| [ |
| { |
| "loss": 1.4019, |
| "grad_norm": 0.6986930966377258, |
| "learning_rate": 2.25e-05, |
| "entropy": 1.199985957145691, |
| "num_tokens": 647929.0, |
| "mean_token_accuracy": 0.6914159089326859, |
| "epoch": 0.19607843137254902, |
| "step": 10 |
| }, |
| { |
| "loss": 1.3734, |
| "grad_norm": 0.18671388924121857, |
| "learning_rate": 4.75e-05, |
| "entropy": 1.325679862499237, |
| "num_tokens": 1300854.0, |
| "mean_token_accuracy": 0.6935183644294739, |
| "epoch": 0.39215686274509803, |
| "step": 20 |
| }, |
| { |
| "loss": 1.3317, |
| "grad_norm": 0.18022498488426208, |
| "learning_rate": 4.9819267987317665e-05, |
| "entropy": 1.4015559554100037, |
| "num_tokens": 1947357.0, |
| "mean_token_accuracy": 0.6976928591728211, |
| "epoch": 0.5882352941176471, |
| "step": 30 |
| }, |
| { |
| "loss": 1.2446, |
| "grad_norm": 0.18924492597579956, |
| "learning_rate": 4.9197869469162815e-05, |
| "entropy": 1.219199287891388, |
| "num_tokens": 2595628.0, |
| "mean_token_accuracy": 0.7132954180240632, |
| "epoch": 0.7843137254901961, |
| "step": 40 |
| }, |
| { |
| "loss": 1.2139, |
| "grad_norm": 0.23499619960784912, |
| "learning_rate": 4.814465939707259e-05, |
| "entropy": 1.2155689418315887, |
| "num_tokens": 3245496.0, |
| "mean_token_accuracy": 0.7180316507816314, |
| "epoch": 0.9803921568627451, |
| "step": 50 |
| }, |
| { |
| "loss": 1.1282, |
| "grad_norm": 0.25000059604644775, |
| "learning_rate": 4.6678432329734434e-05, |
| "entropy": 1.1309684544801712, |
| "num_tokens": 3876930.0, |
| "mean_token_accuracy": 0.734208858013153, |
| "epoch": 1.1764705882352942, |
| "step": 60 |
| }, |
| { |
| "loss": 1.1264, |
| "grad_norm": 0.3251831531524658, |
| "learning_rate": 4.482535312390058e-05, |
| "entropy": 1.1345468521118165, |
| "num_tokens": 4525171.0, |
| "mean_token_accuracy": 0.7357972204685211, |
| "epoch": 1.3725490196078431, |
| "step": 70 |
| }, |
| { |
| "loss": 1.0226, |
| "grad_norm": 0.2027675360441208, |
| "learning_rate": 4.2618490021899384e-05, |
| "entropy": 1.0273457378149033, |
| "num_tokens": 5172588.0, |
| "mean_token_accuracy": 0.7597126334905624, |
| "epoch": 1.5686274509803921, |
| "step": 80 |
| }, |
| { |
| "loss": 1.0134, |
| "grad_norm": 0.2039007991552353, |
| "learning_rate": 4.009722454806761e-05, |
| "entropy": 1.0195463865995407, |
| "num_tokens": 5819977.0, |
| "mean_token_accuracy": 0.761520317196846, |
| "epoch": 1.7647058823529411, |
| "step": 90 |
| }, |
| { |
| "loss": 0.9796, |
| "grad_norm": 0.1867203265428543, |
| "learning_rate": 3.730654874451569e-05, |
| "entropy": 0.9862680763006211, |
| "num_tokens": 6472798.0, |
| "mean_token_accuracy": 0.7683205276727676, |
| "epoch": 1.9607843137254903, |
| "step": 100 |
| }, |
| { |
| "loss": 0.9679, |
| "grad_norm": 0.21607249975204468, |
| "learning_rate": 3.429626228707034e-05, |
| "entropy": 0.9721104234457016, |
| "num_tokens": 7112529.0, |
| "mean_token_accuracy": 0.7711095601320267, |
| "epoch": 2.156862745098039, |
| "step": 110 |
| }, |
| { |
| "loss": 0.9666, |
| "grad_norm": 0.23092280328273773, |
| "learning_rate": 3.112008380887966e-05, |
| "entropy": 0.9724053025245667, |
| "num_tokens": 7758907.0, |
| "mean_token_accuracy": 0.7709219127893447, |
| "epoch": 2.3529411764705883, |
| "step": 120 |
| }, |
| { |
| "loss": 0.9683, |
| "grad_norm": 0.20999926328659058, |
| "learning_rate": 2.7834692290132052e-05, |
| "entropy": 0.9730658024549484, |
| "num_tokens": 8405368.0, |
| "mean_token_accuracy": 0.769457995891571, |
| "epoch": 2.549019607843137, |
| "step": 130 |
| }, |
| { |
| "loss": 0.9385, |
| "grad_norm": 0.7688829898834229, |
