{ "run_info": { "created_at": "2025-06-20T03:31:24+00:00", "total_time": 2742.3845372959986, "experiment_name": "bone/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": "BONE", "auto_mapping": null, "base_model_name_or_path": "meta-llama/Llama-3.2-3B", "revision": null, "inference_mode": false, "r": 64, "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": 14713983755, "accelerator_memory_max": 25251807232, "accelerator_memory_reserved_99th": 20472733368, "train_time": 2430.7548372539895, "file_size": 29367552, "num_trainable_params": 7340032, "num_total_params": 3220089856, "status": "success", "metrics": [ { "step": 250, "valid accuracy": 0.34, "train loss": 0.8741071329116822, "train samples": 1000, "train time": 44.769113782072964, "eval time": 16.53786130100343, "tokens / sec": 4729.130914464948, "mem allocated avg": 6898425409.536, "mem reserved avg": 14773294989.312, "elapsed time": 124.73039968500234 }, { "step": 500, "valid accuracy": 0.42, "train loss": 0.6946564470529556, "train samples": 2000, "train time": 43.747789238033874, "eval time": 16.4541177170031, "tokens / sec": 4754.4116770858745, "mem allocated avg": 6890118709.248, "mem reserved avg": 14662749913.088, "elapsed time": 242.48505929599924 }, { "step": 750, "valid accuracy": 0.42, "train loss": 0.6668610339164733, "train samples": 3000, "train time": 44.788394879076805, "eval time": 8.99262467600056, "tokens / sec": 4786.9766393472355, "mem allocated avg": 6900886024.192, "mem reserved avg": 14820195696.64, "elapsed time": 354.3122298879971 }, { "step": 1000, "valid accuracy": 0.42, "train loss": 0.6476555281877517, "train samples": 4000, "train time": 43.08444309095648, "eval time": 14.581032188005338, "tokens / sec": 4835.527282090601, "mem allocated avg": 6892210176.0, "mem reserved avg": 14677799075.84, "elapsed time": 469.41999823199876 }, { "step": 1250, "valid accuracy": 0.38, "train loss": 0.6442477897405624, "train samples": 5000, "train time": 43.81069704208494, "eval time": 16.504536090003967, "tokens / sec": 4759.979048031958, "mem allocated avg": 6892437598.208, "mem reserved avg": 14675995525.12, "elapsed time": 587.4669312400001 }, { "step": 1500, "valid accuracy": 0.48, "train loss": 0.6370412122011184, "train samples": 6000, "train time": 44.041188616007275, "eval time": 11.50742915799492, "tokens / sec": 4753.07335197389, "mem allocated avg": 6893869041.664, "mem reserved avg": 14704349020.16, "elapsed time": 700.887209352004 }, { "step": 1750, "valid accuracy": 0.44, "train loss": 0.6277673766613007, "train samples": 7000, "train time": 44.32280573899334, "eval time": 16.494074002999696, "tokens / sec": 4723.414876595195, "mem allocated avg": 6895170344.96, "mem reserved avg": 14718215389.184, "elapsed time": 819.4313268580008 }, { "step": 2000, "valid accuracy": 0.48, "train loss": 0.6278820457458496, "train samples": 8000, "train time": 43.325528461049544, "eval time": 16.452074027998606, "tokens / sec": 4793.848047040501, "mem allocated avg": 6891568050.176, "mem reserved avg": 14656710115.328, "elapsed time": 936.9070930559974 }, { "step": 2250, "valid accuracy": 0.44, "train loss": 0.6160005252361298, "train samples": 9000, "train time": 45.04456213898811, "eval time": 16.52133422600309, "tokens / sec": 4771.896757188206, "mem allocated avg": 6903412344.832, "mem reserved avg": 14851812360.192, "elapsed time": 1056.8185863660037 }, { "step": 2500, "valid accuracy": 0.5, "train loss": 0.6121727240085602, "train samples": 10000, "train time": 43.16439942702709, "eval time": 16.356938169003115, "tokens / sec": 4771.686916395162, "mem allocated avg": 6888002562.048, "mem reserved avg": 14598350569.472, "elapsed time": 1173.7929829869972 }, { "step": 2750, "valid accuracy": 0.52, "train loss": 0.6007345867156982, "train samples": 11000, "train time": 44.3066304440581, "eval time": 16.514935120998416, "tokens / sec": 4782.151065798665, "mem allocated avg": 6899352545.28, "mem reserved avg": 14785458470.912, "elapsed time": 1292.7444534430033 }, { "step": 3000, "valid accuracy": 0.52, "train loss": 0.5899704934358597, "train samples": 12000, "train time": 44.07467572299356, "eval time": 16.412788394998643, "tokens / sec": 4735.848796979486, "mem allocated avg": 6894036676.608, "mem reserved avg": 14687865405.44, "elapsed time": 1411.115336062001 }, { "step": 3250, "valid accuracy": 0.48, "train loss": 0.5988378477096558, "train samples": 13000, "train time": 44.070030323957326, "eval time": 10.250203846997465, "tokens / sec": 4785.587812163363, "mem allocated avg": 6895260303.36, "mem reserved avg": 14725043716.096, "elapsed time": 1523.332073521 }, { "step": 3500, "valid accuracy": 0.5, "train loss": 0.5801258901357651, "train samples": 14000, "train time": 43.991991777089424, "eval time": 16.38271237299341, "tokens / sec": 4767.913238909897, "mem allocated avg": 6893688922.112, "mem reserved avg": 14703484993.536, "elapsed time": 1641.7187374700006 }, { "step": 3750, "valid accuracy": 0.5, "train loss": 0.5768071869611741, "train samples": 15000, "train time": 45.04501243098639, "eval time": 16.454509290000715, "tokens / sec": 4810.810083180938, "mem allocated avg": 6905122422.784, "mem reserved avg": 14891314315.264, "elapsed time": 1761.645320085001 }, { "step": 4000, "valid accuracy": 0.52, "train loss": 0.5858320169448853, "train samples": 16000, "train time": 42.547905418032315, "eval time": 16.350580427999375, "tokens / sec": 4803.36218650576, "mem allocated avg": 6886491265.024, "mem reserved avg": 14582730981.376, "elapsed time": 1878.0724109930015 }, { "step": 4250, "valid accuracy": 0.54, "train loss": 0.5723247408866883, "train samples": 17000, "train time": 44.19116178697732, "eval time": 16.508775556001638, "tokens / sec": 4783.513070305705, "mem allocated avg": 6897152284.672, "mem reserved avg": 14738381602.816, "elapsed time": 1996.8971549050038 }, { "step": 4500, "valid accuracy": 0.48, "train loss": 0.5789256048202515, "train samples": 18000, "train time": 43.87211918797402, "eval time": 16.414912490006827, "tokens / sec": 4736.903615473535, "mem allocated avg": 6893093124.096, "mem reserved avg": 14658832433.152, "elapsed time": 2114.9650602839974 }, { "step": 4750, "valid accuracy": 0.48, "train loss": 0.568240401506424, "train samples": 19000, "train time": 43.939464293958736, "eval time": 16.460097985000175, "tokens / sec": 4777.914418698651, "mem allocated avg": 6894218592.256, "mem reserved avg": 14710372040.704, "elapsed time": 2233.517725938 }, { "step": 5000, "valid accuracy": 0.5, "train loss": 0.57634852206707, "train samples": 20000, "train time": 42.787552905057964, "eval time": 16.445046182001533, "tokens / sec": 4867.770785166333, "mem allocated avg": 6890906441.728, "mem reserved avg": 14656718503.936, "elapsed time": 2350.279711092 }, { "step": 5000, "test accuracy": 0.5170583775587566, "train loss": 0.57634852206707, "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" } }