| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 5.0, |
| "eval_steps": 500, |
| "global_step": 190, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02631578947368421, |
| "grad_norm": 9.144888032555306, |
| "learning_rate": 0.0, |
| "loss": 3.0197, |
| "mean_token_accuracy": 0.435130774974823, |
| "num_tokens": 262144.0, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.05263157894736842, |
| "grad_norm": 9.089893864449401, |
| "learning_rate": 5.263157894736843e-07, |
| "loss": 3.0448, |
| "mean_token_accuracy": 0.43275395035743713, |
| "num_tokens": 524288.0, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.07894736842105263, |
| "grad_norm": 8.764265040006972, |
| "learning_rate": 1.0526315789473685e-06, |
| "loss": 3.1011, |
| "mean_token_accuracy": 0.4325822591781616, |
| "num_tokens": 786432.0, |
| "step": 3 |
| }, |
| { |
| "epoch": 0.10526315789473684, |
| "grad_norm": 9.548287094326655, |
| "learning_rate": 1.5789473684210526e-06, |
| "loss": 3.0779, |
| "mean_token_accuracy": 0.43555426597595215, |
| "num_tokens": 1048576.0, |
| "step": 4 |
| }, |
| { |
| "epoch": 0.13157894736842105, |
| "grad_norm": 9.17331273114906, |
| "learning_rate": 2.105263157894737e-06, |
| "loss": 2.9723, |
| "mean_token_accuracy": 0.4433600902557373, |
| "num_tokens": 1310720.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.15789473684210525, |
| "grad_norm": 9.240746915878553, |
| "learning_rate": 2.631578947368421e-06, |
| "loss": 2.9841, |
| "mean_token_accuracy": 0.44807180762290955, |
| "num_tokens": 1572864.0, |
| "step": 6 |
| }, |
| { |
| "epoch": 0.18421052631578946, |
| "grad_norm": 7.341867990328414, |
| "learning_rate": 3.157894736842105e-06, |
| "loss": 2.9011, |
| "mean_token_accuracy": 0.4490027129650116, |
| "num_tokens": 1835008.0, |
| "step": 7 |
| }, |
| { |
| "epoch": 0.21052631578947367, |
| "grad_norm": 6.348735555429503, |
| "learning_rate": 3.6842105263157896e-06, |
| "loss": 2.8076, |
| "mean_token_accuracy": 0.4570488929748535, |
| "num_tokens": 2097152.0, |
| "step": 8 |
| }, |
| { |
| "epoch": 0.23684210526315788, |
| "grad_norm": 6.306969416042576, |
| "learning_rate": 4.210526315789474e-06, |
| "loss": 2.8711, |
| "mean_token_accuracy": 0.44867080450057983, |
| "num_tokens": 2359296.0, |
| "step": 9 |
| }, |
| { |
| "epoch": 0.2631578947368421, |
| "grad_norm": 4.435909523033183, |
| "learning_rate": 4.736842105263158e-06, |
| "loss": 2.721, |
| "mean_token_accuracy": 0.4588572680950165, |
| "num_tokens": 2621440.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.2894736842105263, |
| "grad_norm": 3.9993913214616366, |
| "learning_rate": 5.263157894736842e-06, |
| "loss": 2.6792, |
| "mean_token_accuracy": 0.4573693573474884, |
| "num_tokens": 2883584.0, |
| "step": 11 |
| }, |
| { |
| "epoch": 0.3157894736842105, |
| "grad_norm": 3.923561561790813, |
| "learning_rate": 5.789473684210527e-06, |
| "loss": 2.6165, |
| "mean_token_accuracy": 0.4673383831977844, |
| "num_tokens": 3145728.0, |
| "step": 12 |
| }, |
| { |
| "epoch": 0.34210526315789475, |
| "grad_norm": 3.527199323221065, |
| "learning_rate": 6.31578947368421e-06, |
| "loss": 2.5569, |
| "mean_token_accuracy": 0.46943292021751404, |
| "num_tokens": 3407872.0, |
| "step": 13 |
| }, |
| { |
| "epoch": 0.3684210526315789, |
| "grad_norm": 3.540386269836106, |
| "learning_rate": 6.842105263157896e-06, |
| "loss": 2.4068, |
| "mean_token_accuracy": 0.48950067162513733, |
| "num_tokens": 3670016.0, |
| "step": 14 |
| }, |
| { |
| "epoch": 0.39473684210526316, |
| "grad_norm": 2.8367351336803397, |
| "learning_rate": 7.368421052631579e-06, |
| "loss": 2.38, |
| "mean_token_accuracy": 0.4991988241672516, |
| "num_tokens": 3932160.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.42105263157894735, |
| "grad_norm": 2.7891206971825206, |
| "learning_rate": 7.894736842105265e-06, |
| "loss": 2.4209, |
| "mean_token_accuracy": 0.4855557978153229, |
| "num_tokens": 4194304.0, |
| "step": 16 |
| }, |
| { |
| "epoch": 0.4473684210526316, |
| "grad_norm": 2.215561269771985, |
| "learning_rate": 8.421052631578948e-06, |
| "loss": 2.3677, |
| "mean_token_accuracy": 0.4911602735519409, |
| "num_tokens": 4456448.0, |
| "step": 17 |
| }, |
| { |
| "epoch": 0.47368421052631576, |
| "grad_norm": 2.2769577851472915, |
| "learning_rate": 8.947368421052632e-06, |
| "loss": 2.238, |
| "mean_token_accuracy": 0.5115599632263184, |
| "num_tokens": 4718592.0, |
| "step": 18 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 2.2246857453539426, |
| "learning_rate": 9.473684210526315e-06, |
| "loss": 2.1407, |
| "mean_token_accuracy": 0.5227155089378357, |
| "num_tokens": 4980736.0, |
| "step": 19 |
| }, |
| { |
| "epoch": 0.5263157894736842, |
| "grad_norm": 2.0661055710278697, |
| "learning_rate": 1e-05, |
| "loss": 2.207, |
| "mean_token_accuracy": 0.5132614970207214, |
| "num_tokens": 5242880.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.5526315789473685, |
| "grad_norm": 2.1978917959916364, |
| "learning_rate": 9.999240587349952e-06, |
| "loss": 2.0393, |
| "mean_token_accuracy": 0.5380829572677612, |
| "num_tokens": 5505024.0, |
| "step": 21 |
| }, |
| { |
| "epoch": 0.5789473684210527, |
| "grad_norm": 1.9187572184546404, |
| "learning_rate": 9.996962605714288e-06, |
| "loss": 2.0578, |
| "mean_token_accuracy": 0.5336993336677551, |
| "num_tokens": 5767168.0, |
| "step": 22 |
| }, |
| { |
| "epoch": 0.6052631578947368, |
| "grad_norm": 1.751193492821015, |
| "learning_rate": 9.993166823949924e-06, |
| "loss": 2.007, |
| "mean_token_accuracy": 0.5419515371322632, |
| "num_tokens": 6029312.0, |
| "step": 23 |
| }, |
| { |
| "epoch": 0.631578947368421, |
| "grad_norm": 1.7120403735686438, |
| "learning_rate": 9.987854523196724e-06, |
| "loss": 1.9728, |
