| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7839294463498285, | |
| "eval_steps": 100, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "entropy": 2.6391387462615965, | |
| "epoch": 0.009799118079372856, | |
| "grad_norm": 7.71875, | |
| "learning_rate": 9.000000000000001e-07, | |
| "loss": 0.7942, | |
| "mean_token_accuracy": 0.8082588076591491, | |
| "num_tokens": 43694.0, | |
| "step": 10 | |
| }, | |
| { | |
| "entropy": 2.6823359608650206, | |
| "epoch": 0.019598236158745713, | |
| "grad_norm": 6.375, | |
| "learning_rate": 1.9000000000000002e-06, | |
| "loss": 0.7564, | |
| "mean_token_accuracy": 0.8178666859865189, | |
| "num_tokens": 87464.0, | |
| "step": 20 | |
| }, | |
| { | |
| "entropy": 2.7048767805099487, | |
| "epoch": 0.02939735423811857, | |
| "grad_norm": 6.09375, | |
| "learning_rate": 2.9e-06, | |
| "loss": 0.6878, | |
| "mean_token_accuracy": 0.8223719596862793, | |
| "num_tokens": 128763.0, | |
| "step": 30 | |
| }, | |
| { | |
| "entropy": 2.644310915470123, | |
| "epoch": 0.039196472317491425, | |
| "grad_norm": 7.21875, | |
| "learning_rate": 3.900000000000001e-06, | |
| "loss": 0.6425, | |
| "mean_token_accuracy": 0.8384180337190628, | |
| "num_tokens": 171120.0, | |
| "step": 40 | |
| }, | |
| { | |
| "entropy": 2.7310761332511904, | |
| "epoch": 0.04899559039686428, | |
| "grad_norm": 5.25, | |
| "learning_rate": 4.9000000000000005e-06, | |
| "loss": 0.7267, | |
| "mean_token_accuracy": 0.8242274433374405, | |
| "num_tokens": 214474.0, | |
| "step": 50 | |
| }, | |
| { | |
| "entropy": 2.6951672077178954, | |
| "epoch": 0.05879470847623714, | |
| "grad_norm": 5.28125, | |
| "learning_rate": 5.9e-06, | |
| "loss": 0.7029, | |
| "mean_token_accuracy": 0.8294231444597244, | |
| "num_tokens": 257391.0, | |
| "step": 60 | |
| }, | |
| { | |
| "entropy": 2.7150769352912905, | |
| "epoch": 0.06859382655561, | |
| "grad_norm": 6.28125, | |
| "learning_rate": 6.9e-06, | |
| "loss": 0.695, | |
| "mean_token_accuracy": 0.8301586806774139, | |
| "num_tokens": 299461.0, | |
| "step": 70 | |
| }, | |
| { | |
| "entropy": 2.6886049628257753, | |
| "epoch": 0.07839294463498285, | |
| "grad_norm": 7.34375, | |
| "learning_rate": 7.9e-06, | |
| "loss": 0.6883, | |
| "mean_token_accuracy": 0.8331334680318833, | |
| "num_tokens": 342024.0, | |
| "step": 80 | |
| }, | |
| { | |
| "entropy": 2.782685649394989, | |
| "epoch": 0.0881920627143557, | |
| "grad_norm": 6.03125, | |
| "learning_rate": 8.900000000000001e-06, | |
| "loss": 0.7174, | |
| "mean_token_accuracy": 0.824739807844162, | |
| "num_tokens": 382420.0, | |
| "step": 90 | |
| }, | |
| { | |
| "entropy": 2.801072883605957, | |
| "epoch": 0.09799118079372857, | |
| "grad_norm": 6.125, | |
| "learning_rate": 9.9e-06, | |
| "loss": 0.6569, | |
| "mean_token_accuracy": 0.8391472637653351, | |
| "num_tokens": 422978.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09799118079372857, | |
| "eval_entropy": 2.77619282156229, | |
| "eval_loss": 0.686501145362854, | |
| "eval_mean_token_accuracy": 0.8315936047583818, | |
| "eval_num_tokens": 422978.0, | |
| "eval_runtime": 24.2597, | |
| "eval_samples_per_second": 10.429, | |
| "eval_steps_per_second": 1.319, | |
| "step": 100 | |
| }, | |
| { | |
| "entropy": 2.6945799708366396, | |
| "epoch": 0.10779029887310142, | |
| "grad_norm": 6.84375, | |
| "learning_rate": 9.90228013029316e-06, | |
| "loss": 0.685, | |
| "mean_token_accuracy": 0.8305380374193192, | |
| "num_tokens": 466727.0, | |
| "step": 110 | |
| }, | |
| { | |
| "entropy": 2.8276916265487673, | |
| "epoch": 0.11758941695247428, | |
| "grad_norm": 4.8125, | |
| "learning_rate": 9.79370249728556e-06, | |
| "loss": 0.6494, | |
| "mean_token_accuracy": 0.8401007294654846, | |
| "num_tokens": 508416.0, | |
| "step": 120 | |
| }, | |
| { | |
| "entropy": 2.7822465658187867, | |
| "epoch": 0.12738853503184713, | |
| "grad_norm": 4.03125, | |
| "learning_rate": 9.68512486427796e-06, | |
| "loss": 0.6453, | |
| "mean_token_accuracy": 0.8334015339612961, | |
| "num_tokens": 550543.0, | |
| "step": 130 | |
| }, | |
| { | |
| "entropy": 2.8134533524513246, | |
| "epoch": 0.13718765311122, | |
| "grad_norm": 8.8125, | |
| "learning_rate": 9.576547231270358e-06, | |
| "loss": 0.6241, | |
| "mean_token_accuracy": 0.842885109782219, | |
| "num_tokens": 593308.0, | |
| "step": 140 | |
| }, | |
| { | |
| "entropy": 2.883612883090973, | |
| "epoch": 0.14698677119059284, | |
| "grad_norm": 7.0625, | |
| "learning_rate": 9.467969598262759e-06, | |
| "loss": 0.6454, | |
| "mean_token_accuracy": 0.8373580127954483, | |
| "num_tokens": 634500.0, | |
| "step": 150 | |
| }, | |
| { | |
| "entropy": 2.8347164511680605, | |
| "epoch": 0.1567858892699657, | |
| "grad_norm": 4.65625, | |
| "learning_rate": 9.359391965255158e-06, | |
| "loss": 0.6059, | |
| "mean_token_accuracy": 0.8468258231878281, | |
| "num_tokens": 676536.0, | |
| "step": 160 | |
| }, | |
| { | |
| "entropy": 2.84292151927948, | |
| "epoch": 0.16658500734933857, | |
| "grad_norm": 5.78125, | |
| "learning_rate": 9.250814332247557e-06, | |
| "loss": 0.6551, | |
| "mean_token_accuracy": 0.8365038365125657, | |
| "num_tokens": 717263.0, | |
| "step": 170 | |
| }, | |
| { | |
| "entropy": 2.8945748567581178, | |
| "epoch": 0.1763841254287114, | |
| "grad_norm": 6.15625, | |
| "learning_rate": 9.142236699239957e-06, | |
| "loss": 0.5711, | |
| "mean_token_accuracy": 0.8484834253787994, | |
| "num_tokens": 759038.0, | |
| "step": 180 | |
| }, | |
| { | |
| "entropy": 2.8259629607200623, | |
| "epoch": 0.18618324350808427, | |
| "grad_norm": 6.0625, | |
| "learning_rate": 9.033659066232356e-06, | |
| "loss": 0.6519, | |
| "mean_token_accuracy": 0.8436094433069229, | |
| "num_tokens": 802132.0, | |
| "step": 190 | |
| }, | |
| { | |
| "entropy": 2.8300761938095094, | |
| "epoch": 0.19598236158745713, | |
| "grad_norm": 4.5625, | |
| "learning_rate": 8.925081433224755e-06, | |
| "loss": 0.5844, | |
| "mean_token_accuracy": 0.8484363347291947, | |
| "num_tokens": 845920.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19598236158745713, | |
| "eval_entropy": 2.859285779297352, | |
| "eval_loss": 0.6323556303977966, | |
| "eval_mean_token_accuracy": 0.8397621288895607, | |
| "eval_num_tokens": 845920.0, | |
| "eval_runtime": 19.1952, | |
| "eval_samples_per_second": 13.18, | |
| "eval_steps_per_second": 1.667, | |
| "step": 200 | |
| }, | |
| { | |
| "entropy": 2.8886484503746033, | |
| "epoch": 0.20578147966683, | |
| "grad_norm": 4.5, | |
| "learning_rate": 8.816503800217156e-06, | |
| "loss": 0.5966, | |
| "mean_token_accuracy": 0.843154439330101, | |
| "num_tokens": 889117.0, | |
| "step": 210 | |
| }, | |
| { | |
| "entropy": 2.7586705803871157, | |
| "epoch": 0.21558059774620283, | |
| "grad_norm": 4.34375, | |
| "learning_rate": 8.707926167209557e-06, | |
| "loss": 0.6571, | |
| "mean_token_accuracy": 0.8334330022335052, | |
| "num_tokens": 932966.0, | |
| "step": 220 | |
| }, | |
| { | |
| "entropy": 2.7509212732315063, | |
| "epoch": 0.2253797158255757, | |
| "grad_norm": 11.0, | |
| "learning_rate": 8.599348534201956e-06, | |
| "loss": 0.6431, | |
| "mean_token_accuracy": 0.8375909745693206, | |
| "num_tokens": 976729.0, | |
| "step": 230 | |
| }, | |
| { | |
| "entropy": 2.848712170124054, | |
| "epoch": 0.23517883390494856, | |
| "grad_norm": 9.25, | |
| "learning_rate": 8.490770901194355e-06, | |
| "loss": 0.6426, | |
| "mean_token_accuracy": 0.8388061553239823, | |
| "num_tokens": 1018220.0, | |
| "step": 240 | |
| }, | |
| { | |
| "entropy": 2.7517650604248045, | |
| "epoch": 0.2449779519843214, | |
| "grad_norm": 7.375, | |
| "learning_rate": 8.382193268186755e-06, | |
| "loss": 0.6136, | |
| "mean_token_accuracy": 0.8406570553779602, | |
| "num_tokens": 1060548.0, | |
| "step": 250 | |
| }, | |
| { | |
| "entropy": 2.813077485561371, | |
| "epoch": 0.25477707006369427, | |
| "grad_norm": 5.59375, | |
| "learning_rate": 8.273615635179154e-06, | |
| "loss": 0.6356, | |
| "mean_token_accuracy": 0.8369913339614868, | |
| "num_tokens": 1104019.0, | |
| "step": 260 | |
| }, | |
| { | |
| "entropy": 2.862120878696442, | |
| "epoch": 0.2645761881430671, | |
| "grad_norm": 5.90625, | |
| "learning_rate": 8.165038002171553e-06, | |
| "loss": 0.6128, | |
| "mean_token_accuracy": 0.8411767065525055, | |
| "num_tokens": 1145164.0, | |
| "step": 270 | |
| }, | |
| { | |
| "entropy": 2.7856096982955934, | |
| "epoch": 0.27437530622244, | |
| "grad_norm": 5.96875, | |
| "learning_rate": 8.056460369163954e-06, | |
| "loss": 0.6256, | |
| "mean_token_accuracy": 0.8380905002355575, | |
| "num_tokens": 1186343.0, | |
| "step": 280 | |
| }, | |
| { | |
| "entropy": 2.763255572319031, | |
| "epoch": 0.28417442430181283, | |
| "grad_norm": 5.46875, | |
| "learning_rate": 7.947882736156353e-06, | |
| "loss": 0.5349, | |
| "mean_token_accuracy": 0.8597547829151153, | |
| "num_tokens": 1229330.0, | |
| "step": 290 | |
| }, | |
| { | |
| "entropy": 2.771083080768585, | |
| "epoch": 0.29397354238118567, | |
| "grad_norm": 4.96875, | |
| "learning_rate": 7.839305103148752e-06, | |
| "loss": 0.5939, | |
| "mean_token_accuracy": 0.8486017435789108, | |
| "num_tokens": 1273264.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29397354238118567, | |
| "eval_entropy": 2.8048663288354874, | |
| "eval_loss": 0.610714316368103, | |
| "eval_mean_token_accuracy": 0.8439283277839422, | |
| "eval_num_tokens": 1273264.0, | |
| "eval_runtime": 19.2535, | |
| "eval_samples_per_second": 13.14, | |
| "eval_steps_per_second": 1.662, | |
| "step": 300 | |
| }, | |
| { | |
| "entropy": 2.760388123989105, | |
| "epoch": 0.30377266046055856, | |
| "grad_norm": 15.4375, | |
| "learning_rate": 7.730727470141152e-06, | |
| "loss": 0.5957, | |
| "mean_token_accuracy": 0.847811621427536, | |
| "num_tokens": 1316383.0, | |
| "step": 310 | |
| }, | |
| { | |
| "entropy": 2.7953795909881594, | |
| "epoch": 0.3135717785399314, | |
| "grad_norm": 4.96875, | |
| "learning_rate": 7.622149837133551e-06, | |
| "loss": 0.6305, | |
| "mean_token_accuracy": 0.840548175573349, | |
| "num_tokens": 1359667.0, | |
| "step": 320 | |
| }, | |
| { | |
| "entropy": 2.6983755588531495, | |
| "epoch": 0.32337089661930424, | |
| "grad_norm": 11.375, | |
| "learning_rate": 7.51357220412595e-06, | |
| "loss": 0.5669, | |
| "mean_token_accuracy": 0.8493463039398194, | |
| "num_tokens": 1402893.0, | |
| "step": 330 | |
| }, | |
| { | |
| "entropy": 2.708102560043335, | |
| "epoch": 0.33317001469867713, | |
| "grad_norm": 9.9375, | |
| "learning_rate": 7.40499457111835e-06, | |
| "loss": 0.5803, | |
| "mean_token_accuracy": 0.8529332846403122, | |
| "num_tokens": 1446304.0, | |
| "step": 340 | |
| }, | |
| { | |
| "entropy": 2.731845164299011, | |
| "epoch": 0.34296913277804997, | |
| "grad_norm": 5.375, | |
| "learning_rate": 7.29641693811075e-06, | |
| "loss": 0.5994, | |
| "mean_token_accuracy": 0.8489784836769104, | |
| "num_tokens": 1488947.0, | |
| "step": 350 | |
| }, | |
| { | |
| "entropy": 2.8005746603012085, | |
| "epoch": 0.3527682508574228, | |
| "grad_norm": 6.25, | |
| "learning_rate": 7.187839305103149e-06, | |
| "loss": 0.5211, | |
| "mean_token_accuracy": 0.8627282440662384, | |
| "num_tokens": 1532501.0, | |
| "step": 360 | |
| }, | |
| { | |
| "entropy": 2.706599247455597, | |
| "epoch": 0.3625673689367957, | |
| "grad_norm": 5.03125, | |
| "learning_rate": 7.079261672095549e-06, | |
| "loss": 0.5896, | |
| "mean_token_accuracy": 0.845062306523323, | |
| "num_tokens": 1575209.0, | |
| "step": 370 | |
| }, | |
| { | |
| "entropy": 2.7727012634277344, | |
| "epoch": 0.37236648701616853, | |
| "grad_norm": 15.8125, | |
| "learning_rate": 6.9706840390879485e-06, | |
| "loss": 0.593, | |
| "mean_token_accuracy": 0.8456886291503907, | |
| "num_tokens": 1615520.0, | |
| "step": 380 | |
| }, | |
| { | |
| "entropy": 2.7051647901535034, | |
| "epoch": 0.3821656050955414, | |
| "grad_norm": 5.5, | |
| "learning_rate": 6.8621064060803475e-06, | |
| "loss": 0.5507, | |
| "mean_token_accuracy": 0.8546420007944107, | |
| "num_tokens": 1658292.0, | |
| "step": 390 | |
| }, | |
| { | |
| "entropy": 2.7763685464859007, | |
| "epoch": 0.39196472317491426, | |
| "grad_norm": 4.84375, | |
| "learning_rate": 6.753528773072747e-06, | |
| "loss": 0.6089, | |
| "mean_token_accuracy": 0.8412259608507157, | |
| "num_tokens": 1699630.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.39196472317491426, | |
| "eval_entropy": 2.763894908130169, | |
| "eval_loss": 0.5999214053153992, | |
| "eval_mean_token_accuracy": 0.8466707710176706, | |
| "eval_num_tokens": 1699630.0, | |
| "eval_runtime": 19.051, | |
| "eval_samples_per_second": 13.28, | |
| "eval_steps_per_second": 1.68, | |
| "step": 400 | |
| }, | |
| { | |
| "entropy": 2.809602975845337, | |
| "epoch": 0.4017638412542871, | |
| "grad_norm": 6.875, | |
| "learning_rate": 6.644951140065147e-06, | |
| "loss": 0.6362, | |
| "mean_token_accuracy": 0.8436780005693436, | |
| "num_tokens": 1741483.0, | |
| "step": 410 | |
| }, | |
| { | |
| "entropy": 2.824617338180542, | |
| "epoch": 0.41156295933366, | |
| "grad_norm": 6.25, | |
| "learning_rate": 6.536373507057546e-06, | |
| "loss": 0.5544, | |
| "mean_token_accuracy": 0.862479692697525, | |
| "num_tokens": 1783428.0, | |
| "step": 420 | |
| }, | |
| { | |
| "entropy": 2.780674231052399, | |
| "epoch": 0.42136207741303283, | |
| "grad_norm": 4.46875, | |
| "learning_rate": 6.427795874049946e-06, | |
| "loss": 0.5959, | |
| "mean_token_accuracy": 0.8487352669239044, | |
| "num_tokens": 1825596.0, | |
| "step": 430 | |
| }, | |
| { | |
| "entropy": 2.7935621976852416, | |
| "epoch": 0.43116119549240567, | |
| "grad_norm": 4.84375, | |
| "learning_rate": 6.319218241042345e-06, | |
| "loss": 0.5787, | |
| "mean_token_accuracy": 0.845246434211731, | |
| "num_tokens": 1868684.0, | |
| "step": 440 | |
| }, | |
| { | |
| "entropy": 2.7141441464424134, | |
| "epoch": 0.44096031357177856, | |
| "grad_norm": 6.5, | |
| "learning_rate": 6.2106406080347455e-06, | |
| "loss": 0.5644, | |
| "mean_token_accuracy": 0.852169805765152, | |
| "num_tokens": 1912147.0, | |
| "step": 450 | |
| }, | |
| { | |
| "entropy": 2.7499614357948303, | |
| "epoch": 0.4507594316511514, | |
| "grad_norm": 5.59375, | |
| "learning_rate": 6.102062975027145e-06, | |
| "loss": 0.5641, | |
| "mean_token_accuracy": 0.8543315261602402, | |
| "num_tokens": 1955005.0, | |
| "step": 460 | |
| }, | |
| { | |
| "entropy": 2.7714619636535645, | |
| "epoch": 0.46055854973052424, | |
| "grad_norm": 5.5, | |
| "learning_rate": 5.993485342019545e-06, | |
| "loss": 0.5809, | |
| "mean_token_accuracy": 0.8478847473859787, | |
| "num_tokens": 1996380.0, | |
| "step": 470 | |
| }, | |
| { | |
| "entropy": 2.8680380821228026, | |
| "epoch": 0.47035766780989713, | |
| "grad_norm": 6.4375, | |
| "learning_rate": 5.884907709011944e-06, | |
| "loss": 0.6065, | |
| "mean_token_accuracy": 0.8474393934011459, | |
| "num_tokens": 2037703.0, | |
| "step": 480 | |
| }, | |
| { | |
| "entropy": 2.6957027554512023, | |
| "epoch": 0.48015678588926997, | |
| "grad_norm": 4.71875, | |
| "learning_rate": 5.776330076004344e-06, | |
| "loss": 0.4984, | |
| "mean_token_accuracy": 0.8654001474380493, | |
| "num_tokens": 2079440.0, | |
| "step": 490 | |
| }, | |
| { | |
| "entropy": 2.7772451877593993, | |
| "epoch": 0.4899559039686428, | |
| "grad_norm": 5.09375, | |
| "learning_rate": 5.667752442996744e-06, | |
| "loss": 0.6104, | |
| "mean_token_accuracy": 0.8470509260892868, | |
| "num_tokens": 2121519.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4899559039686428, | |
| "eval_entropy": 2.7685302942991257, | |
| "eval_loss": 0.5946313738822937, | |
| "eval_mean_token_accuracy": 0.8479105345904827, | |
| "eval_num_tokens": 2121519.0, | |
| "eval_runtime": 19.2901, | |
| "eval_samples_per_second": 13.116, | |
| "eval_steps_per_second": 1.659, | |
| "step": 500 | |
| }, | |
| { | |
| "entropy": 2.838483190536499, | |
| "epoch": 0.4997550220480157, | |
| "grad_norm": 6.78125, | |
| "learning_rate": 5.559174809989143e-06, | |
| "loss": 0.5722, | |
| "mean_token_accuracy": 0.8514155298471451, | |
| "num_tokens": 2163668.0, | |
| "step": 510 | |
| }, | |
| { | |
| "entropy": 2.7187190294265746, | |
| "epoch": 0.5095541401273885, | |
| "grad_norm": 8.3125, | |
| "learning_rate": 5.4505971769815425e-06, | |
| "loss": 0.6559, | |
| "mean_token_accuracy": 0.8406646817922592, | |
| "num_tokens": 2208584.0, | |
| "step": 520 | |
| }, | |
| { | |
| "entropy": 2.7912783145904543, | |
| "epoch": 0.5193532582067614, | |
| "grad_norm": 4.03125, | |
| "learning_rate": 5.342019543973942e-06, | |
| "loss": 0.5556, | |
| "mean_token_accuracy": 0.8573304742574692, | |
| "num_tokens": 2250613.0, | |
| "step": 530 | |
| }, | |
| { | |
| "entropy": 2.7244945645332335, | |
| "epoch": 0.5291523762861342, | |
| "grad_norm": 7.46875, | |
| "learning_rate": 5.233441910966341e-06, | |
| "loss": 0.6018, | |
| "mean_token_accuracy": 0.8491319239139556, | |
| "num_tokens": 2292569.0, | |
| "step": 540 | |
| }, | |
| { | |
| "entropy": 2.6968074440956116, | |
| "epoch": 0.5389514943655072, | |
| "grad_norm": 6.34375, | |
| "learning_rate": 5.124864277958741e-06, | |
| "loss": 0.5532, | |
| "mean_token_accuracy": 0.8539753496646881, | |
| "num_tokens": 2336377.0, | |
| "step": 550 | |
| }, | |
| { | |
| "entropy": 2.7182830929756165, | |
| "epoch": 0.54875061244488, | |
| "grad_norm": 5.8125, | |
| "learning_rate": 5.016286644951141e-06, | |
| "loss": 0.547, | |
| "mean_token_accuracy": 0.8532966852188111, | |
| "num_tokens": 2379596.0, | |
| "step": 560 | |
| }, | |
| { | |
| "entropy": 2.750958800315857, | |
| "epoch": 0.5585497305242528, | |
| "grad_norm": 7.6875, | |
| "learning_rate": 4.90770901194354e-06, | |
| "loss": 0.5818, | |
| "mean_token_accuracy": 0.8448131322860718, | |
| "num_tokens": 2422597.0, | |
| "step": 570 | |
| }, | |
| { | |
| "entropy": 2.7499714136123656, | |
| "epoch": 0.5683488486036257, | |
| "grad_norm": 6.5625, | |
| "learning_rate": 4.79913137893594e-06, | |
| "loss": 0.5806, | |
| "mean_token_accuracy": 0.8504658758640289, | |
| "num_tokens": 2468429.0, | |
| "step": 580 | |
| }, | |
| { | |
| "entropy": 2.6956202149391175, | |
| "epoch": 0.5781479666829985, | |
| "grad_norm": 15.1875, | |
| "learning_rate": 4.690553745928339e-06, | |
| "loss": 0.5836, | |
| "mean_token_accuracy": 0.8504124045372009, | |
| "num_tokens": 2511751.0, | |
| "step": 590 | |
| }, | |
| { | |
| "entropy": 2.8232338666915893, | |
| "epoch": 0.5879470847623713, | |
| "grad_norm": 6.1875, | |
| "learning_rate": 4.5819761129207385e-06, | |
| "loss": 0.5789, | |
| "mean_token_accuracy": 0.8494374930858613, | |
| "num_tokens": 2554335.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5879470847623713, | |
| "eval_entropy": 2.7596413865685463, | |
| "eval_loss": 0.591189444065094, | |
| "eval_mean_token_accuracy": 0.8485322501510382, | |
| "eval_num_tokens": 2554335.0, | |
| "eval_runtime": 19.2602, | |
| "eval_samples_per_second": 13.136, | |
| "eval_steps_per_second": 1.661, | |
| "step": 600 | |
| }, | |
| { | |
| "entropy": 2.708176779747009, | |
| "epoch": 0.5977462028417443, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 4.473398479913138e-06, | |
| "loss": 0.5682, | |
| "mean_token_accuracy": 0.8568267792463302, | |
| "num_tokens": 2598711.0, | |
| "step": 610 | |
| }, | |
| { | |
| "entropy": 2.7695399165153503, | |
| "epoch": 0.6075453209211171, | |
| "grad_norm": 4.59375, | |
| "learning_rate": 4.364820846905538e-06, | |
| "loss": 0.5592, | |
| "mean_token_accuracy": 0.8596846997737885, | |
| "num_tokens": 2642176.0, | |
| "step": 620 | |
| }, | |
| { | |
| "entropy": 2.8415961861610413, | |
| "epoch": 0.61734443900049, | |
| "grad_norm": 9.25, | |
| "learning_rate": 4.256243213897938e-06, | |
| "loss": 0.594, | |
| "mean_token_accuracy": 0.8536108046770096, | |
| "num_tokens": 2684207.0, | |
| "step": 630 | |
| }, | |
| { | |
| "entropy": 2.796202802658081, | |
| "epoch": 0.6271435570798628, | |
| "grad_norm": 6.15625, | |
| "learning_rate": 4.147665580890337e-06, | |
| "loss": 0.5432, | |
| "mean_token_accuracy": 0.860190287232399, | |
| "num_tokens": 2724881.0, | |
| "step": 640 | |
| }, | |
| { | |
| "entropy": 2.6983245611190796, | |
| "epoch": 0.6369426751592356, | |
| "grad_norm": 4.5625, | |
| "learning_rate": 4.039087947882737e-06, | |
| "loss": 0.5497, | |
| "mean_token_accuracy": 0.8522610604763031, | |
| "num_tokens": 2768550.0, | |
| "step": 650 | |
| }, | |
| { | |
| "entropy": 2.7941854119300844, | |
| "epoch": 0.6467417932386085, | |
| "grad_norm": 5.0625, | |
| "learning_rate": 3.9305103148751365e-06, | |
| "loss": 0.5893, | |
| "mean_token_accuracy": 0.8445635885000229, | |
| "num_tokens": 2811022.0, | |
| "step": 660 | |
| }, | |
| { | |
| "entropy": 2.760863173007965, | |
| "epoch": 0.6565409113179814, | |
| "grad_norm": 5.8125, | |
| "learning_rate": 3.8219326818675354e-06, | |
| "loss": 0.5791, | |
| "mean_token_accuracy": 0.8516164273023605, | |
| "num_tokens": 2851837.0, | |
| "step": 670 | |
| }, | |
| { | |
| "entropy": 2.806622850894928, | |
| "epoch": 0.6663400293973543, | |
| "grad_norm": 6.09375, | |
| "learning_rate": 3.7133550488599353e-06, | |
| "loss": 0.5868, | |
| "mean_token_accuracy": 0.8492783725261688, | |
| "num_tokens": 2893805.0, | |
| "step": 680 | |
| }, | |
| { | |
| "entropy": 2.8078768491744994, | |
| "epoch": 0.6761391474767271, | |
| "grad_norm": 7.21875, | |
| "learning_rate": 3.6047774158523346e-06, | |
| "loss": 0.555, | |
| "mean_token_accuracy": 0.856513598561287, | |
| "num_tokens": 2936985.0, | |
| "step": 690 | |
| }, | |
| { | |
| "entropy": 2.8118430733680726, | |
| "epoch": 0.6859382655560999, | |
| "grad_norm": 8.125, | |
| "learning_rate": 3.496199782844734e-06, | |
| "loss": 0.5919, | |
| "mean_token_accuracy": 0.8447350233793258, | |
| "num_tokens": 2978790.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.6859382655560999, | |
| "eval_entropy": 2.770235098898411, | |
| "eval_loss": 0.5892172455787659, | |
| "eval_mean_token_accuracy": 0.8493550103157759, | |
| "eval_num_tokens": 2978790.0, | |
| "eval_runtime": 19.2603, | |
| "eval_samples_per_second": 13.136, | |
| "eval_steps_per_second": 1.661, | |
| "step": 700 | |
| }, | |
| { | |
| "entropy": 2.7048792481422423, | |
| "epoch": 0.6957373836354728, | |
| "grad_norm": 4.90625, | |
| "learning_rate": 3.387622149837134e-06, | |