| "learning_rate": 2.449871562031194e-05, |
| "entropy": 0.9452048629522324, |
| "num_tokens": 9053347.0, |
| "mean_token_accuracy": 0.7750110507011414, |
| "epoch": 2.7450980392156863, |
| "step": 140 |
| }, |
| { |
| "loss": 0.9832, |
| "grad_norm": 0.26651284098625183, |
| "learning_rate": 2.1171684382123e-05, |
| "entropy": 0.9904936224222183, |
| "num_tokens": 9703053.0, |
| "mean_token_accuracy": 0.7652157843112946, |
| "epoch": 2.9411764705882355, |
| "step": 150 |
| }, |
| { |
| "loss": 0.9586, |
| "grad_norm": 0.22833150625228882, |
| "learning_rate": 1.7912969526829558e-05, |
| "entropy": 0.9607909858226776, |
| "num_tokens": 10340639.0, |
| "mean_token_accuracy": 0.7710874438285827, |
| "epoch": 3.1372549019607843, |
| "step": 160 |
| }, |
| { |
| "loss": 0.9382, |
| "grad_norm": 0.2782645523548126, |
| "learning_rate": 1.4780722898224707e-05, |
| "entropy": 0.9437746345996857, |
| "num_tokens": 10987284.0, |
| "mean_token_accuracy": 0.7756666958332061, |
| "epoch": 3.3333333333333335, |
| "step": 170 |
| }, |
| { |
| "loss": 0.9441, |
| "grad_norm": 0.2571350038051605, |
| "learning_rate": 1.1830839511600211e-05, |
| "entropy": 0.9467583298683167, |
| "num_tokens": 11638213.0, |
| "mean_token_accuracy": 0.7747708618640899, |
| "epoch": 3.5294117647058822, |
| "step": 180 |
| }, |
| { |
| "loss": 0.9175, |
| "grad_norm": 0.26882949471473694, |
| "learning_rate": 9.11596010587441e-06, |
| "entropy": 0.923730057477951, |
| "num_tokens": 12288333.0, |
| "mean_token_accuracy": 0.7791785061359405, |
| "epoch": 3.7254901960784315, |
| "step": 190 |
| }, |
| { |
| "loss": 0.91, |
| "grad_norm": 0.22646215558052063, |
| "learning_rate": 6.684531768359173e-06, |
| "entropy": 0.9165982186794281, |
| "num_tokens": 12936318.0, |
| "mean_token_accuracy": 0.7812086254358291, |
| "epoch": 3.9215686274509802, |
| "step": 200 |
| }, |
| { |
| "loss": 0.9193, |
| "grad_norm": 0.2461111843585968, |
| "learning_rate": 4.579943395339062e-06, |
| "entropy": 0.9269091933965683, |
| "num_tokens": 13570476.0, |
| "mean_token_accuracy": 0.7785391122102737, |
| "epoch": 4.117647058823529, |
| "step": 210 |
| }, |
| { |
| "loss": 0.9259, |
| "grad_norm": 0.30823367834091187, |
| "learning_rate": 2.8397514161892486e-06, |
| "entropy": 0.9287486761808396, |
| "num_tokens": 14217790.0, |
| "mean_token_accuracy": 0.7776476472616196, |
| "epoch": 4.313725490196078, |
| "step": 220 |
| }, |
| { |
| "loss": 0.8904, |
| "grad_norm": 0.29259830713272095, |
| "learning_rate": 1.4950095980035772e-06, |
| "entropy": 0.8954251229763031, |
| "num_tokens": 14867755.0, |
| "mean_token_accuracy": 0.7863523244857789, |
| "epoch": 4.509803921568627, |
| "step": 230 |
| }, |
| { |
| "loss": 0.9222, |
| "grad_norm": 0.2476108819246292, |
| "learning_rate": 5.697148903850868e-07, |
| "entropy": 0.92759590446949, |
| "num_tokens": 15512923.0, |
| "mean_token_accuracy": 0.7787990540266037, |
| "epoch": 4.705882352941177, |
| "step": 240 |
| }, |
| { |
| "loss": 0.9369, |
| "grad_norm": 0.2094818353652954, |
| "learning_rate": 8.037919931187244e-08, |
| "entropy": 0.9443553179502487, |
| "num_tokens": 16165639.0, |
| "mean_token_accuracy": 0.7741728782653808, |
| "epoch": 4.901960784313726, |
| "step": 250 |
| }, |
| { |
| "train_runtime": 4077.9021, |
| "train_samples_per_second": 1.994, |
| "train_steps_per_second": 0.063, |
| "total_flos": 4.655305572168499e+18, |
| "train_loss": 1.0347469572927437, |
| "entropy": 0.9348823964595795, |
| "num_tokens": 16478955.0, |
| "mean_token_accuracy": 0.7756926536560058, |
| "epoch": 5.0, |
| "step": 255 |
| } |
| ] |
