| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9957924263674616, | |
| "eval_steps": 500, | |
| "global_step": 534, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.296296296296296e-06, | |
| "loss": 1.1525, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 6.48148148148148e-06, | |
| "loss": 1.1069, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.296296296296296e-05, | |
| "loss": 1.0522, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.9444444444444442e-05, | |
| "loss": 0.977, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.592592592592592e-05, | |
| "loss": 0.9307, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.9144, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 3.8888888888888884e-05, | |
| "loss": 0.9218, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.537037037037037e-05, | |
| "loss": 0.8808, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.185185185185184e-05, | |
| "loss": 0.8981, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 5.833333333333333e-05, | |
| "loss": 0.8721, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 6.481481481481482e-05, | |
| "loss": 0.8714, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 6.999925035824452e-05, | |
| "loss": 0.8616, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.99730162684253e-05, | |
| "loss": 0.8686, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 6.99093321968073e-05, | |
| "loss": 0.823, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.980826633788956e-05, | |
| "loss": 0.8186, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 6.96699269155252e-05, | |
| "loss": 0.8349, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 6.949446206703272e-05, | |
| "loss": 0.8436, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 6.928205968456678e-05, | |
| "loss": 0.8353, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 6.903294721391867e-05, | |
| "loss": 0.8553, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 6.874739141096151e-05, | |
| "loss": 0.8244, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 6.842569805600153e-05, | |
| "loss": 0.8424, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 6.806821162634073e-05, | |
| "loss": 0.8328, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 6.767531492740205e-05, | |
| "loss": 0.821, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 6.724742868281173e-05, | |
| "loss": 0.7997, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 6.678501108387784e-05, | |
| "loss": 0.8393, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 6.628855729894762e-05, | |
| "loss": 0.8226, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 6.575859894316877e-05, | |
| "loss": 0.7994, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 6.519570350922271e-05, | |
| "loss": 0.7988, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.460047375963906e-05, | |
| "loss": 0.7932, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 6.397354708134257e-05, | |
| "loss": 0.786, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 6.331559480312315e-05, | |
| "loss": 0.7943, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 6.262732147676025e-05, | |
| "loss": 0.8187, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 6.190946412257108e-05, | |
| "loss": 0.7774, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 6.116279144019089e-05, | |
| "loss": 0.7925, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 6.038810298543006e-05, | |
| "loss": 0.7877, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 5.9586228314089804e-05, | |
| "loss": 0.7614, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7861344218254089, | |
| "eval_runtime": 8.8077, | |
| "eval_samples_per_second": 23.616, | |
| "eval_steps_per_second": 5.904, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 5.8758026093652964e-05, | |
| "loss": 0.7171, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 5.790438318380136e-05, | |
| "loss": 0.5443, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.702621368674431e-05, | |
| "loss": 0.5642, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5.6124457968374994e-05, | |
| "loss": 0.5541, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 5.520008165130318e-05, | |
| "loss": 0.5469, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 5.425407458084227e-05, | |
| "loss": 0.5585, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 5.3287449765058205e-05, | |
| "loss": 0.5547, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 5.230124229001506e-05, | |
| "loss": 0.5497, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 5.1296508211378896e-05, | |
| "loss": 0.5525, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 5.027432342356697e-05, | |
| "loss": 0.5793, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.9235782507653006e-05, | |
| "loss": 0.5504, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.818199755926237e-05, | |
| "loss": 0.5546, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 4.711409699771225e-05, | |
| "loss": 0.5674, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.603322435767194e-05, | |
| "loss": 0.5422, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 4.49405370646373e-05, | |
| "loss": 0.5694, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 4.3837205195530523e-05, | |
| "loss": 0.5359, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 4.2724410225752544e-05, | |
| "loss": 0.5271, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 4.160334376402943e-05, | |
| "loss": 0.5414, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 4.047520627640808e-05, | |
| "loss": 0.5243, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 3.934120580076699e-05, | |
| "loss": 0.5551, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 3.820255665321909e-05, | |
| "loss": 0.5291, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.706047812779161e-05, | |
| "loss": 0.5604, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3.5916193190775565e-05, | |
| "loss": 0.531, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.4770927171142685e-05, | |
| "loss": 0.5531, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.362590644843261e-05, | |
| "loss": 0.5574, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 3.248235713951481e-05, | |
| "loss": 0.5286, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 3.134150378563213e-05, | |
| "loss": 0.5286, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.020456804113112e-05, | |
| "loss": 0.5335, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.9072767365284138e-05, | |
| "loss": 0.5593, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 2.7947313718603398e-05, | |
| "loss": 0.537, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 2.6829412265043318e-05, | |
| "loss": 0.535, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 2.5720260081480713e-05, | |
| "loss": 0.5187, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 2.4621044875855016e-05, | |
| "loss": 0.5157, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 2.353294371534073e-05, | |
| "loss": 0.5321, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 2.245712176591449e-05, | |
| "loss": 0.5577, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 2.139473104466615e-05, | |
| "loss": 0.5033, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7809656858444214, | |
| "eval_runtime": 8.805, | |
| "eval_samples_per_second": 23.623, | |
| "eval_steps_per_second": 5.906, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 2.0346909186190027e-05, | |
| "loss": 0.433, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.931477822437721e-05, | |
| "loss": 0.3788, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.8299443390913702e-05, | |
| "loss": 0.3838, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 1.7301991931770543e-05, | |
| "loss": 0.3723, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.6323491942953597e-05, | |
| "loss": 0.3813, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.5364991226759663e-05, | |
| "loss": 0.3772, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.4427516169763444e-05, | |
| "loss": 0.3917, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.351207064373717e-05, | |
| "loss": 0.3745, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 1.2619634930679534e-05, | |
| "loss": 0.3923, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.175116467310515e-05, | |
| "loss": 0.3817, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 1.0907589850718605e-05, | |
| "loss": 0.4024, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 1.0089813784569004e-05, | |
| "loss": 0.3807, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 9.298712169751012e-06, | |
| "loss": 0.3739, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 8.535132137688704e-06, | |
| "loss": 0.3415, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 7.799891349006018e-06, | |
| "loss": 0.3652, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 7.093777117955255e-06, | |
| "loss": 0.3785, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 6.417545569341238e-06, | |
| "loss": 0.369, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 5.771920828843995e-06, | |
| "loss": 0.3693, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 5.157594247606773e-06, | |
| "loss": 0.3763, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.575223661919969e-06, | |
| "loss": 0.3589, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 4.025432688793538e-06, | |
| "loss": 0.372, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3.5088100581722544e-06, | |
| "loss": 0.3694, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 3.0259089825089655e-06, | |
| "loss": 0.3791, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 2.577246564370866e-06, | |
| "loss": 0.3838, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 2.163303242713058e-06, | |
| "loss": 0.3786, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.7845222784125152e-06, | |
| "loss": 0.4005, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.4413092796132431e-06, | |
| "loss": 0.358, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.1340317673909082e-06, | |
| "loss": 0.3938, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 8.630187822020567e-07, | |
| "loss": 0.3751, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 6.28560531539366e-07, | |
| "loss": 0.3627, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 4.309080791701819e-07, | |
| "loss": 0.3779, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 2.702730762911531e-07, | |
| "loss": 0.3841, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.4682753488684196e-07, | |
| "loss": 0.3566, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 6.070364353494617e-08, | |
| "loss": 0.3684, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 1.1993625855495903e-08, | |
| "loss": 0.3561, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8389958739280701, | |
| "eval_runtime": 8.8169, | |
| "eval_samples_per_second": 23.591, | |
| "eval_steps_per_second": 5.898, | |
| "step": 534 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 534, | |
| "total_flos": 1.0858665717058765e+17, | |
| "train_loss": 0.5910449749075072, | |
| "train_runtime": 1507.9641, | |
| "train_samples_per_second": 5.672, | |
| "train_steps_per_second": 0.354 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 534, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.0858665717058765e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |