| { | |
| "best_metric": 0.588302731513977, | |
| "best_model_checkpoint": "checkpoints/rft-llama-3.2-1b-gsm8k/gsm8k/llama-3.2-1b-instruct-step-1/checkpoint-603", | |
| "epoch": 1.0, | |
| "eval_steps": 67, | |
| "global_step": 664, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007530120481927711, | |
| "grad_norm": 69.0, | |
| "learning_rate": 7.462686567164179e-07, | |
| "loss": 1.987, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015060240963855422, | |
| "grad_norm": 58.5, | |
| "learning_rate": 1.4925373134328358e-06, | |
| "loss": 1.9009, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.022590361445783132, | |
| "grad_norm": 41.5, | |
| "learning_rate": 2.238805970149254e-06, | |
| "loss": 1.752, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.030120481927710843, | |
| "grad_norm": 24.375, | |
| "learning_rate": 2.9850746268656716e-06, | |
| "loss": 1.5653, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03765060240963856, | |
| "grad_norm": 19.0, | |
| "learning_rate": 3.73134328358209e-06, | |
| "loss": 1.3427, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.045180722891566265, | |
| "grad_norm": 13.5625, | |
| "learning_rate": 4.477611940298508e-06, | |
| "loss": 1.0619, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05271084337349398, | |
| "grad_norm": 8.5, | |
| "learning_rate": 5.2238805970149255e-06, | |
| "loss": 0.8756, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.060240963855421686, | |
| "grad_norm": 5.375, | |
| "learning_rate": 5.970149253731343e-06, | |
| "loss": 0.7862, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0677710843373494, | |
| "grad_norm": 3.140625, | |
| "learning_rate": 6.7164179104477625e-06, | |
| "loss": 0.7085, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.07530120481927711, | |
| "grad_norm": 2.90625, | |
| "learning_rate": 7.46268656716418e-06, | |
| "loss": 0.6669, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08283132530120482, | |
| "grad_norm": 2.421875, | |
| "learning_rate": 8.208955223880599e-06, | |
| "loss": 0.6517, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.09036144578313253, | |
| "grad_norm": 2.359375, | |
| "learning_rate": 8.955223880597016e-06, | |
| "loss": 0.6513, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09789156626506024, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 9.701492537313434e-06, | |
| "loss": 0.6549, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.10090361445783133, | |
| "eval_loss": 0.675618052482605, | |
| "eval_runtime": 3.6002, | |
| "eval_samples_per_second": 1722.964, | |
| "eval_steps_per_second": 26.943, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.10542168674698796, | |
| "grad_norm": 2.171875, | |
| "learning_rate": 9.949748743718594e-06, | |
| "loss": 0.6582, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.11295180722891567, | |
| "grad_norm": 2.125, | |
| "learning_rate": 9.865996649916248e-06, | |
| "loss": 0.6411, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.12048192771084337, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 9.782244556113903e-06, | |
| "loss": 0.6239, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1280120481927711, | |
| "grad_norm": 2.234375, | |
| "learning_rate": 9.698492462311559e-06, | |
| "loss": 0.6404, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.1355421686746988, | |
| "grad_norm": 2.328125, | |
| "learning_rate": 9.614740368509213e-06, | |
| "loss": 0.6233, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1430722891566265, | |
| "grad_norm": 2.6875, | |
| "learning_rate": 9.530988274706869e-06, | |
| "loss": 0.6253, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.15060240963855423, | |
| "grad_norm": 3.046875, | |
| "learning_rate": 9.447236180904523e-06, | |
| "loss": 0.6121, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15813253012048192, | |
| "grad_norm": 3.40625, | |
| "learning_rate": 9.363484087102177e-06, | |
| "loss": 0.622, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.16566265060240964, | |
| "grad_norm": 4.0625, | |
| "learning_rate": 9.279731993299833e-06, | |
| "loss": 0.59, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.17319277108433734, | |
| "grad_norm": 3.46875, | |
| "learning_rate": 9.195979899497488e-06, | |
| "loss": 0.5997, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.18072289156626506, | |
| "grad_norm": 2.359375, | |
| "learning_rate": 9.112227805695144e-06, | |
| "loss": 0.5906, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.18825301204819278, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 9.028475711892798e-06, | |
| "loss": 0.5887, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.19578313253012047, | |
| "grad_norm": 2.0, | |
| "learning_rate": 8.944723618090452e-06, | |
| "loss": 0.5751, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.20180722891566266, | |
| "eval_loss": 0.6029960513114929, | |
| "eval_runtime": 3.5961, | |
| "eval_samples_per_second": 1724.903, | |
| "eval_steps_per_second": 26.973, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.2033132530120482, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 8.860971524288108e-06, | |
| "loss": 0.571, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.21084337349397592, | |
| "grad_norm": 2.0, | |
| "learning_rate": 8.777219430485762e-06, | |
| "loss": 0.5661, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.2183734939759036, | |
| "grad_norm": 1.625, | |
| "learning_rate": 8.693467336683418e-06, | |
| "loss": 0.5578, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.22590361445783133, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 8.609715242881073e-06, | |
| "loss": 0.5639, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.23343373493975902, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 8.525963149078727e-06, | |
| "loss": 0.5549, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.24096385542168675, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 8.442211055276383e-06, | |
| "loss": 0.5524, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.24849397590361447, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 8.358458961474037e-06, | |
| "loss": 0.5786, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.2560240963855422, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 8.274706867671693e-06, | |
| "loss": 0.564, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2635542168674699, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 8.190954773869347e-06, | |
| "loss": 0.5666, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.2710843373493976, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 8.107202680067002e-06, | |
| "loss": 0.5531, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.2786144578313253, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 8.023450586264658e-06, | |
| "loss": 0.5663, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.286144578313253, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 7.939698492462312e-06, | |
| "loss": 0.5447, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2936746987951807, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 7.855946398659968e-06, | |
| "loss": 0.5554, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.30120481927710846, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 7.772194304857622e-06, | |
| "loss": 0.5854, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.30271084337349397, | |
| "eval_loss": 0.5955631136894226, | |
| "eval_runtime": 3.5892, | |
| "eval_samples_per_second": 1728.235, | |
| "eval_steps_per_second": 27.025, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 0.30873493975903615, | |
| "grad_norm": 1.890625, | |
| "learning_rate": 7.688442211055276e-06, | |
| "loss": 0.5483, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.31626506024096385, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 7.604690117252932e-06, | |
| "loss": 0.5649, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.32379518072289154, | |
| "grad_norm": 1.9375, | |
| "learning_rate": 7.520938023450587e-06, | |
| "loss": 0.5584, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.3313253012048193, | |
| "grad_norm": 1.7109375, | |
| "learning_rate": 7.437185929648242e-06, | |
| "loss": 0.551, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.338855421686747, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 7.353433835845897e-06, | |
| "loss": 0.5505, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.3463855421686747, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 7.269681742043552e-06, | |
| "loss": 0.5501, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.3539156626506024, | |
| "grad_norm": 1.8359375, | |
| "learning_rate": 7.185929648241206e-06, | |
| "loss": 0.5471, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.3614457831325301, | |
| "grad_norm": 1.8046875, | |
| "learning_rate": 7.102177554438861e-06, | |
| "loss": 0.5542, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.3689759036144578, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 7.0184254606365165e-06, | |
| "loss": 0.5382, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.37650602409638556, | |
| "grad_norm": 1.8125, | |
| "learning_rate": 6.934673366834172e-06, | |
| "loss": 0.5598, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.38403614457831325, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 6.850921273031827e-06, | |
| "loss": 0.5627, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.39156626506024095, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 6.767169179229481e-06, | |
| "loss": 0.5505, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.3990963855421687, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 6.683417085427136e-06, | |
| "loss": 0.5524, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.4036144578313253, | |
| "eval_loss": 0.5925902724266052, | |
| "eval_runtime": 3.5903, | |
| "eval_samples_per_second": 1727.697, | |
| "eval_steps_per_second": 27.017, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.4066265060240964, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 6.599664991624791e-06, | |
| "loss": 0.5574, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.4141566265060241, | |
| "grad_norm": 1.796875, | |
| "learning_rate": 6.515912897822446e-06, | |
| "loss": 0.5494, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.42168674698795183, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 6.4321608040201015e-06, | |
| "loss": 0.5602, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.4292168674698795, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 6.348408710217756e-06, | |
| "loss": 0.5534, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.4367469879518072, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 6.264656616415411e-06, | |
| "loss": 0.5446, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.4442771084337349, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 6.180904522613066e-06, | |
| "loss": 0.5474, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.45180722891566266, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 6.097152428810721e-06, | |
| "loss": 0.5516, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.45933734939759036, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 6.013400335008376e-06, | |
| "loss": 0.5547, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.46686746987951805, | |
| "grad_norm": 1.6875, | |
| "learning_rate": 5.9296482412060305e-06, | |
| "loss": 0.5807, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.4743975903614458, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 5.845896147403686e-06, | |
| "loss": 0.559, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.4819277108433735, | |
| "grad_norm": 1.6875, | |
| "learning_rate": 5.762144053601341e-06, | |
| "loss": 0.5426, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.4894578313253012, | |
| "grad_norm": 1.8828125, | |
| "learning_rate": 5.678391959798996e-06, | |
| "loss": 0.5428, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.49698795180722893, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 5.59463986599665e-06, | |
| "loss": 0.5416, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5045180722891566, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 5.510887772194305e-06, | |
| "loss": 0.556, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.5045180722891566, | |
| "eval_loss": 0.5903183817863464, | |
| "eval_runtime": 3.5762, | |
| "eval_samples_per_second": 1734.545, | |
| "eval_steps_per_second": 27.124, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.5120481927710844, | |
| "grad_norm": 1.875, | |
| "learning_rate": 5.42713567839196e-06, | |
| "loss": 0.5486, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5195783132530121, | |
| "grad_norm": 1.90625, | |
| "learning_rate": 5.3433835845896155e-06, | |
| "loss": 0.5599, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.5271084337349398, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 5.259631490787271e-06, | |
| "loss": 0.5544, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5346385542168675, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 5.175879396984925e-06, | |
| "loss": 0.5309, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.5421686746987951, | |
| "grad_norm": 1.921875, | |
| "learning_rate": 5.09212730318258e-06, | |
| "loss": 0.5532, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5496987951807228, | |
| "grad_norm": 1.7109375, | |
| "learning_rate": 5.008375209380235e-06, | |
| "loss": 0.5515, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.5572289156626506, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 4.92462311557789e-06, | |
| "loss": 0.5429, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.5647590361445783, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 4.840871021775545e-06, | |
| "loss": 0.5585, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.572289156626506, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 4.7571189279732e-06, | |
| "loss": 0.5551, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.5798192771084337, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 4.673366834170855e-06, | |
| "loss": 0.5318, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.5873493975903614, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 4.58961474036851e-06, | |
| "loss": 0.559, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.5948795180722891, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 4.505862646566165e-06, | |
| "loss": 0.5335, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.6024096385542169, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 4.42211055276382e-06, | |
| "loss": 0.5508, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6054216867469879, | |
| "eval_loss": 0.5893600583076477, | |
| "eval_runtime": 3.5911, | |
| "eval_samples_per_second": 1727.335, | |
| "eval_steps_per_second": 27.011, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 0.6099397590361446, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 4.338358458961474e-06, | |
| "loss": 0.5425, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.6174698795180723, | |
| "grad_norm": 1.71875, | |
| "learning_rate": 4.254606365159129e-06, | |
| "loss": 0.535, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 4.170854271356784e-06, | |
| "loss": 0.5483, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.6325301204819277, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 4.087102177554439e-06, | |
| "loss": 0.5447, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.6400602409638554, | |
| "grad_norm": 1.7109375, | |
| "learning_rate": 4.003350083752094e-06, | |
| "loss": 0.5443, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.6475903614457831, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 3.919597989949749e-06, | |
| "loss": 0.5469, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.6551204819277109, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 3.8358458961474034e-06, | |
| "loss": 0.5525, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.6626506024096386, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 3.7520938023450586e-06, | |
| "loss": 0.5524, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.6701807228915663, | |
| "grad_norm": 1.890625, | |
| "learning_rate": 3.6683417085427137e-06, | |
| "loss": 0.5452, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.677710843373494, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 3.5845896147403684e-06, | |
| "loss": 0.5524, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6852409638554217, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 3.5008375209380235e-06, | |
| "loss": 0.5508, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.6927710843373494, | |
| "grad_norm": 1.7109375, | |
| "learning_rate": 3.4170854271356786e-06, | |
| "loss": 0.5393, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.7003012048192772, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.5595, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.7063253012048193, | |
| "eval_loss": 0.588641345500946, | |
| "eval_runtime": 3.5901, | |
| "eval_samples_per_second": 1727.809, | |
| "eval_steps_per_second": 27.019, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 0.7078313253012049, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 3.2495812395309884e-06, | |
| "loss": 0.5581, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.7153614457831325, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 3.165829145728643e-06, | |
| "loss": 0.5469, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.7228915662650602, | |
| "grad_norm": 1.8203125, | |
| "learning_rate": 3.0820770519262983e-06, | |
| "loss": 0.5565, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7304216867469879, | |
| "grad_norm": 1.7421875, | |
| "learning_rate": 2.9983249581239534e-06, | |
| "loss": 0.5511, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 0.7379518072289156, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 2.914572864321608e-06, | |
| "loss": 0.5355, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.7454819277108434, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 2.830820770519263e-06, | |
| "loss": 0.5285, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 0.7530120481927711, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 2.747068676716918e-06, | |
| "loss": 0.5587, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7605421686746988, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 2.663316582914573e-06, | |
| "loss": 0.5516, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 0.7680722891566265, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 2.5795644891122277e-06, | |
| "loss": 0.5484, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.7756024096385542, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 2.495812395309883e-06, | |
| "loss": 0.5419, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 0.7831325301204819, | |
| "grad_norm": 1.75, | |
| "learning_rate": 2.412060301507538e-06, | |
| "loss": 0.5501, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.7906626506024096, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 2.3283082077051927e-06, | |
| "loss": 0.5378, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.7981927710843374, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 2.2445561139028478e-06, | |
| "loss": 0.5258, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.8057228915662651, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 2.1608040201005025e-06, | |
| "loss": 0.5488, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 0.8072289156626506, | |
| "eval_loss": 0.5883608460426331, | |
| "eval_runtime": 3.5937, | |
| "eval_samples_per_second": 1726.072, | |
| "eval_steps_per_second": 26.992, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 0.8132530120481928, | |
| "grad_norm": 1.8125, | |
| "learning_rate": 2.0770519262981576e-06, | |
| "loss": 0.553, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.8207831325301205, | |
| "grad_norm": 1.6796875, | |
| "learning_rate": 1.9932998324958123e-06, | |
| "loss": 0.5254, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 0.8283132530120482, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 1.9095477386934674e-06, | |
| "loss": 0.5468, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.8358433734939759, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 1.8257956448911223e-06, | |
| "loss": 0.5407, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 0.8433734939759037, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 1.7420435510887772e-06, | |
| "loss": 0.5445, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.8509036144578314, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 1.6582914572864323e-06, | |
| "loss": 0.5459, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 0.858433734939759, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 1.5745393634840873e-06, | |
| "loss": 0.5479, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.8659638554216867, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 1.4907872696817422e-06, | |
| "loss": 0.5485, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.8734939759036144, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 1.407035175879397e-06, | |
| "loss": 0.5581, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.8810240963855421, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 1.323283082077052e-06, | |
| "loss": 0.533, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.8885542168674698, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 1.2395309882747069e-06, | |
| "loss": 0.5246, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.8960843373493976, | |
| "grad_norm": 1.7890625, | |
| "learning_rate": 1.155778894472362e-06, | |
| "loss": 0.551, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.9036144578313253, | |
| "grad_norm": 1.7265625, | |
| "learning_rate": 1.072026800670017e-06, | |
| "loss": 0.5425, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9081325301204819, | |
| "eval_loss": 0.588302731513977, | |
| "eval_runtime": 3.591, | |
| "eval_samples_per_second": 1727.393, | |
| "eval_steps_per_second": 27.012, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 0.911144578313253, | |
| "grad_norm": 1.640625, | |
| "learning_rate": 9.882747068676718e-07, | |
| "loss": 0.5317, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 0.9186746987951807, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 9.045226130653267e-07, | |
| "loss": 0.54, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.9262048192771084, | |
| "grad_norm": 1.75, | |
| "learning_rate": 8.207705192629816e-07, | |
| "loss": 0.5274, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 0.9337349397590361, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 7.370184254606367e-07, | |
| "loss": 0.5478, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.9412650602409639, | |
| "grad_norm": 1.765625, | |
| "learning_rate": 6.532663316582916e-07, | |
| "loss": 0.5633, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.9487951807228916, | |
| "grad_norm": 1.875, | |
| "learning_rate": 5.695142378559465e-07, | |
| "loss": 0.5438, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.9563253012048193, | |
| "grad_norm": 1.8046875, | |
| "learning_rate": 4.857621440536014e-07, | |
| "loss": 0.54, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 0.963855421686747, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 4.0201005025125634e-07, | |
| "loss": 0.5439, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.9713855421686747, | |
| "grad_norm": 1.734375, | |
| "learning_rate": 3.1825795644891125e-07, | |
| "loss": 0.5337, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 0.9789156626506024, | |
| "grad_norm": 1.671875, | |
| "learning_rate": 2.3450586264656616e-07, | |
| "loss": 0.5338, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.9864457831325302, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 1.5075376884422112e-07, | |
| "loss": 0.5374, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 0.9939759036144579, | |
| "grad_norm": 1.78125, | |
| "learning_rate": 6.700167504187606e-08, | |
| "loss": 0.5403, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 664, | |
| "total_flos": 7.546108123702886e+16, | |
| "train_loss": 0.6115179532263653, | |
| "train_runtime": 216.1794, | |
| "train_samples_per_second": 196.448, | |
| "train_steps_per_second": 3.072 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 664, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 67, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.546108123702886e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |