CoolFace
Modelpublic

m2g-lab/gemma3-4b-mathwizards-instruct-basic

sourceHugging Faceupdated 1y agoView on Hugging Face
0likes7downloads
trainer_state.json735 linesDownload Raw Back to root
1{2  "best_global_step": null,3  "best_metric": null,4  "best_model_checkpoint": null,5  "epoch": 0.39564787339268054,6  "eval_steps": 500,7  "global_step": 100,8  "is_hyper_param_search": false,9  "is_local_process_zero": true,10  "is_world_process_zero": true,11  "log_history": [12    {13      "epoch": 0.003956478733926805,14      "grad_norm": 23.302045822143555,15      "learning_rate": 0.0,16      "loss": 16.2961,17      "step": 118    },19    {20      "epoch": 0.00791295746785361,21      "grad_norm": 27.37419319152832,22      "learning_rate": 4e-05,23      "loss": 18.4332,24      "step": 225    },26    {27      "epoch": 0.011869436201780416,28      "grad_norm": 27.051715850830078,29      "learning_rate": 8e-05,30      "loss": 18.1618,31      "step": 332    },33    {34      "epoch": 0.01582591493570722,35      "grad_norm": 25.241186141967773,36      "learning_rate": 0.00012,37      "loss": 15.8617,38      "step": 439    },40    {41      "epoch": 0.019782393669634024,42      "grad_norm": 19.574819564819336,43      "learning_rate": 0.00016,44      "loss": 12.3903,45      "step": 546    },47    {48      "epoch": 0.02373887240356083,49      "grad_norm": 14.805221557617188,50      "learning_rate": 0.0002,51      "loss": 11.0495,52      "step": 653    },54    {55      "epoch": 0.027695351137487636,56      "grad_norm": 8.009269714355469,57      "learning_rate": 0.00019789473684210526,58      "loss": 8.2391,59      "step": 760    },61    {62      "epoch": 0.03165182987141444,63      "grad_norm": 6.785006523132324,64      "learning_rate": 0.00019578947368421054,65      "loss": 7.9713,66      "step": 867    },68    {69      "epoch": 0.03560830860534125,70      "grad_norm": 6.116010665893555,71      "learning_rate": 0.0001936842105263158,72      "loss": 7.2536,73      "step": 974    },75    {76      "epoch": 0.03956478733926805,77      "grad_norm": 7.095190048217773,78      "learning_rate": 0.00019157894736842104,79      "loss": 7.2029,80      "step": 1081    },82    {83      "epoch": 0.043521266073194856,84      "grad_norm": 6.16122579574585,85      "learning_rate": 0.00018947368421052632,86      "loss": 6.6535,87      "step": 1188    },89    {90      "epoch": 0.04747774480712166,91      "grad_norm": 5.9373626708984375,92      "learning_rate": 0.0001873684210526316,93      "loss": 6.6532,94      "step": 1295    },96    {97      "epoch": 0.051434223541048464,98      "grad_norm": 3.409684896469116,99      "learning_rate": 0.00018526315789473685,100      "loss": 6.1402,101      "step": 13102    },103    {104      "epoch": 0.05539070227497527,105      "grad_norm": 2.748903512954712,106      "learning_rate": 0.0001831578947368421,107      "loss": 5.544,108      "step": 14109    },110    {111      "epoch": 0.05934718100890208,112      "grad_norm": 2.9191884994506836,113      "learning_rate": 0.00018105263157894739,114      "loss": 6.5131,115      "step": 15116    },117    {118      "epoch": 0.06330365974282888,119      "grad_norm": 2.935926675796509,120      "learning_rate": 0.00017894736842105264,121      "loss": 5.0912,122      "step": 16123    },124    {125      "epoch": 0.06726013847675569,126      "grad_norm": 2.6484129428863525,127      "learning_rate": 0.0001768421052631579,128      "loss": 5.4763,129      "step": 17130    },131    {132      "epoch": 0.0712166172106825,133      "grad_norm": 2.2135961055755615,134      "learning_rate": 0.00017473684210526317,135      "loss": 5.1443,136      "step": 18137    },138    {139      "epoch": 0.0751730959446093,140      "grad_norm": 2.2263529300689697,141      "learning_rate": 0.00017263157894736842,142      "loss": 5.6504,143      "step": 19144    },145    {146      "epoch": 0.0791295746785361,147      "grad_norm": 2.599684476852417,148      "learning_rate": 0.0001705263157894737,149      "loss": 4.7292,150      "step": 20151    },152    {153      "epoch": 0.0830860534124629,154      "grad_norm": 2.9603826999664307,155      "learning_rate": 0.00016842105263157895,156      "loss": 5.4304,157      "step": 21158    },159    {160      "epoch": 0.08704253214638971,161      "grad_norm": 2.178344249725342,162      "learning_rate": 0.00016631578947368423,163      "loss": 6.4425,164      "step": 22165    },166    {167      "epoch": 0.09099901088031652,168      "grad_norm": 2.872128963470459,169      "learning_rate": 0.00016421052631578948,170      "loss": 5.4385,171      "step": 23172    },173    {174      "epoch": 0.09495548961424333,175      "grad_norm": 2.738940715789795,176      "learning_rate": 0.00016210526315789473,177      "loss": 5.4402,178      "step": 24179    },180    {181      "epoch": 0.09891196834817013,182      "grad_norm": 3.214587688446045,183      "learning_rate": 0.00016,184      "loss": 4.7338,185      "step": 25186    },187    {188      "epoch": 0.10286844708209693,189      "grad_norm": 2.9738221168518066,190      "learning_rate": 0.00015789473684210527,191      "loss": 5.4748,192      "step": 26193    },194    {195      "epoch": 0.10682492581602374,196      "grad_norm": 3.050741195678711,197      "learning_rate": 0.00015578947368421052,198      "loss": 5.1201,199      "step": 27200    },201    {202      "epoch": 0.11078140454995054,203      "grad_norm": 3.187633991241455,204      "learning_rate": 0.0001536842105263158,205      "loss": 5.3061,206      "step": 28207    },208    {209      "epoch": 0.11473788328387735,210      "grad_norm": 2.965158700942993,211      "learning_rate": 0.00015157894736842108,212      "loss": 4.9141,213      "step": 29214    },215    {216      "epoch": 0.11869436201780416,217      "grad_norm": 2.6440789699554443,218      "learning_rate": 0.00014947368421052633,219      "loss": 4.4023,220      "step": 30221    },222    {223      "epoch": 0.12265084075173097,224      "grad_norm": 4.033472061157227,225      "learning_rate": 0.00014736842105263158,226      "loss": 7.4582,227      "step": 31228    },229    {230      "epoch": 0.12660731948565776,231      "grad_norm": 3.3675754070281982,232      "learning_rate": 0.00014526315789473686,233      "loss": 4.6167,234      "step": 32235    },236    {237      "epoch": 0.13056379821958458,238      "grad_norm": 3.0074822902679443,239      "learning_rate": 0.0001431578947368421,240      "loss": 4.8099,241      "step": 33242    },243    {244      "epoch": 0.13452027695351138,245      "grad_norm": 2.872875690460205,246      "learning_rate": 0.00014105263157894736,247      "loss": 4.8146,248      "step": 34249    },250    {251      "epoch": 0.13847675568743817,252      "grad_norm": 3.0648305416107178,253      "learning_rate": 0.00013894736842105264,254      "loss": 5.9663,255      "step": 35256    },257    {258      "epoch": 0.142433234421365,259      "grad_norm": 3.0596470832824707,260      "learning_rate": 0.0001368421052631579,261      "loss": 4.8893,262      "step": 36263    },264    {265      "epoch": 0.14638971315529178,266      "grad_norm": 3.099954843521118,267      "learning_rate": 0.00013473684210526317,268      "loss": 5.9515,269      "step": 37270    },271    {272      "epoch": 0.1503461918892186,273      "grad_norm": 2.4886155128479004,274      "learning_rate": 0.00013263157894736842,275      "loss": 4.1192,276      "step": 38277    },278    {279      "epoch": 0.1543026706231454,280      "grad_norm": 3.0190577507019043,281      "learning_rate": 0.0001305263157894737,282      "loss": 6.2374,283      "step": 39284    },285    {286      "epoch": 0.1582591493570722,287      "grad_norm": 3.001729726791382,288      "learning_rate": 0.00012842105263157895,289      "loss": 5.051,290      "step": 40291    },292    {293      "epoch": 0.16221562809099901,294      "grad_norm": 3.3806393146514893,295      "learning_rate": 0.0001263157894736842,296      "loss": 5.2736,297      "step": 41298    },299    {300      "epoch": 0.1661721068249258,301      "grad_norm": 3.381579637527466,302      "learning_rate": 0.00012421052631578949,303      "loss": 5.5759,304      "step": 42305    },306    {307      "epoch": 0.17012858555885263,308      "grad_norm": 3.010310411453247,309      "learning_rate": 0.00012210526315789474,310      "loss": 5.106,311      "step": 43312    },313    {314      "epoch": 0.17408506429277942,315      "grad_norm": 2.5623176097869873,316      "learning_rate": 0.00012,317      "loss": 5.5447,318      "step": 44319    },320    {321      "epoch": 0.17804154302670624,322      "grad_norm": 3.0234758853912354,323      "learning_rate": 0.00011789473684210525,324      "loss": 4.3962,325      "step": 45326    },327    {328      "epoch": 0.18199802176063304,329      "grad_norm": 2.936678886413574,330      "learning_rate": 0.00011578947368421053,331      "loss": 5.5828,332      "step": 46333    },334    {335      "epoch": 0.18595450049455983,336      "grad_norm": 3.5103631019592285,337      "learning_rate": 0.0001136842105263158,338      "loss": 4.3974,339      "step": 47340    },341    {342      "epoch": 0.18991097922848665,343      "grad_norm": 3.447932720184326,344      "learning_rate": 0.00011157894736842105,345      "loss": 5.7814,346      "step": 48347    },348    {349      "epoch": 0.19386745796241345,350      "grad_norm": 2.6578755378723145,351      "learning_rate": 0.00010947368421052633,352      "loss": 4.732,353      "step": 49354    },355    {356      "epoch": 0.19782393669634027,357      "grad_norm": 2.9903833866119385,358      "learning_rate": 0.00010736842105263158,359      "loss": 4.3962,360      "step": 50361    },362    {363      "epoch": 0.20178041543026706,364      "grad_norm": 2.790576696395874,365      "learning_rate": 0.00010526315789473685,366      "loss": 4.8403,367      "step": 51368    },369    {370      "epoch": 0.20573689416419386,371      "grad_norm": 3.302182912826538,372      "learning_rate": 0.00010315789473684211,373      "loss": 4.607,374      "step": 52375    },376    {377      "epoch": 0.20969337289812068,378      "grad_norm": 3.1099631786346436,379      "learning_rate": 0.00010105263157894738,380      "loss": 4.9049,381      "step": 53382    },383    {384      "epoch": 0.21364985163204747,385      "grad_norm": 3.343353509902954,386      "learning_rate": 9.894736842105263e-05,387      "loss": 5.6211,388      "step": 54389    },390    {391      "epoch": 0.2176063303659743,392      "grad_norm": 4.264265060424805,393      "learning_rate": 9.68421052631579e-05,394      "loss": 6.1449,395      "step": 55396    },397    {398      "epoch": 0.2215628090999011,399      "grad_norm": 2.6139657497406006,400      "learning_rate": 9.473684210526316e-05,401      "loss": 5.3725,402      "step": 56403    },404    {405      "epoch": 0.22551928783382788,406      "grad_norm": 3.1093761920928955,407      "learning_rate": 9.263157894736843e-05,408      "loss": 5.158,409      "step": 57410    },411    {412      "epoch": 0.2294757665677547,413      "grad_norm": 2.8728318214416504,414      "learning_rate": 9.052631578947369e-05,415      "loss": 5.2214,416      "step": 58417    },418    {419      "epoch": 0.2334322453016815,420      "grad_norm": 3.1342532634735107,421      "learning_rate": 8.842105263157894e-05,422      "loss": 6.7797,423      "step": 59424    },425    {426      "epoch": 0.23738872403560832,427      "grad_norm": 2.9147534370422363,428      "learning_rate": 8.631578947368421e-05,429      "loss": 4.9591,430      "step": 60431    },432    {433      "epoch": 0.2413452027695351,434      "grad_norm": 2.71091365814209,435      "learning_rate": 8.421052631578948e-05,436      "loss": 5.5919,437      "step": 61438    },439    {440      "epoch": 0.24530168150346193,441      "grad_norm": 4.9715681076049805,442      "learning_rate": 8.210526315789474e-05,443      "loss": 7.9325,444      "step": 62445    },446    {447      "epoch": 0.24925816023738873,448      "grad_norm": 3.243007183074951,449      "learning_rate": 8e-05,450      "loss": 4.9809,451      "step": 63452    },453    {454      "epoch": 0.2532146389713155,455      "grad_norm": 2.425530433654785,456      "learning_rate": 7.789473684210526e-05,457      "loss": 4.7267,458      "step": 64459    },460    {461      "epoch": 0.2571711177052423,462      "grad_norm": 2.584341526031494,463      "learning_rate": 7.578947368421054e-05,464      "loss": 5.1655,465      "step": 65466    },467    {468      "epoch": 0.26112759643916916,469      "grad_norm": 2.864600658416748,470      "learning_rate": 7.368421052631579e-05,471      "loss": 4.7094,472      "step": 66473    },474    {475      "epoch": 0.26508407517309596,476      "grad_norm": 2.7184255123138428,477      "learning_rate": 7.157894736842105e-05,478      "loss": 5.0806,479      "step": 67480    },481    {482      "epoch": 0.26904055390702275,483      "grad_norm": 3.3327622413635254,484      "learning_rate": 6.947368421052632e-05,485      "loss": 4.8204,486      "step": 68487    },488    {489      "epoch": 0.27299703264094954,490      "grad_norm": 2.6390984058380127,491      "learning_rate": 6.736842105263159e-05,492      "loss": 4.4369,493      "step": 69494    },495    {496      "epoch": 0.27695351137487634,497      "grad_norm": 2.8620619773864746,498      "learning_rate": 6.526315789473685e-05,499      "loss": 4.8152,500      "step": 70501    },502    {503      "epoch": 0.2809099901088032,504      "grad_norm": 2.886209011077881,505      "learning_rate": 6.31578947368421e-05,506      "loss": 4.5341,507      "step": 71508    },509    {510      "epoch": 0.28486646884273,511      "grad_norm": 3.60684871673584,512      "learning_rate": 6.105263157894737e-05,513      "loss": 5.147,514      "step": 72515    },516    {517      "epoch": 0.2888229475766568,518      "grad_norm": 2.6427412033081055,519      "learning_rate": 5.894736842105263e-05,520      "loss": 5.1367,521      "step": 73522    },523    {524      "epoch": 0.29277942631058357,525      "grad_norm": 3.7594962120056152,526      "learning_rate": 5.68421052631579e-05,527      "loss": 5.821,528      "step": 74529    },530    {531      "epoch": 0.29673590504451036,532      "grad_norm": 2.2990944385528564,533      "learning_rate": 5.4736842105263165e-05,534      "loss": 4.6194,535      "step": 75536    },537    {538      "epoch": 0.3006923837784372,539      "grad_norm": 2.6638574600219727,540      "learning_rate": 5.2631578947368424e-05,541      "loss": 5.8379,542      "step": 76543    },544    {545      "epoch": 0.304648862512364,546      "grad_norm": 3.256946086883545,547      "learning_rate": 5.052631578947369e-05,548      "loss": 5.1437,549      "step": 77550    },551    {552      "epoch": 0.3086053412462908,553      "grad_norm": 2.8275022506713867,554      "learning_rate": 4.842105263157895e-05,555      "loss": 5.0719,556      "step": 78557    },558    {559      "epoch": 0.3125618199802176,560      "grad_norm": 2.6694302558898926,561      "learning_rate": 4.6315789473684214e-05,562      "loss": 4.9186,563      "step": 79564    },565    {566      "epoch": 0.3165182987141444,567      "grad_norm": 3.2759788036346436,568      "learning_rate": 4.421052631578947e-05,569      "loss": 4.7457,570      "step": 80571    },572    {573      "epoch": 0.32047477744807124,574      "grad_norm": 3.3160855770111084,575      "learning_rate": 4.210526315789474e-05,576      "loss": 4.2814,577      "step": 81578    },579    {580      "epoch": 0.32443125618199803,581      "grad_norm": 2.8146486282348633,582      "learning_rate": 4e-05,583      "loss": 4.8129,584      "step": 82585    },586    {587      "epoch": 0.3283877349159248,588      "grad_norm": 2.767481565475464,589      "learning_rate": 3.789473684210527e-05,590      "loss": 5.5585,591      "step": 83592    },593    {594      "epoch": 0.3323442136498516,595      "grad_norm": 3.330084800720215,596      "learning_rate": 3.578947368421053e-05,597      "loss": 5.271,598      "step": 84599    },600    {601      "epoch": 0.3363006923837784,602      "grad_norm": 2.786287546157837,603      "learning_rate": 3.368421052631579e-05,604      "loss": 4.7267,605      "step": 85606    },607    {608      "epoch": 0.34025717111770526,609      "grad_norm": 3.373779296875,610      "learning_rate": 3.157894736842105e-05,611      "loss": 5.3886,612      "step": 86613    },614    {615      "epoch": 0.34421364985163205,616      "grad_norm": 2.731201171875,617      "learning_rate": 2.9473684210526314e-05,618      "loss": 5.1014,619      "step": 87620    },621    {622      "epoch": 0.34817012858555885,623      "grad_norm": 3.213326930999756,624      "learning_rate": 2.7368421052631583e-05,625      "loss": 5.2954,626      "step": 88627    },628    {629      "epoch": 0.35212660731948564,630      "grad_norm": 2.9587507247924805,631      "learning_rate": 2.5263157894736845e-05,632      "loss": 4.9975,633      "step": 89634    },635    {636      "epoch": 0.3560830860534125,637      "grad_norm": 2.7448644638061523,638      "learning_rate": 2.3157894736842107e-05,639      "loss": 4.8342,640      "step": 90641    },642    {643      "epoch": 0.3600395647873393,644      "grad_norm": 3.670548915863037,645      "learning_rate": 2.105263157894737e-05,646      "loss": 5.6434,647      "step": 91648    },649    {650      "epoch": 0.3639960435212661,651      "grad_norm": 3.459158182144165,652      "learning_rate": 1.8947368421052634e-05,653      "loss": 4.0876,654      "step": 92655    },656    {657      "epoch": 0.36795252225519287,658      "grad_norm": 2.871955633163452,659      "learning_rate": 1.6842105263157896e-05,660      "loss": 5.4301,661      "step": 93662    },663    {664      "epoch": 0.37190900098911966,665      "grad_norm": 3.1002166271209717,666      "learning_rate": 1.4736842105263157e-05,667      "loss": 4.7507,668      "step": 94669    },670    {671      "epoch": 0.3758654797230465,672      "grad_norm": 2.540811061859131,673      "learning_rate": 1.2631578947368422e-05,674      "loss": 4.8875,675      "step": 95676    },677    {678      "epoch": 0.3798219584569733,679      "grad_norm": 3.154196262359619,680      "learning_rate": 1.0526315789473684e-05,681      "loss": 5.8064,682      "step": 96683    },684    {685      "epoch": 0.3837784371909001,686      "grad_norm": 2.2360923290252686,687      "learning_rate": 8.421052631578948e-06,688      "loss": 5.1602,689      "step": 97690    },691    {692      "epoch": 0.3877349159248269,693      "grad_norm": 4.483743667602539,694      "learning_rate": 6.315789473684211e-06,695      "loss": 6.253,696      "step": 98697    },698    {699      "epoch": 0.3916913946587537,700      "grad_norm": 2.8893954753875732,701      "learning_rate": 4.210526315789474e-06,702      "loss": 4.7195,703      "step": 99704    },705    {706      "epoch": 0.39564787339268054,707      "grad_norm": 3.337634325027466,708      "learning_rate": 2.105263157894737e-06,709      "loss": 5.5186,710      "step": 100711    }712  ],713  "logging_steps": 1,714  "max_steps": 100,715  "num_input_tokens_seen": 0,716  "num_train_epochs": 1,717  "save_steps": 500,718  "stateful_callbacks": {719    "TrainerControl": {720      "args": {721        "should_epoch_stop": false,722        "should_evaluate": false,723        "should_log": false,724        "should_save": true,725        "should_training_stop": true726      },727      "attributes": {}728    }729  },730  "total_flos": 2480523648539904.0,731  "train_batch_size": 2,732  "trial_name": null,733  "trial_params": null734}735