{ "run_info": { "created_at": "2025-06-19T19:15:35+00:00", "total_time": 1993.494420946001, "experiment_name": "lora/llama-3.2-3B-rank32", "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": "CAUSAL_LM", "peft_type": "LORA", "auto_mapping": null, "base_model_name_or_path": "meta-llama/Llama-3.2-3B", "revision": null, "inference_mode": false, "r": 32, "target_modules": [ "v_proj", "q_proj" ], "exclude_modules": null, "lora_alpha": 64, "lora_dropout": 0.0, "fan_in_fan_out": false, "bias": "none", "use_rslora": false, "modules_to_save": null, "init_lora_weights": true, "layers_to_transform": null, "layers_pattern": null, "rank_pattern": {}, "alpha_pattern": {}, "megatron_config": null, "megatron_core": "megatron.core", "trainable_token_indices": null, "loftq_config": {}, "eva_config": null, "corda_config": null, "use_dora": false, "layer_replication": null, "lora_bias": false }, "error_msg": "" }, "train_info": { "accelerator_memory_reserved_avg": 11868689976, "accelerator_memory_max": 22273851392, "accelerator_memory_reserved_99th": 17710763212, "train_time": 1796.1857790169925, "file_size": 36715216, "num_trainable_params": 9175040, "num_total_params": 3221924864, "status": "success", "metrics": [ { "step": 250, "valid accuracy": 0.34, "train loss": 0.9827028260231018, "train samples": 1000, "train time": 31.395267726013117, "eval time": 11.27943390099972, "tokens / sec": 6743.659644748829, "mem allocated avg": 6925580957.696, "mem reserved avg": 11920245522.432, "elapsed time": 94.68654379600048 }, { "step": 500, "valid accuracy": 0.44, "train loss": 0.7164744178056717, "train samples": 2000, "train time": 30.728173206967767, "eval time": 11.244831023999723, "tokens / sec": 6768.869681873444, "mem allocated avg": 6918363699.2, "mem reserved avg": 11811654991.872, "elapsed time": 182.6767855429971 }, { "step": 750, "valid accuracy": 0.38, "train loss": 0.6791989279985428, "train samples": 3000, "train time": 31.248708018982143, "eval time": 6.873092081001232, "tokens / sec": 6861.115661798283, "mem allocated avg": 6929003134.976, "mem reserved avg": 11970174517.248, "elapsed time": 267.2763524209986 }, { "step": 1000, "valid accuracy": 0.42, "train loss": 0.6590347054004669, "train samples": 4000, "train time": 31.016855426081747, "eval time": 7.663122134003061, "tokens / sec": 6716.864012746194, "mem allocated avg": 6919503566.848, "mem reserved avg": 11835008876.544, "elapsed time": 351.92747904299904 }, { "step": 1250, "valid accuracy": 0.4, "train loss": 0.6547032891511917, "train samples": 5000, "train time": 30.914218463025463, "eval time": 11.249955232000502, "tokens / sec": 6745.698593332356, "mem allocated avg": 6919763681.28, "mem reserved avg": 11832551014.4, "elapsed time": 440.29597954699784 }, { "step": 1500, "valid accuracy": 0.42, "train loss": 0.647298491358757, "train samples": 6000, "train time": 31.093457819981268, "eval time": 11.25276822899832, "tokens / sec": 6732.316528188762, "mem allocated avg": 6920362313.728, "mem reserved avg": 11859000295.424, "elapsed time": 529.2981231249978 }, { "step": 1750, "valid accuracy": 0.46, "train loss": 0.6378061240911483, "train samples": 7000, "train time": 31.079548971014447, "eval time": 11.2527706639994, "tokens / sec": 6736.101614449091, "mem allocated avg": 6922653980.672, "mem reserved avg": 11870048092.16, "elapsed time": 617.7172930779998 }, { "step": 2000, "valid accuracy": 0.4, "train loss": 0.641120473742485, "train samples": 8000, "train time": 30.851661891996628, "eval time": 7.384566520999215, "tokens / sec": 6732.084667823985, "mem allocated avg": 6919747647.488, "mem reserved avg": 11816562327.552, "elapsed time": 702.0775224069985 }, { "step": 2250, "valid accuracy": 0.46, "train loss": 0.6332860335111618, "train samples": 9000, "train time": 31.288193090975255, "eval time": 11.258606130999397, "tokens / sec": 6869.939704571801, "mem allocated avg": 6930711803.904, "mem reserved avg": 12003997384.704, "elapsed time": 791.1291831710005 }, { "step": 2500, "valid accuracy": 0.44, "train loss": 0.6298432033061981, "train samples": 10000, "train time": 30.668521790059458, "eval time": 11.22552015600013, "tokens / sec": 6715.908950876132, "mem allocated avg": 6916224055.296, "mem reserved avg": 11759050031.104, "elapsed time": 878.9048607999976 }, { "step": 2750, "valid accuracy": 0.4, "train loss": 0.6213459351062774, "train samples": 11000, "train time": 31.198134894020768, "eval time": 7.820672179997928, "tokens / sec": 6791.463679471677, "mem allocated avg": 6926273599.488, "mem reserved avg": 11930135691.264, "elapsed time": 964.4001106439973 }, { "step": 3000, "valid accuracy": 0.46, "train loss": 0.6136174714565277, "train samples": 12000, "train time": 30.652901480014407, "eval time": 8.59450396900138, "tokens / sec": 6809.502197894445, "mem allocated avg": 6921910312.96, "mem reserved avg": 11851509268.48, "elapsed time": 1049.6233134680006 }, { "step": 3250, "valid accuracy": 0.46, "train loss": 0.6227310271263122, "train samples": 13000, "train time": 30.898520497004938, "eval time": 11.247846516002028, "tokens / sec": 6825.601893153528, "mem allocated avg": 6923552774.144, "mem reserved avg": 11884266782.72, "elapsed time": 1137.9473550990006 }, { "step": 3500, "valid accuracy": 0.52, "train loss": 0.6058980323076248, "train samples": 14000, "train time": 31.043968706952, "eval time": 7.071496761000162, "tokens / sec": 6756.545916535101, "mem allocated avg": 6922457063.424, "mem reserved avg": 11865602129.92, "elapsed time": 1222.3722963839973 }, { "step": 3750, "valid accuracy": 0.5, "train loss": 0.6032638043165207, "train samples": 15000, "train time": 31.41906641800597, "eval time": 6.834270917999675, "tokens / sec": 6897.18138397039, "mem allocated avg": 6932064409.6, "mem reserved avg": 12041553182.72, "elapsed time": 1307.517348808 }, { "step": 4000, "valid accuracy": 0.48, "train loss": 0.6166473155021668, "train samples": 16000, "train time": 30.82234557695483, "eval time": 6.627715251001064, "tokens / sec": 6630.676419149782, "mem allocated avg": 6914480900.096, "mem reserved avg": 11738338557.952, "elapsed time": 1390.9289551410002 }, { "step": 4250, "valid accuracy": 0.44, "train loss": 0.601645546555519, "train samples": 17000, "train time": 30.811621871023817, "eval time": 11.241402788000414, "tokens / sec": 6860.690452611215, "mem allocated avg": 6925075550.208, "mem reserved avg": 11899366277.12, "elapsed time": 1479.325017957999 }, { "step": 4500, "valid accuracy": 0.46, "train loss": 0.6076700875759125, "train samples": 18000, "train time": 30.499847401017178, "eval time": 11.232504903000518, "tokens / sec": 6813.73900884072, "mem allocated avg": 6919328847.872, "mem reserved avg": 11814020579.328, "elapsed time": 1567.0791362639975 }, { "step": 4750, "valid accuracy": 0.46, "train loss": 0.5997640329599381, "train samples": 19000, "train time": 30.974938084971654, "eval time": 11.246996836001927, "tokens / sec": 6777.705234602477, "mem allocated avg": 6921498724.352, "mem reserved avg": 11864662605.824, "elapsed time": 1655.6881185989987 }, { "step": 5000, "valid accuracy": 0.5, "train loss": 0.6069052599668503, "train samples": 20000, "train time": 30.736502733019734, "eval time": 11.28520023999954, "tokens / sec": 6776.307695418065, "mem allocated avg": 6918408683.52, "mem reserved avg": 11806051401.728, "elapsed time": 1743.785376189 }, { "step": 5000, "test accuracy": 0.48218347232752085, "train loss": 0.6069052599668503, "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": "0.15.2.dev0", "peft-commit-hash": "5fe7f8f8abe914d313fc3751f2ea92de7718fbaf", "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": "6.8.0-1029-aws", "version": "#31-Ubuntu SMP Wed Apr 23 18:42:41 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" } }