| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 1100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 2.3833, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 2.4734, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 2.2655, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 2.2091, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 2.1358, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 2.0997, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00010909090909090909, | |
| "loss": 1.931, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00012727272727272725, | |
| "loss": 2.0453, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00014545454545454546, | |
| "loss": 1.9392, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00016363636363636363, | |
| "loss": 1.8909, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 1.7724, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.854, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.00021818181818181818, | |
| "loss": 1.9588, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00023636363636363636, | |
| "loss": 1.8768, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "gpt4_scores": 0.67, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "std": 0.12413702106946178, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.84342360496521, | |
| "eval_runtime": 4.9467, | |
| "eval_samples_per_second": 4.65, | |
| "eval_steps_per_second": 1.213, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002545454545454545, | |
| "loss": 1.7451, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.00027272727272727274, | |
| "loss": 1.7394, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.0002909090909090909, | |
| "loss": 1.6971, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0003090909090909091, | |
| "loss": 1.6419, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00032727272727272726, | |
| "loss": 1.65, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00034545454545454544, | |
| "loss": 1.594, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00036363636363636367, | |
| "loss": 1.5985, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00038181818181818184, | |
| "loss": 1.719, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0004, | |
| "loss": 1.655, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00041818181818181814, | |
| "loss": 1.6052, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00043636363636363637, | |
| "loss": 1.6288, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00045454545454545455, | |
| "loss": 1.6272, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0004727272727272727, | |
| "loss": 1.6124, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.0004909090909090909, | |
| "loss": 1.5109, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "gpt4_scores": 0.675, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "std": 0.11688669727560959, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.89065682888031, | |
| "eval_runtime": 4.9546, | |
| "eval_samples_per_second": 4.642, | |
| "eval_steps_per_second": 1.211, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0004999949650182266, | |
| "loss": 1.3323, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0004999546863808815, | |
| "loss": 1.0715, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0004998741355957963, | |
| "loss": 0.8812, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.0004997533256411359, | |
| "loss": 0.9464, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0004995922759815339, | |
| "loss": 0.9049, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0004993910125649561, | |
| "loss": 0.9246, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.0004991495678185201, | |
| "loss": 1.0853, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0004988679806432712, | |
| "loss": 1.1289, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0004985462964079136, | |
| "loss": 0.9598, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0004981845669415021, | |
| "loss": 0.8825, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.0004977828505250904, | |
| "loss": 0.985, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0004973412118823412, | |
| "loss": 0.9869, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0004968597221690986, | |
| "loss": 0.985, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 0.0004963384589619233, | |
| "loss": 1.1215, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "gpt4_scores": 0.67, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "std": 0.11920570456148481, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.164942741394043, | |
| "eval_runtime": 4.9498, | |
| "eval_samples_per_second": 4.647, | |
| "eval_steps_per_second": 1.212, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 0.0004957775062455933, | |
| "loss": 0.8362, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 0.0004951769543995731, | |
| "loss": 0.3812, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0004945369001834514, | |
| "loss": 0.4813, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 0.0004938574467213517, | |
| "loss": 0.4693, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.0004931387034853173, | |
| "loss": 0.447, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.0004923807862776728, | |
| "loss": 0.6197, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.0004915838172123671, | |
| "loss": 0.5158, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 0.0004907479246952981, | |
| "loss": 0.5785, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 0.0004898732434036243, | |
| "loss": 0.5258, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.0004889599142640663, | |
| "loss": 0.4577, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.0004880080844302004, | |
| "loss": 0.5965, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 0.0004870179072587499, | |
| "loss": 0.5373, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.0004859895422848767, | |
| "loss": 0.4731, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0004849231551964771, | |
| "loss": 0.4876, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "gpt4_scores": 0.61, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "std": 0.12445882853377657, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.6900253295898438, | |
| "eval_runtime": 4.9381, | |
| "eval_samples_per_second": 4.658, | |
| "eval_steps_per_second": 1.215, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 0.00048381891780748665, | |
| "loss": 0.288, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 0.00048267700803019775, | |
| "loss": 0.2556, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.0004814976098465951, | |
| "loss": 0.2897, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 0.00048028091327871256, | |
| "loss": 0.2375, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.0004790271143580174, | |
| "loss": 0.2367, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 0.00047773641509382626, | |
| "loss": 0.2878, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 0.0004764090234407577, | |
| "loss": 0.3138, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.00047504515326522696, | |
| "loss": 0.277, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 0.0004736450243109884, | |
| "loss": 0.2624, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.0004722088621637309, | |
| "loss": 0.2888, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.00047073689821473173, | |
| "loss": 0.2256, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 0.00046922936962357577, | |
| "loss": 0.2243, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.00046768651927994433, | |
| "loss": 0.2916, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "gpt4_scores": 0.63, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "std": 0.12573782247199924, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.821211099624634, | |
| "eval_runtime": 4.9351, | |
| "eval_samples_per_second": 4.661, | |
| "eval_steps_per_second": 1.216, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 0.0004661085957644817, | |
| "loss": 0.233, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 0.0004644958533087443, | |
| "loss": 0.1537, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.0004628485517542392, | |
| "loss": 0.1743, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 0.0004611669565105596, | |
| "loss": 0.1827, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 0.00045945133851262184, | |
| "loss": 0.1928, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.00045770197417701366, | |
| "loss": 0.1718, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 0.0004559191453574582, | |
| "loss": 0.166, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 0.00045410313929940244, | |
| "loss": 0.1484, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.0004522542485937369, | |
| "loss": 0.1521, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 0.00045037277112965383, | |
| "loss": 0.195, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 0.0004484590100466523, | |
| "loss": 0.153, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 0.0004465132736856969, | |
| "loss": 0.1751, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 0.0004445358755395382, | |
| "loss": 0.2153, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 0.00044252713420220394, | |
| "loss": 0.1983, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "gpt4_scores": 0.7500000000000001, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "std": 0.0972111104761179, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.8983585834503174, | |
| "eval_runtime": 4.9684, | |
| "eval_samples_per_second": 4.629, | |
| "eval_steps_per_second": 1.208, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 0.0004404873733176677, | |
| "loss": 0.1353, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.00043841692152770415, | |
| "loss": 0.102, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 0.0004363161124189387, | |
| "loss": 0.1055, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 0.00043418528446910123, | |
| "loss": 0.0998, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 0.00043202478099249104, | |
| "loss": 0.1168, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.0004298349500846628, | |
| "loss": 0.1156, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 0.00042761614456634226, | |
| "loss": 0.1345, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 0.00042536872192658034, | |
| "loss": 0.0968, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.0004230930442651557, | |
| "loss": 0.1179, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 0.00042078947823423365, | |
| "loss": 0.0959, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 0.00041845839497929203, | |
| "loss": 0.1078, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 0.0004161001700793231, | |
| "loss": 0.1175, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 0.0004137151834863213, | |
| "loss": 0.1095, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 0.00041130381946406574, | |
| "loss": 0.1199, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "gpt4_scores": 0.49000000000000005, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "std": 0.12525973016097391, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.949794292449951, | |
| "eval_runtime": 4.9646, | |
| "eval_samples_per_second": 4.633, | |
| "eval_steps_per_second": 1.209, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 0.0004088664665262091, | |
| "loss": 0.076, | |
| "step": 388 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 0.0004064035173736804, | |
| "loss": 0.0709, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 0.00040391536883141455, | |
| "loss": 0.073, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 0.00040140242178441667, | |
| "loss": 0.0819, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 0.000398865081113172, | |
| "loss": 0.0643, | |
| "step": 404 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.0003963037556284129, | |
| "loss": 0.1603, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 0.0003937188580052518, | |
| "loss": 0.119, | |
| "step": 412 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 0.0003911108047166924, | |
| "loss": 0.0754, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 0.0003884800159665276, | |
| "loss": 0.0622, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "learning_rate": 0.00038582691562163827, | |
| "loss": 0.0711, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 0.00038315193114369994, | |
| "loss": 0.0643, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 0.0003804554935203115, | |
| "loss": 0.0672, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 0.00037773803719555514, | |
| "loss": 0.0688, | |
| "step": 436 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.000375, | |
| "loss": 0.0743, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "gpt4_scores": 0.51, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "std": 0.12763228431709586, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 3.250723123550415, | |
| "eval_runtime": 4.9396, | |
| "eval_samples_per_second": 4.656, | |
| "eval_steps_per_second": 1.215, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.00037224182308015974, | |
| "loss": 0.0528, | |
| "step": 444 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 0.0003694639508274158, | |
| "loss": 0.0502, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 0.00036666683080641843, | |
| "loss": 0.0659, | |
| "step": 452 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 0.0003638509136829758, | |
| "loss": 0.065, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "learning_rate": 0.00036101665315144355, | |
| "loss": 0.0524, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 0.00035816450586162706, | |
| "loss": 0.0565, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 0.00035529493134520666, | |
| "loss": 0.0432, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "learning_rate": 0.00035240839194169884, | |
| "loss": 0.0582, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "learning_rate": 0.0003495053527239656, | |
| "loss": 0.0507, | |
| "step": 476 | |
| }, | |
| { | |
| "epoch": 8.73, | |
| "learning_rate": 0.00034658628142328216, | |
| "loss": 0.0571, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 0.00034365164835397803, | |
| "loss": 0.0925, | |
| "step": 484 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 0.00034070192633766023, | |
| "loss": 0.0601, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 0.00033773759062703394, | |
| "loss": 0.062, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "gpt4_scores": 0.53, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "std": 0.1192057045614848, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 3.1474077701568604, | |
| "eval_runtime": 4.9704, | |
| "eval_samples_per_second": 4.627, | |
| "eval_steps_per_second": 1.207, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 0.0003347591188293301, | |
| "loss": 0.0535, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 0.00033176699082935546, | |
| "loss": 0.0484, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 0.00032876168871217323, | |
| "loss": 0.0541, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 0.00032574369668543187, | |
| "loss": 0.0438, | |
| "step": 508 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 0.00032271350100134975, | |
| "loss": 0.0451, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 0.00031967158987837195, | |
| "loss": 0.0403, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 0.0003166184534225087, | |
| "loss": 0.0517, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 0.0003135545835483718, | |
| "loss": 0.0498, | |
| "step": 524 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 0.0003104804738999169, | |
| "loss": 0.0516, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 0.00030739661977091025, | |
| "loss": 0.0487, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "learning_rate": 0.00030430351802512693, | |
| "loss": 0.045, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 0.0003012016670162977, | |
| "loss": 0.0397, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 0.00029809156650781527, | |
| "loss": 0.0484, | |
| "step": 544 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "learning_rate": 0.0002949737175922135, | |
| "loss": 0.0404, | |
| "step": 548 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "gpt4_scores": 0.6749999999999999, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "std": 0.11602801385872293, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 3.3666398525238037, | |
| "eval_runtime": 4.9475, | |
| "eval_samples_per_second": 4.649, | |
| "eval_steps_per_second": 1.213, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 0.0002918486226104327, | |
| "loss": 0.0403, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 0.0002887167850708831, | |
| "loss": 0.0401, | |
| "step": 556 | |
| }, | |
| { | |
| "epoch": 10.18, | |
| "learning_rate": 0.00028557870956832135, | |
| "loss": 0.0442, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 10.25, | |
| "learning_rate": 0.00028243490170255044, | |
| "loss": 0.041, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 10.33, | |
| "learning_rate": 0.0002792858679969596, | |
| "loss": 0.036, | |
| "step": 568 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 0.0002761321158169134, | |
| "loss": 0.0533, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "learning_rate": 0.0002729741532880069, | |
| "loss": 0.0506, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 0.0002698124892141971, | |
| "loss": 0.0532, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 0.000266647632995826, | |
| "loss": 0.0468, | |
| "step": 584 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 0.0002634800945475465, | |
| "loss": 0.0396, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 0.00026031038421616684, | |
| "loss": 0.0383, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 0.00025713901269842405, | |
| "loss": 0.0475, | |
| "step": 596 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 0.000253966490958702, | |
| "loss": 0.0491, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "learning_rate": 0.00025079333014670557, | |
| "loss": 0.038, | |
| "step": 604 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "gpt4_scores": 0.63, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "std": 0.11580155439371269, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 3.3349289894104004, | |
| "eval_runtime": 4.9384, | |
| "eval_samples_per_second": 4.657, | |
| "eval_steps_per_second": 1.215, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 0.00024762004151510585, | |
| "loss": 0.0482, | |
| "step": 608 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 0.00024444713633716764, | |
| "loss": 0.0356, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 0.00024127512582437484, | |
| "loss": 0.0444, | |
| "step": 616 | |
| }, | |
| { | |
| "epoch": 11.27, | |
| "learning_rate": 0.00023810452104406444, | |
| "loss": 0.0411, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 0.00023493583283708543, | |
| "loss": 0.0441, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 11.42, | |
| "learning_rate": 0.00023176957173549233, | |
| "loss": 0.0393, | |
| "step": 628 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 0.00022860624788029015, | |
| "loss": 0.0363, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 0.00022544637093924072, | |
| "loss": 0.0393, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "learning_rate": 0.00022229045002474727, | |
| "loss": 0.0452, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "learning_rate": 0.00021913899361182632, | |
| "loss": 0.0417, | |
| "step": 644 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 0.000215992509456184, | |
| "loss": 0.0366, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 0.00021285150451240712, | |
| "loss": 0.0443, | |
| "step": 652 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 0.000209716484852284, | |
| "loss": 0.0447, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.00020658795558326743, | |
| "loss": 0.0374, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "gpt4_scores": 0.6100000000000001, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "std": 0.12841339493993606, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 3.4455862045288086, | |
| "eval_runtime": 4.9572, | |
| "eval_samples_per_second": 4.64, | |
| "eval_steps_per_second": 1.21, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 0.0002034664207670925, | |
| "loss": 0.0442, | |
| "step": 664 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 0.00020035238333856371, | |
| "loss": 0.0391, | |
| "step": 668 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 0.0001972463450245226, | |
| "loss": 0.0392, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "learning_rate": 0.00019414880626301146, | |
| "loss": 0.0426, | |
| "step": 676 | |
| }, | |
| { | |
| "epoch": 12.36, | |
| "learning_rate": 0.00019106026612264316, | |
| "loss": 0.043, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "learning_rate": 0.0001879812222221929, | |
| "loss": 0.0395, | |
| "step": 684 | |
| }, | |
| { | |
| "epoch": 12.51, | |
| "learning_rate": 0.00018491217065042198, | |
| "loss": 0.0396, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 12.58, | |
| "learning_rate": 0.00018185360588615057, | |
| "loss": 0.0394, | |
| "step": 692 | |
| }, | |
| { | |
| "epoch": 12.65, | |
| "learning_rate": 0.00017880602071858692, | |
| "loss": 0.0403, | |
| "step": 696 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 0.00017576990616793137, | |
| "loss": 0.0349, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.00017274575140626317, | |
| "loss": 0.0404, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 12.87, | |
| "learning_rate": 0.0001697340436787273, | |
| "loss": 0.0344, | |
| "step": 708 | |
| }, | |
| { | |
| "epoch": 12.95, | |
| "learning_rate": 0.00016673526822502983, | |
| "loss": 0.0384, | |
| "step": 712 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "gpt4_scores": 0.53, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "std": 0.13787675656179324, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 3.4822309017181396, | |
| "eval_runtime": 4.946, | |
| "eval_samples_per_second": 4.65, | |
| "eval_steps_per_second": 1.213, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 0.0001637499082012574, | |
| "loss": 0.0522, | |
| "step": 716 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "learning_rate": 0.00016077844460203207, | |
| "loss": 0.0384, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 0.00015782135618301485, | |
| "loss": 0.0336, | |
| "step": 724 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 0.00015487911938376925, | |
| "loss": 0.0401, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 0.00015195220825099862, | |
| "loss": 0.0395, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 0.00014904109436216883, | |
| "loss": 0.0414, | |
| "step": 736 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 0.0001461462467495284, | |
| "loss": 0.0393, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "learning_rate": 0.00014326813182453956, | |
| "loss": 0.0383, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 0.00014040721330273062, | |
| "loss": 0.038, | |
| "step": 748 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 0.0001375639521289836, | |
| "loss": 0.0383, | |
| "step": 752 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 0.00013473880640326724, | |
| "loss": 0.0405, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "learning_rate": 0.00013193223130682935, | |
| "loss": 0.0422, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 0.000129144679028859, | |
| "loss": 0.0408, | |
| "step": 764 | |
| }, | |
| { | |
| "epoch": 13.96, | |
| "learning_rate": 0.00012637659869363084, | |
| "loss": 0.0408, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "gpt4_scores": 0.6500000000000001, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "std": 0.11423659658795862, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 3.471827983856201, | |
| "eval_runtime": 4.9416, | |
| "eval_samples_per_second": 4.654, | |
| "eval_steps_per_second": 1.214, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 0.00012362843628814266, | |
| "loss": 0.0343, | |
| "step": 772 | |
| }, | |
| { | |
| "epoch": 14.11, | |
| "learning_rate": 0.00012090063459025954, | |
| "loss": 0.0386, | |
| "step": 776 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 0.00011819363309737438, | |
| "loss": 0.0412, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 0.0001155078679555969, | |
| "loss": 0.0402, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 14.33, | |
| "learning_rate": 0.00011284377188948258, | |
| "loss": 0.038, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 0.00011020177413231333, | |
| "loss": 0.0367, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 0.0001075823003569403, | |
| "loss": 0.0449, | |
| "step": 796 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 0.00010498577260720049, | |
| "loss": 0.036, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 0.00010241260922991761, | |
| "loss": 0.0385, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 9.986322480749927e-05, | |
| "loss": 0.0428, | |
| "step": 808 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 9.733803009114044e-05, | |
| "loss": 0.0375, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 14.84, | |
| "learning_rate": 9.483743193464408e-05, | |
| "loss": 0.0435, | |
| "step": 816 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 9.236183322886945e-05, | |
| "loss": 0.0394, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "learning_rate": 8.991163283681945e-05, | |
| "loss": 0.0347, | |
| "step": 824 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "gpt4_scores": 0.67, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "std": 0.11644741302407709, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 3.502847671508789, | |
| "eval_runtime": 4.9425, | |
| "eval_samples_per_second": 4.654, | |
| "eval_steps_per_second": 1.214, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "learning_rate": 8.748722552937688e-05, | |
| "loss": 0.0395, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 15.13, | |
| "learning_rate": 8.508900192169963e-05, | |
| "loss": 0.0389, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "learning_rate": 8.271734841028553e-05, | |
| "loss": 0.0425, | |
| "step": 836 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "learning_rate": 8.037264711071699e-05, | |
| "loss": 0.0403, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 15.35, | |
| "learning_rate": 7.805527579609576e-05, | |
| "loss": 0.0405, | |
| "step": 844 | |
| }, | |
| { | |
| "epoch": 15.42, | |
| "learning_rate": 7.576560783617667e-05, | |
| "loss": 0.0392, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 7.35040121372109e-05, | |
| "loss": 0.0383, | |
| "step": 852 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 7.127085308250913e-05, | |
| "loss": 0.0404, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 15.64, | |
| "learning_rate": 6.906649047373245e-05, | |
| "loss": 0.0347, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "learning_rate": 6.689127947292231e-05, | |
| "loss": 0.0363, | |
| "step": 864 | |
| }, | |
| { | |
| "epoch": 15.78, | |
| "learning_rate": 6.474557054527707e-05, | |
| "loss": 0.0341, | |
| "step": 868 | |
| }, | |
| { | |
| "epoch": 15.85, | |
| "learning_rate": 6.262970940268654e-05, | |
| "loss": 0.0377, | |
| "step": 872 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 6.054403694803079e-05, | |
| "loss": 0.0442, | |
| "step": 876 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 5.848888922025553e-05, | |
| "loss": 0.0377, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "gpt4_scores": 0.5700000000000001, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "std": 0.1319469590403659, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 3.5217506885528564, | |
| "eval_runtime": 4.9355, | |
| "eval_samples_per_second": 4.66, | |
| "eval_steps_per_second": 1.216, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 5.646459734022938e-05, | |
| "loss": 0.041, | |
| "step": 884 | |
| }, | |
| { | |
| "epoch": 16.15, | |
| "learning_rate": 5.4471487457395216e-05, | |
| "loss": 0.0333, | |
| "step": 888 | |
| }, | |
| { | |
| "epoch": 16.22, | |
| "learning_rate": 5.2509880697220956e-05, | |
| "loss": 0.0407, | |
| "step": 892 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "learning_rate": 5.058009310946118e-05, | |
| "loss": 0.0391, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 4.8682435617235344e-05, | |
| "loss": 0.0365, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 16.44, | |
| "learning_rate": 4.6817213966933034e-05, | |
| "loss": 0.0425, | |
| "step": 904 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 4.498472867895223e-05, | |
| "loss": 0.0363, | |
| "step": 908 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "learning_rate": 4.318527499928074e-05, | |
| "loss": 0.0391, | |
| "step": 912 | |
| }, | |
| { | |
| "epoch": 16.65, | |
| "learning_rate": 4.141914285192619e-05, | |
| "loss": 0.0375, | |
| "step": 916 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "learning_rate": 3.968661679220467e-05, | |
| "loss": 0.0388, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "learning_rate": 3.798797596089351e-05, | |
| "loss": 0.0417, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 16.87, | |
| "learning_rate": 3.632349403925664e-05, | |
| "loss": 0.0378, | |
| "step": 928 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 3.4693439204949856e-05, | |
| "loss": 0.0395, | |
| "step": 932 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "gpt4_scores": 0.595, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "std": 0.1132364782214636, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 3.5319790840148926, | |
| "eval_runtime": 4.9408, | |
| "eval_samples_per_second": 4.655, | |
| "eval_steps_per_second": 1.214, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 3.309807408881269e-05, | |
| "loss": 0.0416, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 17.09, | |
| "learning_rate": 3.1537655732553766e-05, | |
| "loss": 0.0409, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 3.0012435547336736e-05, | |
| "loss": 0.0412, | |
| "step": 944 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.8522659273273603e-05, | |
| "loss": 0.0411, | |
| "step": 948 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 2.7068566939831645e-05, | |
| "loss": 0.0332, | |
| "step": 952 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 2.5650392827160445e-05, | |
| "loss": 0.0315, | |
| "step": 956 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 2.4268365428344735e-05, | |
| "loss": 0.0448, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 2.29227074125907e-05, | |
| "loss": 0.0349, | |
| "step": 964 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 2.1613635589349755e-05, | |
| "loss": 0.0367, | |
| "step": 968 | |
| }, | |
| { | |
| "epoch": 17.67, | |
| "learning_rate": 2.0341360873386672e-05, | |
| "loss": 0.0386, | |
| "step": 972 | |
| }, | |
| { | |
| "epoch": 17.75, | |
| "learning_rate": 1.9106088250797264e-05, | |
| "loss": 0.0371, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 17.82, | |
| "learning_rate": 1.7908016745981858e-05, | |
| "loss": 0.0369, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 1.674733938957873e-05, | |
| "loss": 0.0464, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "learning_rate": 1.562424318736344e-05, | |
| "loss": 0.0408, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "gpt4_scores": 0.59, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "std": 0.13224976370489286, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 3.537071466445923, | |
| "eval_runtime": 4.9404, | |
| "eval_samples_per_second": 4.656, | |
| "eval_steps_per_second": 1.214, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 18.04, | |
| "learning_rate": 1.4538909090118846e-05, | |
| "loss": 0.0424, | |
| "step": 992 | |
| }, | |
| { | |
| "epoch": 18.11, | |
| "learning_rate": 1.3491511964480702e-05, | |
| "loss": 0.0406, | |
| "step": 996 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 1.2482220564763668e-05, | |
| "loss": 0.0404, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 18.25, | |
| "learning_rate": 1.1511197505771842e-05, | |
| "loss": 0.039, | |
| "step": 1004 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "learning_rate": 1.0578599236598707e-05, | |
| "loss": 0.0398, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 9.684576015420277e-06, | |
| "loss": 0.0347, | |
| "step": 1012 | |
| }, | |
| { | |
| "epoch": 18.47, | |
| "learning_rate": 8.829271885286095e-06, | |
| "loss": 0.0386, | |
| "step": 1016 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 8.012824650910938e-06, | |
| "loss": 0.034, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 7.235365856472442e-06, | |
| "loss": 0.0349, | |
| "step": 1024 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 6.497020764416634e-06, | |
| "loss": 0.0376, | |
| "step": 1028 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 5.797908335276214e-06, | |
| "loss": 0.0422, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 5.1381412085036995e-06, | |
| "loss": 0.0404, | |
| "step": 1036 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "learning_rate": 4.517825684323323e-06, | |
| "loss": 0.0356, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "learning_rate": 3.937061706604072e-06, | |
| "loss": 0.0468, | |
| "step": 1044 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "gpt4_scores": 0.575, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "std": 0.12791598805466028, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 3.5391297340393066, | |
| "eval_runtime": 4.9485, | |
| "eval_samples_per_second": 4.648, | |
| "eval_steps_per_second": 1.212, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "learning_rate": 3.3959428467570664e-06, | |
| "loss": 0.0362, | |
| "step": 1048 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 2.8945562886593944e-06, | |
| "loss": 0.0408, | |
| "step": 1052 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 2.4329828146074094e-06, | |
| "loss": 0.0391, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 19.27, | |
| "learning_rate": 2.011296792301165e-06, | |
| "loss": 0.0376, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 1.6295661628624448e-06, | |
| "loss": 0.0446, | |
| "step": 1064 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "learning_rate": 1.2878524298882698e-06, | |
| "loss": 0.0332, | |
| "step": 1068 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 9.862106495415469e-07, | |
| "loss": 0.0375, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 19.56, | |
| "learning_rate": 7.246894216806354e-07, | |
| "loss": 0.0451, | |
| "step": 1076 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 5.033308820289185e-07, | |
| "loss": 0.0451, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 19.71, | |
| "learning_rate": 3.221706953860093e-07, | |
| "loss": 0.0354, | |
| "step": 1084 | |
| }, | |
| { | |
| "epoch": 19.78, | |
| "learning_rate": 1.8123804988159908e-07, | |
| "loss": 0.0397, | |
| "step": 1088 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "learning_rate": 8.0555652272718e-08, | |
| "loss": 0.0394, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 2.0139724285161975e-08, | |
| "loss": 0.0321, | |
| "step": 1096 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0358, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "gpt4_scores": 0.58, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "std": 0.12692517480783708, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 3.5393826961517334, | |
| "eval_runtime": 4.9612, | |
| "eval_samples_per_second": 4.636, | |
| "eval_steps_per_second": 1.209, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 1100, | |
| "total_flos": 3.792163606268314e+16, | |
| "train_loss": 0.31520286625081845, | |
| "train_runtime": 10333.7168, | |
| "train_samples_per_second": 0.42, | |
| "train_steps_per_second": 0.106 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 1100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 55, | |
| "total_flos": 3.792163606268314e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |