| { |
| "best_global_step": 600, |
| "best_metric": 0.35182714462280273, |
| "best_model_checkpoint": "/workspace/anchor/qwen-urban-replaychat/train/checkpoint-600", |
| "epoch": 1.0, |
| "eval_steps": 50, |
| "global_step": 615, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.008141664970486465, |
| "grad_norm": 35.977874755859375, |
| "learning_rate": 2.5806451612903226e-05, |
| "loss": 3.7574, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.01628332994097293, |
| "grad_norm": 11.635550498962402, |
| "learning_rate": 5.8064516129032266e-05, |
| "loss": 3.072, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.024424994911459393, |
| "grad_norm": 9.978185653686523, |
| "learning_rate": 9.032258064516129e-05, |
| "loss": 2.067, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.03256665988194586, |
| "grad_norm": 3.755309581756592, |
| "learning_rate": 0.00012258064516129034, |
| "loss": 1.4303, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.04070832485243232, |
| "grad_norm": 2.346052646636963, |
| "learning_rate": 0.00015483870967741937, |
| "loss": 1.0914, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.048849989822918785, |
| "grad_norm": 3.1284399032592773, |
| "learning_rate": 0.0001870967741935484, |
| "loss": 0.8777, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.056991654793405254, |
| "grad_norm": 3.3274853229522705, |
| "learning_rate": 0.000199986978016549, |
| "loss": 0.8015, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.06513331976389172, |
| "grad_norm": 4.444795608520508, |
| "learning_rate": 0.00019990741151022301, |
| "loss": 0.6687, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.07327498473437818, |
| "grad_norm": 2.6133811473846436, |
| "learning_rate": 0.00019975557042268953, |
| "loss": 0.649, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.08141664970486465, |
| "grad_norm": 1.8493469953536987, |
| "learning_rate": 0.00019953156459816179, |
| "loss": 0.6611, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08141664970486465, |
| "eval_loss": 0.5115963816642761, |
| "eval_runtime": 201.2047, |
| "eval_samples_per_second": 1.252, |
| "eval_steps_per_second": 0.626, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08955831467535111, |
| "grad_norm": 1.652091145515442, |
| "learning_rate": 0.0001992355560859503, |
| "loss": 0.4561, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.09769997964583757, |
| "grad_norm": 1.2801644802093506, |
| "learning_rate": 0.00019886775902323405, |
| "loss": 0.5446, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.10584164461632403, |
| "grad_norm": 1.5553380250930786, |
| "learning_rate": 0.0001984284394801501, |
| "loss": 0.5921, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.11398330958681051, |
| "grad_norm": 2.0229711532592773, |
| "learning_rate": 0.00019791791526731445, |
| "loss": 0.6623, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.12212497455729697, |
| "grad_norm": 1.2393938302993774, |
| "learning_rate": 0.00019733655570591276, |
| "loss": 0.6631, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.13026663952778345, |
| "grad_norm": 1.3820499181747437, |
| "learning_rate": 0.00019668478136052774, |
| "loss": 0.6028, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.1384083044982699, |
| "grad_norm": 1.1327757835388184, |
| "learning_rate": 0.0001959630637348962, |
| "loss": 0.6096, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.14654996946875637, |
| "grad_norm": 1.4019911289215088, |
| "learning_rate": 0.00019517192493081565, |
| "loss": 0.5804, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.15469163443924283, |
| "grad_norm": 1.2397918701171875, |
| "learning_rate": 0.00019431193727044796, |
| "loss": 0.4769, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.1628332994097293, |
| "grad_norm": 1.1968052387237549, |
| "learning_rate": 0.0001933837228822925, |
| "loss": 0.4679, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.1628332994097293, |
| "eval_loss": 0.4188096225261688, |
| "eval_runtime": 116.2096, |
| "eval_samples_per_second": 2.168, |
| "eval_steps_per_second": 1.084, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17097496438021575, |
| "grad_norm": 0.9613081812858582, |
| "learning_rate": 0.0001923879532511287, |
| "loss": 0.5382, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.17911662935070222, |
| "grad_norm": 1.1866436004638672, |
| "learning_rate": 0.00019132534873225323, |
| "loss": 0.5187, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.18725829432118868, |
| "grad_norm": 1.5536378622055054, |
| "learning_rate": 0.00019019667803036417, |
| "loss": 0.4902, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.19539995929167514, |
| "grad_norm": 1.0800343751907349, |
| "learning_rate": 0.00018900275764346768, |
| "loss": 0.5605, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.2035416242621616, |
| "grad_norm": 1.1118768453598022, |
| "learning_rate": 0.0001877444512722107, |
| "loss": 0.5416, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.21168328923264806, |
| "grad_norm": 0.8304303288459778, |
| "learning_rate": 0.00018642266919506644, |
| "loss": 0.3332, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.21982495420313455, |
| "grad_norm": 6.846231460571289, |
| "learning_rate": 0.00018503836760982477, |
| "loss": 0.407, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.22796661917362102, |
| "grad_norm": 1.3753745555877686, |
| "learning_rate": 0.0001835925479418637, |
| "loss": 0.6531, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.23610828414410748, |
| "grad_norm": 1.116352915763855, |
| "learning_rate": 0.00018208625611970268, |
| "loss": 0.5036, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.24424994911459394, |
| "grad_norm": 0.9815019369125366, |
| "learning_rate": 0.00018052058181836151, |
| "loss": 0.5161, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.24424994911459394, |
| "eval_loss": 0.3987187445163727, |
| "eval_runtime": 116.2527, |
| "eval_samples_per_second": 2.168, |
| "eval_steps_per_second": 1.084, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.2523916140850804, |
| "grad_norm": 1.0172415971755981, |
| "learning_rate": 0.00017889665767107266, |
| "loss": 0.5213, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.2605332790555669, |
| "grad_norm": 0.8838318586349487, |
| "learning_rate": 0.00017721565844991643, |
| "loss": 0.4496, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.26867494402605335, |
| "grad_norm": 1.0285552740097046, |
| "learning_rate": 0.00017547880021597255, |
| "loss": 0.4549, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.2768166089965398, |
| "grad_norm": 1.4849443435668945, |
| "learning_rate": 0.00017368733943960276, |
| "loss": 0.4415, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.2849582739670263, |
| "grad_norm": 1.0443873405456543, |
| "learning_rate": 0.00017184257209150027, |
| "loss": 0.4202, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.29309993893751274, |
| "grad_norm": 1.1385072469711304, |
| "learning_rate": 0.0001699458327051647, |
| "loss": 0.5685, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.3012416039079992, |
| "grad_norm": 1.0990793704986572, |
| "learning_rate": 0.00016799849341147939, |
| "loss": 0.4092, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.30938326887848566, |
| "grad_norm": 1.1561559438705444, |
| "learning_rate": 0.00016600196294609045, |
| "loss": 0.7216, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.3175249338489721, |
| "grad_norm": 1.2006827592849731, |
| "learning_rate": 0.00016395768563030584, |
| "loss": 0.439, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.3256665988194586, |
| "grad_norm": 0.9248276948928833, |
| "learning_rate": 0.00016186714032625035, |
| "loss": 0.3376, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3256665988194586, |
| "eval_loss": 0.38663938641548157, |
| "eval_runtime": 116.1378, |
| "eval_samples_per_second": 2.17, |
| "eval_steps_per_second": 1.085, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.33380826378994505, |
| "grad_norm": 1.1307693719863892, |
| "learning_rate": 0.0001597318393670338, |
| "loss": 0.5049, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.3419499287604315, |
| "grad_norm": 1.1373472213745117, |
| "learning_rate": 0.00015755332746270572, |
| "loss": 0.3912, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.35009159373091797, |
| "grad_norm": 0.8620793223381042, |
| "learning_rate": 0.00015533318058278786, |
| "loss": 0.4972, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.35823325870140443, |
| "grad_norm": 1.1621710062026978, |
| "learning_rate": 0.00015307300481619333, |
| "loss": 0.612, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.3663749236718909, |
| "grad_norm": 0.8007318377494812, |
| "learning_rate": 0.0001507744352093569, |
| "loss": 0.4753, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.37451658864237736, |
| "grad_norm": 0.9540786743164062, |
| "learning_rate": 0.00014843913458341645, |
| "loss": 0.5342, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.3826582536128638, |
| "grad_norm": 0.9736375212669373, |
| "learning_rate": 0.00014606879233130293, |
| "loss": 0.5131, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.3907999185833503, |
| "grad_norm": 0.7898627519607544, |
| "learning_rate": 0.0001436651231956064, |
| "loss": 0.4659, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.39894158355383674, |
| "grad_norm": 1.122058629989624, |
| "learning_rate": 0.00014122986602810487, |
| "loss": 0.4947, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.4070832485243232, |
| "grad_norm": 1.218949317932129, |
| "learning_rate": 0.00013876478253185183, |
| "loss": 0.489, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.4070832485243232, |
| "eval_loss": 0.37968578934669495, |
| "eval_runtime": 116.531, |
| "eval_samples_per_second": 2.163, |
| "eval_steps_per_second": 1.081, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.41522491349480967, |
| "grad_norm": 0.6106458306312561, |
| "learning_rate": 0.00013627165598673247, |
| "loss": 0.3834, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.42336657846529613, |
| "grad_norm": 1.096251368522644, |
| "learning_rate": 0.00013375228995941133, |
| "loss": 0.3748, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.4315082434357826, |
| "grad_norm": 0.9184795022010803, |
| "learning_rate": 0.00013120850699860416, |
| "loss": 0.4812, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.4396499084062691, |
| "grad_norm": 1.1240248680114746, |
| "learning_rate": 0.00012864214731661742, |
| "loss": 0.4922, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.44779157337675557, |
| "grad_norm": 0.9654471278190613, |
| "learning_rate": 0.0001260550674581103, |
| "loss": 0.4458, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.45593323834724203, |
| "grad_norm": 1.2910221815109253, |
| "learning_rate": 0.00012344913895704097, |
| "loss": 0.516, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.4640749033177285, |
| "grad_norm": 1.061428189277649, |
| "learning_rate": 0.00012082624698277005, |
| "loss": 0.3993, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.47221656828821496, |
| "grad_norm": 0.8641824126243591, |
| "learning_rate": 0.0001181882889762994, |
| "loss": 0.4291, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.4803582332587014, |
| "grad_norm": 1.2518894672393799, |
| "learning_rate": 0.00011553717327763385, |
| "loss": 0.5343, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.4884998982291879, |
| "grad_norm": 1.23012375831604, |
| "learning_rate": 0.0001128748177452581, |
| "loss": 0.4389, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.4884998982291879, |
| "eval_loss": 0.37130653858184814, |
| "eval_runtime": 117.0691, |
| "eval_samples_per_second": 2.153, |
| "eval_steps_per_second": 1.076, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.49664156319967434, |
| "grad_norm": 1.0429818630218506, |
| "learning_rate": 0.00011020314836872811, |
| "loss": 0.5072, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.5047832281701607, |
| "grad_norm": 1.0350611209869385, |
| "learning_rate": 0.00010752409787538, |
| "loss": 0.4678, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.5129248931406473, |
| "grad_norm": 0.7658055424690247, |
| "learning_rate": 0.00010483960433216522, |
| "loss": 0.4008, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.5210665581111338, |
| "grad_norm": 0.9374896287918091, |
| "learning_rate": 0.00010215160974362223, |
| "loss": 0.4806, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.5292082230816202, |
| "grad_norm": 0.6456654071807861, |
| "learning_rate": 9.946205864700052e-05, |
| "loss": 0.3545, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.5373498880521067, |
| "grad_norm": 1.1381654739379883, |
| "learning_rate": 9.677289670555169e-05, |
| "loss": 0.5195, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.5454915530225931, |
| "grad_norm": 1.141062617301941, |
| "learning_rate": 9.408606930100654e-05, |
| "loss": 0.5135, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.5536332179930796, |
| "grad_norm": 0.8046601414680481, |
| "learning_rate": 9.140352012625537e-05, |
| "loss": 0.5015, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.561774882963566, |
| "grad_norm": 0.9050662517547607, |
| "learning_rate": 8.872718977925022e-05, |
| "loss": 0.54, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.5699165479340526, |
| "grad_norm": 1.1043193340301514, |
| "learning_rate": 8.605901435914607e-05, |
| "loss": 0.4573, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5699165479340526, |
| "eval_loss": 0.36485499143600464, |
| "eval_runtime": 118.595, |
| "eval_samples_per_second": 2.125, |
| "eval_steps_per_second": 1.062, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.578058212904539, |
| "grad_norm": 0.858796238899231, |
| "learning_rate": 8.340092406569649e-05, |
| "loss": 0.4916, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.5861998778750255, |
| "grad_norm": 1.0917872190475464, |
| "learning_rate": 8.075484180291701e-05, |
| "loss": 0.4989, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.5943415428455119, |
| "grad_norm": 0.9984243512153625, |
| "learning_rate": 7.812268178802626e-05, |
| "loss": 0.4327, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.6024832078159984, |
| "grad_norm": 1.0273386240005493, |
| "learning_rate": 7.550634816667142e-05, |
| "loss": 0.4777, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.6106248727864848, |
| "grad_norm": 1.1594866514205933, |
| "learning_rate": 7.290773363543945e-05, |
| "loss": 0.5688, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.6187665377569713, |
| "grad_norm": 0.5220950841903687, |
| "learning_rate": 7.032871807265096e-05, |
| "loss": 0.404, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.6269082027274577, |
| "grad_norm": 1.069298267364502, |
| "learning_rate": 6.777116717842686e-05, |
| "loss": 0.437, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.6350498676979442, |
| "grad_norm": 0.5795661211013794, |
| "learning_rate": 6.52369311250116e-05, |
| "loss": 0.522, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.6431915326684307, |
| "grad_norm": 0.7598361372947693, |
| "learning_rate": 6.27278432183299e-05, |
| "loss": 0.4256, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.6513331976389172, |
| "grad_norm": 0.9439622759819031, |
| "learning_rate": 6.024571857174443e-05, |
| "loss": 0.4, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6513331976389172, |
| "eval_loss": 0.36053991317749023, |
| "eval_runtime": 117.7368, |
| "eval_samples_per_second": 2.14, |
| "eval_steps_per_second": 1.07, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6594748626094036, |
| "grad_norm": 1.073849081993103, |
| "learning_rate": 5.779235279297467e-05, |
| "loss": 0.5748, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.6676165275798901, |
| "grad_norm": 0.7306942343711853, |
| "learning_rate": 5.536952068512608e-05, |
| "loss": 0.2809, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.6757581925503765, |
| "grad_norm": 0.7313353419303894, |
| "learning_rate": 5.2978974962769975e-05, |
| "loss": 0.4336, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.683899857520863, |
| "grad_norm": 0.9219333529472351, |
| "learning_rate": 5.062244498400228e-05, |
| "loss": 0.4336, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.6920415224913494, |
| "grad_norm": 1.1247750520706177, |
| "learning_rate": 4.830163549939899e-05, |
| "loss": 0.4364, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.7001831874618359, |
| "grad_norm": 0.9264132380485535, |
| "learning_rate": 4.601822541877291e-05, |
| "loss": 0.5146, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.7083248524323225, |
| "grad_norm": 1.1259379386901855, |
| "learning_rate": 4.377386659662425e-05, |
| "loss": 0.4121, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.7164665174028089, |
| "grad_norm": 0.7834416627883911, |
| "learning_rate": 4.1570182637163155e-05, |
| "loss": 0.3537, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.7246081823732954, |
| "grad_norm": 1.1130311489105225, |
| "learning_rate": 3.940876771976904e-05, |
| "loss": 0.4026, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.7327498473437818, |
| "grad_norm": 0.9526257514953613, |
| "learning_rate": 3.7291185445736444e-05, |
| "loss": 0.3395, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.7327498473437818, |
| "eval_loss": 0.35684067010879517, |
| "eval_runtime": 115.9943, |
| "eval_samples_per_second": 2.173, |
| "eval_steps_per_second": 1.086, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.7408915123142683, |
| "grad_norm": 0.6844274997711182, |
| "learning_rate": 3.5218967707141335e-05, |
| "loss": 0.2771, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.7490331772847547, |
| "grad_norm": 1.021323800086975, |
| "learning_rate": 3.319361357864663e-05, |
| "loss": 0.4614, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.7571748422552412, |
| "grad_norm": 0.5935538411140442, |
| "learning_rate": 3.121658823304806e-05, |
| "loss": 0.2239, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.7653165072257276, |
| "grad_norm": 0.967462420463562, |
| "learning_rate": 2.9289321881345254e-05, |
| "loss": 0.5165, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.7734581721962142, |
| "grad_norm": 0.55368971824646, |
| "learning_rate": 2.7413208738104868e-05, |
| "loss": 0.387, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.7815998371667006, |
| "grad_norm": 1.1664959192276, |
| "learning_rate": 2.5589606012863963e-05, |
| "loss": 0.5815, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.7897415021371871, |
| "grad_norm": 1.2442911863327026, |
| "learning_rate": 2.381983292830343e-05, |
| "loss": 0.3994, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.7978831671076735, |
| "grad_norm": 0.9416378736495972, |
| "learning_rate": 2.210516976590179e-05, |
| "loss": 0.51, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.80602483207816, |
| "grad_norm": 1.1915277242660522, |
| "learning_rate": 2.0446856939759472e-05, |
| "loss": 0.5335, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.8141664970486464, |
| "grad_norm": 0.4900799095630646, |
| "learning_rate": 1.8846094099263912e-05, |
| "loss": 0.3594, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8141664970486464, |
| "eval_loss": 0.353680819272995, |
| "eval_runtime": 116.1356, |
| "eval_samples_per_second": 2.17, |
| "eval_steps_per_second": 1.085, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8223081620191329, |
| "grad_norm": 0.9394781589508057, |
| "learning_rate": 1.73040392612445e-05, |
| "loss": 0.4975, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.8304498269896193, |
| "grad_norm": 0.5942463874816895, |
| "learning_rate": 1.582180797224507e-05, |
| "loss": 0.4867, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.8385914919601059, |
| "grad_norm": 1.0589011907577515, |
| "learning_rate": 1.4400472501520223e-05, |
| "loss": 0.4071, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.8467331569305923, |
| "grad_norm": 0.7804487943649292, |
| "learning_rate": 1.30410610653389e-05, |
| "loss": 0.4039, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.8548748219010788, |
| "grad_norm": 1.17083740234375, |
| "learning_rate": 1.1744557083156704e-05, |
| "loss": 0.5449, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.8630164868715652, |
| "grad_norm": 0.941277027130127, |
| "learning_rate": 1.0511898466194903e-05, |
| "loss": 0.484, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.8711581518420517, |
| "grad_norm": 1.004425048828125, |
| "learning_rate": 9.343976938940869e-06, |
| "loss": 0.5177, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.8792998168125382, |
| "grad_norm": 1.049841284751892, |
| "learning_rate": 8.24163739406062e-06, |
| "loss": 0.4859, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.8874414817830246, |
| "grad_norm": 1.0460737943649292, |
| "learning_rate": 7.205677281190304e-06, |
| "loss": 0.452, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.8955831467535111, |
| "grad_norm": 0.8967907428741455, |
| "learning_rate": 6.236846030048604e-06, |
| "loss": 0.4042, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.8955831467535111, |
| "eval_loss": 0.3522235155105591, |
| "eval_runtime": 116.6127, |
| "eval_samples_per_second": 2.161, |
| "eval_steps_per_second": 1.081, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.9037248117239975, |
| "grad_norm": 0.8039125800132751, |
| "learning_rate": 5.335844508287679e-06, |
| "loss": 0.3912, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.9118664766944841, |
| "grad_norm": 0.6862779855728149, |
| "learning_rate": 4.503324514474483e-06, |
| "loss": 0.493, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.9200081416649705, |
| "grad_norm": 0.8961853384971619, |
| "learning_rate": 3.739888306569572e-06, |
| "loss": 0.351, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.928149806635457, |
| "grad_norm": 0.49110013246536255, |
| "learning_rate": 3.0460881662442763e-06, |
| "loss": 0.2817, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.9362914716059434, |
| "grad_norm": 0.5299960374832153, |
| "learning_rate": 2.422425999351463e-06, |
| "loss": 0.4803, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.9444331365764299, |
| "grad_norm": 1.0633695125579834, |
| "learning_rate": 1.869352972839067e-06, |
| "loss": 0.3545, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.9525748015469163, |
| "grad_norm": 1.0999653339385986, |
| "learning_rate": 1.387269188368867e-06, |
| "loss": 0.6437, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.9607164665174028, |
| "grad_norm": 0.5522617697715759, |
| "learning_rate": 9.765233928766493e-07, |
| "loss": 0.4132, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.9688581314878892, |
| "grad_norm": 0.6470587253570557, |
| "learning_rate": 6.374127262833484e-07, |
| "loss": 0.3911, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.9769997964583758, |
| "grad_norm": 1.0086009502410889, |
| "learning_rate": 3.701825065392184e-07, |
| "loss": 0.4381, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9769997964583758, |
| "eval_loss": 0.35182714462280273, |
| "eval_runtime": 116.1792, |
| "eval_samples_per_second": 2.169, |
| "eval_steps_per_second": 1.085, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9851414614288622, |
| "grad_norm": 0.8223856091499329, |
| "learning_rate": 1.7502605215715672e-07, |
| "loss": 0.4936, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.9932831263993487, |
| "grad_norm": 0.5851192474365234, |
| "learning_rate": 5.208454236296234e-08, |
| "loss": 0.3419, |
| "step": 610 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 2.273397445678711, |
| "learning_rate": 1.4469149641538337e-09, |
| "loss": 0.5031, |
| "step": 615 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 615, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 2, |
| "early_stopping_threshold": 0.001 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 1 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.537514438683775e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
