{ "run_info": { "created_at": "2025-08-14T11:55:49+00:00", "total_time": 2808.721444314, "experiment_name": "miss/llama-3.2-3B-bat", "peft_branch": "main", "train_config": { "model_id": "meta-llama/Llama-3.2-3B", "dtype": "bfloat16", "max_seq_length": 768, "batch_size": 4, "batch_size_eval": 50, "max_steps": 5000, "eval_steps": 250, "compile": false, "query_template": "Question: {query} Think step by step.\nAnswer:", "seed": 0, "grad_norm_clip": 1.0, "optimizer_type": "AdamW", "optimizer_kwargs": { "lr": 0.0001, "weight_decay": 0.1 }, "lr_scheduler": "cosine", "use_amp": false, "autocast_adapter_dtype": true, "generation_kwargs": { "max_length": 800, "max_new_tokens": 300 }, "attn_implementation": null }, "peft_config": { "task_type": null, "peft_type": "MISS", "auto_mapping": null, "base_model_name_or_path": "meta-llama/Llama-3.2-3B", "revision": null, "inference_mode": false, "r": 64, "miss_dropout": 0.0, "mini_r": 1, "target_modules": [ "v_proj", "q_proj" ], "exclude_modules": null, "init_weights": "bat", "layers_to_transform": null, "layers_pattern": null, "bias": "none", "modules_to_save": null }, "error_msg": "" }, "train_info": { "accelerator_memory_reserved_avg": 14713719934, "accelerator_memory_max": 25251807232, "accelerator_memory_reserved_99th": 20472733368, "train_time": 2466.149786608999, "file_size": 29367552, "num_trainable_params": 7340032, "num_total_params": 3220089856, "status": "success", "metrics": [ { "step": 250, "valid accuracy": 0.32, "train loss": 0.8741402707099915, "train samples": 1000, "train time": 44.507981576001725, "eval time": 16.603345405999903, "tokens / sec": 4756.8771376088835, "mem allocated avg": 6898417197.056, "mem reserved avg": 14772422574.08, "elapsed time": 128.87205576299993 }, { "step": 500, "valid accuracy": 0.42, "train loss": 0.6949697629213333, "train samples": 2000, "train time": 43.6579733309992, "eval time": 12.170993550999924, "tokens / sec": 4764.192749467687, "mem allocated avg": 6890132037.632, "mem reserved avg": 14662515032.064, "elapsed time": 244.05737383899998 }, { "step": 750, "valid accuracy": 0.38, "train loss": 0.667268633723259, "train samples": 3000, "train time": 44.76929137299828, "eval time": 8.243386759000032, "tokens / sec": 4789.0192903368525, "mem allocated avg": 6900972326.912, "mem reserved avg": 14823525974.016, "elapsed time": 357.2643382499999 }, { "step": 1000, "valid accuracy": 0.48, "train loss": 0.6478440872430802, "train samples": 4000, "train time": 43.91589877199954, "eval time": 9.950706549000074, "tokens / sec": 4743.976687842116, "mem allocated avg": 6892131758.08, "mem reserved avg": 14678444998.656, "elapsed time": 470.61746281599994 }, { "step": 1250, "valid accuracy": 0.4, "train loss": 0.6435494017601013, "train samples": 5000, "train time": 44.14956537599949, "eval time": 16.547810228000117, "tokens / sec": 4723.444007296278, "mem allocated avg": 6892566360.064, "mem reserved avg": 14674737233.92, "elapsed time": 591.057877963 }, { "step": 1500, "valid accuracy": 0.44, "train loss": 0.6368351166248322, "train samples": 6000, "train time": 44.08414804900008, "eval time": 16.39257521799982, "tokens / sec": 4748.441543371237, "mem allocated avg": 6893236697.088, "mem reserved avg": 14706580389.888, "elapsed time": 711.4482007859999 }, { "step": 1750, "valid accuracy": 0.48, "train loss": 0.6278127529621125, "train samples": 7000, "train time": 44.35628801999951, "eval time": 16.51757288099998, "tokens / sec": 4719.849413584954, "mem allocated avg": 6894834587.648, "mem reserved avg": 14716881600.512, "elapsed time": 832.303061434 }, { "step": 2000, "valid accuracy": 0.44, "train loss": 0.6281237225532532, "train samples": 8000, "train time": 43.95804043099747, "eval time": 16.465996583000106, "tokens / sec": 4724.869397352412, "mem allocated avg": 6891602710.528, "mem reserved avg": 14655669927.936, "elapsed time": 952.480474365 }, { "step": 2250, "valid accuracy": 0.42, "train loss": 0.6159191156625747, "train samples": 9000, "train time": 44.99231110500091, "eval time": 16.5404373570002, "tokens / sec": 4777.4385160692145, "mem allocated avg": 6903352731.648, "mem reserved avg": 14850520514.56, "elapsed time": 1074.326083797 }, { "step": 2500, "valid accuracy": 0.44, "train loss": 0.6119081476926803, "train samples": 10000, "train time": 43.74939265700118, "eval time": 16.33099729599985, "tokens / sec": 4707.882498273705, "mem allocated avg": 6887975004.16, "mem reserved avg": 14597494931.456, "elapsed time": 1194.094911997 }, { "step": 2750, "valid accuracy": 0.44, "train loss": 0.6010881408452987, "train samples": 11000, "train time": 43.686495668999896, "eval time": 11.229614545000004, "tokens / sec": 4850.0342441142875, "mem allocated avg": 6899207546.88, "mem reserved avg": 14785391362.048, "elapsed time": 1308.783695182 }, { "step": 3000, "valid accuracy": 0.5, "train loss": 0.5899516706466674, "train samples": 12000, "train time": 43.49030302700089, "eval time": 16.45857661900004, "tokens / sec": 4799.483688821613, "mem allocated avg": 6894123913.216, "mem reserved avg": 14693427052.544, "elapsed time": 1428.4006117669999 }, { "step": 3250, "valid accuracy": 0.52, "train loss": 0.5989595657587051, "train samples": 13000, "train time": 44.46332806799887, "eval time": 16.496417500999996, "tokens / sec": 4743.257177633304, "mem allocated avg": 6895596777.472, "mem reserved avg": 14723995140.096, "elapsed time": 1549.445265484 }, { "step": 3500, "valid accuracy": 0.46, "train loss": 0.579978278040886, "train samples": 14000, "train time": 43.63575344299579, "eval time": 10.30441635599982, "tokens / sec": 4806.838050224342, "mem allocated avg": 6893774680.064, "mem reserved avg": 14699450073.088, "elapsed time": 1663.316950223 }, { "step": 3750, "valid accuracy": 0.44, "train loss": 0.5772325273752212, "train samples": 15000, "train time": 45.25726027099972, "eval time": 16.524598716000128, "tokens / sec": 4788.2483098266675, "mem allocated avg": 6905177583.616, "mem reserved avg": 14889795977.216, "elapsed time": 1785.1977310290001 }, { "step": 4000, "valid accuracy": 0.4, "train loss": 0.5859311088323593, "train samples": 16000, "train time": 43.383903580999686, "eval time": 16.386461492000308, "tokens / sec": 4710.802466597467, "mem allocated avg": 6886734053.376, "mem reserved avg": 14584660361.216, "elapsed time": 1904.6209389110002 }, { "step": 4250, "valid accuracy": 0.5, "train loss": 0.5724418247938157, "train samples": 17000, "train time": 44.42285394400233, "eval time": 9.048803244000283, "tokens / sec": 4758.564145078759, "mem allocated avg": 6896789555.2, "mem reserved avg": 14740688470.016, "elapsed time": 2018.321323589 }, { "step": 4500, "valid accuracy": 0.46, "train loss": 0.5792494393587112, "train samples": 18000, "train time": 43.636566284001674, "eval time": 16.3964514889999, "tokens / sec": 4762.4737163655245, "mem allocated avg": 6892818855.936, "mem reserved avg": 14655921586.176, "elapsed time": 2137.859151554 }, { "step": 4750, "valid accuracy": 0.46, "train loss": 0.5680228790044785, "train samples": 19000, "train time": 43.96985955700529, "eval time": 16.500367100000403, "tokens / sec": 4774.61156608476, "mem allocated avg": 6894185185.28, "mem reserved avg": 14706722996.224, "elapsed time": 2258.0618387639997 }, { "step": 5000, "valid accuracy": 0.44, "train loss": 0.5760680929422378, "train samples": 20000, "train time": 43.83249596400128, "eval time": 16.474086973999874, "tokens / sec": 4751.7257555001215, "mem allocated avg": 6891346642.944, "mem reserved avg": 14655552487.424, "elapsed time": 2377.7959423069997 }, { "step": 5000, "test accuracy": 0.5049279757391963, "train loss": 0.5760680929422378, "train samples": 20000, "train total tokens": 4198051 } ] }, "meta_info": { "model_info": { "sha": "13afe5124825b4f3751f836b40dafda64c1ed062", "created_at": "2024-09-18T15:23:48+00:00" }, "dataset_info": { "metamath": { "sha": "aa4f34d3d2d3231299b5b03d9b3e5a20da45aa18", "created_at": "2023-09-21T17:22:46+00:00" }, "gsm8k": { "sha": "e53f048856ff4f594e959d75785d2c2d37b678ee", "created_at": "2022-04-12T10:22:10+00:00" } }, "package_info": { "transformers-version": "4.52.4", "transformers-commit-hash": null, "peft-version": "0.17.1.dev0", "peft-commit-hash": "47961bb54706e45fd3b5460baa4921a48bcdce35", "datasets-version": "3.6.0", "datasets-commit-hash": null, "bitsandbytes-version": "0.46.0", "bitsandbytes-commit-hash": null, "torch-version": "2.7.1+cu126", "torch-commit-hash": null }, "system_info": { "system": "Linux", "release": "6.14.0-1010-aws", "version": "#10~24.04.1-Ubuntu SMP Fri Jul 18 20:44:30 UTC 2025", "machine": "x86_64", "processor": "x86_64", "accelerator": "NVIDIA L40S" }, "pytorch_info": "PyTorch built with:\n - GCC 11.2\n - C++ Version: 201703\n - Intel(R) oneAPI Math Kernel Library Version 2024.2-Product Build 20240605 for Intel(R) 64 architecture applications\n - Intel(R) MKL-DNN v3.7.1 (Git Hash 8d263e693366ef8db40acc569cc7d8edf644556d)\n - OpenMP 201511 (a.k.a. OpenMP 4.5)\n - LAPACK is enabled (usually provided by MKL)\n - NNPACK is enabled\n - CPU capability usage: AVX2\n - CUDA Runtime 12.6\n - NVCC architecture flags: -gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_90,code=sm_90\n - CuDNN 90.7.1 (built against CUDA 12.8)\n - Built with CuDNN 90.5.1\n - Magma 2.6.1\n - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, COMMIT_SHA=e2d141dbde55c2a4370fac5165b0561b6af4798b, CUDA_VERSION=12.6, CUDNN_VERSION=9.5.1, CXX_COMPILER=/opt/rh/gcc-toolset-11/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=1 -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOROCTRACER -DLIBKINETO_NOXPUPTI=ON -DUSE_FBGEMM -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Werror=range-loop-construct -Werror=bool-operation -Wnarrowing -Wno-missing-field-initializers -Wno-unknown-pragmas -Wno-unused-parameter -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wsuggest-override -Wno-psabi -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, TORCH_VERSION=2.7.1, USE_CUDA=ON, USE_CUDNN=ON, USE_CUSPARSELT=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_GLOO=ON, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=1, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, USE_ROCM_KERNEL_ASSERT=OFF, \n" } }