| "loss": 0.5567, | |
| "mean_token_accuracy": 0.849539065361023, | |
| "num_tokens": 3021835.0, | |
| "step": 710 | |
| }, | |
| { | |
| "entropy": 2.728425168991089, | |
| "epoch": 0.7055365017148456, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 3.2790445168295332e-06, | |
| "loss": 0.5378, | |
| "mean_token_accuracy": 0.8606635153293609, | |
| "num_tokens": 3064935.0, | |
| "step": 720 | |
| }, | |
| { | |
| "entropy": 2.8092617750167848, | |
| "epoch": 0.7153356197942186, | |
| "grad_norm": 4.40625, | |
| "learning_rate": 3.1704668838219326e-06, | |
| "loss": 0.5941, | |
| "mean_token_accuracy": 0.8491881370544434, | |
| "num_tokens": 3106511.0, | |
| "step": 730 | |
| }, | |
| { | |
| "entropy": 2.728269863128662, | |
| "epoch": 0.7251347378735914, | |
| "grad_norm": 4.1875, | |
| "learning_rate": 3.061889250814333e-06, | |
| "loss": 0.5555, | |
| "mean_token_accuracy": 0.8544395923614502, | |
| "num_tokens": 3147924.0, | |
| "step": 740 | |
| }, | |
| { | |
| "entropy": 2.747904062271118, | |
| "epoch": 0.7349338559529642, | |
| "grad_norm": 5.03125, | |
| "learning_rate": 2.9533116178067322e-06, | |
| "loss": 0.5753, | |
| "mean_token_accuracy": 0.8510884076356888, | |
| "num_tokens": 3189628.0, | |
| "step": 750 | |
| }, | |
| { | |
| "entropy": 2.784381794929504, | |
| "epoch": 0.7447329740323371, | |
| "grad_norm": 5.28125, | |
| "learning_rate": 2.8447339847991316e-06, | |
| "loss": 0.5406, | |
| "mean_token_accuracy": 0.8581822723150253, | |
| "num_tokens": 3231421.0, | |
| "step": 760 | |
| }, | |
| { | |
| "entropy": 2.761149501800537, | |
| "epoch": 0.7545320921117099, | |
| "grad_norm": 4.71875, | |
| "learning_rate": 2.7361563517915314e-06, | |
| "loss": 0.5812, | |
| "mean_token_accuracy": 0.8436756283044815, | |
| "num_tokens": 3274633.0, | |
| "step": 770 | |
| }, | |
| { | |
| "entropy": 2.732284414768219, | |
| "epoch": 0.7643312101910829, | |
| "grad_norm": 4.3125, | |
| "learning_rate": 2.627578718783931e-06, | |
| "loss": 0.519, | |
| "mean_token_accuracy": 0.8654556035995483, | |
| "num_tokens": 3316474.0, | |
| "step": 780 | |
| }, | |
| { | |
| "entropy": 2.7518323063850403, | |
| "epoch": 0.7741303282704557, | |
| "grad_norm": 6.625, | |
| "learning_rate": 2.5190010857763302e-06, | |
| "loss": 0.5715, | |
| "mean_token_accuracy": 0.8510609477758407, | |
| "num_tokens": 3357965.0, | |
| "step": 790 | |
| }, | |
| { | |
| "entropy": 2.7155523538589477, | |
| "epoch": 0.7839294463498285, | |
| "grad_norm": 6.125, | |
| "learning_rate": 2.4104234527687296e-06, | |
| "loss": 0.5508, | |
| "mean_token_accuracy": 0.8542987465858459, | |
| "num_tokens": 3401349.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7839294463498285, | |
| "eval_entropy": 2.7648734971880913, | |
| "eval_loss": 0.5884459614753723, | |
| "eval_mean_token_accuracy": 0.8490999434143305, | |
| "eval_num_tokens": 3401349.0, | |
| "eval_runtime": 19.2717, | |
| "eval_samples_per_second": 13.128, | |
| "eval_steps_per_second": 1.66, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1021, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.109110482665472e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |