| { | |
| "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": 4.545454545454545e-07, | |
| "loss": 2.3833, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 2.4784, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 2.3152, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 2.3229, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 2.2926, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 2.3494, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 2.3118, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.2727272727272727e-05, | |
| "loss": 2.5019, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 2.4156, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.6363636363636366e-05, | |
| "loss": 2.2803, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 2.1552, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2e-05, | |
| "loss": 2.2168, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 2.2977, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.3636363636363637e-05, | |
| "loss": 2.0998, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "gpt4_scores": 0.19999999999999998, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.0215020179748535, | |
| "eval_runtime": 4.9485, | |
| "eval_samples_per_second": 4.648, | |
| "eval_steps_per_second": 1.212, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 1.9844, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 1.9635, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 1.949, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.090909090909091e-05, | |
| "loss": 1.9375, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 2.0071, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 3.454545454545455e-05, | |
| "loss": 1.8369, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 1.8529, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 3.818181818181819e-05, | |
| "loss": 1.979, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4e-05, | |
| "loss": 1.9027, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.181818181818182e-05, | |
| "loss": 1.7632, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 1.8529, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 1.7896, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 1.7356, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 4.909090909090909e-05, | |
| "loss": 1.7439, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "gpt4_scores": 0.7333333333333334, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.8371074199676514, | |
| "eval_runtime": 4.9302, | |
| "eval_samples_per_second": 4.665, | |
| "eval_steps_per_second": 1.217, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 4.999949650182266e-05, | |
| "loss": 1.709, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.999546863808815e-05, | |
| "loss": 1.7374, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 4.9987413559579636e-05, | |
| "loss": 1.636, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 4.99753325641136e-05, | |
| "loss": 1.7623, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.995922759815339e-05, | |
| "loss": 1.7131, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.65, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 4.991495678185202e-05, | |
| "loss": 1.7748, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.988679806432712e-05, | |
| "loss": 1.7411, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.985462964079137e-05, | |
| "loss": 1.6517, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 4.981845669415022e-05, | |
| "loss": 1.5844, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 4.977828505250903e-05, | |
| "loss": 1.6811, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 4.973412118823412e-05, | |
| "loss": 1.6825, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 4.968597221690986e-05, | |
| "loss": 1.6176, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 4.963384589619233e-05, | |
| "loss": 1.7017, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "gpt4_scores": 0.6666666666666666, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.818211555480957, | |
| "eval_runtime": 4.9115, | |
| "eval_samples_per_second": 4.683, | |
| "eval_steps_per_second": 1.222, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.957775062455933e-05, | |
| "loss": 1.59, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.951769543995731e-05, | |
| "loss": 1.326, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 4.9453690018345144e-05, | |
| "loss": 1.4622, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 4.938574467213518e-05, | |
| "loss": 1.4188, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 4.931387034853173e-05, | |
| "loss": 1.4488, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 4.923807862776728e-05, | |
| "loss": 1.5268, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 4.915838172123671e-05, | |
| "loss": 1.3807, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 4.9074792469529815e-05, | |
| "loss": 1.4319, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 4.898732434036244e-05, | |
| "loss": 1.4653, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 4.889599142640663e-05, | |
| "loss": 1.417, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.880080844302004e-05, | |
| "loss": 1.4848, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 4.870179072587499e-05, | |
| "loss": 1.4331, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 4.859895422848767e-05, | |
| "loss": 1.3679, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 1.3662, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "gpt4_scores": 0.75, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.9195555448532104, | |
| "eval_runtime": 4.9614, | |
| "eval_samples_per_second": 4.636, | |
| "eval_steps_per_second": 1.209, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 4.838189178074867e-05, | |
| "loss": 1.1767, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 4.826770080301978e-05, | |
| "loss": 1.1217, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 4.8149760984659506e-05, | |
| "loss": 1.1938, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "learning_rate": 4.802809132787125e-05, | |
| "loss": 1.0658, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 4.790271143580174e-05, | |
| "loss": 0.9068, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 4.777364150938263e-05, | |
| "loss": 1.0878, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 4.764090234407577e-05, | |
| "loss": 1.0284, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 4.75045153265227e-05, | |
| "loss": 1.0695, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 4.7364502431098844e-05, | |
| "loss": 1.1011, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 4.722088621637309e-05, | |
| "loss": 1.0877, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.9747, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 4.692293696235758e-05, | |
| "loss": 0.9563, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 4.6768651927994434e-05, | |
| "loss": 1.0344, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "gpt4_scores": 0.6, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.226499557495117, | |
| "eval_runtime": 4.9165, | |
| "eval_samples_per_second": 4.678, | |
| "eval_steps_per_second": 1.22, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 4.6610859576448176e-05, | |
| "loss": 0.9582, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 4.644958533087443e-05, | |
| "loss": 0.7354, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.628485517542392e-05, | |
| "loss": 0.7346, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 4.611669565105596e-05, | |
| "loss": 0.6712, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 4.594513385126218e-05, | |
| "loss": 0.7341, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 4.5770197417701365e-05, | |
| "loss": 0.6719, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 4.559191453574582e-05, | |
| "loss": 0.6348, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 4.5410313929940244e-05, | |
| "loss": 0.7058, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 0.6653, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 4.503727711296538e-05, | |
| "loss": 0.6575, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 4.4845901004665234e-05, | |
| "loss": 0.6561, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 4.465132736856969e-05, | |
| "loss": 0.6728, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 4.445358755395382e-05, | |
| "loss": 0.703, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 4.425271342022039e-05, | |
| "loss": 0.6959, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "gpt4_scores": 0.7666666666666666, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.5187010765075684, | |
| "eval_runtime": 4.9193, | |
| "eval_samples_per_second": 4.675, | |
| "eval_steps_per_second": 1.22, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 4.404873733176678e-05, | |
| "loss": 0.6332, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 4.384169215277041e-05, | |
| "loss": 0.3578, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 4.3631611241893874e-05, | |
| "loss": 0.3878, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 4.341852844691012e-05, | |
| "loss": 0.3941, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 4.3202478099249105e-05, | |
| "loss": 0.3927, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 4.2983495008466276e-05, | |
| "loss": 0.4278, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 4.276161445663423e-05, | |
| "loss": 0.4638, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 4.2536872192658036e-05, | |
| "loss": 0.4473, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 4.230930442651557e-05, | |
| "loss": 0.4563, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 4.2078947823423364e-05, | |
| "loss": 0.3735, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 4.1845839497929204e-05, | |
| "loss": 0.4068, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 4.161001700793231e-05, | |
| "loss": 0.455, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 4.137151834863213e-05, | |
| "loss": 0.4346, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 4.113038194640658e-05, | |
| "loss": 0.4353, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "gpt4_scores": 0.75, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.8303465843200684, | |
| "eval_runtime": 4.9148, | |
| "eval_samples_per_second": 4.68, | |
| "eval_steps_per_second": 1.221, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 4.088664665262091e-05, | |
| "loss": 0.3461, | |
| "step": 388 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 4.064035173736804e-05, | |
| "loss": 0.2204, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.2423, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 4.014024217844167e-05, | |
| "loss": 0.2308, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "learning_rate": 3.98865081113172e-05, | |
| "loss": 0.2553, | |
| "step": 404 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 3.9630375562841295e-05, | |
| "loss": 0.2258, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 3.937188580052518e-05, | |
| "loss": 0.264, | |
| "step": 412 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 3.911108047166924e-05, | |
| "loss": 0.2548, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 3.884800159665276e-05, | |
| "loss": 0.272, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "learning_rate": 3.858269156216383e-05, | |
| "loss": 0.2596, | |
| "step": 424 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 3.8315193114369996e-05, | |
| "loss": 0.24, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 3.804554935203115e-05, | |
| "loss": 0.2434, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 3.7773803719555514e-05, | |
| "loss": 0.2694, | |
| "step": 436 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2736, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "gpt4_scores": 0.5833333333333334, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 3.1702194213867188, | |
| "eval_runtime": 4.9672, | |
| "eval_samples_per_second": 4.63, | |
| "eval_steps_per_second": 1.208, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 3.7224182308015975e-05, | |
| "loss": 0.1419, | |
| "step": 444 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 3.694639508274158e-05, | |
| "loss": 0.148, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 3.6666683080641846e-05, | |
| "loss": 0.1635, | |
| "step": 452 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 3.638509136829758e-05, | |
| "loss": 0.1667, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "learning_rate": 3.610166531514436e-05, | |
| "loss": 0.164, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 8.44, | |
| "learning_rate": 3.581645058616271e-05, | |
| "loss": 0.1428, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 3.552949313452067e-05, | |
| "loss": 0.1392, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "learning_rate": 3.5240839194169885e-05, | |
| "loss": 0.1557, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "learning_rate": 3.495053527239656e-05, | |
| "loss": 0.1714, | |
| "step": 476 | |
| }, | |
| { | |
| "epoch": 8.73, | |
| "learning_rate": 3.465862814232822e-05, | |
| "loss": 0.1622, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 3.436516483539781e-05, | |
| "loss": 0.18, | |
| "step": 484 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 3.4070192633766025e-05, | |
| "loss": 0.1654, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 3.3773759062703396e-05, | |
| "loss": 0.1598, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "gpt4_scores": 0.5499999999999999, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 3.373944044113159, | |
| "eval_runtime": 4.9104, | |
| "eval_samples_per_second": 4.684, | |
| "eval_steps_per_second": 1.222, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 3.3475911882933015e-05, | |
| "loss": 0.1388, | |
| "step": 496 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 3.3176699082935545e-05, | |
| "loss": 0.1064, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 3.2876168871217325e-05, | |
| "loss": 0.1153, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 3.257436966854319e-05, | |
| "loss": 0.1225, | |
| "step": 508 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 3.2271350100134975e-05, | |
| "loss": 0.1146, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 3.19671589878372e-05, | |
| "loss": 0.1202, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 3.166184534225087e-05, | |
| "loss": 0.1132, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 3.135545835483718e-05, | |
| "loss": 0.106, | |
| "step": 524 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.1127, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 3.073966197709103e-05, | |
| "loss": 0.1059, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "learning_rate": 3.0430351802512698e-05, | |
| "loss": 0.1089, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 3.012016670162977e-05, | |
| "loss": 0.1139, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 2.9809156650781528e-05, | |
| "loss": 0.1247, | |
| "step": 544 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "learning_rate": 2.9497371759221347e-05, | |
| "loss": 0.117, | |
| "step": 548 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "gpt4_scores": 0.7000000000000001, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 3.4963889122009277, | |
| "eval_runtime": 4.9532, | |
| "eval_samples_per_second": 4.643, | |
| "eval_steps_per_second": 1.211, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.918486226104327e-05, | |
| "loss": 0.0894, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 2.8871678507088312e-05, | |
| "loss": 0.0835, | |
| "step": 556 | |
| }, | |
| { | |
| "epoch": 10.18, | |
| "learning_rate": 2.8557870956832132e-05, | |
| "loss": 0.0913, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 10.25, | |
| "learning_rate": 2.8243490170255043e-05, | |
| "loss": 0.0737, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 10.33, | |
| "learning_rate": 2.792858679969596e-05, | |
| "loss": 0.0897, | |
| "step": 568 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 2.761321158169134e-05, | |
| "loss": 0.0825, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "learning_rate": 2.7297415328800692e-05, | |
| "loss": 0.0845, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 2.698124892141971e-05, | |
| "loss": 0.0819, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 2.6664763299582602e-05, | |
| "loss": 0.0868, | |
| "step": 584 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 2.6348009454754653e-05, | |
| "loss": 0.083, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 2.6031038421616683e-05, | |
| "loss": 0.0937, | |
| "step": 592 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 2.5713901269842404e-05, | |
| "loss": 0.0872, | |
| "step": 596 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 2.5396649095870202e-05, | |
| "loss": 0.0869, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "learning_rate": 2.507933301467056e-05, | |
| "loss": 0.0856, | |
| "step": 604 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "gpt4_scores": 0.48333333333333334, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 3.6468489170074463, | |
| "eval_runtime": 4.9449, | |
| "eval_samples_per_second": 4.651, | |
| "eval_steps_per_second": 1.213, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.4762004151510584e-05, | |
| "loss": 0.0808, | |
| "step": 608 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 2.4444713633716765e-05, | |
| "loss": 0.0641, | |
| "step": 612 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 2.4127512582437485e-05, | |
| "loss": 0.0699, | |
| "step": 616 | |
| }, | |
| { | |
| "epoch": 11.27, | |
| "learning_rate": 2.3810452104406444e-05, | |
| "loss": 0.0692, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 2.349358328370854e-05, | |
| "loss": 0.0689, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 11.42, | |
| "learning_rate": 2.3176957173549235e-05, | |
| "loss": 0.0763, | |
| "step": 628 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 2.2860624788029013e-05, | |
| "loss": 0.0609, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 11.56, | |
| "learning_rate": 2.2544637093924074e-05, | |
| "loss": 0.0679, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "learning_rate": 2.222904500247473e-05, | |
| "loss": 0.0675, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "learning_rate": 2.1913899361182632e-05, | |
| "loss": 0.0741, | |
| "step": 644 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 2.1599250945618402e-05, | |
| "loss": 0.0644, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 2.1285150451240713e-05, | |
| "loss": 0.0672, | |
| "step": 652 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 2.09716484852284e-05, | |
| "loss": 0.0722, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.0681, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "gpt4_scores": 0.7166666666666667, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 3.8001902103424072, | |
| "eval_runtime": 4.9149, | |
| "eval_samples_per_second": 4.68, | |
| "eval_steps_per_second": 1.221, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 2.034664207670925e-05, | |
| "loss": 0.0662, | |
| "step": 664 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 2.003523833385637e-05, | |
| "loss": 0.0572, | |
| "step": 668 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 1.972463450245226e-05, | |
| "loss": 0.061, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "learning_rate": 1.9414880626301147e-05, | |
| "loss": 0.0611, | |
| "step": 676 | |
| }, | |
| { | |
| "epoch": 12.36, | |
| "learning_rate": 1.9106026612264316e-05, | |
| "loss": 0.0631, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "learning_rate": 1.879812222221929e-05, | |
| "loss": 0.0586, | |
| "step": 684 | |
| }, | |
| { | |
| "epoch": 12.51, | |
| "learning_rate": 1.84912170650422e-05, | |
| "loss": 0.0597, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 12.58, | |
| "learning_rate": 1.8185360588615058e-05, | |
| "loss": 0.0566, | |
| "step": 692 | |
| }, | |
| { | |
| "epoch": 12.65, | |
| "learning_rate": 1.7880602071858692e-05, | |
| "loss": 0.0569, | |
| "step": 696 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 1.7576990616793137e-05, | |
| "loss": 0.0548, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 0.058, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 12.87, | |
| "learning_rate": 1.697340436787273e-05, | |
| "loss": 0.0512, | |
| "step": 708 | |
| }, | |
| { | |
| "epoch": 12.95, | |
| "learning_rate": 1.667352682250298e-05, | |
| "loss": 0.058, | |
| "step": 712 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "gpt4_scores": 0.5, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 3.918207883834839, | |
| "eval_runtime": 4.9544, | |
| "eval_samples_per_second": 4.642, | |
| "eval_steps_per_second": 1.211, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 1.637499082012574e-05, | |
| "loss": 0.0681, | |
| "step": 716 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "learning_rate": 1.6077844460203206e-05, | |
| "loss": 0.0547, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 1.5782135618301486e-05, | |
| "loss": 0.0449, | |
| "step": 724 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 1.5487911938376924e-05, | |
| "loss": 0.0555, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 1.5195220825099862e-05, | |
| "loss": 0.0508, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 1.4904109436216884e-05, | |
| "loss": 0.0527, | |
| "step": 736 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 1.4614624674952842e-05, | |
| "loss": 0.0521, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "learning_rate": 1.4326813182453958e-05, | |
| "loss": 0.0554, | |
| "step": 744 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 1.4040721330273062e-05, | |
| "loss": 0.05, | |
| "step": 748 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 1.3756395212898359e-05, | |
| "loss": 0.0509, | |
| "step": 752 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "learning_rate": 1.3473880640326725e-05, | |
| "loss": 0.0517, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 13.82, | |
| "learning_rate": 1.3193223130682936e-05, | |
| "loss": 0.0557, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 1.2914467902885901e-05, | |
| "loss": 0.0524, | |
| "step": 764 | |
| }, | |
| { | |
| "epoch": 13.96, | |
| "learning_rate": 1.2637659869363083e-05, | |
| "loss": 0.0532, | |
| "step": 768 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "gpt4_scores": 0.5166666666666667, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 4.011674880981445, | |
| "eval_runtime": 4.9088, | |
| "eval_samples_per_second": 4.685, | |
| "eval_steps_per_second": 1.222, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 1.2362843628814267e-05, | |
| "loss": 0.0456, | |
| "step": 772 | |
| }, | |
| { | |
| "epoch": 14.11, | |
| "learning_rate": 1.2090063459025955e-05, | |
| "loss": 0.0476, | |
| "step": 776 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 1.181936330973744e-05, | |
| "loss": 0.0501, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 14.25, | |
| "learning_rate": 1.155078679555969e-05, | |
| "loss": 0.0497, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 14.33, | |
| "learning_rate": 1.1284377188948258e-05, | |
| "loss": 0.0496, | |
| "step": 788 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.0467, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 1.0758230035694031e-05, | |
| "loss": 0.0539, | |
| "step": 796 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 1.049857726072005e-05, | |
| "loss": 0.0446, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 1.024126092299176e-05, | |
| "loss": 0.0477, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 9.986322480749927e-06, | |
| "loss": 0.0529, | |
| "step": 808 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 9.733803009114045e-06, | |
| "loss": 0.0466, | |
| "step": 812 | |
| }, | |
| { | |
| "epoch": 14.84, | |
| "learning_rate": 9.483743193464408e-06, | |
| "loss": 0.0518, | |
| "step": 816 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "learning_rate": 9.236183322886945e-06, | |
| "loss": 0.0479, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "learning_rate": 8.991163283681944e-06, | |
| "loss": 0.0443, | |
| "step": 824 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "gpt4_scores": 0.6666666666666666, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 4.091893672943115, | |
| "eval_runtime": 4.9566, | |
| "eval_samples_per_second": 4.64, | |
| "eval_steps_per_second": 1.211, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "learning_rate": 8.748722552937689e-06, | |
| "loss": 0.0471, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 15.13, | |
| "learning_rate": 8.508900192169964e-06, | |
| "loss": 0.0467, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "learning_rate": 8.271734841028553e-06, | |
| "loss": 0.0498, | |
| "step": 836 | |
| }, | |
| { | |
| "epoch": 15.27, | |
| "learning_rate": 8.0372647110717e-06, | |
| "loss": 0.0476, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 15.35, | |
| "learning_rate": 7.805527579609576e-06, | |
| "loss": 0.0481, | |
| "step": 844 | |
| }, | |
| { | |
| "epoch": 15.42, | |
| "learning_rate": 7.576560783617668e-06, | |
| "loss": 0.0466, | |
| "step": 848 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 7.350401213721089e-06, | |
| "loss": 0.0455, | |
| "step": 852 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 7.127085308250914e-06, | |
| "loss": 0.048, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 15.64, | |
| "learning_rate": 6.906649047373246e-06, | |
| "loss": 0.0422, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "learning_rate": 6.689127947292231e-06, | |
| "loss": 0.0442, | |
| "step": 864 | |
| }, | |
| { | |
| "epoch": 15.78, | |
| "learning_rate": 6.4745570545277075e-06, | |
| "loss": 0.0414, | |
| "step": 868 | |
| }, | |
| { | |
| "epoch": 15.85, | |
| "learning_rate": 6.2629709402686535e-06, | |
| "loss": 0.0455, | |
| "step": 872 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 6.054403694803079e-06, | |
| "loss": 0.0519, | |
| "step": 876 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 5.848888922025553e-06, | |
| "loss": 0.0464, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 5.646459734022938e-06, | |
| "loss": 0.0443, | |
| "step": 884 | |
| }, | |
| { | |
| "epoch": 16.15, | |
| "learning_rate": 5.4471487457395225e-06, | |
| "loss": 0.0405, | |
| "step": 888 | |
| }, | |
| { | |
| "epoch": 16.22, | |
| "learning_rate": 5.250988069722096e-06, | |
| "loss": 0.0441, | |
| "step": 892 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "learning_rate": 5.058009310946119e-06, | |
| "loss": 0.0412, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 4.868243561723535e-06, | |
| "loss": 0.0468, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 16.44, | |
| "learning_rate": 4.681721396693303e-06, | |
| "loss": 0.0476, | |
| "step": 904 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 4.498472867895223e-06, | |
| "loss": 0.0481, | |
| "step": 908 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "learning_rate": 4.318527499928074e-06, | |
| "loss": 0.0485, | |
| "step": 912 | |
| }, | |
| { | |
| "epoch": 16.65, | |
| "learning_rate": 4.141914285192619e-06, | |
| "loss": 0.0441, | |
| "step": 916 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "learning_rate": 3.968661679220468e-06, | |
| "loss": 0.0469, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.0436, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 16.87, | |
| "learning_rate": 3.632349403925664e-06, | |
| "loss": 0.0528, | |
| "step": 928 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 3.4693439204949858e-06, | |
| "loss": 0.0461, | |
| "step": 932 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "gpt4_scores": 0.4833333333333334, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 4.148024082183838, | |
| "eval_runtime": 4.957, | |
| "eval_samples_per_second": 4.64, | |
| "eval_steps_per_second": 1.21, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 3.3098074088812686e-06, | |
| "loss": 0.0472, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 17.09, | |
| "learning_rate": 3.1537655732553768e-06, | |
| "loss": 0.0422, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 3.0012435547336737e-06, | |
| "loss": 0.0415, | |
| "step": 944 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.8522659273273604e-06, | |
| "loss": 0.0446, | |
| "step": 948 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 2.7068566939831645e-06, | |
| "loss": 0.0477, | |
| "step": 952 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 2.565039282716045e-06, | |
| "loss": 0.0448, | |
| "step": 956 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 2.4268365428344736e-06, | |
| "loss": 0.0408, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 2.29227074125907e-06, | |
| "loss": 0.0472, | |
| "step": 964 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 2.1613635589349756e-06, | |
| "loss": 0.0485, | |
| "step": 968 | |
| }, | |
| { | |
| "epoch": 17.67, | |
| "learning_rate": 2.0341360873386674e-06, | |
| "loss": 0.0419, | |
| "step": 972 | |
| }, | |
| { | |
| "epoch": 17.75, | |
| "learning_rate": 1.9106088250797267e-06, | |
| "loss": 0.0515, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 17.82, | |
| "learning_rate": 1.790801674598186e-06, | |
| "loss": 0.0502, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 1.674733938957873e-06, | |
| "loss": 0.0422, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "learning_rate": 1.5624243187363441e-06, | |
| "loss": 0.0465, | |
| "step": 988 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "gpt4_scores": 0.4000000000000001, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 4.16222620010376, | |
| "eval_runtime": 4.9609, | |
| "eval_samples_per_second": 4.636, | |
| "eval_steps_per_second": 1.209, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 18.04, | |
| "learning_rate": 1.4538909090118846e-06, | |
| "loss": 0.0463, | |
| "step": 992 | |
| }, | |
| { | |
| "epoch": 18.11, | |
| "learning_rate": 1.3491511964480702e-06, | |
| "loss": 0.0433, | |
| "step": 996 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 1.248222056476367e-06, | |
| "loss": 0.0388, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 18.25, | |
| "learning_rate": 1.1511197505771843e-06, | |
| "loss": 0.0491, | |
| "step": 1004 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "learning_rate": 1.0578599236598707e-06, | |
| "loss": 0.0486, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 9.684576015420278e-07, | |
| "loss": 0.0537, | |
| "step": 1012 | |
| }, | |
| { | |
| "epoch": 18.47, | |
| "learning_rate": 8.829271885286094e-07, | |
| "loss": 0.0464, | |
| "step": 1016 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 8.012824650910938e-07, | |
| "loss": 0.0418, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 7.235365856472442e-07, | |
| "loss": 0.0506, | |
| "step": 1024 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 6.497020764416633e-07, | |
| "loss": 0.0398, | |
| "step": 1028 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 5.797908335276214e-07, | |
| "loss": 0.0418, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 5.1381412085037e-07, | |
| "loss": 0.041, | |
| "step": 1036 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "learning_rate": 4.517825684323324e-07, | |
| "loss": 0.0446, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "learning_rate": 3.9370617066040726e-07, | |
| "loss": 0.049, | |
| "step": 1044 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "gpt4_scores": 0.5333333333333333, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 4.166710376739502, | |
| "eval_runtime": 4.9226, | |
| "eval_samples_per_second": 4.672, | |
| "eval_steps_per_second": 1.219, | |
| "step": 1045 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "learning_rate": 3.395942846757066e-07, | |
| "loss": 0.0449, | |
| "step": 1048 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "learning_rate": 2.894556288659395e-07, | |
| "loss": 0.0444, | |
| "step": 1052 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.0465, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 19.27, | |
| "learning_rate": 2.011296792301165e-07, | |
| "loss": 0.0464, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 1.6295661628624447e-07, | |
| "loss": 0.0446, | |
| "step": 1064 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "learning_rate": 1.2878524298882698e-07, | |
| "loss": 0.0425, | |
| "step": 1068 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 9.862106495415469e-08, | |
| "loss": 0.054, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 19.56, | |
| "learning_rate": 7.246894216806355e-08, | |
| "loss": 0.0421, | |
| "step": 1076 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 5.033308820289184e-08, | |
| "loss": 0.045, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 19.71, | |
| "learning_rate": 3.221706953860093e-08, | |
| "loss": 0.0455, | |
| "step": 1084 | |
| }, | |
| { | |
| "epoch": 19.78, | |
| "learning_rate": 1.812380498815991e-08, | |
| "loss": 0.0411, | |
| "step": 1088 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "learning_rate": 8.055565227271799e-09, | |
| "loss": 0.0504, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 2.0139724285161977e-09, | |
| "loss": 0.0432, | |
| "step": 1096 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0416, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "gpt4_scores": 0.43333333333333335, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 4.167160511016846, | |
| "eval_runtime": 4.9522, | |
| "eval_samples_per_second": 4.644, | |
| "eval_steps_per_second": 1.212, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 1100, | |
| "total_flos": 3.795619210798694e+16, | |
| "train_loss": 0.009085019393400713, | |
| "train_runtime": 1593.5865, | |
| "train_samples_per_second": 2.723, | |
| "train_steps_per_second": 0.69 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 1100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 55, | |
| "total_flos": 3.795619210798694e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |