CoolFace
Modelpublic

SASVAAI/GLM-4.7-Flash-sql-create-context

sourceHugging Facemitupdated 22d agoView on Hugging Face
0likes5downloads
trainer_state.json725 linesDownload Raw Back to root
1{2  "best_global_step": 675,3  "best_metric": 0.5975516438484192,4  "best_model_checkpoint": "/home/svadouser1/pooja/llm_autotuning/checkpoints/exp_20260827_221617/checkpoint-675",5  "epoch": 3.0,6  "eval_steps": 9999,7  "global_step": 675,8  "is_hyper_param_search": false,9  "is_local_process_zero": true,10  "is_world_process_zero": true,11  "log_history": [12    {13      "entropy": 1.461071628332138,14      "epoch": 0.044444444444444446,15      "grad_norm": 1.6171875,16      "learning_rate": 5.294117647058824e-05,17      "loss": 2.388203811645508,18      "mean_token_accuracy": 0.6123090386390686,19      "num_tokens": 19486.0,20      "step": 1021    },22    {23      "entropy": 0.9562640145421029,24      "epoch": 0.08888888888888889,25      "grad_norm": 0.6875,26      "learning_rate": 0.00011176470588235294,27      "loss": 1.1077078819274901,28      "mean_token_accuracy": 0.8430786401033401,29      "num_tokens": 38826.0,30      "step": 2031    },32    {33      "entropy": 0.77465368360281,34      "epoch": 0.13333333333333333,35      "grad_norm": 0.455078125,36      "learning_rate": 0.00017058823529411766,37      "loss": 0.8623821258544921,38      "mean_token_accuracy": 0.864073084294796,39      "num_tokens": 58244.0,40      "step": 3041    },42    {43      "entropy": 0.7483784720301628,44      "epoch": 0.17777777777777778,45      "grad_norm": 0.423828125,46      "learning_rate": 0.00019996997576394573,47      "loss": 0.7713714599609375,48      "mean_token_accuracy": 0.8697837173938752,49      "num_tokens": 77711.0,50      "step": 4051    },52    {53      "entropy": 0.7391687169671058,54      "epoch": 0.2222222222222222,55      "grad_norm": 0.376953125,56      "learning_rate": 0.00019972989003925543,57      "loss": 0.7115508556365967,58      "mean_token_accuracy": 0.8751530349254608,59      "num_tokens": 97104.0,60      "step": 5061    },62    {63      "entropy": 0.7042164742946625,64      "epoch": 0.26666666666666666,65      "grad_norm": 0.330078125,66      "learning_rate": 0.00019925029517454195,67      "loss": 0.6672511577606202,68      "mean_token_accuracy": 0.8725973218679428,69      "num_tokens": 116498.0,70      "step": 6071    },72    {73      "entropy": 0.6745826601982117,74      "epoch": 0.3111111111111111,75      "grad_norm": 0.361328125,76      "learning_rate": 0.00019853234295442638,77      "loss": 0.6365277290344238,78      "mean_token_accuracy": 0.872233435511589,79      "num_tokens": 136089.0,80      "step": 7081    },82    {83      "entropy": 0.6878040120005607,84      "epoch": 0.35555555555555557,85      "grad_norm": 0.37109375,86      "learning_rate": 0.00019757775759738272,87      "loss": 0.6395976066589355,88      "mean_token_accuracy": 0.8741081401705741,89      "num_tokens": 155486.0,90      "step": 8091    },92    {93      "entropy": 0.6898319900035859,94      "epoch": 0.4,95      "grad_norm": 0.330078125,96      "learning_rate": 0.00019638883161489224,97      "loss": 0.6368544578552247,98      "mean_token_accuracy": 0.8765213042497635,99      "num_tokens": 175116.0,100      "step": 90101    },102    {103      "entropy": 0.656438185274601,104      "epoch": 0.4444444444444444,105      "grad_norm": 0.33203125,106      "learning_rate": 0.0001949684203057978,107      "loss": 0.6183670043945313,108      "mean_token_accuracy": 0.8779006212949753,109      "num_tokens": 194495.0,110      "step": 100111    },112    {113      "entropy": 0.6711793914437294,114      "epoch": 0.4888888888888889,115      "grad_norm": 0.388671875,116      "learning_rate": 0.00019331993489907977,117      "loss": 0.633416748046875,118      "mean_token_accuracy": 0.8727847129106522,119      "num_tokens": 214311.0,120      "step": 110121    },122    {123      "entropy": 0.6290019288659096,124      "epoch": 0.5333333333333333,125      "grad_norm": 0.33203125,126      "learning_rate": 0.00019144733436152284,127      "loss": 0.5889303684234619,128      "mean_token_accuracy": 0.8814716726541519,129      "num_tokens": 233676.0,130      "step": 120131    },132    {133      "entropy": 0.6458568304777146,134      "epoch": 0.5777777777777777,135      "grad_norm": 0.318359375,136      "learning_rate": 0.00018935511588994715,137      "loss": 0.6043062686920166,138      "mean_token_accuracy": 0.8794952735304833,139      "num_tokens": 253188.0,140      "step": 130141    },142    {143      "entropy": 0.6591948792338371,144      "epoch": 0.6222222222222222,145      "grad_norm": 0.32421875,146      "learning_rate": 0.000187048304110838,147      "loss": 0.6233312129974365,148      "mean_token_accuracy": 0.873698017001152,149      "num_tokens": 273185.0,150      "step": 140151    },152    {153      "entropy": 0.6109217047691345,154      "epoch": 0.6666666666666666,155      "grad_norm": 0.291015625,156      "learning_rate": 0.00018453243901331195,157      "loss": 0.5925732135772706,158      "mean_token_accuracy": 0.884858050942421,159      "num_tokens": 292431.0,160      "step": 150161    },162    {163      "entropy": 0.6377103522419929,164      "epoch": 0.7111111111111111,165      "grad_norm": 0.3671875,166      "learning_rate": 0.00018181356264439905,167      "loss": 0.5884737968444824,168      "mean_token_accuracy": 0.8802034169435501,169      "num_tokens": 311687.0,170      "step": 160171    },172    {173      "entropy": 0.6447647094726563,174      "epoch": 0.7555555555555555,175      "grad_norm": 0.3046875,176      "learning_rate": 0.0001788982045985939,177      "loss": 0.5939778327941895,178      "mean_token_accuracy": 0.8817022785544395,179      "num_tokens": 330813.0,180      "step": 170181    },182    {183      "entropy": 0.6029744669795036,184      "epoch": 0.8,185      "grad_norm": 0.3125,186      "learning_rate": 0.00017579336633652317,187      "loss": 0.5752908706665039,188      "mean_token_accuracy": 0.8855120778083801,189      "num_tokens": 350308.0,190      "step": 180191    },192    {193      "entropy": 0.6441277623176574,194      "epoch": 0.8444444444444444,195      "grad_norm": 0.259765625,196      "learning_rate": 0.00017250650437038964,197      "loss": 0.5958236217498779,198      "mean_token_accuracy": 0.8808144956827164,199      "num_tokens": 370083.0,200      "step": 190201    },202    {203      "entropy": 0.5766214028000831,204      "epoch": 0.8888888888888888,205      "grad_norm": 0.2734375,206      "learning_rate": 0.0001690455123565743,207      "loss": 0.5667627811431885,208      "mean_token_accuracy": 0.8866458177566529,209      "num_tokens": 389516.0,210      "step": 200211    },212    {213      "entropy": 0.6425503984093666,214      "epoch": 0.9333333333333333,215      "grad_norm": 0.3046875,216      "learning_rate": 0.00016541870213840243,217      "loss": 0.6137699127197266,218      "mean_token_accuracy": 0.8819929376244545,219      "num_tokens": 408940.0,220      "step": 210221    },222    {223      "entropy": 0.6159977808594703,224      "epoch": 0.9777777777777777,225      "grad_norm": 0.265625,226      "learning_rate": 0.0001616347837846011,227      "loss": 0.5788507461547852,228      "mean_token_accuracy": 0.8852896198630333,229      "num_tokens": 428166.0,230      "step": 220231    },232    {233      "entropy": 0.6020576372742653,234      "epoch": 1.0222222222222221,235      "grad_norm": 0.2353515625,236      "learning_rate": 0.000157702844671387,237      "loss": 0.5500371932983399,238      "mean_token_accuracy": 0.8901642486453056,239      "num_tokens": 447438.0,240      "step": 230241    },242    {243      "entropy": 0.5630205765366554,244      "epoch": 1.0666666666666667,245      "grad_norm": 0.275390625,246      "learning_rate": 0.00015363232765842089,247      "loss": 0.5382141590118408,248      "mean_token_accuracy": 0.8885036528110504,249      "num_tokens": 467009.0,250      "step": 240251    },252    {253      "entropy": 0.5631943918764591,254      "epoch": 1.1111111111111112,255      "grad_norm": 0.306640625,256      "learning_rate": 0.00014943300841104094,257      "loss": 0.5194426536560058,258      "mean_token_accuracy": 0.8929360061883926,259      "num_tokens": 486591.0,260      "step": 250261    },262    {263      "entropy": 0.5655099518597126,264      "epoch": 1.1555555555555554,265      "grad_norm": 0.259765625,266      "learning_rate": 0.0001451149719232366,267      "loss": 0.5290531158447266,268      "mean_token_accuracy": 0.8912677451968193,269      "num_tokens": 506180.0,270      "step": 260271    },272    {273      "entropy": 0.5660845704376698,274      "epoch": 1.2,275      "grad_norm": 0.326171875,276      "learning_rate": 0.00014068858829774608,277      "loss": 0.5366004943847656,278      "mean_token_accuracy": 0.8909024119377136,279      "num_tokens": 525696.0,280      "step": 270281    },282    {283      "entropy": 0.5818345256149768,284      "epoch": 1.2444444444444445,285      "grad_norm": 0.26171875,286      "learning_rate": 0.0001361644878414428,287      "loss": 0.5424720764160156,288      "mean_token_accuracy": 0.8896975666284561,289      "num_tokens": 545392.0,290      "step": 280291    },292    {293      "entropy": 0.5546154774725437,294      "epoch": 1.2888888888888888,295      "grad_norm": 0.3359375,296      "learning_rate": 0.00013155353553582057,297      "loss": 0.5176132202148438,298      "mean_token_accuracy": 0.8928953528404235,299      "num_tokens": 565036.0,300      "step": 290301    },302    {303      "entropy": 0.5676225990056991,304      "epoch": 1.3333333333333333,305      "grad_norm": 0.275390625,306      "learning_rate": 0.0001268668049438902,307      "loss": 0.5130613803863525,308      "mean_token_accuracy": 0.8928169339895249,309      "num_tokens": 584507.0,310      "step": 300311    },312    {313      "entropy": 0.5529272347688675,314      "epoch": 1.3777777777777778,315      "grad_norm": 0.326171875,316      "learning_rate": 0.0001221155516161506,317      "loss": 0.5317890644073486,318      "mean_token_accuracy": 0.8913554430007935,319      "num_tokens": 604008.0,320      "step": 310321    },322    {323      "entropy": 0.5658974438905716,324      "epoch": 1.4222222222222223,325      "grad_norm": 0.26953125,326      "learning_rate": 0.0001173111860595032,327      "loss": 0.5273444652557373,328      "mean_token_accuracy": 0.8927861481904984,329      "num_tokens": 623480.0,330      "step": 320331    },332    {333      "entropy": 0.5697685681283474,334      "epoch": 1.4666666666666668,335      "grad_norm": 0.322265625,336      "learning_rate": 0.00011246524633402573,337      "loss": 0.5287697792053223,338      "mean_token_accuracy": 0.8912984907627106,339      "num_tokens": 642979.0,340      "step": 330341    },342    {343      "entropy": 0.5488456651568413,344      "epoch": 1.511111111111111,345      "grad_norm": 0.326171875,346      "learning_rate": 0.00010758937034341787,347      "loss": 0.513739824295044,348      "mean_token_accuracy": 0.8952319726347924,349      "num_tokens": 662396.0,350      "step": 340351    },352    {353      "entropy": 0.55315260887146,354      "epoch": 1.5555555555555556,355      "grad_norm": 0.26953125,356      "learning_rate": 0.00010269526788566408,357      "loss": 0.5225533485412598,358      "mean_token_accuracy": 0.8937178790569306,359      "num_tokens": 681644.0,360      "step": 350361    },362    {363      "entropy": 0.5738558314740658,364      "epoch": 1.6,365      "grad_norm": 0.29296875,366      "learning_rate": 9.779469253103684e-05,367      "loss": 0.5239317417144775,368      "mean_token_accuracy": 0.8908430561423302,369      "num_tokens": 701030.0,370      "step": 360371    },372    {373      "entropy": 0.5420261971652508,374      "epoch": 1.6444444444444444,375      "grad_norm": 0.259765625,376      "learning_rate": 9.289941339497719e-05,377      "loss": 0.5274893283843994,378      "mean_token_accuracy": 0.8941561266779899,379      "num_tokens": 720294.0,380      "step": 370381    },382    {383      "entropy": 0.577882957458496,384      "epoch": 1.6888888888888889,385      "grad_norm": 0.33203125,386      "learning_rate": 8.802118687364284e-05,387      "loss": 0.5202207088470459,388      "mean_token_accuracy": 0.8931182771921158,389      "num_tokens": 739607.0,390      "step": 380391    },392    {393      "entropy": 0.5461296208202839,394      "epoch": 1.7333333333333334,395      "grad_norm": 0.30859375,396      "learning_rate": 8.317172841000173e-05,397      "loss": 0.5105932235717774,398      "mean_token_accuracy": 0.8917569547891617,399      "num_tokens": 759370.0,400      "step": 390401    },402    {403      "entropy": 0.5631573006510735,404      "epoch": 1.7777777777777777,405      "grad_norm": 0.298828125,406      "learning_rate": 7.836268435827875e-05,407      "loss": 0.5263056755065918,408      "mean_token_accuracy": 0.8906426534056664,409      "num_tokens": 778969.0,410      "step": 400411    },412    {413      "entropy": 0.5476421102881431,414      "epoch": 1.8222222222222222,415      "grad_norm": 0.255859375,416      "learning_rate": 7.360560401432401e-05,417      "loss": 0.5025547504425049,418      "mean_token_accuracy": 0.8946620702743531,419      "num_tokens": 798105.0,420      "step": 410421    },422    {423      "entropy": 0.5343898519873619,424      "epoch": 1.8666666666666667,425      "grad_norm": 0.27734375,426      "learning_rate": 6.891191187907455e-05,427      "loss": 0.49821176528930666,428      "mean_token_accuracy": 0.8958980679512024,429      "num_tokens": 817238.0,430      "step": 420431    },432    {433      "entropy": 0.5404686734080315,434      "epoch": 1.911111111111111,435      "grad_norm": 0.296875,436      "learning_rate": 6.429288022172068e-05,437      "loss": 0.5022043704986572,438      "mean_token_accuracy": 0.8958747044205666,439      "num_tokens": 836646.0,440      "step": 430441    },442    {443      "entropy": 0.5426375091075897,444      "epoch": 1.9555555555555557,445      "grad_norm": 0.279296875,446      "learning_rate": 5.9759602008468996e-05,447      "loss": 0.5035938262939453,448      "mean_token_accuracy": 0.8944751426577568,449      "num_tokens": 856282.0,450      "step": 440451    },452    {453      "entropy": 0.5425564736127854,454      "epoch": 2.0,455      "grad_norm": 0.263671875,456      "learning_rate": 5.532296426191539e-05,457      "loss": 0.5004886627197266,458      "mean_token_accuracy": 0.8947419315576554,459      "num_tokens": 875656.0,460      "step": 450461    },462    {463      "entropy": 0.526330380141735,464      "epoch": 2.0444444444444443,465      "grad_norm": 0.279296875,466      "learning_rate": 5.0993621915007785e-05,467      "loss": 0.44887552261352537,468      "mean_token_accuracy": 0.9024429768323898,469      "num_tokens": 894973.0,470      "step": 460471    },472    {473      "entropy": 0.4774711772799492,474      "epoch": 2.088888888888889,475      "grad_norm": 0.365234375,476      "learning_rate": 4.678197222239035e-05,477      "loss": 0.44640169143676756,478      "mean_token_accuracy": 0.9045588001608849,479      "num_tokens": 914579.0,480      "step": 470481    },482    {483      "entropy": 0.4789727419614792,484      "epoch": 2.1333333333333333,485      "grad_norm": 0.3828125,486      "learning_rate": 4.269812979058235e-05,487      "loss": 0.44276885986328124,488      "mean_token_accuracy": 0.9066318318247795,489      "num_tokens": 933807.0,490      "step": 480491    },492    {493      "entropy": 0.47537712305784224,494      "epoch": 2.1777777777777776,495      "grad_norm": 0.359375,496      "learning_rate": 3.875190228695862e-05,497      "loss": 0.4367781639099121,498      "mean_token_accuracy": 0.9070042923092843,499      "num_tokens": 953241.0,500      "step": 490501    },502    {503      "entropy": 0.4877164676785469,504      "epoch": 2.2222222222222223,505      "grad_norm": 0.328125,506      "learning_rate": 3.4952766885868346e-05,507      "loss": 0.4422135353088379,508      "mean_token_accuracy": 0.905805604159832,509      "num_tokens": 972758.0,510      "step": 500511    },512    {513      "entropy": 0.4915403999388218,514      "epoch": 2.2666666666666666,515      "grad_norm": 0.376953125,516      "learning_rate": 3.130984750845885e-05,517      "loss": 0.4357293128967285,518      "mean_token_accuracy": 0.9067289993166924,519      "num_tokens": 992534.0,520      "step": 510521    },522    {523      "entropy": 0.4957615494728088,524      "epoch": 2.311111111111111,525      "grad_norm": 0.33984375,526      "learning_rate": 2.7831892910864434e-05,527      "loss": 0.44167218208312986,528      "mean_token_accuracy": 0.9066730305552483,529      "num_tokens": 1012003.0,530      "step": 520531    },532    {533      "entropy": 0.48269262462854384,534      "epoch": 2.3555555555555556,535      "grad_norm": 0.375,536      "learning_rate": 2.4527255673383565e-05,537      "loss": 0.437894344329834,538      "mean_token_accuracy": 0.904815036058426,539      "num_tokens": 1031505.0,540      "step": 530541    },542    {543      "entropy": 0.47741171419620515,544      "epoch": 2.4,545      "grad_norm": 0.4140625,546      "learning_rate": 2.140387214110322e-05,547      "loss": 0.4400351047515869,548      "mean_token_accuracy": 0.9068154886364936,549      "num_tokens": 1050914.0,550      "step": 540551    },552    {553      "entropy": 0.4727069653570652,554      "epoch": 2.4444444444444446,555      "grad_norm": 0.35546875,556      "learning_rate": 1.846924336414474e-05,557      "loss": 0.43111190795898435,558      "mean_token_accuracy": 0.9075597256422043,559      "num_tokens": 1070218.0,560      "step": 550561    },562    {563      "entropy": 0.4883471392095089,564      "epoch": 2.488888888888889,565      "grad_norm": 0.314453125,566      "learning_rate": 1.5730417083304573e-05,567      "loss": 0.4493571758270264,568      "mean_token_accuracy": 0.9073263511061669,569      "num_tokens": 1089667.0,570      "step": 560571    },572    {573      "entropy": 0.4878625735640526,574      "epoch": 2.533333333333333,575      "grad_norm": 0.375,576      "learning_rate": 1.3193970804352952e-05,577      "loss": 0.44793548583984377,578      "mean_token_accuracy": 0.9050837978720665,579      "num_tokens": 1109201.0,580      "step": 570581    },582    {583      "entropy": 0.48537297546863556,584      "epoch": 2.5777777777777775,585      "grad_norm": 0.337890625,586      "learning_rate": 1.08659960016387e-05,587      "loss": 0.4368610382080078,588      "mean_token_accuracy": 0.9097044110298157,589      "num_tokens": 1128649.0,590      "step": 580591    },592    {593      "entropy": 0.4859867602586746,594      "epoch": 2.6222222222222222,595      "grad_norm": 0.328125,596      "learning_rate": 8.75208348893667e-06,597      "loss": 0.4249687194824219,598      "mean_token_accuracy": 0.9087634190917016,599      "num_tokens": 1148353.0,600      "step": 590601    },602    {603      "entropy": 0.4833585321903229,604      "epoch": 2.6666666666666665,605      "grad_norm": 0.3828125,606      "learning_rate": 6.857309992670624e-06,607      "loss": 0.4455591678619385,608      "mean_token_accuracy": 0.9075253725051879,609      "num_tokens": 1167638.0,610      "step": 600611    },612    {613      "entropy": 0.4794360339641571,614      "epoch": 2.7111111111111112,615      "grad_norm": 0.3203125,616      "learning_rate": 5.186225959757207e-06,617      "loss": 0.43627562522888186,618      "mean_token_accuracy": 0.907138803601265,619      "num_tokens": 1186894.0,620      "step": 610621    },622    {623      "entropy": 0.4800324112176895,624      "epoch": 2.7555555555555555,625      "grad_norm": 0.365234375,626      "learning_rate": 3.7428446293514386e-06,627      "loss": 0.42800016403198243,628      "mean_token_accuracy": 0.9089159920811654,629      "num_tokens": 1206224.0,630      "step": 620631    },632    {633      "entropy": 0.48999542444944383,634      "epoch": 2.8,635      "grad_norm": 0.31640625,636      "learning_rate": 2.5306323947385746e-06,637      "loss": 0.44663453102111816,638      "mean_token_accuracy": 0.9060632780194282,639      "num_tokens": 1225569.0,640      "step": 630641    },642    {643      "entropy": 0.4874999448657036,644      "epoch": 2.8444444444444446,645      "grad_norm": 0.326171875,646      "learning_rate": 1.5525004785192143e-06,647      "loss": 0.44841961860656737,648      "mean_token_accuracy": 0.9045136958360672,649      "num_tokens": 1244959.0,650      "step": 640651    },652    {653      "entropy": 0.49936808943748473,654      "epoch": 2.888888888888889,655      "grad_norm": 0.4296875,656      "learning_rate": 8.107979410802769e-07,657      "loss": 0.4453989028930664,658      "mean_token_accuracy": 0.9050952970981598,659      "num_tokens": 1264904.0,660      "step": 650661    },662    {663      "entropy": 0.49325661584734914,664      "epoch": 2.9333333333333336,665      "grad_norm": 0.373046875,666      "learning_rate": 3.0730603914255193e-07,667      "loss": 0.44756379127502444,668      "mean_token_accuracy": 0.9043820068240166,669      "num_tokens": 1284311.0,670      "step": 660671    },672    {673      "entropy": 0.5068465434014797,674      "epoch": 2.977777777777778,675      "grad_norm": 0.373046875,676      "learning_rate": 4.323394793315228e-08,677      "loss": 0.4593667030334473,678      "mean_token_accuracy": 0.9011617928743363,679      "num_tokens": 1303914.0,680      "step": 670681    },682    {683      "epoch": 3.0,684      "eval_entropy": 0.5207519015669823,685      "eval_loss": 0.5975516438484192,686      "eval_mean_token_accuracy": 0.8822531878948212,687      "eval_num_tokens": 1313484.0,688      "eval_runtime": 57.0793,689      "eval_samples_per_second": 7.008,690      "eval_steps_per_second": 1.752,691      "step": 675692    },693    {694      "epoch": 3.0,695      "step": 675,696      "total_flos": 2.3441226939026637e+17,697      "train_loss": 0.5646523337894016,698      "train_runtime": 6392.6844,699      "train_samples_per_second": 1.689,700      "train_steps_per_second": 0.106701    }702  ],703  "logging_steps": 10,704  "max_steps": 675,705  "num_input_tokens_seen": 0,706  "num_train_epochs": 3,707  "save_steps": 9999,708  "stateful_callbacks": {709    "TrainerControl": {710      "args": {711        "should_epoch_stop": false,712        "should_evaluate": false,713        "should_log": false,714        "should_save": true,715        "should_training_stop": true716      },717      "attributes": {}718    }719  },720  "total_flos": 2.3441226939026637e+17,721  "train_batch_size": 1,722  "trial_name": null,723  "trial_params": null724}725