| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 200, | |
| "global_step": 533, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.018788163457022077, | |
| "grad_norm": 39.5539436340332, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 1.7708, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03757632691404415, | |
| "grad_norm": 15.056221961975098, | |
| "learning_rate": 3.5185185185185187e-06, | |
| "loss": 0.8157, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05636449037106623, | |
| "grad_norm": 8.456525802612305, | |
| "learning_rate": 5.370370370370371e-06, | |
| "loss": 0.5049, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0751526538280883, | |
| "grad_norm": 9.841835021972656, | |
| "learning_rate": 7.222222222222223e-06, | |
| "loss": 0.4008, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09394081728511038, | |
| "grad_norm": 8.734881401062012, | |
| "learning_rate": 9.074074074074075e-06, | |
| "loss": 0.3481, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11272898074213246, | |
| "grad_norm": 5.559770107269287, | |
| "learning_rate": 9.997311749002358e-06, | |
| "loss": 0.3233, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.13151714419915453, | |
| "grad_norm": 9.457917213439941, | |
| "learning_rate": 9.975823081977089e-06, | |
| "loss": 0.2962, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1503053076561766, | |
| "grad_norm": 8.458663940429688, | |
| "learning_rate": 9.93293815016874e-06, | |
| "loss": 0.3505, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.1690934711131987, | |
| "grad_norm": 5.281663417816162, | |
| "learning_rate": 9.868841360727856e-06, | |
| "loss": 0.341, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.18788163457022075, | |
| "grad_norm": 4.8133440017700195, | |
| "learning_rate": 9.783808332754242e-06, | |
| "loss": 0.3644, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.20666979802724283, | |
| "grad_norm": 5.978756904602051, | |
| "learning_rate": 9.678204712122328e-06, | |
| "loss": 0.2572, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.22545796148426492, | |
| "grad_norm": 3.6604723930358887, | |
| "learning_rate": 9.552484599187344e-06, | |
| "loss": 0.2801, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.244246124941287, | |
| "grad_norm": 5.864062786102295, | |
| "learning_rate": 9.407188596133212e-06, | |
| "loss": 0.2438, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.26303428839830906, | |
| "grad_norm": 4.652751445770264, | |
| "learning_rate": 9.242941482358646e-06, | |
| "loss": 0.2426, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.28182245185533117, | |
| "grad_norm": 7.59506368637085, | |
| "learning_rate": 9.060449527897424e-06, | |
| "loss": 0.2699, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3006106153123532, | |
| "grad_norm": 5.299258232116699, | |
| "learning_rate": 8.860497456425226e-06, | |
| "loss": 0.2438, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3193987787693753, | |
| "grad_norm": 6.822812080383301, | |
| "learning_rate": 8.643945070912269e-06, | |
| "loss": 0.2703, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.3381869422263974, | |
| "grad_norm": 4.299904823303223, | |
| "learning_rate": 8.411723556431555e-06, | |
| "loss": 0.2858, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.35697510568341945, | |
| "grad_norm": 4.70183801651001, | |
| "learning_rate": 8.164831476020856e-06, | |
| "loss": 0.1675, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.3757632691404415, | |
| "grad_norm": 6.74561071395874, | |
| "learning_rate": 7.904330476816391e-06, | |
| "loss": 0.2649, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3757632691404415, | |
| "eval_loss": 0.22606511414051056, | |
| "eval_runtime": 222.2942, | |
| "eval_samples_per_second": 4.256, | |
| "eval_steps_per_second": 1.066, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3945514325974636, | |
| "grad_norm": 5.191539287567139, | |
| "learning_rate": 7.631340724922023e-06, | |
| "loss": 0.2252, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.41333959605448567, | |
| "grad_norm": 6.738773822784424, | |
| "learning_rate": 7.347036088644232e-06, | |
| "loss": 0.2396, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.4321277595115077, | |
| "grad_norm": 5.1947808265686035, | |
| "learning_rate": 7.0526390908052e-06, | |
| "loss": 0.2902, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.45091592296852984, | |
| "grad_norm": 4.540430545806885, | |
| "learning_rate": 6.7494156518392625e-06, | |
| "loss": 0.2185, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4697040864255519, | |
| "grad_norm": 7.267935752868652, | |
| "learning_rate": 6.43866964627766e-06, | |
| "loss": 0.2483, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.488492249882574, | |
| "grad_norm": 6.775832653045654, | |
| "learning_rate": 6.121737296028959e-06, | |
| "loss": 0.2132, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5072804133395961, | |
| "grad_norm": 4.134191036224365, | |
| "learning_rate": 5.799981424564275e-06, | |
| "loss": 0.2062, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5260685767966181, | |
| "grad_norm": 3.4992332458496094, | |
| "learning_rate": 5.474785596714581e-06, | |
| "loss": 0.1665, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.5448567402536402, | |
| "grad_norm": 3.856785297393799, | |
| "learning_rate": 5.1475481692792235e-06, | |
| "loss": 0.1902, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5636449037106623, | |
| "grad_norm": 5.150210857391357, | |
| "learning_rate": 4.819676278028305e-06, | |
| "loss": 0.2333, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5824330671676844, | |
| "grad_norm": 5.201061725616455, | |
| "learning_rate": 4.4925797869550865e-06, | |
| "loss": 0.2023, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6012212306247064, | |
| "grad_norm": 5.331509590148926, | |
| "learning_rate": 4.167665225796925e-06, | |
| "loss": 0.1776, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6200093940817285, | |
| "grad_norm": 3.550278902053833, | |
| "learning_rate": 3.846329741893646e-06, | |
| "loss": 0.2364, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.6387975575387506, | |
| "grad_norm": 5.495516300201416, | |
| "learning_rate": 3.52995509239065e-06, | |
| "loss": 0.2192, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.6575857209957726, | |
| "grad_norm": 4.215266704559326, | |
| "learning_rate": 3.2199017026205744e-06, | |
| "loss": 0.1601, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6763738844527948, | |
| "grad_norm": 3.173532247543335, | |
| "learning_rate": 2.917502816212685e-06, | |
| "loss": 0.1688, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6951620479098168, | |
| "grad_norm": 6.164701461791992, | |
| "learning_rate": 2.6240587620848512e-06, | |
| "loss": 0.2083, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.7139502113668389, | |
| "grad_norm": 2.307567596435547, | |
| "learning_rate": 2.340831362970257e-06, | |
| "loss": 0.181, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.732738374823861, | |
| "grad_norm": 5.143068313598633, | |
| "learning_rate": 2.0690385095224557e-06, | |
| "loss": 0.2234, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.751526538280883, | |
| "grad_norm": 2.648653030395508, | |
| "learning_rate": 1.8098489233303595e-06, | |
| "loss": 0.1551, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.751526538280883, | |
| "eval_loss": 0.17390167713165283, | |
| "eval_runtime": 222.2168, | |
| "eval_samples_per_second": 4.257, | |
| "eval_steps_per_second": 1.067, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7703147017379052, | |
| "grad_norm": 3.061445713043213, | |
| "learning_rate": 1.5643771313624394e-06, | |
| "loss": 0.1899, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.7891028651949272, | |
| "grad_norm": 3.341491460800171, | |
| "learning_rate": 1.3336786734502294e-06, | |
| "loss": 0.1615, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.8078910286519493, | |
| "grad_norm": 3.6715078353881836, | |
| "learning_rate": 1.1187455634192307e-06, | |
| "loss": 0.1671, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.8266791921089713, | |
| "grad_norm": 2.944624423980713, | |
| "learning_rate": 9.205020233844736e-07, | |
| "loss": 0.1275, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.8454673555659934, | |
| "grad_norm": 4.107908725738525, | |
| "learning_rate": 7.398005095535565e-07, | |
| "loss": 0.1797, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8642555190230155, | |
| "grad_norm": 5.744738578796387, | |
| "learning_rate": 5.774180466262985e-07, | |
| "loss": 0.2238, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.8830436824800376, | |
| "grad_norm": 3.282698631286621, | |
| "learning_rate": 4.340528865533161e-07, | |
| "loss": 0.1766, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.9018318459370597, | |
| "grad_norm": 6.375194549560547, | |
| "learning_rate": 3.103215060209902e-07, | |
| "loss": 0.2085, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.9206200093940817, | |
| "grad_norm": 2.820821523666382, | |
| "learning_rate": 2.0675595557376916e-07, | |
| "loss": 0.1595, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.9394081728511038, | |
| "grad_norm": 3.918117046356201, | |
| "learning_rate": 1.2380157177271369e-07, | |
| "loss": 0.2025, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9581963363081258, | |
| "grad_norm": 2.6554410457611084, | |
| "learning_rate": 6.181506222809885e-08, | |
| "loss": 0.1641, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.976984499765148, | |
| "grad_norm": 4.8615899085998535, | |
| "learning_rate": 2.1062971740523076e-08, | |
| "loss": 0.1658, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.9957726632221701, | |
| "grad_norm": 3.5291943550109863, | |
| "learning_rate": 1.7205361461825054e-09, | |
| "loss": 0.1645, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 533, | |
| "total_flos": 78811436154880.0, | |
| "train_loss": 0.27306876374901345, | |
| "train_runtime": 11552.2832, | |
| "train_samples_per_second": 0.737, | |
| "train_steps_per_second": 0.046 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 533, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 78811436154880.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |