| "learning_rate": 1.698e-05, |
| "loss": 2.8811, |
| "mean_token_accuracy": 0.4669422248378396, |
| "num_tokens": 8336300.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 2.826403010636568, |
| "epoch": 2.7488, |
| "grad_norm": 20.375, |
| "learning_rate": 1.718e-05, |
| "loss": 2.7594, |
| "mean_token_accuracy": 0.47525308579206466, |
| "num_tokens": 8440110.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 2.8751538641750813, |
| "epoch": 2.7808, |
| "grad_norm": 22.125, |
| "learning_rate": 1.7380000000000003e-05, |
| "loss": 2.8221, |
| "mean_token_accuracy": 0.4738822041079402, |
| "num_tokens": 8533528.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 2.885574182868004, |
| "epoch": 2.8128, |
| "grad_norm": 22.375, |
| "learning_rate": 1.758e-05, |
| "loss": 2.8067, |
| "mean_token_accuracy": 0.47314695566892623, |
| "num_tokens": 8632702.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 2.9287849619984625, |
| "epoch": 2.8448, |
| "grad_norm": 20.25, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 2.8323, |
| "mean_token_accuracy": 0.4664028905332088, |
| "num_tokens": 8729727.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 2.885370475798845, |
| "epoch": 2.8768000000000002, |
| "grad_norm": 20.75, |
| "learning_rate": 1.798e-05, |
| "loss": 2.8442, |
| "mean_token_accuracy": 0.46528117191046475, |
| "num_tokens": 8831917.0, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.8768000000000002, |
| "eval_coding_entropy": 1.8902995553016662, |
| "eval_coding_loss": 2.1872222423553467, |
| "eval_coding_mean_token_accuracy": 0.5130699964761734, |
| "eval_coding_num_tokens": 8831917.0, |
| "eval_coding_runtime": 78.9145, |
| "eval_coding_samples_per_second": 6.336, |
| "eval_coding_steps_per_second": 3.168, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.8768000000000002, |
| "eval_biology_entropy": 2.868830183506012, |
| "eval_biology_loss": 2.891908884048462, |
| "eval_biology_mean_token_accuracy": 0.4745277794599533, |
| "eval_biology_num_tokens": 8831917.0, |
| "eval_biology_runtime": 34.562, |
| "eval_biology_samples_per_second": 14.467, |
| "eval_biology_steps_per_second": 7.233, |
| "step": 900 |
| }, |
| { |
| "entropy": 2.884679964929819, |
| "epoch": 2.9088000000000003, |
| "grad_norm": 22.0, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 2.7851, |
| "mean_token_accuracy": 0.4735592044889927, |
| "num_tokens": 8931811.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 2.8508158676326274, |
| "epoch": 2.9408, |
| "grad_norm": 23.25, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 2.7834, |
| "mean_token_accuracy": 0.47639975920319555, |
| "num_tokens": 9033255.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 2.8332293681800365, |
| "epoch": 2.9728, |
| "grad_norm": 19.875, |
| "learning_rate": 1.858e-05, |
| "loss": 2.7458, |
| "mean_token_accuracy": 0.47748742904514074, |
| "num_tokens": 9129675.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 2.8580648820651207, |
| "epoch": 3.0032, |
| "grad_norm": 20.375, |
| "learning_rate": 1.878e-05, |
| "loss": 2.7791, |
| "mean_token_accuracy": 0.4723091986226408, |
| "num_tokens": 9224886.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 2.7373731434345245, |
| "epoch": 3.0352, |
| "grad_norm": 18.625, |
| "learning_rate": 1.898e-05, |
| "loss": 2.6356, |
| "mean_token_accuracy": 0.4919461887329817, |
| "num_tokens": 9325062.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 2.727712345123291, |
| "epoch": 3.0672, |
| "grad_norm": 20.75, |
| "learning_rate": 1.918e-05, |
| "loss": 2.6412, |
| "mean_token_accuracy": 0.4901068951934576, |
| "num_tokens": 9424019.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 2.707367330044508, |
| "epoch": 3.0992, |
| "grad_norm": 20.625, |
| "learning_rate": 1.938e-05, |
| "loss": 2.6608, |
| "mean_token_accuracy": 0.4915515352040529, |
| "num_tokens": 9526246.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 2.718513052165508, |
| "epoch": 3.1312, |
| "grad_norm": 18.25, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 2.6232, |
| "mean_token_accuracy": 0.48776335902512075, |
| "num_tokens": 9624382.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 2.7374294199049474, |
| "epoch": 3.1632, |
| "grad_norm": 21.375, |
| "learning_rate": 1.978e-05, |
| "loss": 2.6429, |
| "mean_token_accuracy": 0.48988417275249957, |
| "num_tokens": 9719411.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 2.7222474955022333, |
| "epoch": 3.1952, |
| "grad_norm": 20.375, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 2.6263, |
| "mean_token_accuracy": 0.4935696581378579, |
| "num_tokens": 9819220.0, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.1952, |
| "eval_coding_entropy": 1.7593334603309632, |
| "eval_coding_loss": 2.205096960067749, |
| "eval_coding_mean_token_accuracy": 0.5136193619966507, |
| "eval_coding_num_tokens": 9819220.0, |
| "eval_coding_runtime": 78.8943, |
| "eval_coding_samples_per_second": 6.338, |
| "eval_coding_steps_per_second": 3.169, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.1952, |
| "eval_biology_entropy": 2.6804689655303955, |
| "eval_biology_loss": 2.818235397338867, |
| "eval_biology_mean_token_accuracy": 0.48389568710327147, |
| "eval_biology_num_tokens": 9819220.0, |
| "eval_biology_runtime": 34.6726, |
| "eval_biology_samples_per_second": 14.421, |
| "eval_biology_steps_per_second": 7.21, |
| "step": 1000 |
| }, |
| { |
| "entropy": 2.6821524411439897, |
| "epoch": 3.2272, |
| "grad_norm": 25.0, |
| "learning_rate": 1.9980000000000002e-05, |
| "loss": 2.5881, |
| "mean_token_accuracy": 0.4957658048719168, |
| "num_tokens": 9917642.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 2.643439619243145, |
| "epoch": 3.2592, |
| "grad_norm": 17.25, |
| "learning_rate": 1.995777777777778e-05, |
| "loss": 2.5861, |
| "mean_token_accuracy": 0.4964464308694005, |
| "num_tokens": 10016171.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 2.69493540674448, |
| "epoch": 3.2912, |
| "grad_norm": 21.25, |
| "learning_rate": 1.9935555555555557e-05, |
| "loss": 2.6244, |
| "mean_token_accuracy": 0.49275039993226527, |
| "num_tokens": 10113317.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 2.6980527356266975, |
| "epoch": 3.3232, |
| "grad_norm": 21.0, |
| "learning_rate": 1.9913333333333335e-05, |
| "loss": 2.6358, |
| "mean_token_accuracy": 0.4931097852066159, |
| "num_tokens": 10209948.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 2.704365950077772, |
| "epoch": 3.3552, |
| "grad_norm": 17.375, |
| "learning_rate": 1.9891111111111112e-05, |
| "loss": 2.6278, |
| "mean_token_accuracy": 0.4901678072288632, |
| "num_tokens": 10304993.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 2.6926429100334643, |
| "epoch": 3.3872, |
| "grad_norm": 23.375, |
| "learning_rate": 1.986888888888889e-05, |
| "loss": 2.6281, |
| "mean_token_accuracy": 0.49370026849210263, |
| "num_tokens": 10403340.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 2.7046832494437694, |
| "epoch": 3.4192, |
| "grad_norm": 23.625, |
| "learning_rate": 1.9846666666666668e-05, |
| "loss": 2.6136, |
| "mean_token_accuracy": 0.49537604544311764, |
| "num_tokens": 10499289.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 2.6765974409878255, |
| "epoch": 3.4512, |
| "grad_norm": 22.75, |
| "learning_rate": 1.9824444444444445e-05, |
| "loss": 2.6201, |
| "mean_token_accuracy": 0.49774726927280427, |
| "num_tokens": 10595192.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 2.661435030400753, |
| "epoch": 3.4832, |
| "grad_norm": 20.375, |
| "learning_rate": 1.9802222222222226e-05, |
| "loss": 2.6017, |
| "mean_token_accuracy": 0.49518649335950615, |
| "num_tokens": 10692762.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 2.6369900465011598, |
| "epoch": 3.5152, |
| "grad_norm": 20.25, |
| "learning_rate": 1.978e-05, |
| "loss": 2.5617, |
| "mean_token_accuracy": 0.4994138855487108, |
| "num_tokens": 10795865.0, |
| "step": 1100 |
| }, |
| { |
| "epoch": 3.5152, |
| "eval_coding_entropy": 1.8151787004470825, |
| "eval_coding_loss": 2.212179183959961, |
| "eval_coding_mean_token_accuracy": 0.5126535563468934, |
| "eval_coding_num_tokens": 10795865.0, |
| "eval_coding_runtime": 79.0002, |
| "eval_coding_samples_per_second": 6.329, |
| "eval_coding_steps_per_second": 3.165, |
| "step": 1100 |
| }, |
| { |
| "epoch": 3.5152, |
| "eval_biology_entropy": 2.6696902551651003, |
| "eval_biology_loss": 2.7315852642059326, |
| "eval_biology_mean_token_accuracy": 0.4940973354578018, |
| "eval_biology_num_tokens": 10795865.0, |
| "eval_biology_runtime": 34.6872, |
| "eval_biology_samples_per_second": 14.415, |
| "eval_biology_steps_per_second": 7.207, |
| "step": 1100 |
| }, |
| { |
| "entropy": 2.64646837413311, |
| "epoch": 3.5472, |
| "grad_norm": 19.75, |
| "learning_rate": 1.975777777777778e-05, |
| "loss": 2.5694, |
| "mean_token_accuracy": 0.5007585499435663, |
| "num_tokens": 10896150.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 2.5886894077062608, |
| "epoch": 3.5792, |
| "grad_norm": 20.125, |
| "learning_rate": 1.9735555555555556e-05, |
| "loss": 2.5421, |
| "mean_token_accuracy": 0.503319687396288, |
| "num_tokens": 10994009.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 2.664226660132408, |
| "epoch": 3.6112, |
| "grad_norm": 21.0, |
| "learning_rate": 1.9713333333333337e-05, |
| "loss": 2.5748, |
| "mean_token_accuracy": 0.4976232260465622, |
| "num_tokens": 11088408.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 2.591143436729908, |
| "epoch": 3.6432, |
| "grad_norm": 19.0, |
| "learning_rate": 1.969111111111111e-05, |
| "loss": 2.5241, |
| "mean_token_accuracy": 0.5072892276570201, |
| "num_tokens": 11189792.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 2.607444966584444, |
| "epoch": 3.6752000000000002, |
| "grad_norm": 18.375, |
| "learning_rate": 1.9668888888888892e-05, |
| "loss": 2.5651, |
| "mean_token_accuracy": 0.5035420592874289, |
| "num_tokens": 11287885.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 2.5825724259018896, |
| "epoch": 3.7072000000000003, |
| "grad_norm": 18.875, |
| "learning_rate": 1.9646666666666666e-05, |
| "loss": 2.519, |
| "mean_token_accuracy": 0.5080195646733046, |
| "num_tokens": 11386948.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 2.638581147044897, |
| "epoch": 3.7392, |
| "grad_norm": 15.8125, |
| "learning_rate": 1.9624444444444447e-05, |
| "loss": 2.5488, |
| "mean_token_accuracy": 0.4984665771946311, |
| "num_tokens": 11486738.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 2.623228520154953, |
| "epoch": 3.7712, |
| "grad_norm": 18.5, |
| "learning_rate": 1.9602222222222225e-05, |
| "loss": 2.5644, |
| "mean_token_accuracy": 0.4989262154325843, |
| "num_tokens": 11587111.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 2.595949812233448, |
| "epoch": 3.8032, |
| "grad_norm": 20.25, |
| "learning_rate": 1.9580000000000002e-05, |
| "loss": 2.5348, |
| "mean_token_accuracy": 0.5062878143042326, |
| "num_tokens": 11683518.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 2.552212902158499, |
| "epoch": 3.8352, |
| "grad_norm": 18.5, |
| "learning_rate": 1.955777777777778e-05, |
| "loss": 2.4693, |
| "mean_token_accuracy": 0.5114001492038369, |
| "num_tokens": 11787112.0, |
| "step": 1200 |
| }, |
| { |
| "epoch": 3.8352, |
| "eval_coding_entropy": 1.821248462677002, |
| "eval_coding_loss": 2.2172763347625732, |
| "eval_coding_mean_token_accuracy": 0.5105675404071808, |
| "eval_coding_num_tokens": 11787112.0, |
| "eval_coding_runtime": 79.0098, |
| "eval_coding_samples_per_second": 6.328, |
| "eval_coding_steps_per_second": 3.164, |
| "step": 1200 |
| }, |
| { |
| "epoch": 3.8352, |
| "eval_biology_entropy": 2.618246515274048, |
| "eval_biology_loss": 2.6723146438598633, |
| "eval_biology_mean_token_accuracy": 0.499475853562355, |
| "eval_biology_num_tokens": 11787112.0, |
| "eval_biology_runtime": 34.6592, |
| "eval_biology_samples_per_second": 14.426, |
| "eval_biology_steps_per_second": 7.213, |
| "step": 1200 |
| }, |
| { |
| "entropy": 2.606553990393877, |
| "epoch": 3.8672, |
| "grad_norm": 19.5, |
| "learning_rate": 1.9535555555555557e-05, |
| "loss": 2.5365, |
| "mean_token_accuracy": 0.5034115230664611, |
| "num_tokens": 11882695.0, |
| "step": 1210 |
| }, |
| { |
| "entropy": 2.623321808129549, |
| "epoch": 3.8992, |
| "grad_norm": 17.5, |
| "learning_rate": 1.9513333333333335e-05, |
| "loss": 2.5545, |
| "mean_token_accuracy": 0.5018050175160169, |
| "num_tokens": 11983604.0, |
| "step": 1220 |
| }, |
| { |
| "entropy": 2.588004069030285, |
| "epoch": 3.9312, |
| "grad_norm": 17.375, |
| "learning_rate": 1.9491111111111113e-05, |
| "loss": 2.5174, |
| "mean_token_accuracy": 0.5051012465730309, |
| "num_tokens": 12076708.0, |
| "step": 1230 |
| }, |
| { |
| "entropy": 2.6090625591576098, |
| "epoch": 3.9632, |
| "grad_norm": 16.875, |
| "learning_rate": 1.946888888888889e-05, |
| "loss": 2.5337, |
| "mean_token_accuracy": 0.5051190486177802, |
| "num_tokens": 12176747.0, |
| "step": 1240 |
| }, |
| { |
| "entropy": 2.584036312997341, |
| "epoch": 3.9952, |
| "grad_norm": 19.75, |
| "learning_rate": 1.9446666666666668e-05, |
| "loss": 2.4902, |
| "mean_token_accuracy": 0.5087237313389779, |
| "num_tokens": 12272241.0, |
| "step": 1250 |
| }, |
| { |
| "entropy": 2.472858669726472, |
| "epoch": 4.0256, |
| "grad_norm": 18.125, |
| "learning_rate": 1.9424444444444446e-05, |
| "loss": 2.3856, |
| "mean_token_accuracy": 0.521043097110171, |
| "num_tokens": 12359739.0, |
| "step": 1260 |
| }, |
| { |
| "entropy": 2.34643474444747, |
| "epoch": 4.0576, |
| "grad_norm": 21.0, |
| "learning_rate": 1.9402222222222223e-05, |
| "loss": 2.2626, |
| "mean_token_accuracy": 0.5364229042083025, |
| "num_tokens": 12460555.0, |
| "step": 1270 |
| }, |
| { |
| "entropy": 2.4080297112464906, |
| "epoch": 4.0896, |
| "grad_norm": 17.25, |
| "learning_rate": 1.938e-05, |
| "loss": 2.355, |
| "mean_token_accuracy": 0.5255401773378253, |
| "num_tokens": 12554461.0, |
| "step": 1280 |
| }, |
| { |
| "entropy": 2.394774776697159, |
| "epoch": 4.1216, |
| "grad_norm": 16.5, |
| "learning_rate": 1.935777777777778e-05, |
| "loss": 2.3188, |
| "mean_token_accuracy": 0.5288082377985119, |
| "num_tokens": 12653465.0, |
| "step": 1290 |
| }, |
| { |
| "entropy": 2.391275504231453, |
| "epoch": 4.1536, |
| "grad_norm": 21.625, |
| "learning_rate": 1.9335555555555556e-05, |
| "loss": 2.2848, |
| "mean_token_accuracy": 0.5329740956425667, |
| "num_tokens": 12748340.0, |
| "step": 1300 |
| }, |
| { |
| "epoch": 4.1536, |
| "eval_coding_entropy": 1.6686400065422058, |
| "eval_coding_loss": 2.2645514011383057, |
| "eval_coding_mean_token_accuracy": 0.5102254449129104, |
| "eval_coding_num_tokens": 12748340.0, |
| "eval_coding_runtime": 79.0388, |
| "eval_coding_samples_per_second": 6.326, |
| "eval_coding_steps_per_second": 3.163, |
| "step": 1300 |
| }, |
| { |
| "epoch": 4.1536, |
| "eval_biology_entropy": 2.3491821846961973, |
| "eval_biology_loss": 2.634176015853882, |
| "eval_biology_mean_token_accuracy": 0.5061246186494828, |
| "eval_biology_num_tokens": 12748340.0, |
| "eval_biology_runtime": 34.6148, |
| "eval_biology_samples_per_second": 14.445, |
| "eval_biology_steps_per_second": 7.222, |
| "step": 1300 |
| }, |
| { |
| "entropy": 2.3668478064239027, |
| "epoch": 4.1856, |
| "grad_norm": 20.125, |
| "learning_rate": 1.9313333333333334e-05, |
| "loss": 2.2853, |
| "mean_token_accuracy": 0.5337071912363172, |
| "num_tokens": 12847096.0, |
| "step": 1310 |
| }, |
| { |
| "entropy": 2.311310113221407, |
| "epoch": 4.2176, |
| "grad_norm": 15.0625, |
| "learning_rate": 1.9291111111111115e-05, |
| "loss": 2.2384, |
| "mean_token_accuracy": 0.5418357422575355, |
| "num_tokens": 12946388.0, |
| "step": 1320 |
| }, |
| { |
| "entropy": 2.352651536464691, |
| "epoch": 4.2496, |
| "grad_norm": 16.375, |
| "learning_rate": 1.926888888888889e-05, |
| "loss": 2.2586, |
| "mean_token_accuracy": 0.5339978907257319, |
| "num_tokens": 13042597.0, |
| "step": 1330 |
| }, |
| { |
| "entropy": 2.3916583247482777, |
| "epoch": 4.2816, |
| "grad_norm": 20.875, |
| "learning_rate": 1.924666666666667e-05, |
| "loss": 2.3214, |
| "mean_token_accuracy": 0.5277318973094225, |
| "num_tokens": 13137019.0, |
| "step": 1340 |
| }, |
| { |
| "entropy": 2.407442457228899, |
| "epoch": 4.3136, |
| "grad_norm": 22.375, |
| "learning_rate": 1.9224444444444444e-05, |
| "loss": 2.3233, |
| "mean_token_accuracy": 0.529756323620677, |
| "num_tokens": 13234780.0, |
| "step": 1350 |
| }, |
| { |
| "entropy": 2.370780614763498, |
| "epoch": 4.3456, |
| "grad_norm": 16.75, |
| "learning_rate": 1.9202222222222225e-05, |
| "loss": 2.2856, |
| "mean_token_accuracy": 0.5306474581360817, |
| "num_tokens": 13334469.0, |
| "step": 1360 |
| }, |
| { |
| "entropy": 2.375711131095886, |
| "epoch": 4.3776, |
| "grad_norm": 17.125, |
| "learning_rate": 1.918e-05, |
| "loss": 2.3092, |
| "mean_token_accuracy": 0.5324419310316444, |
| "num_tokens": 13432836.0, |
| "step": 1370 |
| }, |
| { |
| "entropy": 2.4337742544710634, |
| "epoch": 4.4096, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.915777777777778e-05, |
| "loss": 2.3333, |
| "mean_token_accuracy": 0.5236062532290816, |
| "num_tokens": 13530508.0, |
| "step": 1380 |
| }, |
| { |
| "entropy": 2.330331479758024, |
| "epoch": 4.4416, |
| "grad_norm": 18.125, |
| "learning_rate": 1.9135555555555555e-05, |
| "loss": 2.2633, |
| "mean_token_accuracy": 0.5389384172856808, |
| "num_tokens": 13633161.0, |
| "step": 1390 |
| }, |
| { |
| "entropy": 2.3866625733673574, |
| "epoch": 4.4736, |
| "grad_norm": 15.375, |
| "learning_rate": 1.9113333333333336e-05, |
| "loss": 2.2892, |
| "mean_token_accuracy": 0.5333944335579872, |
| "num_tokens": 13731330.0, |
| "step": 1400 |
| }, |
| { |
| "epoch": 4.4736, |
| "eval_coding_entropy": 1.7261567716598512, |
| "eval_coding_loss": 2.27376389503479, |
| "eval_coding_mean_token_accuracy": 0.5079633537530899, |
| "eval_coding_num_tokens": 13731330.0, |
| "eval_coding_runtime": 78.9748, |
| "eval_coding_samples_per_second": 6.331, |
| "eval_coding_steps_per_second": 3.166, |
| "step": 1400 |
| }, |
| { |
| "epoch": 4.4736, |
| "eval_biology_entropy": 2.3869504251480103, |
| "eval_biology_loss": 2.5860044956207275, |
| "eval_biology_mean_token_accuracy": 0.5117661204338074, |
| "eval_biology_num_tokens": 13731330.0, |
| "eval_biology_runtime": 34.6224, |
| "eval_biology_samples_per_second": 14.442, |
| "eval_biology_steps_per_second": 7.221, |
| "step": 1400 |
| }, |
| { |
| "entropy": 2.3977055683732034, |
| "epoch": 4.5056, |
| "grad_norm": 17.0, |
| "learning_rate": 1.9091111111111113e-05, |
| "loss": 2.3212, |
| "mean_token_accuracy": 0.526662690937519, |
| "num_tokens": 13822911.0, |
| "step": 1410 |
| }, |
| { |
| "entropy": 2.313427411019802, |
| "epoch": 4.5376, |
| "grad_norm": 17.375, |
| "learning_rate": 1.906888888888889e-05, |
| "loss": 2.2541, |
| "mean_token_accuracy": 0.5411791836842894, |
| "num_tokens": 13926058.0, |
| "step": 1420 |
| }, |
| { |
| "entropy": 2.2551471143960953, |
| "epoch": 4.5696, |
| "grad_norm": 17.0, |
| "learning_rate": 1.904666666666667e-05, |
| "loss": 2.1854, |
| "mean_token_accuracy": 0.5493083782494068, |
| "num_tokens": 14035552.0, |
| "step": 1430 |
| }, |
| { |
| "entropy": 2.3578838713467123, |
| "epoch": 4.6016, |
| "grad_norm": 16.625, |
| "learning_rate": 1.9024444444444446e-05, |
| "loss": 2.2836, |
| "mean_token_accuracy": 0.5363319680094719, |
| "num_tokens": 14134769.0, |
| "step": 1440 |
| }, |
| { |
| "entropy": 2.3566885352134705, |
| "epoch": 4.6336, |
| "grad_norm": 21.625, |
| "learning_rate": 1.9002222222222224e-05, |
| "loss": 2.2625, |
| "mean_token_accuracy": 0.5347570797428489, |
| "num_tokens": 14232004.0, |
| "step": 1450 |
| }, |
| { |
| "entropy": 2.3043198555707933, |
| "epoch": 4.6655999999999995, |
| "grad_norm": 16.875, |
| "learning_rate": 1.898e-05, |
| "loss": 2.2346, |
| "mean_token_accuracy": 0.5425456454977393, |
| "num_tokens": 14333953.0, |
| "step": 1460 |
| }, |
| { |
| "entropy": 2.310309498012066, |
| "epoch": 4.6975999999999996, |
| "grad_norm": 16.625, |
| "learning_rate": 1.895777777777778e-05, |
| "loss": 2.2344, |
| "mean_token_accuracy": 0.5393823731690646, |
| "num_tokens": 14437164.0, |
| "step": 1470 |
| }, |
| { |
| "entropy": 2.373570270091295, |
| "epoch": 4.7296, |
| "grad_norm": 19.25, |
| "learning_rate": 1.8935555555555556e-05, |
| "loss": 2.2902, |
| "mean_token_accuracy": 0.5355667784810066, |
| "num_tokens": 14532641.0, |
| "step": 1480 |
| }, |
| { |
| "entropy": 2.345385494083166, |
| "epoch": 4.7616, |
| "grad_norm": 18.25, |
| "learning_rate": 1.8913333333333334e-05, |
| "loss": 2.2528, |
| "mean_token_accuracy": 0.5387666182592511, |
| "num_tokens": 14633165.0, |
| "step": 1490 |
| }, |
| { |
| "entropy": 2.3080789014697074, |
| "epoch": 4.7936, |
| "grad_norm": 16.625, |
| "learning_rate": 1.8891111111111115e-05, |
| "loss": 2.2308, |
| "mean_token_accuracy": 0.5400335062295198, |
| "num_tokens": 14732212.0, |
| "step": 1500 |
| }, |
| { |
| "epoch": 4.7936, |
| "eval_coding_entropy": 1.73171142911911, |
| "eval_coding_loss": 2.283193826675415, |
| "eval_coding_mean_token_accuracy": 0.508676981806755, |
| "eval_coding_num_tokens": 14732212.0, |
| "eval_coding_runtime": 78.9397, |
| "eval_coding_samples_per_second": 6.334, |
| "eval_coding_steps_per_second": 3.167, |
| "step": 1500 |
| }, |
| { |
| "epoch": 4.7936, |
| "eval_biology_entropy": 2.392736424446106, |
| "eval_biology_loss": 2.536526679992676, |
| "eval_biology_mean_token_accuracy": 0.5183096739053726, |
| "eval_biology_num_tokens": 14732212.0, |
| "eval_biology_runtime": 34.5523, |
| "eval_biology_samples_per_second": 14.471, |
| "eval_biology_steps_per_second": 7.235, |
| "step": 1500 |
| }, |
| { |
| "entropy": 2.379144427925348, |
| "epoch": 4.8256, |
| "grad_norm": 17.25, |
| "learning_rate": 1.886888888888889e-05, |
| "loss": 2.287, |
| "mean_token_accuracy": 0.5354940524324775, |
| "num_tokens": 14829637.0, |
| "step": 1510 |
| }, |
| { |
| "entropy": 2.378316131234169, |
| "epoch": 4.8576, |
| "grad_norm": 16.25, |
| "learning_rate": 1.884666666666667e-05, |
| "loss": 2.299, |
| "mean_token_accuracy": 0.5294545406475664, |
| "num_tokens": 14920973.0, |
| "step": 1520 |
| }, |
| { |
| "entropy": 2.388844187557697, |
| "epoch": 4.8896, |
| "grad_norm": 17.375, |
| "learning_rate": 1.8824444444444445e-05, |
| "loss": 2.3172, |
| "mean_token_accuracy": 0.5300939682871103, |
| "num_tokens": 15017913.0, |
| "step": 1530 |
| }, |
| { |
| "entropy": 2.319888374954462, |
| "epoch": 4.9216, |
| "grad_norm": 15.875, |
| "learning_rate": 1.8802222222222226e-05, |
| "loss": 2.25, |
| "mean_token_accuracy": 0.5391564985737205, |
| "num_tokens": 15115005.0, |
| "step": 1540 |
| }, |
| { |
| "entropy": 2.328956404328346, |
| "epoch": 4.9536, |
| "grad_norm": 17.0, |
| "learning_rate": 1.878e-05, |
| "loss": 2.2304, |
| "mean_token_accuracy": 0.5401576224714517, |
| "num_tokens": 15210403.0, |
| "step": 1550 |
| }, |
| { |
| "entropy": 2.322893363237381, |
| "epoch": 4.9856, |
| "grad_norm": 14.75, |
| "learning_rate": 1.875777777777778e-05, |
| "loss": 2.2403, |
| "mean_token_accuracy": 0.5404085170477628, |
| "num_tokens": 15315187.0, |
| "step": 1560 |
| }, |
| { |
| "entropy": 2.2824014707615503, |
| "epoch": 5.016, |
| "grad_norm": 14.1875, |
| "learning_rate": 1.873555555555556e-05, |
| "loss": 2.2134, |
| "mean_token_accuracy": 0.5456873845112952, |
| "num_tokens": 15409643.0, |
| "step": 1570 |
| }, |
| { |
| "entropy": 2.0921800576150416, |
| "epoch": 5.048, |
| "grad_norm": 21.125, |
| "learning_rate": 1.8713333333333336e-05, |
| "loss": 1.9933, |
| "mean_token_accuracy": 0.5728920195251703, |
| "num_tokens": 15512455.0, |
| "step": 1580 |
| }, |
| { |
| "entropy": 2.205993994325399, |
| "epoch": 5.08, |
| "grad_norm": 15.375, |
| "learning_rate": 1.8691111111111114e-05, |
| "loss": 2.1219, |
| "mean_token_accuracy": 0.5554239621385932, |
| "num_tokens": 15605993.0, |
| "step": 1590 |
| }, |
| { |
| "entropy": 2.1570269420742987, |
| "epoch": 5.112, |
| "grad_norm": 16.125, |
| "learning_rate": 1.866888888888889e-05, |
| "loss": 2.0662, |
| "mean_token_accuracy": 0.5643594058230519, |
| "num_tokens": 15704177.0, |
| "step": 1600 |
| }, |
| { |
| "epoch": 5.112, |
| "eval_coding_entropy": 1.606255702495575, |
| "eval_coding_loss": 2.350952625274658, |
| "eval_coding_mean_token_accuracy": 0.5046476621627808, |
| "eval_coding_num_tokens": 15704177.0, |
| "eval_coding_runtime": 78.9543, |
| "eval_coding_samples_per_second": 6.333, |
| "eval_coding_steps_per_second": 3.166, |
| "step": 1600 |
| }, |
| { |
| "epoch": 5.112, |
| "eval_biology_entropy": 2.1643831424713134, |
| "eval_biology_loss": 2.545173168182373, |
| "eval_biology_mean_token_accuracy": 0.518920934677124, |
| "eval_biology_num_tokens": 15704177.0, |
| "eval_biology_runtime": 34.572, |
| "eval_biology_samples_per_second": 14.463, |
| "eval_biology_steps_per_second": 7.231, |
| "step": 1600 |
| }, |
| { |
| "entropy": 2.1564591567963363, |
| "epoch": 5.144, |
| "grad_norm": 18.125, |
| "learning_rate": 1.864666666666667e-05, |
| "loss": 2.0704, |
| "mean_token_accuracy": 0.5592444071546196, |
| "num_tokens": 15798327.0, |
| "step": 1610 |
| }, |
| { |
| "entropy": 2.141941160708666, |
| "epoch": 5.176, |
| "grad_norm": 17.375, |
| "learning_rate": 1.8624444444444446e-05, |
| "loss": 2.0211, |
| "mean_token_accuracy": 0.5634208859875798, |
| "num_tokens": 15892964.0, |
| "step": 1620 |
| }, |
| { |
| "entropy": 2.1816854923963547, |
| "epoch": 5.208, |
| "grad_norm": 17.0, |
| "learning_rate": 1.8602222222222224e-05, |
| "loss": 2.0997, |
| "mean_token_accuracy": 0.5580162599682807, |
| "num_tokens": 15992896.0, |
| "step": 1630 |
| }, |
| { |
| "entropy": 2.1366577565670015, |
| "epoch": 5.24, |
| "grad_norm": 16.875, |
| "learning_rate": 1.858e-05, |
| "loss": 2.0527, |
| "mean_token_accuracy": 0.5645586859434844, |
| "num_tokens": 16092490.0, |
| "step": 1640 |
| }, |
| { |
| "entropy": 2.1900168932974338, |
| "epoch": 5.272, |
| "grad_norm": 15.3125, |
| "learning_rate": 1.855777777777778e-05, |
| "loss": 2.1034, |
| "mean_token_accuracy": 0.5599051252007484, |
| "num_tokens": 16192809.0, |
| "step": 1650 |
| }, |
| { |
| "entropy": 2.0789007574319838, |
| "epoch": 5.304, |
| "grad_norm": 17.75, |
| "learning_rate": 1.8535555555555557e-05, |
| "loss": 1.9994, |
| "mean_token_accuracy": 0.575955624319613, |
| "num_tokens": 16295883.0, |
| "step": 1660 |
| }, |
| { |
| "entropy": 2.1156801864504815, |
| "epoch": 5.336, |
| "grad_norm": 14.0, |
| "learning_rate": 1.8513333333333335e-05, |
| "loss": 2.025, |
| "mean_token_accuracy": 0.5662754917517304, |
| "num_tokens": 16391035.0, |
| "step": 1670 |
| }, |
| { |
| "entropy": 2.167324597388506, |
| "epoch": 5.368, |
| "grad_norm": 19.5, |
| "learning_rate": 1.8491111111111112e-05, |
| "loss": 2.1154, |
| "mean_token_accuracy": 0.5577886115759612, |
| "num_tokens": 16484283.0, |
| "step": 1680 |
| }, |
| { |
| "entropy": 2.09282573312521, |
| "epoch": 5.4, |
| "grad_norm": 14.0625, |
| "learning_rate": 1.846888888888889e-05, |
| "loss": 2.0044, |
| "mean_token_accuracy": 0.5713933141902089, |
| "num_tokens": 16582622.0, |
| "step": 1690 |
| }, |
| { |
| "entropy": 2.1480760030448436, |
| "epoch": 5.432, |
| "grad_norm": 15.125, |
| "learning_rate": 1.8446666666666667e-05, |
| "loss": 2.0442, |
| "mean_token_accuracy": 0.562722404487431, |
| "num_tokens": 16680224.0, |
| "step": 1700 |
| }, |
| { |
| "epoch": 5.432, |
| "eval_coding_entropy": 1.5985201392173767, |
| "eval_coding_loss": 2.355341911315918, |
| "eval_coding_mean_token_accuracy": 0.5027512958049775, |
| "eval_coding_num_tokens": 16680224.0, |
| "eval_coding_runtime": 78.9432, |
| "eval_coding_samples_per_second": 6.334, |
| "eval_coding_steps_per_second": 3.167, |
| "step": 1700 |
| }, |
| { |
| "epoch": 5.432, |
| "eval_biology_entropy": 2.1854574666023256, |
| "eval_biology_loss": 2.508324146270752, |
| "eval_biology_mean_token_accuracy": 0.5225892701148986, |
| "eval_biology_num_tokens": 16680224.0, |
| "eval_biology_runtime": 34.6309, |
| "eval_biology_samples_per_second": 14.438, |
| "eval_biology_steps_per_second": 7.219, |
| "step": 1700 |
| }, |
| { |
| "entropy": 2.0809989593923093, |
| "epoch": 5.464, |
| "grad_norm": 15.875, |
| "learning_rate": 1.842444444444445e-05, |
| "loss": 2.0057, |
| "mean_token_accuracy": 0.5693793009966612, |
| "num_tokens": 16786599.0, |
| "step": 1710 |
| }, |
| { |
| "entropy": 2.103484392911196, |
| "epoch": 5.496, |
| "grad_norm": 14.5, |
| "learning_rate": 1.8402222222222223e-05, |
| "loss": 2.0262, |
| "mean_token_accuracy": 0.569659705273807, |
| "num_tokens": 16886909.0, |
| "step": 1720 |
| }, |
| { |
| "entropy": 2.161328002810478, |
| "epoch": 5.5280000000000005, |
| "grad_norm": 17.25, |
| "learning_rate": 1.8380000000000004e-05, |
| "loss": 2.0827, |
| "mean_token_accuracy": 0.5596394496038556, |
| "num_tokens": 16976330.0, |
| "step": 1730 |
| }, |
| { |
| "entropy": 2.111246530711651, |
| "epoch": 5.5600000000000005, |
| "grad_norm": 15.375, |
| "learning_rate": 1.8357777777777778e-05, |
| "loss": 1.9911, |
| "mean_token_accuracy": 0.5712085694074631, |
| "num_tokens": 17077377.0, |
| "step": 1740 |
| }, |
| { |
| "entropy": 2.050750946253538, |
| "epoch": 5.592, |
| "grad_norm": 18.625, |
| "learning_rate": 1.833555555555556e-05, |
| "loss": 1.9786, |
| "mean_token_accuracy": 0.572969813272357, |
| "num_tokens": 17175724.0, |
| "step": 1750 |
| }, |
| { |
| "entropy": 2.1586254596710206, |
| "epoch": 5.624, |
| "grad_norm": 15.1875, |
| "learning_rate": 1.8313333333333333e-05, |
| "loss": 2.0624, |
| "mean_token_accuracy": 0.5622333124279976, |
| "num_tokens": 17273208.0, |
| "step": 1760 |
| }, |
| { |
| "entropy": 2.1071687392890452, |
| "epoch": 5.656, |
| "grad_norm": 16.625, |
| "learning_rate": 1.8291111111111114e-05, |
| "loss": 2.0157, |
| "mean_token_accuracy": 0.5673394788056612, |
| "num_tokens": 17369608.0, |
| "step": 1770 |
| }, |
| { |
| "entropy": 2.1135758481919766, |
| "epoch": 5.688, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.8268888888888888e-05, |
| "loss": 2.0291, |
| "mean_token_accuracy": 0.5678149970248342, |
| "num_tokens": 17463192.0, |
| "step": 1780 |
| }, |
| { |
| "entropy": 2.130234959721565, |
| "epoch": 5.72, |
| "grad_norm": 15.5625, |
| "learning_rate": 1.824666666666667e-05, |
| "loss": 2.0629, |
| "mean_token_accuracy": 0.5624657543376088, |
| "num_tokens": 17561538.0, |
| "step": 1790 |
| }, |
| { |
| "entropy": 2.1406769305467606, |
| "epoch": 5.752, |
| "grad_norm": 17.625, |
| "learning_rate": 1.8224444444444447e-05, |
| "loss": 2.0362, |
| "mean_token_accuracy": 0.5657033832743764, |
| "num_tokens": 17659648.0, |
| "step": 1800 |
| }, |
| { |
| "epoch": 5.752, |
| "eval_coding_entropy": 1.5892016682624817, |
| "eval_coding_loss": 2.3968281745910645, |
| "eval_coding_mean_token_accuracy": 0.49928104650974275, |
| "eval_coding_num_tokens": 17659648.0, |
| "eval_coding_runtime": 78.9507, |
| "eval_coding_samples_per_second": 6.333, |
| "eval_coding_steps_per_second": 3.167, |
| "step": 1800 |
| }, |
| { |
| "epoch": 5.752, |
| "eval_biology_entropy": 2.13502747964859, |
| "eval_biology_loss": 2.4895498752593994, |
| "eval_biology_mean_token_accuracy": 0.5265302859544754, |
| "eval_biology_num_tokens": 17659648.0, |
| "eval_biology_runtime": 34.6074, |
| "eval_biology_samples_per_second": 14.448, |
| "eval_biology_steps_per_second": 7.224, |
| "step": 1800 |
| }, |
| { |
| "entropy": 2.1306553721427917, |
| "epoch": 5.784, |
| "grad_norm": 15.375, |
| "learning_rate": 1.8202222222222225e-05, |
| "loss": 2.0563, |
| "mean_token_accuracy": 0.5620468180626631, |
| "num_tokens": 17758427.0, |
| "step": 1810 |
| }, |
| { |
| "entropy": 2.055387606471777, |
| "epoch": 5.816, |
| "grad_norm": 16.0, |
| "learning_rate": 1.8180000000000002e-05, |
| "loss": 1.9563, |
| "mean_token_accuracy": 0.5758049864321947, |
| "num_tokens": 17862480.0, |
| "step": 1820 |
| }, |
| { |
| "entropy": 2.126579797267914, |
| "epoch": 5.848, |
| "grad_norm": 14.0, |
| "learning_rate": 1.815777777777778e-05, |
| "loss": 2.028, |
| "mean_token_accuracy": 0.5658794036135077, |
| "num_tokens": 17964710.0, |
| "step": 1830 |
| }, |
| { |
| "entropy": 2.07228102684021, |
| "epoch": 5.88, |
| "grad_norm": 15.25, |
| "learning_rate": 1.8135555555555557e-05, |
| "loss": 2.0038, |
| "mean_token_accuracy": 0.5738436767831445, |
| "num_tokens": 18068440.0, |
| "step": 1840 |
| }, |
| { |
| "entropy": 2.161037439852953, |
| "epoch": 5.912, |
| "grad_norm": 17.125, |
| "learning_rate": 1.8113333333333335e-05, |
| "loss": 2.0765, |
| "mean_token_accuracy": 0.5592521218582988, |
| "num_tokens": 18164947.0, |
| "step": 1850 |
| }, |
| { |
| "entropy": 2.1283591516315936, |
| "epoch": 5.944, |
| "grad_norm": 15.4375, |
| "learning_rate": 1.8091111111111113e-05, |
| "loss": 2.0441, |
| "mean_token_accuracy": 0.5676074722781778, |
| "num_tokens": 18261099.0, |
| "step": 1860 |
| }, |
| { |
| "entropy": 2.1420124895870685, |
| "epoch": 5.976, |
| "grad_norm": 15.25, |
| "learning_rate": 1.806888888888889e-05, |
| "loss": 2.0512, |
| "mean_token_accuracy": 0.5624487655237317, |
| "num_tokens": 18356946.0, |
| "step": 1870 |
| }, |
| { |
| "entropy": 2.110038713405007, |
| "epoch": 6.0064, |
| "grad_norm": 17.25, |
| "learning_rate": 1.8046666666666668e-05, |
| "loss": 2.0047, |
| "mean_token_accuracy": 0.573460263641257, |
| "num_tokens": 18449252.0, |
| "step": 1880 |
| }, |
| { |
| "entropy": 1.8439034026116132, |
| "epoch": 6.0384, |
| "grad_norm": 14.9375, |
| "learning_rate": 1.8024444444444445e-05, |
| "loss": 1.7534, |
| "mean_token_accuracy": 0.607002929598093, |
| "num_tokens": 18550959.0, |
| "step": 1890 |
| }, |
| { |
| "entropy": 1.9470315918326377, |
| "epoch": 6.0704, |
| "grad_norm": 14.75, |
| "learning_rate": 1.8002222222222223e-05, |
| "loss": 1.848, |
| "mean_token_accuracy": 0.5916370095685124, |
| "num_tokens": 18646263.0, |
| "step": 1900 |
| }, |
| { |
| "epoch": 6.0704, |
| "eval_coding_entropy": 1.4729241604804992, |
| "eval_coding_loss": 2.447042465209961, |
| "eval_coding_mean_token_accuracy": 0.49919502437114716, |
| "eval_coding_num_tokens": 18646263.0, |
| "eval_coding_runtime": 78.8811, |
| "eval_coding_samples_per_second": 6.339, |
| "eval_coding_steps_per_second": 3.169, |
| "step": 1900 |
| }, |
| { |
| "epoch": 6.0704, |
| "eval_biology_entropy": 1.9375120058059692, |
| "eval_biology_loss": 2.524658203125, |
| "eval_biology_mean_token_accuracy": 0.5277164667844773, |
| "eval_biology_num_tokens": 18646263.0, |
| "eval_biology_runtime": 34.5905, |
| "eval_biology_samples_per_second": 14.455, |
| "eval_biology_steps_per_second": 7.227, |
| "step": 1900 |
| }, |
| { |
| "entropy": 1.9363722302019597, |
| "epoch": 6.1024, |
| "grad_norm": 15.9375, |
| "learning_rate": 1.798e-05, |
| "loss": 1.8454, |
| "mean_token_accuracy": 0.5952178355306387, |
| "num_tokens": 18744274.0, |
| "step": 1910 |
| }, |
| { |
| "entropy": 1.9612058866769075, |
| "epoch": 6.1344, |
| "grad_norm": 16.125, |
| "learning_rate": 1.7957777777777778e-05, |
| "loss": 1.8713, |
| "mean_token_accuracy": 0.5913926368579269, |
| "num_tokens": 18831480.0, |
| "step": 1920 |
| }, |
| { |
| "entropy": 1.9057672590017318, |
| "epoch": 6.1664, |
| "grad_norm": 16.75, |
| "learning_rate": 1.7935555555555556e-05, |
| "loss": 1.8167, |
| "mean_token_accuracy": 0.5998800182715058, |
| "num_tokens": 18932857.0, |
| "step": 1930 |
| }, |
| { |
| "entropy": 1.857564640045166, |
| "epoch": 6.1984, |
| "grad_norm": 17.125, |
| "learning_rate": 1.7913333333333337e-05, |
| "loss": 1.7559, |
| "mean_token_accuracy": 0.6085444940254092, |
| "num_tokens": 19035806.0, |
| "step": 1940 |
| }, |
| { |
| "entropy": 1.839936225861311, |
| "epoch": 6.2304, |
| "grad_norm": 14.6875, |
| "learning_rate": 1.789111111111111e-05, |
| "loss": 1.7591, |
| "mean_token_accuracy": 0.6074719944968819, |
| "num_tokens": 19138641.0, |
| "step": 1950 |
| }, |
| { |
| "entropy": 1.9588794164359569, |
| "epoch": 6.2624, |
| "grad_norm": 17.125, |
| "learning_rate": 1.7868888888888892e-05, |
| "loss": 1.8805, |
| "mean_token_accuracy": 0.59009426869452, |
| "num_tokens": 19235701.0, |
| "step": 1960 |
| }, |
| { |
| "entropy": 1.9073541566729546, |
| "epoch": 6.2943999999999996, |
| "grad_norm": 18.5, |
| "learning_rate": 1.7846666666666666e-05, |
| "loss": 1.8107, |
| "mean_token_accuracy": 0.5988917088136076, |
| "num_tokens": 19334685.0, |
| "step": 1970 |
| }, |
| { |
| "entropy": 1.915497824549675, |
| "epoch": 6.3264, |
| "grad_norm": 17.625, |
| "learning_rate": 1.7824444444444447e-05, |
| "loss": 1.8412, |
| "mean_token_accuracy": 0.5978883042931556, |
| "num_tokens": 19430275.0, |
| "step": 1980 |
| }, |
| { |
| "entropy": 1.8486388459801675, |
| "epoch": 6.3584, |
| "grad_norm": 21.25, |
| "learning_rate": 1.780222222222222e-05, |
| "loss": 1.7726, |
| "mean_token_accuracy": 0.6066146129742265, |
| "num_tokens": 19525540.0, |
| "step": 1990 |
| }, |
| { |
| "entropy": 1.9017979100346565, |
| "epoch": 6.3904, |
| "grad_norm": 13.875, |
| "learning_rate": 1.7780000000000003e-05, |
| "loss": 1.7975, |
| "mean_token_accuracy": 0.5972068969160318, |
| "num_tokens": 19626609.0, |
| "step": 2000 |
| }, |
| { |
| "epoch": 6.3904, |
| "eval_coding_entropy": 1.4967151761054993, |
| "eval_coding_loss": 2.457455635070801, |
| "eval_coding_mean_token_accuracy": 0.4975039474964142, |
| "eval_coding_num_tokens": 19626609.0, |
| "eval_coding_runtime": 78.947, |
| "eval_coding_samples_per_second": 6.333, |
| "eval_coding_steps_per_second": 3.167, |
| "step": 2000 |
| }, |
| { |
| "epoch": 6.3904, |
| "eval_biology_entropy": 1.9981313619613648, |
| "eval_biology_loss": 2.497177839279175, |
| "eval_biology_mean_token_accuracy": 0.5299569643735885, |
| "eval_biology_num_tokens": 19626609.0, |
| "eval_biology_runtime": 34.5154, |
| "eval_biology_samples_per_second": 14.486, |
| "eval_biology_steps_per_second": 7.243, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 32, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.3069778923930337e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
