| { |
| "best_global_step": 270, |
| "best_metric": 0.12460343539714813, |
| "best_model_checkpoint": "saves_stability/lntuning/llama-3-8b-instruct/train_copa_1757340207/checkpoint-270", |
| "epoch": 10.0, |
| "eval_steps": 45, |
| "global_step": 900, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.05555555555555555, |
| "grad_norm": 9.265876770019531, |
| "learning_rate": 2.2222222222222225e-06, |
| "loss": 0.5372, |
| "num_input_tokens_seen": 1536, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1111111111111111, |
| "grad_norm": 10.255990028381348, |
| "learning_rate": 5e-06, |
| "loss": 0.6204, |
| "num_input_tokens_seen": 3168, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.16666666666666666, |
| "grad_norm": 11.147568702697754, |
| "learning_rate": 7.777777777777777e-06, |
| "loss": 0.5905, |
| "num_input_tokens_seen": 4736, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 8.53930377960205, |
| "learning_rate": 1.0555555555555555e-05, |
| "loss": 0.4626, |
| "num_input_tokens_seen": 6304, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.2777777777777778, |
| "grad_norm": 9.899728775024414, |
| "learning_rate": 1.3333333333333333e-05, |
| "loss": 0.3369, |
| "num_input_tokens_seen": 7840, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3333333333333333, |
| "grad_norm": 4.038922309875488, |
| "learning_rate": 1.6111111111111115e-05, |
| "loss": 0.2416, |
| "num_input_tokens_seen": 9408, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.3888888888888889, |
| "grad_norm": 4.957484245300293, |
| "learning_rate": 1.888888888888889e-05, |
| "loss": 0.3098, |
| "num_input_tokens_seen": 10912, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 3.828603506088257, |
| "learning_rate": 2.1666666666666667e-05, |
| "loss": 0.2185, |
| "num_input_tokens_seen": 12448, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 4.001208305358887, |
| "learning_rate": 2.4444444444444445e-05, |
| "loss": 0.1596, |
| "num_input_tokens_seen": 14016, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_loss": 0.19575616717338562, |
| "eval_runtime": 0.5579, |
| "eval_samples_per_second": 71.698, |
| "eval_steps_per_second": 17.925, |
| "num_input_tokens_seen": 14016, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 1.5252408981323242, |
| "learning_rate": 2.7222222222222223e-05, |
| "loss": 0.1403, |
| "num_input_tokens_seen": 15584, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6111111111111112, |
| "grad_norm": 0.27813827991485596, |
| "learning_rate": 3e-05, |
| "loss": 0.1846, |
| "num_input_tokens_seen": 17184, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 8.89137077331543, |
| "learning_rate": 3.277777777777778e-05, |
| "loss": 0.1133, |
| "num_input_tokens_seen": 18752, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.7222222222222222, |
| "grad_norm": 10.673196792602539, |
| "learning_rate": 3.555555555555556e-05, |
| "loss": 0.2668, |
| "num_input_tokens_seen": 20352, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7777777777777778, |
| "grad_norm": 0.1354873925447464, |
| "learning_rate": 3.8333333333333334e-05, |
| "loss": 0.0427, |
| "num_input_tokens_seen": 21952, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 4.400477409362793, |
| "learning_rate": 4.111111111111111e-05, |
| "loss": 0.0673, |
| "num_input_tokens_seen": 23456, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 0.6681251525878906, |
| "learning_rate": 4.388888888888889e-05, |
| "loss": 0.2793, |
| "num_input_tokens_seen": 25056, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.9444444444444444, |
| "grad_norm": 0.12160417437553406, |
| "learning_rate": 4.666666666666667e-05, |
| "loss": 0.0663, |
| "num_input_tokens_seen": 26560, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.1885559856891632, |
| "learning_rate": 4.9444444444444446e-05, |
| "loss": 0.2404, |
| "num_input_tokens_seen": 28096, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.13765892386436462, |
| "eval_runtime": 0.5571, |
| "eval_samples_per_second": 71.797, |
| "eval_steps_per_second": 17.949, |
| "num_input_tokens_seen": 28096, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.0555555555555556, |
| "grad_norm": 3.6679036617279053, |
| "learning_rate": 4.9996991493233693e-05, |
| "loss": 0.1721, |
| "num_input_tokens_seen": 29696, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.9191456437110901, |
| "learning_rate": 4.99847706754774e-05, |
| "loss": 0.1093, |
| "num_input_tokens_seen": 31232, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.1666666666666667, |
| "grad_norm": 6.504665374755859, |
| "learning_rate": 4.9963154107272295e-05, |
| "loss": 0.0872, |
| "num_input_tokens_seen": 32768, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.2222222222222223, |
| "grad_norm": 0.039618778973817825, |
| "learning_rate": 4.993214991772563e-05, |
| "loss": 0.053, |
| "num_input_tokens_seen": 34304, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.2777777777777777, |
| "grad_norm": 0.03970110043883324, |
| "learning_rate": 4.989176976624511e-05, |
| "loss": 0.0143, |
| "num_input_tokens_seen": 35872, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 0.7861797213554382, |
| "learning_rate": 4.9842028838154285e-05, |
| "loss": 0.1648, |
| "num_input_tokens_seen": 37408, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.3888888888888888, |
| "grad_norm": 3.8421483039855957, |
| "learning_rate": 4.978294583898196e-05, |
| "loss": 0.0997, |
| "num_input_tokens_seen": 38976, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.4444444444444444, |
| "grad_norm": 4.874511241912842, |
| "learning_rate": 4.971454298742779e-05, |
| "loss": 0.0817, |
| "num_input_tokens_seen": 40576, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 0.09447989612817764, |
| "learning_rate": 4.963684600700679e-05, |
| "loss": 0.1623, |
| "num_input_tokens_seen": 42144, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5, |
| "eval_loss": 0.12774308025836945, |
| "eval_runtime": 0.5763, |
| "eval_samples_per_second": 69.413, |
| "eval_steps_per_second": 17.353, |
| "num_input_tokens_seen": 42144, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 0.29610562324523926, |
| "learning_rate": 4.9549884116375714e-05, |
| "loss": 0.0454, |
| "num_input_tokens_seen": 43680, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.6111111111111112, |
| "grad_norm": 1.8017239570617676, |
| "learning_rate": 4.9453690018345144e-05, |
| "loss": 0.1939, |
| "num_input_tokens_seen": 45248, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 0.028127163648605347, |
| "learning_rate": 4.934829988758131e-05, |
| "loss": 0.0218, |
| "num_input_tokens_seen": 46816, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.7222222222222223, |
| "grad_norm": 5.569133758544922, |
| "learning_rate": 4.923375335700223e-05, |
| "loss": 0.134, |
| "num_input_tokens_seen": 48384, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 0.02653215453028679, |
| "learning_rate": 4.9110093502873476e-05, |
| "loss": 0.0358, |
| "num_input_tokens_seen": 49952, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.8333333333333335, |
| "grad_norm": 0.017496682703495026, |
| "learning_rate": 4.897736682860885e-05, |
| "loss": 0.0138, |
| "num_input_tokens_seen": 51520, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.8888888888888888, |
| "grad_norm": 1.7326111793518066, |
| "learning_rate": 4.883562324728241e-05, |
| "loss": 0.0933, |
| "num_input_tokens_seen": 53024, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.9444444444444444, |
| "grad_norm": 4.243280410766602, |
| "learning_rate": 4.868491606285823e-05, |
| "loss": 0.0705, |
| "num_input_tokens_seen": 54592, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.1312873661518097, |
| "learning_rate": 4.8525301950144894e-05, |
| "loss": 0.1, |
| "num_input_tokens_seen": 56128, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.12832966446876526, |
| "eval_runtime": 0.5953, |
| "eval_samples_per_second": 67.196, |
| "eval_steps_per_second": 16.799, |
| "num_input_tokens_seen": 56128, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.0555555555555554, |
| "grad_norm": 0.019096579402685165, |
| "learning_rate": 4.835684093348244e-05, |
| "loss": 0.0266, |
| "num_input_tokens_seen": 57696, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.111111111111111, |
| "grad_norm": 3.8375511169433594, |
| "learning_rate": 4.817959636416969e-05, |
| "loss": 0.171, |
| "num_input_tokens_seen": 59264, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.1666666666666665, |
| "grad_norm": 0.015614030882716179, |
| "learning_rate": 4.7993634896640394e-05, |
| "loss": 0.0205, |
| "num_input_tokens_seen": 60864, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 3.2604098320007324, |
| "learning_rate": 4.779902646339722e-05, |
| "loss": 0.0879, |
| "num_input_tokens_seen": 62464, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.2777777777777777, |
| "grad_norm": 3.7999112606048584, |
| "learning_rate": 4.759584424871302e-05, |
| "loss": 0.1829, |
| "num_input_tokens_seen": 64032, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.3333333333333335, |
| "grad_norm": 0.16668812930583954, |
| "learning_rate": 4.7384164661109176e-05, |
| "loss": 0.0406, |
| "num_input_tokens_seen": 65568, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.388888888888889, |
| "grad_norm": 0.12282378226518631, |
| "learning_rate": 4.7164067304621536e-05, |
| "loss": 0.013, |
| "num_input_tokens_seen": 67104, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 0.18318073451519012, |
| "learning_rate": 4.693563494886455e-05, |
| "loss": 0.1203, |
| "num_input_tokens_seen": 68704, |
| "step": 220 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 1.7672929763793945, |
| "learning_rate": 4.669895349790502e-05, |
| "loss": 0.0681, |
| "num_input_tokens_seen": 70272, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.12674304842948914, |
| "eval_runtime": 0.5595, |
| "eval_samples_per_second": 71.496, |
| "eval_steps_per_second": 17.874, |
| "num_input_tokens_seen": 70272, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.5555555555555554, |
| "grad_norm": 1.3705925941467285, |
| "learning_rate": 4.645411195795709e-05, |
| "loss": 0.0121, |
| "num_input_tokens_seen": 71808, |
| "step": 230 |
| }, |
| { |
| "epoch": 2.611111111111111, |
| "grad_norm": 0.883948564529419, |
| "learning_rate": 4.620120240391065e-05, |
| "loss": 0.0753, |
| "num_input_tokens_seen": 73408, |
| "step": 235 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 0.8193748593330383, |
| "learning_rate": 4.5940319944705736e-05, |
| "loss": 0.0347, |
| "num_input_tokens_seen": 74912, |
| "step": 240 |
| }, |
| { |
| "epoch": 2.7222222222222223, |
| "grad_norm": 5.69434118270874, |
| "learning_rate": 4.567156268756594e-05, |
| "loss": 0.1271, |
| "num_input_tokens_seen": 76544, |
| "step": 245 |
| }, |
| { |
| "epoch": 2.7777777777777777, |
| "grad_norm": 0.6252865791320801, |
| "learning_rate": 4.539503170110431e-05, |
| "loss": 0.0205, |
| "num_input_tokens_seen": 78112, |
| "step": 250 |
| }, |
| { |
| "epoch": 2.8333333333333335, |
| "grad_norm": 0.019671587273478508, |
| "learning_rate": 4.5110830977315556e-05, |
| "loss": 0.0045, |
| "num_input_tokens_seen": 79712, |
| "step": 255 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.9472150206565857, |
| "learning_rate": 4.4819067392468944e-05, |
| "loss": 0.0293, |
| "num_input_tokens_seen": 81280, |
| "step": 260 |
| }, |
| { |
| "epoch": 2.9444444444444446, |
| "grad_norm": 0.03704234957695007, |
| "learning_rate": 4.4519850666916484e-05, |
| "loss": 0.0186, |
| "num_input_tokens_seen": 82848, |
| "step": 265 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.06520653516054153, |
| "learning_rate": 4.4213293323831585e-05, |
| "loss": 0.0289, |
| "num_input_tokens_seen": 84352, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.12460343539714813, |
| "eval_runtime": 0.5631, |
| "eval_samples_per_second": 71.042, |
| "eval_steps_per_second": 17.76, |
| "num_input_tokens_seen": 84352, |
| "step": 270 |
| }, |
| { |
| "epoch": 3.0555555555555554, |
| "grad_norm": 4.020796775817871, |
| "learning_rate": 4.38995106468937e-05, |
| "loss": 0.026, |
| "num_input_tokens_seen": 85920, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.20490235090255737, |
| "learning_rate": 4.357862063693486e-05, |
| "loss": 0.0158, |
| "num_input_tokens_seen": 87520, |
| "step": 280 |
| }, |
| { |
| "epoch": 3.1666666666666665, |
| "grad_norm": 0.23053014278411865, |
| "learning_rate": 4.325074396756437e-05, |
| "loss": 0.0022, |
| "num_input_tokens_seen": 89088, |
| "step": 285 |
| }, |
| { |
| "epoch": 3.2222222222222223, |
| "grad_norm": 0.022983640432357788, |
| "learning_rate": 4.2916003939788403e-05, |
| "loss": 0.0403, |
| "num_input_tokens_seen": 90688, |
| "step": 290 |
| }, |
| { |
| "epoch": 3.2777777777777777, |
| "grad_norm": 1.0208379030227661, |
| "learning_rate": 4.257452643564155e-05, |
| "loss": 0.0512, |
| "num_input_tokens_seen": 92160, |
| "step": 295 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 6.755514144897461, |
| "learning_rate": 4.22264398708477e-05, |
| "loss": 0.0857, |
| "num_input_tokens_seen": 93760, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.388888888888889, |
| "grad_norm": 0.6758346557617188, |
| "learning_rate": 4.1871875146528195e-05, |
| "loss": 0.1237, |
| "num_input_tokens_seen": 95360, |
| "step": 305 |
| }, |
| { |
| "epoch": 3.4444444444444446, |
| "grad_norm": 0.02263474650681019, |
| "learning_rate": 4.1510965599975196e-05, |
| "loss": 0.0318, |
| "num_input_tokens_seen": 96928, |
| "step": 310 |
| }, |
| { |
| "epoch": 3.5, |
| "grad_norm": 1.6407543420791626, |
| "learning_rate": 4.114384695450906e-05, |
| "loss": 0.0638, |
| "num_input_tokens_seen": 98464, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5, |
| "eval_loss": 0.13137976825237274, |
| "eval_runtime": 0.5668, |
| "eval_samples_per_second": 70.572, |
| "eval_steps_per_second": 17.643, |
| "num_input_tokens_seen": 98464, |
| "step": 315 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 0.7610914707183838, |
| "learning_rate": 4.077065726843828e-05, |
| "loss": 0.0127, |
| "num_input_tokens_seen": 100064, |
| "step": 320 |
| }, |
| { |
| "epoch": 3.611111111111111, |
| "grad_norm": 0.09962353110313416, |
| "learning_rate": 4.039153688314145e-05, |
| "loss": 0.0677, |
| "num_input_tokens_seen": 101600, |
| "step": 325 |
| }, |
| { |
| "epoch": 3.6666666666666665, |
| "grad_norm": 0.023464810103178024, |
| "learning_rate": 4.000662837029062e-05, |
| "loss": 0.0808, |
| "num_input_tokens_seen": 103200, |
| "step": 330 |
| }, |
| { |
| "epoch": 3.7222222222222223, |
| "grad_norm": 1.1725422143936157, |
| "learning_rate": 3.961607647823583e-05, |
| "loss": 0.0887, |
| "num_input_tokens_seen": 104768, |
| "step": 335 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 0.22920739650726318, |
| "learning_rate": 3.9220028077571295e-05, |
| "loss": 0.0037, |
| "num_input_tokens_seen": 106304, |
| "step": 340 |
| }, |
| { |
| "epoch": 3.8333333333333335, |
| "grad_norm": 0.010305184870958328, |
| "learning_rate": 3.881863210590332e-05, |
| "loss": 0.0032, |
| "num_input_tokens_seen": 107904, |
| "step": 345 |
| }, |
| { |
| "epoch": 3.888888888888889, |
| "grad_norm": 6.484604358673096, |
| "learning_rate": 3.841203951184095e-05, |
| "loss": 0.1634, |
| "num_input_tokens_seen": 109408, |
| "step": 350 |
| }, |
| { |
| "epoch": 3.9444444444444446, |
| "grad_norm": 0.034305017441511154, |
| "learning_rate": 3.8000403198230387e-05, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 111008, |
| "step": 355 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.008833297528326511, |
| "learning_rate": 3.75838779646545e-05, |
| "loss": 0.0061, |
| "num_input_tokens_seen": 112576, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.1305050104856491, |
| "eval_runtime": 0.5704, |
| "eval_samples_per_second": 70.13, |
| "eval_steps_per_second": 17.533, |
| "num_input_tokens_seen": 112576, |
| "step": 360 |
| }, |
| { |
| "epoch": 4.055555555555555, |
| "grad_norm": 0.042020078748464584, |
| "learning_rate": 3.7162620449219e-05, |
| "loss": 0.0477, |
| "num_input_tokens_seen": 114144, |
| "step": 365 |
| }, |
| { |
| "epoch": 4.111111111111111, |
| "grad_norm": 1.0857365131378174, |
| "learning_rate": 3.673678906964727e-05, |
| "loss": 0.0079, |
| "num_input_tokens_seen": 115712, |
| "step": 370 |
| }, |
| { |
| "epoch": 4.166666666666667, |
| "grad_norm": 0.008365446701645851, |
| "learning_rate": 3.630654396370594e-05, |
| "loss": 0.0716, |
| "num_input_tokens_seen": 117216, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 0.018131185322999954, |
| "learning_rate": 3.5872046928983626e-05, |
| "loss": 0.0048, |
| "num_input_tokens_seen": 118816, |
| "step": 380 |
| }, |
| { |
| "epoch": 4.277777777777778, |
| "grad_norm": 1.3139055967330933, |
| "learning_rate": 3.543346136204545e-05, |
| "loss": 0.0115, |
| "num_input_tokens_seen": 120352, |
| "step": 385 |
| }, |
| { |
| "epoch": 4.333333333333333, |
| "grad_norm": 5.382805347442627, |
| "learning_rate": 3.499095219698631e-05, |
| "loss": 0.0637, |
| "num_input_tokens_seen": 121920, |
| "step": 390 |
| }, |
| { |
| "epoch": 4.388888888888889, |
| "grad_norm": 4.375858783721924, |
| "learning_rate": 3.454468584340588e-05, |
| "loss": 0.0274, |
| "num_input_tokens_seen": 123456, |
| "step": 395 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 5.618959426879883, |
| "learning_rate": 3.409483012382879e-05, |
| "loss": 0.0655, |
| "num_input_tokens_seen": 125056, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.5, |
| "grad_norm": 6.164790630340576, |
| "learning_rate": 3.364155421059342e-05, |
| "loss": 0.1354, |
| "num_input_tokens_seen": 126624, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.5, |
| "eval_loss": 0.13557614386081696, |
| "eval_runtime": 0.5605, |
| "eval_samples_per_second": 71.363, |
| "eval_steps_per_second": 17.841, |
| "num_input_tokens_seen": 126624, |
| "step": 405 |
| }, |
| { |
| "epoch": 4.555555555555555, |
| "grad_norm": 0.45653727650642395, |
| "learning_rate": 3.318502856223311e-05, |
| "loss": 0.0173, |
| "num_input_tokens_seen": 128224, |
| "step": 410 |
| }, |
| { |
| "epoch": 4.611111111111111, |
| "grad_norm": 0.5255909562110901, |
| "learning_rate": 3.272542485937369e-05, |
| "loss": 0.0078, |
| "num_input_tokens_seen": 129728, |
| "step": 415 |
| }, |
| { |
| "epoch": 4.666666666666667, |
| "grad_norm": 0.3994658589363098, |
| "learning_rate": 3.2262915940171376e-05, |
| "loss": 0.0671, |
| "num_input_tokens_seen": 131328, |
| "step": 420 |
| }, |
| { |
| "epoch": 4.722222222222222, |
| "grad_norm": 2.863457679748535, |
| "learning_rate": 3.1797675735315455e-05, |
| "loss": 0.0198, |
| "num_input_tokens_seen": 132896, |
| "step": 425 |
| }, |
| { |
| "epoch": 4.777777777777778, |
| "grad_norm": 0.02107965014874935, |
| "learning_rate": 3.132987920262005e-05, |
| "loss": 0.0024, |
| "num_input_tokens_seen": 134496, |
| "step": 430 |
| }, |
| { |
| "epoch": 4.833333333333333, |
| "grad_norm": 0.007149787154048681, |
| "learning_rate": 3.085970226122962e-05, |
| "loss": 0.0037, |
| "num_input_tokens_seen": 136064, |
| "step": 435 |
| }, |
| { |
| "epoch": 4.888888888888889, |
| "grad_norm": 0.4444464445114136, |
| "learning_rate": 3.0387321725463e-05, |
| "loss": 0.0843, |
| "num_input_tokens_seen": 137664, |
| "step": 440 |
| }, |
| { |
| "epoch": 4.944444444444445, |
| "grad_norm": 0.009568980894982815, |
| "learning_rate": 2.9912915238320754e-05, |
| "loss": 0.0546, |
| "num_input_tokens_seen": 139232, |
| "step": 445 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.20437009632587433, |
| "learning_rate": 2.9436661204680882e-05, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 140832, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.14013271033763885, |
| "eval_runtime": 0.5664, |
| "eval_samples_per_second": 70.616, |
| "eval_steps_per_second": 17.654, |
| "num_input_tokens_seen": 140832, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.055555555555555, |
| "grad_norm": 1.0058284997940063, |
| "learning_rate": 2.8958738724208072e-05, |
| "loss": 0.108, |
| "num_input_tokens_seen": 142368, |
| "step": 455 |
| }, |
| { |
| "epoch": 5.111111111111111, |
| "grad_norm": 0.10319157689809799, |
| "learning_rate": 2.8479327524001636e-05, |
| "loss": 0.0042, |
| "num_input_tokens_seen": 144000, |
| "step": 460 |
| }, |
| { |
| "epoch": 5.166666666666667, |
| "grad_norm": 0.1945868283510208, |
| "learning_rate": 2.7998607891007495e-05, |
| "loss": 0.0058, |
| "num_input_tokens_seen": 145632, |
| "step": 465 |
| }, |
| { |
| "epoch": 5.222222222222222, |
| "grad_norm": 7.325711250305176, |
| "learning_rate": 2.7516760604219617e-05, |
| "loss": 0.0685, |
| "num_input_tokens_seen": 147168, |
| "step": 470 |
| }, |
| { |
| "epoch": 5.277777777777778, |
| "grad_norm": 4.848036289215088, |
| "learning_rate": 2.7033966866696457e-05, |
| "loss": 0.0724, |
| "num_input_tokens_seen": 148736, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "grad_norm": 0.009390565566718578, |
| "learning_rate": 2.6550408237417885e-05, |
| "loss": 0.0094, |
| "num_input_tokens_seen": 150304, |
| "step": 480 |
| }, |
| { |
| "epoch": 5.388888888888889, |
| "grad_norm": 0.17524394392967224, |
| "learning_rate": 2.6066266563008267e-05, |
| "loss": 0.0719, |
| "num_input_tokens_seen": 151872, |
| "step": 485 |
| }, |
| { |
| "epoch": 5.444444444444445, |
| "grad_norm": 4.543068885803223, |
| "learning_rate": 2.5581723909351406e-05, |
| "loss": 0.0208, |
| "num_input_tokens_seen": 153472, |
| "step": 490 |
| }, |
| { |
| "epoch": 5.5, |
| "grad_norm": 0.07598326355218887, |
| "learning_rate": 2.5096962493123012e-05, |
| "loss": 0.0111, |
| "num_input_tokens_seen": 154976, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.5, |
| "eval_loss": 0.13528624176979065, |
| "eval_runtime": 0.563, |
| "eval_samples_per_second": 71.042, |
| "eval_steps_per_second": 17.76, |
| "num_input_tokens_seen": 154976, |
| "step": 495 |
| }, |
| { |
| "epoch": 5.555555555555555, |
| "grad_norm": 0.19703620672225952, |
| "learning_rate": 2.461216461326642e-05, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 156544, |
| "step": 500 |
| }, |
| { |
| "epoch": 5.611111111111111, |
| "grad_norm": 0.005900075193494558, |
| "learning_rate": 2.4127512582437485e-05, |
| "loss": 0.0417, |
| "num_input_tokens_seen": 158112, |
| "step": 505 |
| }, |
| { |
| "epoch": 5.666666666666667, |
| "grad_norm": 0.3932000398635864, |
| "learning_rate": 2.364318865844416e-05, |
| "loss": 0.0678, |
| "num_input_tokens_seen": 159680, |
| "step": 510 |
| }, |
| { |
| "epoch": 5.722222222222222, |
| "grad_norm": 0.0053927903063595295, |
| "learning_rate": 2.3159374975706884e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 161312, |
| "step": 515 |
| }, |
| { |
| "epoch": 5.777777777777778, |
| "grad_norm": 0.32272854447364807, |
| "learning_rate": 2.2676253476765196e-05, |
| "loss": 0.002, |
| "num_input_tokens_seen": 162880, |
| "step": 520 |
| }, |
| { |
| "epoch": 5.833333333333333, |
| "grad_norm": 0.006594807840883732, |
| "learning_rate": 2.2194005843856636e-05, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 164448, |
| "step": 525 |
| }, |
| { |
| "epoch": 5.888888888888889, |
| "grad_norm": 0.12406452000141144, |
| "learning_rate": 2.1712813430593436e-05, |
| "loss": 0.0127, |
| "num_input_tokens_seen": 166016, |
| "step": 530 |
| }, |
| { |
| "epoch": 5.944444444444445, |
| "grad_norm": 0.18200954794883728, |
| "learning_rate": 2.1232857193762924e-05, |
| "loss": 0.0355, |
| "num_input_tokens_seen": 167552, |
| "step": 535 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 0.36632028222084045, |
| "learning_rate": 2.0754317625276983e-05, |
| "loss": 0.0039, |
| "num_input_tokens_seen": 169056, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.14131739735603333, |
| "eval_runtime": 0.5687, |
| "eval_samples_per_second": 70.342, |
| "eval_steps_per_second": 17.585, |
| "num_input_tokens_seen": 169056, |
| "step": 540 |
| }, |
| { |
| "epoch": 6.055555555555555, |
| "grad_norm": 0.005900369491428137, |
| "learning_rate": 2.02773746842965e-05, |
| "loss": 0.033, |
| "num_input_tokens_seen": 170592, |
| "step": 545 |
| }, |
| { |
| "epoch": 6.111111111111111, |
| "grad_norm": 0.022186147049069405, |
| "learning_rate": 1.980220772955602e-05, |
| "loss": 0.0262, |
| "num_input_tokens_seen": 172192, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.166666666666667, |
| "grad_norm": 0.6460637450218201, |
| "learning_rate": 1.932899545191433e-05, |
| "loss": 0.107, |
| "num_input_tokens_seen": 173792, |
| "step": 555 |
| }, |
| { |
| "epoch": 6.222222222222222, |
| "grad_norm": 0.1838725209236145, |
| "learning_rate": 1.8857915807156092e-05, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 175360, |
| "step": 560 |
| }, |
| { |
| "epoch": 6.277777777777778, |
| "grad_norm": 0.007739254739135504, |
| "learning_rate": 1.838914594906995e-05, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 176992, |
| "step": 565 |
| }, |
| { |
| "epoch": 6.333333333333333, |
| "grad_norm": 0.03323983773589134, |
| "learning_rate": 1.792286216282824e-05, |
| "loss": 0.0067, |
| "num_input_tokens_seen": 178592, |
| "step": 570 |
| }, |
| { |
| "epoch": 6.388888888888889, |
| "grad_norm": 0.006993795745074749, |
| "learning_rate": 1.7459239798693364e-05, |
| "loss": 0.0043, |
| "num_input_tokens_seen": 180128, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.444444444444445, |
| "grad_norm": 5.127136707305908, |
| "learning_rate": 1.699845320607571e-05, |
| "loss": 0.0604, |
| "num_input_tokens_seen": 181632, |
| "step": 580 |
| }, |
| { |
| "epoch": 6.5, |
| "grad_norm": 9.503324508666992, |
| "learning_rate": 1.6540675667967974e-05, |
| "loss": 0.1049, |
| "num_input_tokens_seen": 183200, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.5, |
| "eval_loss": 0.13735271990299225, |
| "eval_runtime": 0.572, |
| "eval_samples_per_second": 69.928, |
| "eval_steps_per_second": 17.482, |
| "num_input_tokens_seen": 183200, |
| "step": 585 |
| }, |
| { |
| "epoch": 6.555555555555555, |
| "grad_norm": 0.00709711667150259, |
| "learning_rate": 1.60860793357805e-05, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 184800, |
| "step": 590 |
| }, |
| { |
| "epoch": 6.611111111111111, |
| "grad_norm": 0.05299042537808418, |
| "learning_rate": 1.56348351646022e-05, |
| "loss": 0.0147, |
| "num_input_tokens_seen": 186336, |
| "step": 595 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.27443385124206543, |
| "learning_rate": 1.5187112848911323e-05, |
| "loss": 0.0114, |
| "num_input_tokens_seen": 187904, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.722222222222222, |
| "grad_norm": 6.114893436431885, |
| "learning_rate": 1.47430807587603e-05, |
| "loss": 0.0357, |
| "num_input_tokens_seen": 189472, |
| "step": 605 |
| }, |
| { |
| "epoch": 6.777777777777778, |
| "grad_norm": 0.12536771595478058, |
| "learning_rate": 1.430290587645865e-05, |
| "loss": 0.0216, |
| "num_input_tokens_seen": 191072, |
| "step": 610 |
| }, |
| { |
| "epoch": 6.833333333333333, |
| "grad_norm": 0.10304760187864304, |
| "learning_rate": 1.3866753733777765e-05, |
| "loss": 0.0021, |
| "num_input_tokens_seen": 192608, |
| "step": 615 |
| }, |
| { |
| "epoch": 6.888888888888889, |
| "grad_norm": 0.01537319179624319, |
| "learning_rate": 1.343478834970121e-05, |
| "loss": 0.0124, |
| "num_input_tokens_seen": 194208, |
| "step": 620 |
| }, |
| { |
| "epoch": 6.944444444444445, |
| "grad_norm": 0.23381663858890533, |
| "learning_rate": 1.3007172168743854e-05, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 195776, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 0.19626721739768982, |
| "learning_rate": 1.2584065999863102e-05, |
| "loss": 0.0106, |
| "num_input_tokens_seen": 197344, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.1402236372232437, |
| "eval_runtime": 0.5618, |
| "eval_samples_per_second": 71.204, |
| "eval_steps_per_second": 17.801, |
| "num_input_tokens_seen": 197344, |
| "step": 630 |
| }, |
| { |
| "epoch": 7.055555555555555, |
| "grad_norm": 0.5308299660682678, |
| "learning_rate": 1.2165628955985314e-05, |
| "loss": 0.0937, |
| "num_input_tokens_seen": 198944, |
| "step": 635 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "grad_norm": 0.15454865992069244, |
| "learning_rate": 1.175201839416988e-05, |
| "loss": 0.0617, |
| "num_input_tokens_seen": 200512, |
| "step": 640 |
| }, |
| { |
| "epoch": 7.166666666666667, |
| "grad_norm": 0.017149705439805984, |
| "learning_rate": 1.1343389856433658e-05, |
| "loss": 0.0013, |
| "num_input_tokens_seen": 202016, |
| "step": 645 |
| }, |
| { |
| "epoch": 7.222222222222222, |
| "grad_norm": 0.33661720156669617, |
| "learning_rate": 1.0939897011258001e-05, |
| "loss": 0.0281, |
| "num_input_tokens_seen": 203648, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.277777777777778, |
| "grad_norm": 0.030632294714450836, |
| "learning_rate": 1.0541691595800337e-05, |
| "loss": 0.1214, |
| "num_input_tokens_seen": 205184, |
| "step": 655 |
| }, |
| { |
| "epoch": 7.333333333333333, |
| "grad_norm": 0.32496336102485657, |
| "learning_rate": 1.0148923358832022e-05, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 206720, |
| "step": 660 |
| }, |
| { |
| "epoch": 7.388888888888889, |
| "grad_norm": 0.3806283473968506, |
| "learning_rate": 9.761740004423927e-06, |
| "loss": 0.0021, |
| "num_input_tokens_seen": 208320, |
| "step": 665 |
| }, |
| { |
| "epoch": 7.444444444444445, |
| "grad_norm": 0.005406851880252361, |
| "learning_rate": 9.380287136401e-06, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 209856, |
| "step": 670 |
| }, |
| { |
| "epoch": 7.5, |
| "grad_norm": 1.4966259002685547, |
| "learning_rate": 9.00470820358663e-06, |
| "loss": 0.018, |
| "num_input_tokens_seen": 211392, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.5, |
| "eval_loss": 0.1403760015964508, |
| "eval_runtime": 0.5668, |
| "eval_samples_per_second": 70.566, |
| "eval_steps_per_second": 17.641, |
| "num_input_tokens_seen": 211392, |
| "step": 675 |
| }, |
| { |
| "epoch": 7.555555555555555, |
| "grad_norm": 0.24374495446681976, |
| "learning_rate": 8.635144445857406e-06, |
| "loss": 0.0012, |
| "num_input_tokens_seen": 212960, |
| "step": 680 |
| }, |
| { |
| "epoch": 7.611111111111111, |
| "grad_norm": 0.06474851071834564, |
| "learning_rate": 8.271734841028553e-06, |
| "loss": 0.0021, |
| "num_input_tokens_seen": 214528, |
| "step": 685 |
| }, |
| { |
| "epoch": 7.666666666666667, |
| "grad_norm": 0.008336671628057957, |
| "learning_rate": 7.914616052590071e-06, |
| "loss": 0.0053, |
| "num_input_tokens_seen": 216000, |
| "step": 690 |
| }, |
| { |
| "epoch": 7.722222222222222, |
| "grad_norm": 0.02428458444774151, |
| "learning_rate": 7.563922378313218e-06, |
| "loss": 0.0193, |
| "num_input_tokens_seen": 217632, |
| "step": 695 |
| }, |
| { |
| "epoch": 7.777777777777778, |
| "grad_norm": 0.007317886222153902, |
| "learning_rate": 7.219785699746573e-06, |
| "loss": 0.0322, |
| "num_input_tokens_seen": 219232, |
| "step": 700 |
| }, |
| { |
| "epoch": 7.833333333333333, |
| "grad_norm": 0.024397918954491615, |
| "learning_rate": 6.882335432620779e-06, |
| "loss": 0.0112, |
| "num_input_tokens_seen": 220800, |
| "step": 705 |
| }, |
| { |
| "epoch": 7.888888888888889, |
| "grad_norm": 0.006021441426128149, |
| "learning_rate": 6.55169847818059e-06, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 222368, |
| "step": 710 |
| }, |
| { |
| "epoch": 7.944444444444445, |
| "grad_norm": 0.417515367269516, |
| "learning_rate": 6.22799917546252e-06, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 223968, |
| "step": 715 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.19296586513519287, |
| "learning_rate": 5.9113592545359945e-06, |
| "loss": 0.0021, |
| "num_input_tokens_seen": 225536, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.14400552213191986, |
| "eval_runtime": 0.5746, |
| "eval_samples_per_second": 69.613, |
| "eval_steps_per_second": 17.403, |
| "num_input_tokens_seen": 225536, |
| "step": 720 |
| }, |
| { |
| "epoch": 8.055555555555555, |
| "grad_norm": 0.19193291664123535, |
| "learning_rate": 5.601897790725643e-06, |
| "loss": 0.0054, |
| "num_input_tokens_seen": 227168, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.11111111111111, |
| "grad_norm": 0.014541651122272015, |
| "learning_rate": 5.299731159831953e-06, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 228704, |
| "step": 730 |
| }, |
| { |
| "epoch": 8.166666666666666, |
| "grad_norm": 0.059697527438402176, |
| "learning_rate": 5.004972994367102e-06, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 230336, |
| "step": 735 |
| }, |
| { |
| "epoch": 8.222222222222221, |
| "grad_norm": 0.006615983322262764, |
| "learning_rate": 4.7177341408224e-06, |
| "loss": 0.1106, |
| "num_input_tokens_seen": 231936, |
| "step": 740 |
| }, |
| { |
| "epoch": 8.277777777777779, |
| "grad_norm": 0.03018399514257908, |
| "learning_rate": 4.438122617983443e-06, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 233472, |
| "step": 745 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.01191117987036705, |
| "learning_rate": 4.166243576308712e-06, |
| "loss": 0.024, |
| "num_input_tokens_seen": 235040, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.38888888888889, |
| "grad_norm": 0.05099299177527428, |
| "learning_rate": 3.9021992583867325e-06, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 236608, |
| "step": 755 |
| }, |
| { |
| "epoch": 8.444444444444445, |
| "grad_norm": 0.152078777551651, |
| "learning_rate": 3.6460889604868626e-06, |
| "loss": 0.0193, |
| "num_input_tokens_seen": 238144, |
| "step": 760 |
| }, |
| { |
| "epoch": 8.5, |
| "grad_norm": 0.332498699426651, |
| "learning_rate": 3.398008995217988e-06, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 239680, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.5, |
| "eval_loss": 0.1408601701259613, |
| "eval_runtime": 0.576, |
| "eval_samples_per_second": 69.449, |
| "eval_steps_per_second": 17.362, |
| "num_input_tokens_seen": 239680, |
| "step": 765 |
| }, |
| { |
| "epoch": 8.555555555555555, |
| "grad_norm": 0.005812807474285364, |
| "learning_rate": 3.158052655309332e-06, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 241280, |
| "step": 770 |
| }, |
| { |
| "epoch": 8.61111111111111, |
| "grad_norm": 0.28903236985206604, |
| "learning_rate": 2.9263101785268254e-06, |
| "loss": 0.1272, |
| "num_input_tokens_seen": 242816, |
| "step": 775 |
| }, |
| { |
| "epoch": 8.666666666666666, |
| "grad_norm": 0.23313328623771667, |
| "learning_rate": 2.7028687137384267e-06, |
| "loss": 0.028, |
| "num_input_tokens_seen": 244352, |
| "step": 780 |
| }, |
| { |
| "epoch": 8.722222222222221, |
| "grad_norm": 0.009992318227887154, |
| "learning_rate": 2.487812288140945e-06, |
| "loss": 0.0095, |
| "num_input_tokens_seen": 245856, |
| "step": 785 |
| }, |
| { |
| "epoch": 8.777777777777779, |
| "grad_norm": 1.103724479675293, |
| "learning_rate": 2.281221775660894e-06, |
| "loss": 0.007, |
| "num_input_tokens_seen": 247456, |
| "step": 790 |
| }, |
| { |
| "epoch": 8.833333333333334, |
| "grad_norm": 0.08995617181062698, |
| "learning_rate": 2.0831748665410765e-06, |
| "loss": 0.0066, |
| "num_input_tokens_seen": 248992, |
| "step": 795 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.01152830757200718, |
| "learning_rate": 1.893746038124497e-06, |
| "loss": 0.0171, |
| "num_input_tokens_seen": 250528, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.944444444444445, |
| "grad_norm": 0.13026604056358337, |
| "learning_rate": 1.713006526846439e-06, |
| "loss": 0.002, |
| "num_input_tokens_seen": 252128, |
| "step": 805 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 0.07047366350889206, |
| "learning_rate": 1.541024301445404e-06, |
| "loss": 0.0019, |
| "num_input_tokens_seen": 253696, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.14213652908802032, |
| "eval_runtime": 0.5631, |
| "eval_samples_per_second": 71.037, |
| "eval_steps_per_second": 17.759, |
| "num_input_tokens_seen": 253696, |
| "step": 810 |
| }, |
| { |
| "epoch": 9.055555555555555, |
| "grad_norm": 0.1695806384086609, |
| "learning_rate": 1.3778640374027985e-06, |
| "loss": 0.0143, |
| "num_input_tokens_seen": 255296, |
| "step": 815 |
| }, |
| { |
| "epoch": 9.11111111111111, |
| "grad_norm": 0.011573059484362602, |
| "learning_rate": 1.2235870926211619e-06, |
| "loss": 0.0061, |
| "num_input_tokens_seen": 256896, |
| "step": 820 |
| }, |
| { |
| "epoch": 9.166666666666666, |
| "grad_norm": 0.9394159913063049, |
| "learning_rate": 1.0782514843499653e-06, |
| "loss": 0.0041, |
| "num_input_tokens_seen": 258432, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.222222222222221, |
| "grad_norm": 0.023374371230602264, |
| "learning_rate": 9.419118673676924e-07, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 260000, |
| "step": 830 |
| }, |
| { |
| "epoch": 9.277777777777779, |
| "grad_norm": 0.10246945917606354, |
| "learning_rate": 8.146195134284052e-07, |
| "loss": 0.1567, |
| "num_input_tokens_seen": 261568, |
| "step": 835 |
| }, |
| { |
| "epoch": 9.333333333333334, |
| "grad_norm": 0.05011824518442154, |
| "learning_rate": 6.964222919805391e-07, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 263200, |
| "step": 840 |
| }, |
| { |
| "epoch": 9.38888888888889, |
| "grad_norm": 3.851982831954956, |
| "learning_rate": 5.87364652165176e-07, |
| "loss": 0.012, |
| "num_input_tokens_seen": 264736, |
| "step": 845 |
| }, |
| { |
| "epoch": 9.444444444444445, |
| "grad_norm": 0.12149734050035477, |
| "learning_rate": 4.874876061005173e-07, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 266304, |
| "step": 850 |
| }, |
| { |
| "epoch": 9.5, |
| "grad_norm": 0.10646462440490723, |
| "learning_rate": 3.9682871345891883e-07, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 267840, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.5, |
| "eval_loss": 0.14488616585731506, |
| "eval_runtime": 0.5732, |
| "eval_samples_per_second": 69.781, |
| "eval_steps_per_second": 17.445, |
| "num_input_tokens_seen": 267840, |
| "step": 855 |
| }, |
| { |
| "epoch": 9.555555555555555, |
| "grad_norm": 0.011037622578442097, |
| "learning_rate": 3.1542206734221924e-07, |
| "loss": 0.0043, |
| "num_input_tokens_seen": 269376, |
| "step": 860 |
| }, |
| { |
| "epoch": 9.61111111111111, |
| "grad_norm": 0.4150460362434387, |
| "learning_rate": 2.4329828146074095e-07, |
| "loss": 0.0163, |
| "num_input_tokens_seen": 270944, |
| "step": 865 |
| }, |
| { |
| "epoch": 9.666666666666666, |
| "grad_norm": 0.08698780089616776, |
| "learning_rate": 1.8048447862070718e-07, |
| "loss": 0.0264, |
| "num_input_tokens_seen": 272448, |
| "step": 870 |
| }, |
| { |
| "epoch": 9.722222222222221, |
| "grad_norm": 0.22865882515907288, |
| "learning_rate": 1.2700428052447033e-07, |
| "loss": 0.0237, |
| "num_input_tokens_seen": 273984, |
| "step": 875 |
| }, |
| { |
| "epoch": 9.777777777777779, |
| "grad_norm": 0.22310896217823029, |
| "learning_rate": 8.28777988873486e-08, |
| "loss": 0.0033, |
| "num_input_tokens_seen": 275520, |
| "step": 880 |
| }, |
| { |
| "epoch": 9.833333333333334, |
| "grad_norm": 0.003800946054980159, |
| "learning_rate": 4.8121627874450625e-08, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 277152, |
| "step": 885 |
| }, |
| { |
| "epoch": 9.88888888888889, |
| "grad_norm": 0.13200801610946655, |
| "learning_rate": 2.2748837860270267e-08, |
| "loss": 0.0624, |
| "num_input_tokens_seen": 278688, |
| "step": 890 |
| }, |
| { |
| "epoch": 9.944444444444445, |
| "grad_norm": 0.006874218117445707, |
| "learning_rate": 6.768970513457151e-09, |
| "loss": 0.002, |
| "num_input_tokens_seen": 280256, |
| "step": 895 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.13232700526714325, |
| "learning_rate": 1.8803520859811406e-10, |
| "loss": 0.0235, |
| "num_input_tokens_seen": 281856, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.14712795615196228, |
| "eval_runtime": 0.5625, |
| "eval_samples_per_second": 71.109, |
| "eval_steps_per_second": 17.777, |
| "num_input_tokens_seen": 281856, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.0, |
| "num_input_tokens_seen": 281856, |
| "step": 900, |
| "total_flos": 1.2692298538156032e+16, |
| "train_loss": 0.06281121766349922, |
| "train_runtime": 159.7285, |
| "train_samples_per_second": 22.538, |
| "train_steps_per_second": 5.635 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 900, |
| "num_input_tokens_seen": 281856, |
| "num_train_epochs": 10, |
| "save_steps": 45, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.2692298538156032e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