| "mean_token_accuracy": 0.5452325940132141, |
| "num_tokens": 6291456.0, |
| "step": 24 |
| }, |
| { |
| "epoch": 0.6578947368421053, |
| "grad_norm": 1.6167100976334166, |
| "learning_rate": 9.981027496445084e-06, |
| "loss": 1.97, |
| "mean_token_accuracy": 0.5456637144088745, |
| "num_tokens": 6553600.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.6842105263157895, |
| "grad_norm": 1.6494993317134798, |
| "learning_rate": 9.972688047930773e-06, |
| "loss": 1.9057, |
| "mean_token_accuracy": 0.5569413304328918, |
| "num_tokens": 6815744.0, |
| "step": 26 |
| }, |
| { |
| "epoch": 0.7105263157894737, |
| "grad_norm": 1.4116871328104919, |
| "learning_rate": 9.962838992357212e-06, |
| "loss": 1.8411, |
| "mean_token_accuracy": 0.5720035433769226, |
| "num_tokens": 7077888.0, |
| "step": 27 |
| }, |
| { |
| "epoch": 0.7368421052631579, |
| "grad_norm": 1.3548679701598223, |
| "learning_rate": 9.951483653945461e-06, |
| "loss": 1.8332, |
| "mean_token_accuracy": 0.5701990127563477, |
| "num_tokens": 7340032.0, |
| "step": 28 |
| }, |
| { |
| "epoch": 0.7631578947368421, |
| "grad_norm": 1.5289726872863745, |
| "learning_rate": 9.938625865312252e-06, |
| "loss": 1.8215, |
| "mean_token_accuracy": 0.5682075023651123, |
| "num_tokens": 7602176.0, |
| "step": 29 |
| }, |
| { |
| "epoch": 0.7894736842105263, |
| "grad_norm": 1.4048047900834215, |
| "learning_rate": 9.9242699661764e-06, |
| "loss": 1.764, |
| "mean_token_accuracy": 0.5825715661048889, |
| "num_tokens": 7864320.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.8157894736842105, |
| "grad_norm": 1.3199591889964046, |
| "learning_rate": 9.908420801894083e-06, |
| "loss": 1.698, |
| "mean_token_accuracy": 0.5920604467391968, |
| "num_tokens": 8118799.0, |
| "step": 31 |
| }, |
| { |
| "epoch": 0.8421052631578947, |
| "grad_norm": 1.2695821423340337, |
| "learning_rate": 9.89108372182346e-06, |
| "loss": 1.7723, |
| "mean_token_accuracy": 0.581972599029541, |
| "num_tokens": 8380943.0, |
| "step": 32 |
| }, |
| { |
| "epoch": 0.868421052631579, |
| "grad_norm": 1.2217610450324474, |
| "learning_rate": 9.872264577519163e-06, |
| "loss": 1.6117, |
| "mean_token_accuracy": 0.6087397933006287, |
| "num_tokens": 8643087.0, |
| "step": 33 |
| }, |
| { |
| "epoch": 0.8947368421052632, |
| "grad_norm": 1.5890070323202978, |
| "learning_rate": 9.8519697207573e-06, |
| "loss": 1.6977, |
| "mean_token_accuracy": 0.5938796997070312, |
| "num_tokens": 8905231.0, |
| "step": 34 |
| }, |
| { |
| "epoch": 0.9210526315789473, |
| "grad_norm": 18.086842934362615, |
| "learning_rate": 9.830206001391627e-06, |
| "loss": 1.6338, |
| "mean_token_accuracy": 0.610822856426239, |
| "num_tokens": 9167375.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.9473684210526315, |
| "grad_norm": 1.47252943858889, |
| "learning_rate": 9.806980765041615e-06, |
| "loss": 1.6437, |
| "mean_token_accuracy": 0.6038067936897278, |
| "num_tokens": 9429519.0, |
| "step": 36 |
| }, |
| { |
| "epoch": 0.9736842105263158, |
| "grad_norm": 1.1229751841198494, |
| "learning_rate": 9.782301850613165e-06, |
| "loss": 1.591, |
| "mean_token_accuracy": 0.6120895147323608, |
| "num_tokens": 9691663.0, |
| "step": 37 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 1.0329182091569846, |
| "learning_rate": 9.756177587652857e-06, |
| "loss": 1.6121, |
| "mean_token_accuracy": 0.6088923811912537, |
| "num_tokens": 9953807.0, |
| "step": 38 |
| }, |
| { |
| "epoch": 1.0263157894736843, |
| "grad_norm": 1.037629173374562, |
| "learning_rate": 9.728616793536588e-06, |
| "loss": 1.5069, |
| "mean_token_accuracy": 0.6265947222709656, |
| "num_tokens": 10215951.0, |
| "step": 39 |
| }, |
| { |
| "epoch": 1.0526315789473684, |
| "grad_norm": 1.1247130552032012, |
| "learning_rate": 9.699628770493557e-06, |
| "loss": 1.6151, |
| "mean_token_accuracy": 0.6127304434776306, |
| "num_tokens": 10478095.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.0789473684210527, |
| "grad_norm": 0.9326470270807049, |
| "learning_rate": 9.669223302466609e-06, |
| "loss": 1.5202, |
| "mean_token_accuracy": 0.6245765089988708, |
| "num_tokens": 10740239.0, |
| "step": 41 |
| }, |
| { |
| "epoch": 1.1052631578947367, |
| "grad_norm": 1.081738987272881, |
| "learning_rate": 9.637410651809985e-06, |
| "loss": 1.4707, |
| "mean_token_accuracy": 0.639268696308136, |
| "num_tokens": 11002383.0, |
| "step": 42 |
| }, |
| { |
| "epoch": 1.131578947368421, |
| "grad_norm": 1.0128599536731275, |
| "learning_rate": 9.604201555825618e-06, |
| "loss": 1.5681, |
| "mean_token_accuracy": 0.6182166337966919, |
| "num_tokens": 11264527.0, |
| "step": 43 |
| }, |
| { |
| "epoch": 1.1578947368421053, |
| "grad_norm": 0.9486897801531551, |
| "learning_rate": 9.5696072231391e-06, |
| "loss": 1.4973, |
| "mean_token_accuracy": 0.6312530636787415, |
| "num_tokens": 11526671.0, |
| "step": 44 |
| }, |
| { |
| "epoch": 1.1842105263157894, |
| "grad_norm": 0.9528661635292972, |
| "learning_rate": 9.533639329916592e-06, |
| "loss": 1.3808, |
| "mean_token_accuracy": 0.6557158827781677, |
| "num_tokens": 11788815.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 1.2105263157894737, |
| "grad_norm": 15.454966425556524, |
| "learning_rate": 9.496310015923921e-06, |
| "loss": 1.5053, |
| "mean_token_accuracy": 0.6316651105880737, |
| "num_tokens": 12050959.0, |
| "step": 46 |
| }, |
| { |
| "epoch": 1.236842105263158, |
| "grad_norm": 0.9538351658576188, |
| "learning_rate": 9.4576318804292e-06, |
| "loss": 1.4802, |
| "mean_token_accuracy": 0.6355336904525757, |
| "num_tokens": 12313103.0, |
| "step": 47 |
| }, |
| { |
| "epoch": 1.263157894736842, |
| "grad_norm": 0.8318020814980096, |
| "learning_rate": 9.417617977950383e-06, |
| "loss": 1.4011, |
| "mean_token_accuracy": 0.6495734453201294, |
| "num_tokens": 12575247.0, |
| "step": 48 |
| }, |
| { |
| "epoch": 1.2894736842105263, |
| "grad_norm": 0.87491229088957, |
| "learning_rate": 9.376281813849114e-06, |
| "loss": 1.4425, |
| "mean_token_accuracy": 0.6401004195213318, |
| "num_tokens": 12837391.0, |
| "step": 49 |
| }, |
| { |
| "epoch": 1.3157894736842106, |
| "grad_norm": 13.352523689341348, |
| "learning_rate": 9.333637339772472e-06, |
| "loss": 1.4498, |
| "mean_token_accuracy": 0.6377197504043579, |
| "num_tokens": 13099535.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.3421052631578947, |
| "grad_norm": 2.6904758771681445, |
| "learning_rate": 9.289698948944046e-06, |
| "loss": 1.3891, |
| "mean_token_accuracy": 0.6509392857551575, |
| "num_tokens": 13361679.0, |
| "step": 51 |
| }, |
| { |
| "epoch": 1.368421052631579, |
| "grad_norm": 1.035669421403492, |
| "learning_rate": 9.244481471305977e-06, |
| "loss": 1.3858, |
| "mean_token_accuracy": 0.6511223912239075, |
| "num_tokens": 13623823.0, |
| "step": 52 |
| }, |
| { |
| "epoch": 1.3947368421052633, |
| "grad_norm": 88.70855789773888, |
| "learning_rate": 9.198000168513604e-06, |
| "loss": 1.4689, |
| "mean_token_accuracy": 0.6373191475868225, |
| "num_tokens": 13885967.0, |
| "step": 53 |
| }, |
| { |
| "epoch": 1.4210526315789473, |
| "grad_norm": 0.8796467760984398, |
| "learning_rate": 9.150270728784411e-06, |
| "loss": 1.441, |
| "mean_token_accuracy": 0.6398676633834839, |
| "num_tokens": 14148111.0, |
| "step": 54 |
| }, |
| { |
| "epoch": 1.4473684210526316, |
| "grad_norm": 0.8537471648823969, |
| "learning_rate": 9.101309261602978e-06, |
| "loss": 1.4113, |
| "mean_token_accuracy": 0.6479100584983826, |
| "num_tokens": 14410255.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.4736842105263157, |
| "grad_norm": 0.8164741757465972, |
| "learning_rate": 9.051132292283772e-06, |
| "loss": 1.4088, |
| "mean_token_accuracy": 0.6451835036277771, |
| "num_tokens": 14664734.0, |
| "step": 56 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 0.8363413113484846, |
| "learning_rate": 8.999756756393562e-06, |
| "loss": 1.4238, |
| "mean_token_accuracy": 0.6428511738777161, |
| "num_tokens": 14926878.0, |
| "step": 57 |
| }, |
| { |
| "epoch": 1.526315789473684, |
| "grad_norm": 0.7766725489164594, |
| "learning_rate": 8.947199994035402e-06, |
| "loss": 1.3153, |
| "mean_token_accuracy": 0.6633500456809998, |
| "num_tokens": 15189022.0, |
| "step": 58 |
| }, |
| { |
| "epoch": 1.5526315789473686, |
| "grad_norm": 0.8059596602743602, |
| "learning_rate": 8.893479743996034e-06, |
| "loss": 1.3448, |
| "mean_token_accuracy": 0.65765780210495, |
| "num_tokens": 15451166.0, |
| "step": 59 |
| }, |
| { |
| "epoch": 1.5789473684210527, |
| "grad_norm": 0.7324537445066931, |
| "learning_rate": 8.838614137758767e-06, |
| "loss": 1.3363, |
| "mean_token_accuracy": 0.6614806056022644, |
| "num_tokens": 15713310.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.6052631578947367, |
| "grad_norm": 0.77781086842083, |
| "learning_rate": 8.782621693383793e-06, |
| "loss": 1.4157, |
| "mean_token_accuracy": 0.6485319137573242, |
| "num_tokens": 15975454.0, |
| "step": 61 |
| }, |
| { |
| "epoch": 1.631578947368421, |
| "grad_norm": 0.7061723115441305, |
| "learning_rate": 8.725521309258031e-06, |
| "loss": 1.3202, |
| "mean_token_accuracy": 0.6643381714820862, |
| "num_tokens": 16237598.0, |
| "step": 62 |
| }, |
| { |
| "epoch": 1.6578947368421053, |
| "grad_norm": 0.7141045939197578, |
| "learning_rate": 8.667332257716603e-06, |
| "loss": 1.3325, |
| "mean_token_accuracy": 0.6641473770141602, |
| "num_tokens": 16499742.0, |
| "step": 63 |
| }, |
| { |
| "epoch": 1.6842105263157894, |
| "grad_norm": 0.740270282620103, |
| "learning_rate": 8.608074178538101e-06, |
| "loss": 1.428, |
| "mean_token_accuracy": 0.6470096707344055, |
| "num_tokens": 16761886.0, |
| "step": 64 |
| }, |
| { |
| "epoch": 1.7105263157894737, |
| "grad_norm": 0.6639377114211265, |
| "learning_rate": 8.547767072315835e-06, |
| "loss": 1.4018, |
| "mean_token_accuracy": 0.6485319137573242, |
| "num_tokens": 17024030.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.736842105263158, |
| "grad_norm": 0.6739541270775268, |
| "learning_rate": 8.486431293707298e-06, |
| "loss": 1.3142, |
| "mean_token_accuracy": 0.6693703532218933, |
| "num_tokens": 17286174.0, |
| "step": 66 |
| }, |
| { |
| "epoch": 1.763157894736842, |
| "grad_norm": 1.5441747838886493, |
| "learning_rate": 8.42408754456412e-06, |
| "loss": 1.3586, |
| "mean_token_accuracy": 0.6568832993507385, |
| "num_tokens": 17548318.0, |
| "step": 67 |
| }, |
| { |
| "epoch": 1.7894736842105263, |
| "grad_norm": 0.7004893065739863, |
| "learning_rate": 8.360756866944858e-06, |
| "loss": 1.3285, |
| "mean_token_accuracy": 0.6602864265441895, |
| "num_tokens": 17810462.0, |
| "step": 68 |
| }, |
| { |
| "epoch": 1.8157894736842106, |
| "grad_norm": 0.6598266465075178, |
| "learning_rate": 8.296460636012942e-06, |
| "loss": 1.3476, |
| "mean_token_accuracy": 0.6597141623497009, |
| "num_tokens": 18072606.0, |
| "step": 69 |
| }, |
| { |
| "epoch": 1.8421052631578947, |
| "grad_norm": 3.373108959827941, |
| "learning_rate": 8.231220552822206e-06, |
| "loss": 1.3502, |
| "mean_token_accuracy": 0.6596264243125916, |
| "num_tokens": 18334750.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.868421052631579, |
| "grad_norm": 0.7335650187416184, |
| "learning_rate": 8.165058636992411e-06, |
| "loss": 1.3047, |
| "mean_token_accuracy": 0.670862078666687, |
| "num_tokens": 18596894.0, |
| "step": 71 |
| }, |
| { |
| "epoch": 1.8947368421052633, |
| "grad_norm": 0.6500785746933061, |
| "learning_rate": 8.09799721927726e-06, |
| "loss": 1.3302, |
| "mean_token_accuracy": 0.6620299816131592, |
| "num_tokens": 18859038.0, |
| "step": 72 |
| }, |
| { |
| "epoch": 1.9210526315789473, |
| "grad_norm": 0.6274604158335885, |
| "learning_rate": 8.030058934027395e-06, |
| "loss": 1.2792, |
| "mean_token_accuracy": 0.6728650331497192, |
| "num_tokens": 19121182.0, |
| "step": 73 |
| }, |
| { |
| "epoch": 1.9473684210526314, |
| "grad_norm": 0.6291519836042733, |
| "learning_rate": 7.961266711550922e-06, |
| "loss": 1.3297, |
| "mean_token_accuracy": 0.6599164009094238, |
| "num_tokens": 19383326.0, |
| "step": 74 |
| }, |
| { |
| "epoch": 1.973684210526316, |
| "grad_norm": 0.6444596016200881, |
| "learning_rate": 7.891643770374052e-06, |
| "loss": 1.3164, |
| "mean_token_accuracy": 0.6649523973464966, |
| "num_tokens": 19645470.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 28.209700826902587, |
| "learning_rate": 7.821213609404459e-06, |
| "loss": 1.3146, |
| "mean_token_accuracy": 0.6623695492744446, |
| "num_tokens": 19907614.0, |
| "step": 76 |
| }, |
| { |
| "epoch": 2.026315789473684, |
| "grad_norm": 0.708893212650595, |
| "learning_rate": 7.75e-06, |
| "loss": 1.2725, |
| "mean_token_accuracy": 0.6715068221092224, |
| "num_tokens": 20169758.0, |
| "step": 77 |
| }, |
| { |
| "epoch": 2.0526315789473686, |
| "grad_norm": 0.9273454935271076, |
| "learning_rate": 7.678026977945497e-06, |
| "loss": 1.2723, |
| "mean_token_accuracy": 0.6746886968612671, |
| "num_tokens": 20431902.0, |
| "step": 78 |
| }, |
| { |
| "epoch": 2.0789473684210527, |
| "grad_norm": 0.6077828732079009, |
| "learning_rate": 7.605318835340237e-06, |
| "loss": 1.1928, |
| "mean_token_accuracy": 0.6885720491409302, |
| "num_tokens": 20694046.0, |
| "step": 79 |
| }, |
| { |
| "epoch": 2.1052631578947367, |
| "grad_norm": 0.6625126850722808, |
| "learning_rate": 7.531900112399004e-06, |
| "loss": 1.2694, |
| "mean_token_accuracy": 0.6737463474273682, |
| "num_tokens": 20956190.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 2.1315789473684212, |
| "grad_norm": 0.6763431442698962, |
| "learning_rate": 7.4577955891693295e-06, |
| "loss": 1.2715, |
| "mean_token_accuracy": 0.6732046008110046, |
| "num_tokens": 21218334.0, |
| "step": 81 |
| }, |
| { |
| "epoch": 2.1578947368421053, |
| "grad_norm": 0.5832389341094574, |
| "learning_rate": 7.383030277167825e-06, |
| "loss": 1.2205, |
| "mean_token_accuracy": 0.683753490447998, |
| "num_tokens": 21480478.0, |
| "step": 82 |
| }, |
| { |
| "epoch": 2.1842105263157894, |
| "grad_norm": 2.2405291609094293, |
| "learning_rate": 7.307629410938364e-06, |
| "loss": 1.278, |
| "mean_token_accuracy": 0.6732503771781921, |
| "num_tokens": 21742622.0, |
| "step": 83 |
| }, |
| { |
| "epoch": 2.2105263157894735, |
| "grad_norm": 2.138457688507137, |
| "learning_rate": 7.231618439535006e-06, |
| "loss": 1.2778, |
| "mean_token_accuracy": 0.668744683265686, |
| "num_tokens": 22004766.0, |
| "step": 84 |
| }, |
| { |
| "epoch": 2.236842105263158, |
| "grad_norm": 0.6375029873512914, |
| "learning_rate": 7.155023017932521e-06, |
| "loss": 1.2534, |
| "mean_token_accuracy": 0.6742499470710754, |
| "num_tokens": 22266910.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 2.263157894736842, |
| "grad_norm": 0.6682751703287256, |
| "learning_rate": 7.0778689983673955e-06, |
| "loss": 1.2666, |
| "mean_token_accuracy": 0.6781108975410461, |
| "num_tokens": 22529054.0, |
| "step": 86 |
| }, |
| { |
| "epoch": 2.2894736842105265, |
| "grad_norm": 0.5844308031288254, |
| "learning_rate": 7.000182421612281e-06, |
| "loss": 1.2878, |
| "mean_token_accuracy": 0.6716785430908203, |
| "num_tokens": 22791198.0, |
| "step": 87 |
| }, |
| { |
| "epoch": 2.3157894736842106, |
| "grad_norm": 0.6304946926699229, |
| "learning_rate": 6.921989508186787e-06, |
| "loss": 1.2245, |
| "mean_token_accuracy": 0.6867827773094177, |
| "num_tokens": 23053342.0, |
| "step": 88 |
| }, |
| { |
| "epoch": 2.3421052631578947, |
| "grad_norm": 0.588447827602647, |
| "learning_rate": 6.843316649507627e-06, |
| "loss": 1.2488, |
| "mean_token_accuracy": 0.6804381608963013, |
| "num_tokens": 23315486.0, |
| "step": 89 |
| }, |
| { |
| "epoch": 2.3684210526315788, |
| "grad_norm": 0.5752511341385269, |
| "learning_rate": 6.764190398981063e-06, |
| "loss": 1.2183, |
| "mean_token_accuracy": 0.6857373714447021, |
| "num_tokens": 23577630.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 2.3947368421052633, |
| "grad_norm": 0.601133537795538, |
| "learning_rate": 6.684637463040698e-06, |
| "loss": 1.1987, |
| "mean_token_accuracy": 0.6845928430557251, |
| "num_tokens": 23839774.0, |
| "step": 91 |
| }, |
| { |
| "epoch": 2.4210526315789473, |
| "grad_norm": 1.599039072518535, |
| "learning_rate": 6.604684692133597e-06, |
| "loss": 1.2244, |
| "mean_token_accuracy": 0.6837954521179199, |
| "num_tokens": 24101918.0, |
| "step": 92 |
| }, |
| { |
| "epoch": 2.4473684210526314, |
| "grad_norm": 17.31027379733536, |
| "learning_rate": 6.524359071657822e-06, |
| "loss": 1.2287, |
| "mean_token_accuracy": 0.6827691793441772, |
| "num_tokens": 24364062.0, |
| "step": 93 |
| }, |
| { |
| "epoch": 2.473684210526316, |
| "grad_norm": 0.6809165019277617, |
| "learning_rate": 6.443687712854397e-06, |
| "loss": 1.2708, |
| "mean_token_accuracy": 0.676279604434967, |
| "num_tokens": 24626206.0, |
| "step": 94 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 0.6442727822424891, |
| "learning_rate": 6.362697843656823e-06, |
| "loss": 1.2795, |
| "mean_token_accuracy": 0.6701677441596985, |
| "num_tokens": 24888350.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 2.526315789473684, |
| "grad_norm": 0.5769629627925874, |
| "learning_rate": 6.281416799501188e-06, |
| "loss": 1.2311, |
| "mean_token_accuracy": 0.6823189854621887, |
| "num_tokens": 25150494.0, |
| "step": 96 |
| }, |
| { |
| "epoch": 2.5526315789473686, |
| "grad_norm": 1.233425967241537, |
| "learning_rate": 6.199872014099995e-06, |
| "loss": 1.2665, |
| "mean_token_accuracy": 0.6751083731651306, |
| "num_tokens": 25412638.0, |
| "step": 97 |
| }, |
| { |
| "epoch": 2.5789473684210527, |
| "grad_norm": 0.7476662613111043, |
| "learning_rate": 6.118091010182837e-06, |
| "loss": 1.2243, |
| "mean_token_accuracy": 0.6817924976348877, |
| "num_tokens": 25674782.0, |
| "step": 98 |
| }, |
| { |
| "epoch": 2.6052631578947367, |
| "grad_norm": 1.58371089124971, |
| "learning_rate": 6.036101390207006e-06, |
| "loss": 1.1538, |
| "mean_token_accuracy": 0.7012002468109131, |
| "num_tokens": 25936926.0, |
| "step": 99 |
| }, |
| { |
| "epoch": 2.6315789473684212, |
| "grad_norm": 1.4370365700790926, |
| "learning_rate": 5.953930827041209e-06, |
| "loss": 1.2446, |
| "mean_token_accuracy": 0.6795377731323242, |
| "num_tokens": 26199070.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 2.6578947368421053, |
| "grad_norm": 0.7891205485960927, |
| "learning_rate": 5.871607054625497e-06, |
| "loss": 1.2842, |
| "mean_token_accuracy": 0.6718769073486328, |
| "num_tokens": 26461214.0, |
| "step": 101 |
| }, |
| { |
| "epoch": 2.6842105263157894, |
| "grad_norm": 7.060494358042542, |
| "learning_rate": 5.7891578586106165e-06, |
| "loss": 1.1096, |
| "mean_token_accuracy": 0.7096241116523743, |
| "num_tokens": 26723358.0, |
| "step": 102 |
| }, |
| { |
| "epoch": 2.7105263157894735, |
| "grad_norm": 2.0263515651724164, |
| "learning_rate": 5.706611066979874e-06, |
| "loss": 1.269, |
| "mean_token_accuracy": 0.6746886968612671, |
| "num_tokens": 26985502.0, |
| "step": 103 |
| }, |
| { |
| "epoch": 2.736842105263158, |
| "grad_norm": 1.570123901253544, |
| "learning_rate": 5.62399454065673e-06, |
| "loss": 1.2304, |
| "mean_token_accuracy": 0.681018054485321, |
| "num_tokens": 27247646.0, |
| "step": 104 |
| }, |
| { |
| "epoch": 2.763157894736842, |
| "grad_norm": 0.7111522615328816, |
| "learning_rate": 5.54133616410127e-06, |
| "loss": 1.2478, |
| "mean_token_accuracy": 0.6804953813552856, |
| "num_tokens": 27509790.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 2.7894736842105265, |
| "grad_norm": 0.9119519373287289, |
| "learning_rate": 5.458663835898732e-06, |
| "loss": 1.2073, |
| "mean_token_accuracy": 0.6871718764305115, |
| "num_tokens": 27771934.0, |
| "step": 106 |
| }, |
| { |
| "epoch": 2.8157894736842106, |
| "grad_norm": 0.9240457779811717, |
| "learning_rate": 5.376005459343272e-06, |
| "loss": 1.2718, |
| "mean_token_accuracy": 0.6782024502754211, |
| "num_tokens": 28034078.0, |
| "step": 107 |
| }, |
| { |
| "epoch": 2.8421052631578947, |
| "grad_norm": 0.5926290558432593, |
| "learning_rate": 5.293388933020127e-06, |
| "loss": 1.2398, |
| "mean_token_accuracy": 0.6791562438011169, |
| "num_tokens": 28296222.0, |
| "step": 108 |
| }, |
| { |
| "epoch": 2.8684210526315788, |
| "grad_norm": 0.5937415300824982, |
| "learning_rate": 5.210842141389385e-06, |
| "loss": 1.2589, |
| "mean_token_accuracy": 0.6773592829704285, |
| "num_tokens": 28558366.0, |
| "step": 109 |
| }, |
| { |
| "epoch": 2.8947368421052633, |
| "grad_norm": 0.8205931665334268, |
| "learning_rate": 5.1283929453745055e-06, |
| "loss": 1.2793, |
| "mean_token_accuracy": 0.6735174059867859, |
| "num_tokens": 28820510.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.9210526315789473, |
| "grad_norm": 0.5510081067492951, |
| "learning_rate": 5.046069172958794e-06, |
| "loss": 1.2383, |
| "mean_token_accuracy": 0.6795263290405273, |
| "num_tokens": 29082654.0, |
| "step": 111 |
| }, |
| { |
| "epoch": 2.9473684210526314, |
| "grad_norm": 0.5717504770508175, |
| "learning_rate": 4.963898609792994e-06, |
| "loss": 1.3184, |
| "mean_token_accuracy": 0.6652461290359497, |
| "num_tokens": 29344798.0, |
| "step": 112 |
| }, |
| { |
| "epoch": 2.973684210526316, |
| "grad_norm": 0.6116330301476128, |
| "learning_rate": 4.881908989817163e-06, |
| "loss": 1.1489, |
| "mean_token_accuracy": 0.6971180438995361, |
| "num_tokens": 29606942.0, |
| "step": 113 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.5899271048276149, |
| "learning_rate": 4.8001279859000074e-06, |
| "loss": 1.1643, |
| "mean_token_accuracy": 0.6946829557418823, |
| "num_tokens": 29861421.0, |
| "step": 114 |
| }, |
| { |
| "epoch": 3.026315789473684, |
| "grad_norm": 0.6052609643580157, |
| "learning_rate": 4.718583200498814e-06, |
| "loss": 1.2267, |
| "mean_token_accuracy": 0.6843944787979126, |
| "num_tokens": 30123565.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 3.0526315789473686, |
| "grad_norm": 0.5571230863291556, |
| "learning_rate": 4.637302156343178e-06, |
| "loss": 1.137, |
| "mean_token_accuracy": 0.7024783492088318, |
| "num_tokens": 30385709.0, |
| "step": 116 |
| }, |
| { |
| "epoch": 3.0789473684210527, |
| "grad_norm": 1.9895691560023896, |
| "learning_rate": 4.556312287145604e-06, |
| "loss": 1.151, |
| "mean_token_accuracy": 0.6993651390075684, |
| "num_tokens": 30647853.0, |
| "step": 117 |
| }, |
| { |
| "epoch": 3.1052631578947367, |
| "grad_norm": 1.091284264386462, |
| "learning_rate": 4.475640928342182e-06, |
| "loss": 1.132, |
| "mean_token_accuracy": 0.7014902234077454, |
| "num_tokens": 30909997.0, |
| "step": 118 |
| }, |
| { |
| "epoch": 3.1315789473684212, |
| "grad_norm": 0.6626943880051851, |
| "learning_rate": 4.395315307866404e-06, |
| "loss": 1.2015, |
| "mean_token_accuracy": 0.6918988823890686, |
| "num_tokens": 31172141.0, |
| "step": 119 |
| }, |
| { |
| "epoch": 3.1578947368421053, |
| "grad_norm": 0.6523541542451436, |
| "learning_rate": 4.315362536959305e-06, |
| "loss": 1.2112, |
| "mean_token_accuracy": 0.6867483854293823, |
| "num_tokens": 31434285.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 3.1842105263157894, |
| "grad_norm": 0.8841056815148688, |
| "learning_rate": 4.2358096010189385e-06, |
| "loss": 1.2752, |
| "mean_token_accuracy": 0.6752685904502869, |
| "num_tokens": 31696429.0, |
| "step": 121 |
| }, |
| { |
| "epoch": 3.2105263157894735, |
| "grad_norm": 0.6012402010833562, |
| "learning_rate": 4.156683350492376e-06, |
| "loss": 1.1933, |
| "mean_token_accuracy": 0.6922765970230103, |
| "num_tokens": 31958573.0, |
| "step": 122 |
| }, |
| { |
| "epoch": 3.236842105263158, |
| "grad_norm": 0.6163214035759135, |
| "learning_rate": 4.078010491813215e-06, |
| "loss": 1.2009, |
| "mean_token_accuracy": 0.6913189888000488, |
| "num_tokens": 32220717.0, |
| "step": 123 |
| }, |
| { |
| "epoch": 3.263157894736842, |
| "grad_norm": 0.5669681844522956, |
| "learning_rate": 3.999817578387723e-06, |
| "loss": 1.1801, |
| "mean_token_accuracy": 0.6925512552261353, |
| "num_tokens": 32482861.0, |
| "step": 124 |
| }, |
| { |
| "epoch": 3.2894736842105265, |
| "grad_norm": 0.5503406805003092, |
| "learning_rate": 3.922131001632607e-06, |
| "loss": 1.0835, |
| "mean_token_accuracy": 0.7118712663650513, |
| "num_tokens": 32745005.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 3.3157894736842106, |
| "grad_norm": 0.7031631763052849, |
| "learning_rate": 3.8449769820674824e-06, |
| "loss": 1.1845, |
| "mean_token_accuracy": 0.6908764243125916, |
| "num_tokens": 33007149.0, |
| "step": 126 |
| }, |
| { |
| "epoch": 3.3421052631578947, |
| "grad_norm": 0.5523930786032378, |
| "learning_rate": 3.7683815604649955e-06, |
| "loss": 1.1769, |
| "mean_token_accuracy": 0.694538950920105, |
| "num_tokens": 33269293.0, |
| "step": 127 |
| }, |
| { |
| "epoch": 3.3684210526315788, |
| "grad_norm": 0.5731302432477412, |
| "learning_rate": 3.692370589061639e-06, |
| "loss": 1.1629, |
| "mean_token_accuracy": 0.6966297030448914, |
| "num_tokens": 33531437.0, |
| "step": 128 |
| }, |
| { |
| "epoch": 3.3947368421052633, |
| "grad_norm": 0.5520020236878893, |
| "learning_rate": 3.616969722832174e-06, |
| "loss": 1.1374, |
| "mean_token_accuracy": 0.7009102702140808, |
| "num_tokens": 33793581.0, |
| "step": 129 |
| }, |
| { |
| "epoch": 3.4210526315789473, |
| "grad_norm": 2.5073145307120246, |
| "learning_rate": 3.542204410830672e-06, |
| "loss": 1.1954, |
| "mean_token_accuracy": 0.6902927160263062, |
| "num_tokens": 34055725.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 3.4473684210526314, |
| "grad_norm": 0.7447715458686521, |
| "learning_rate": 3.468099887600999e-06, |
| "loss": 1.1337, |
| "mean_token_accuracy": 0.6988958716392517, |
| "num_tokens": 34317869.0, |
| "step": 131 |
| }, |
| { |
| "epoch": 3.473684210526316, |
| "grad_norm": 0.55360781640575, |
| "learning_rate": 3.394681164659766e-06, |
| "loss": 1.1612, |
| "mean_token_accuracy": 0.693768322467804, |
| "num_tokens": 34580013.0, |
| "step": 132 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 0.6209441329712366, |
| "learning_rate": 3.3219730220545065e-06, |
| "loss": 1.1428, |
| "mean_token_accuracy": 0.6991400718688965, |
| "num_tokens": 34842157.0, |
| "step": 133 |
| }, |
| { |
| "epoch": 3.526315789473684, |
| "grad_norm": 0.5778728779594863, |
| "learning_rate": 3.2500000000000015e-06, |
| "loss": 1.1229, |
| "mean_token_accuracy": 0.7043516039848328, |
| "num_tokens": 35104301.0, |
| "step": 134 |
| }, |
| { |
| "epoch": 3.5526315789473686, |
| "grad_norm": 1.0328742077694102, |
| "learning_rate": 3.178786390595544e-06, |
| "loss": 1.0963, |
| "mean_token_accuracy": 0.709181547164917, |
| "num_tokens": 35366445.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 3.5789473684210527, |
| "grad_norm": 0.7214124791993152, |
| "learning_rate": 3.1083562296259494e-06, |
| "loss": 1.1063, |
| "mean_token_accuracy": 0.7070298194885254, |
| "num_tokens": 35628589.0, |
| "step": 136 |
| }, |
| { |
| "epoch": 3.6052631578947367, |
| "grad_norm": 0.5288818064176912, |
| "learning_rate": 3.0387332884490806e-06, |
| "loss": 1.095, |
| "mean_token_accuracy": 0.7074266076087952, |
| "num_tokens": 35890733.0, |
| "step": 137 |
| }, |
| { |
| "epoch": 3.6315789473684212, |
| "grad_norm": 0.5389789995349765, |
| "learning_rate": 2.9699410659726076e-06, |
| "loss": 1.1101, |
| "mean_token_accuracy": 0.7031764984130859, |
| "num_tokens": 36152877.0, |
| "step": 138 |
| }, |
| { |
| "epoch": 3.6578947368421053, |
| "grad_norm": 0.5522512649782974, |
| "learning_rate": 2.9020027807227426e-06, |
| "loss": 1.2377, |
| "mean_token_accuracy": 0.6814453601837158, |
| "num_tokens": 36415021.0, |
| "step": 139 |
| }, |
| { |
| "epoch": 3.6842105263157894, |
| "grad_norm": 0.5402428472844846, |
| "learning_rate": 2.8349413630075907e-06, |
| "loss": 1.1458, |
| "mean_token_accuracy": 0.6977704167366028, |
| "num_tokens": 36677165.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 3.7105263157894735, |
| "grad_norm": 0.5940564905132736, |
| "learning_rate": 2.7687794471777958e-06, |
| "loss": 1.1514, |
| "mean_token_accuracy": 0.6953821182250977, |
| "num_tokens": 36939309.0, |
| "step": 141 |
| }, |
| { |
| "epoch": 3.736842105263158, |
| "grad_norm": 0.5734295274738652, |
| "learning_rate": 2.70353936398706e-06, |
| "loss": 1.1403, |
| "mean_token_accuracy": 0.6989117860794067, |
| "num_tokens": 37193788.0, |
| "step": 142 |
| }, |
| { |
| "epoch": 3.763157894736842, |
| "grad_norm": 0.532059870308926, |
| "learning_rate": 2.639243133055145e-06, |
| "loss": 1.2474, |
| "mean_token_accuracy": 0.680209219455719, |
| "num_tokens": 37455932.0, |
| "step": 143 |
| }, |
| { |
| "epoch": 3.7894736842105265, |
| "grad_norm": 0.5510330430152282, |
| "learning_rate": 2.575912455435883e-06, |
| "loss": 1.2086, |
| "mean_token_accuracy": 0.688400387763977, |
| "num_tokens": 37718076.0, |
| "step": 144 |
| }, |
| { |
| "epoch": 3.8157894736842106, |
| "grad_norm": 0.5658132485612524, |
| "learning_rate": 2.513568706292705e-06, |
| "loss": 1.2624, |
| "mean_token_accuracy": 0.6760048866271973, |
| "num_tokens": 37980220.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 3.8421052631578947, |
| "grad_norm": 0.5251146314121506, |
| "learning_rate": 2.4522329276841664e-06, |
| "loss": 1.2014, |
| "mean_token_accuracy": 0.6878548264503479, |
| "num_tokens": 38242364.0, |
| "step": 146 |
| }, |
| { |
| "epoch": 3.8684210526315788, |
| "grad_norm": 0.5357295896942552, |
| "learning_rate": 2.391925821461901e-06, |
| "loss": 1.1545, |
| "mean_token_accuracy": 0.6958895325660706, |
| "num_tokens": 38504508.0, |
| "step": 147 |
| }, |
| { |
| "epoch": 3.8947368421052633, |
| "grad_norm": 0.5705951115408219, |
| "learning_rate": 2.3326677422833992e-06, |
| "loss": 1.1662, |
| "mean_token_accuracy": 0.6927115321159363, |
| "num_tokens": 38766652.0, |
| "step": 148 |
| }, |
| { |
| "epoch": 3.9210526315789473, |
| "grad_norm": 0.5394182117597536, |
| "learning_rate": 2.2744786907419704e-06, |
| "loss": 1.1902, |
| "mean_token_accuracy": 0.6899531483650208, |
| "num_tokens": 39028796.0, |
| "step": 149 |
| }, |
| { |
| "epoch": 3.9473684210526314, |
| "grad_norm": 0.547923373437002, |
| "learning_rate": 2.2173783066162076e-06, |
| "loss": 1.1707, |
| "mean_token_accuracy": 0.6980451345443726, |
| "num_tokens": 39290940.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 3.973684210526316, |
| "grad_norm": 0.5518777404038198, |
| "learning_rate": 2.1613858622412336e-06, |
| "loss": 1.1786, |
| "mean_token_accuracy": 0.6942414045333862, |
| "num_tokens": 39553084.0, |
| "step": 151 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.5521786822039609, |
| "learning_rate": 2.1065202560039678e-06, |
| "loss": 1.1497, |
| "mean_token_accuracy": 0.699094295501709, |
| "num_tokens": 39815228.0, |
| "step": 152 |
| }, |
| { |
| "epoch": 4.026315789473684, |
| "grad_norm": 0.5822084229656587, |
| "learning_rate": 2.0528000059646e-06, |
| "loss": 1.1643, |
| "mean_token_accuracy": 0.6988691687583923, |
| "num_tokens": 40077372.0, |
| "step": 153 |
| }, |
| { |
| "epoch": 4.052631578947368, |
| "grad_norm": 0.5350433596377399, |
| "learning_rate": 2.0002432436064385e-06, |
| "loss": 1.1162, |
| "mean_token_accuracy": 0.7054846882820129, |
| "num_tokens": 40339516.0, |
| "step": 154 |
| }, |
| { |
| "epoch": 4.078947368421052, |
| "grad_norm": 0.5518812481108508, |
| "learning_rate": 1.94886770771623e-06, |
| "loss": 1.1493, |
| "mean_token_accuracy": 0.6981443166732788, |
| "num_tokens": 40601660.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 4.105263157894737, |
| "grad_norm": 0.5480415544046374, |
| "learning_rate": 1.8986907383970228e-06, |
| "loss": 1.1315, |
| "mean_token_accuracy": 0.7005326151847839, |
| "num_tokens": 40863804.0, |
| "step": 156 |
| }, |
| { |
| "epoch": 4.131578947368421, |
| "grad_norm": 0.7656041762600009, |
| "learning_rate": 1.8497292712155907e-06, |
| "loss": 1.1118, |
| "mean_token_accuracy": 0.7078768014907837, |
| "num_tokens": 41125948.0, |
| "step": 157 |
| }, |
| { |
| "epoch": 4.157894736842105, |
| "grad_norm": 0.5497688927341485, |
| "learning_rate": 1.8019998314863974e-06, |
| "loss": 1.2162, |
| "mean_token_accuracy": 0.6840701699256897, |
| "num_tokens": 41388092.0, |
| "step": 158 |
| }, |
| { |
| "epoch": 4.184210526315789, |
| "grad_norm": 0.5634861823726028, |
| "learning_rate": 1.7555185286940252e-06, |
| "loss": 1.1001, |
| "mean_token_accuracy": 0.7082964777946472, |
| "num_tokens": 41650236.0, |
| "step": 159 |
| }, |
| { |
| "epoch": 4.2105263157894735, |
| "grad_norm": 0.5562593279549117, |
| "learning_rate": 1.710301051055955e-06, |
| "loss": 1.1233, |
| "mean_token_accuracy": 0.706102728843689, |
| "num_tokens": 41912380.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 4.2368421052631575, |
| "grad_norm": 0.5649313028818561, |
| "learning_rate": 1.666362660227529e-06, |
| "loss": 1.1423, |
| "mean_token_accuracy": 0.6994147300720215, |
| "num_tokens": 42174524.0, |
| "step": 161 |
| }, |
| { |
| "epoch": 4.2631578947368425, |
| "grad_norm": 0.8616736787357053, |
| "learning_rate": 1.6237181861508889e-06, |
| "loss": 1.1505, |
| "mean_token_accuracy": 0.6955958008766174, |
| "num_tokens": 42436668.0, |
| "step": 162 |
| }, |
| { |
| "epoch": 4.2894736842105265, |
| "grad_norm": 0.8986161306924315, |
| "learning_rate": 1.5823820220496197e-06, |
| "loss": 1.1581, |
| "mean_token_accuracy": 0.6982167959213257, |
| "num_tokens": 42698812.0, |
| "step": 163 |
| }, |
| { |
| "epoch": 4.315789473684211, |
| "grad_norm": 0.5467975548359034, |
| "learning_rate": 1.5423681195707997e-06, |
| "loss": 1.1762, |
| "mean_token_accuracy": 0.6948594450950623, |
| "num_tokens": 42960956.0, |
| "step": 164 |
| }, |
| { |
| "epoch": 4.342105263157895, |
| "grad_norm": 0.5361701527040977, |
| "learning_rate": 1.5036899840760815e-06, |
| "loss": 1.0706, |
| "mean_token_accuracy": 0.7150683403015137, |
| "num_tokens": 43223100.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 4.368421052631579, |
| "grad_norm": 0.6682056700117512, |
| "learning_rate": 1.4663606700834102e-06, |
| "loss": 1.1245, |
| "mean_token_accuracy": 0.7043439149856567, |
| "num_tokens": 43485244.0, |
| "step": 166 |
| }, |
| { |
| "epoch": 4.394736842105263, |
| "grad_norm": 0.5358865574721129, |
| "learning_rate": 1.4303927768609016e-06, |
| "loss": 1.1313, |
| "mean_token_accuracy": 0.7052977085113525, |
| "num_tokens": 43747388.0, |
| "step": 167 |
| }, |
| { |
| "epoch": 4.421052631578947, |
| "grad_norm": 0.525962123793606, |
| "learning_rate": 1.3957984441743832e-06, |
| "loss": 1.1384, |
| "mean_token_accuracy": 0.7020624876022339, |
| "num_tokens": 44009532.0, |
| "step": 168 |
| }, |
| { |
| "epoch": 4.447368421052632, |
| "grad_norm": 0.5796984218177971, |
| "learning_rate": 1.3625893481900155e-06, |
| "loss": 1.0882, |
| "mean_token_accuracy": 0.7103375792503357, |
| "num_tokens": 44271676.0, |
| "step": 169 |
| }, |
| { |
| "epoch": 4.473684210526316, |
| "grad_norm": 0.5393374011777778, |
| "learning_rate": 1.3307766975333922e-06, |
| "loss": 1.0466, |
| "mean_token_accuracy": 0.7210162281990051, |
| "num_tokens": 44533820.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 0.5349090159384385, |
| "learning_rate": 1.3003712295064433e-06, |
| "loss": 1.1418, |
| "mean_token_accuracy": 0.6994032859802246, |
| "num_tokens": 44795964.0, |
| "step": 171 |
| }, |
| { |
| "epoch": 4.526315789473684, |
| "grad_norm": 0.5057581166787403, |
| "learning_rate": 1.2713832064634127e-06, |
| "loss": 1.0858, |
| "mean_token_accuracy": 0.7137597799301147, |
| "num_tokens": 45058108.0, |
| "step": 172 |
| }, |
| { |
| "epoch": 4.552631578947368, |
| "grad_norm": 0.7631542771888906, |
| "learning_rate": 1.2438224123471442e-06, |
| "loss": 1.1672, |
| "mean_token_accuracy": 0.6955614686012268, |
| "num_tokens": 45320252.0, |
| "step": 173 |
| }, |
| { |
| "epoch": 4.578947368421053, |
| "grad_norm": 0.5440066367064722, |
| "learning_rate": 1.2176981493868374e-06, |
| "loss": 1.139, |
| "mean_token_accuracy": 0.7027553915977478, |
| "num_tokens": 45574731.0, |
| "step": 174 |
| }, |
| { |
| "epoch": 4.605263157894737, |
| "grad_norm": 0.518246804782426, |
| "learning_rate": 1.1930192349583866e-06, |
| "loss": 1.1575, |
| "mean_token_accuracy": 0.6979191899299622, |
| "num_tokens": 45836875.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 4.631578947368421, |
| "grad_norm": 0.5364733989201356, |
| "learning_rate": 1.1697939986083732e-06, |
| "loss": 1.0587, |
| "mean_token_accuracy": 0.7159839868545532, |
| "num_tokens": 46099019.0, |
| "step": 176 |
| }, |
| { |
| "epoch": 4.657894736842105, |
| "grad_norm": 0.5170966910701931, |
| "learning_rate": 1.1480302792427021e-06, |
| "loss": 1.0986, |
| "mean_token_accuracy": 0.709250271320343, |
| "num_tokens": 46361163.0, |
| "step": 177 |
| }, |
| { |
| "epoch": 4.684210526315789, |
| "grad_norm": 0.5157074509948011, |
| "learning_rate": 1.127735422480838e-06, |
| "loss": 1.0903, |
| "mean_token_accuracy": 0.7117949724197388, |
| "num_tokens": 46623307.0, |
| "step": 178 |
| }, |
| { |
| "epoch": 4.7105263157894735, |
| "grad_norm": 0.5766154191021017, |
| "learning_rate": 1.10891627817654e-06, |
| "loss": 1.0934, |
| "mean_token_accuracy": 0.709181547164917, |
| "num_tokens": 46885451.0, |
| "step": 179 |
| }, |
| { |
| "epoch": 4.7368421052631575, |
| "grad_norm": 0.5135486472483117, |
| "learning_rate": 1.0915791981059174e-06, |
| "loss": 1.1199, |
| "mean_token_accuracy": 0.7034855484962463, |
| "num_tokens": 47147595.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 4.7631578947368425, |
| "grad_norm": 0.7547407302261836, |
| "learning_rate": 1.0757300338236027e-06, |
| "loss": 1.1165, |
| "mean_token_accuracy": 0.7076631188392639, |
| "num_tokens": 47409739.0, |
| "step": 181 |
| }, |
| { |
| "epoch": 4.7894736842105265, |
| "grad_norm": 0.5690959613109913, |
| "learning_rate": 1.0613741346877498e-06, |
| "loss": 1.1585, |
| "mean_token_accuracy": 0.6951303482055664, |
| "num_tokens": 47671883.0, |
| "step": 182 |
| }, |
| { |
| "epoch": 4.815789473684211, |
| "grad_norm": 0.5277151826124948, |
| "learning_rate": 1.0485163460545396e-06, |
| "loss": 1.1409, |
| "mean_token_accuracy": 0.6986250281333923, |
| "num_tokens": 47934027.0, |
| "step": 183 |
| }, |
| { |
| "epoch": 4.842105263157895, |
| "grad_norm": 0.5192139515319969, |
| "learning_rate": 1.03716100764279e-06, |
| "loss": 1.165, |
| "mean_token_accuracy": 0.6956835389137268, |
| "num_tokens": 48196171.0, |
| "step": 184 |
| }, |
| { |
| "epoch": 4.868421052631579, |
| "grad_norm": 0.539511614407237, |
| "learning_rate": 1.0273119520692274e-06, |
| "loss": 1.1842, |
| "mean_token_accuracy": 0.6949090361595154, |
| "num_tokens": 48458315.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 4.894736842105263, |
| "grad_norm": 0.5375538776614278, |
| "learning_rate": 1.0189725035549165e-06, |
| "loss": 1.1754, |
| "mean_token_accuracy": 0.6964083909988403, |
| "num_tokens": 48720459.0, |
| "step": 186 |
| }, |
| { |
| "epoch": 4.921052631578947, |
| "grad_norm": 0.5066546944417655, |
| "learning_rate": 1.0121454768032772e-06, |
| "loss": 1.1248, |
| "mean_token_accuracy": 0.7042332887649536, |
| "num_tokens": 48982603.0, |
| "step": 187 |
| }, |
| { |
| "epoch": 4.947368421052632, |
| "grad_norm": 0.516984174228912, |
| "learning_rate": 1.0068331760500773e-06, |
| "loss": 1.1127, |
| "mean_token_accuracy": 0.705129861831665, |
| "num_tokens": 49244747.0, |
| "step": 188 |
| }, |
| { |
| "epoch": 4.973684210526316, |
| "grad_norm": 0.5078294598491031, |
| "learning_rate": 1.0030373942857136e-06, |
| "loss": 1.0498, |
| "mean_token_accuracy": 0.7175978422164917, |
| "num_tokens": 49506891.0, |
| "step": 189 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.5042595032615071, |
| "learning_rate": 1.000759412650048e-06, |
| "loss": 1.1777, |
| "mean_token_accuracy": 0.694607675075531, |
| "num_tokens": 49769035.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 5.0, |
| "step": 190, |
| "total_flos": 39782134579200.0, |
| "train_loss": 1.4404595375061036, |
| "train_runtime": 2942.5505, |
| "train_samples_per_second": 2.048, |
| "train_steps_per_second": 0.065 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 190, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 39782134579200.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
