{ "run_info": { "created_at": "2025-06-20T10:18:57+00:00", "total_time": 2823.832106703994, "experiment_name": "fourierft/llama-3.2-3B-default", "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": "FOURIERFT", "auto_mapping": null, "base_model_name_or_path": "meta-llama/Llama-3.2-3B", "revision": null, "inference_mode": false, "n_frequency": 1000, "scaling": 300, "random_loc_seed": 777, "fan_in_fan_out": false, "target_modules": [ "q_proj", "v_proj" ], "exclude_modules": null, "bias": "none", "modules_to_save": null, "layers_to_transform": null, "layers_pattern": null, "n_frequency_pattern": {}, "init_weights": false }, "error_msg": "" }, "train_info": { "accelerator_memory_reserved_avg": 13104129350, "accelerator_memory_max": 23653777408, "accelerator_memory_reserved_99th": 19017267937, "train_time": 2424.3862988609762, "file_size": 231416, "num_trainable_params": 56000, "num_total_params": 3212805824, "status": "success", "metrics": [ { "step": 250, "valid accuracy": 0.0, "train loss": 1.3263031902313231, "train samples": 1000, "train time": 53.55340486107161, "eval time": 19.578013352002017, "tokens / sec": 3953.4180982374883, "mem allocated avg": 6781303625.728, "mem reserved avg": 13152850804.736, "elapsed time": 119.84825310099404 }, { "step": 500, "valid accuracy": 0.0, "train loss": 1.3399862418174744, "train samples": 2000, "train time": 52.85717789203045, "eval time": 19.544192551999004, "tokens / sec": 3935.03793231005, "mem allocated avg": 6774035257.344, "mem reserved avg": 13043463356.416, "elapsed time": 233.5829256769939 }, { "step": 750, "valid accuracy": 0.0, "train loss": 1.3045952091217041, "train samples": 3000, "train time": 53.35706212905643, "eval time": 19.607110917990212, "tokens / sec": 4018.2309790861696, "mem allocated avg": 6783920330.752, "mem reserved avg": 13205673869.312, "elapsed time": 348.1469791559939 }, { "step": 1000, "valid accuracy": 0.0, "train loss": 1.3111453976631164, "train samples": 4000, "train time": 52.95546973698947, "eval time": 19.472347582006478, "tokens / sec": 3934.1733919976355, "mem allocated avg": 6776025266.176, "mem reserved avg": 13077269446.656, "elapsed time": 461.81266678999236 }, { "step": 1250, "valid accuracy": 0.0, "train loss": 1.299716483592987, "train samples": 5000, "train time": 52.12036712520057, "eval time": 19.626158429004136, "tokens / sec": 4001.0846335572023, "mem allocated avg": 6775331573.76, "mem reserved avg": 13063344357.376, "elapsed time": 574.6407375999988 }, { "step": 1500, "valid accuracy": 0.0, "train loss": 1.2867344057559966, "train samples": 6000, "train time": 52.594848359090975, "eval time": 19.54386943600548, "tokens / sec": 3980.0666135738998, "mem allocated avg": 6776458844.16, "mem reserved avg": 13093568512.0, "elapsed time": 688.0431025519938 }, { "step": 1750, "valid accuracy": 0.0, "train loss": 1.2803141210079194, "train samples": 7000, "train time": 52.98738884186605, "eval time": 19.568909612993593, "tokens / sec": 3951.0344739725274, "mem allocated avg": 6778496358.4, "mem reserved avg": 13108768669.696, "elapsed time": 801.9154772249894 }, { "step": 2000, "valid accuracy": 0.0, "train loss": 1.2766506419181824, "train samples": 8000, "train time": 52.03297274692159, "eval time": 19.525613270001486, "tokens / sec": 3991.62279292005, "mem allocated avg": 6774647097.344, "mem reserved avg": 13051189264.384, "elapsed time": 914.5343848449993 }, { "step": 2250, "valid accuracy": 0.0, "train loss": 1.2596003375053406, "train samples": 9000, "train time": 53.934016149127274, "eval time": 19.535415460006334, "tokens / sec": 3985.388356870549, "mem allocated avg": 6785830477.824, "mem reserved avg": 13237223424.0, "elapsed time": 1029.9007452719961 }, { "step": 2500, "valid accuracy": 0.0, "train loss": 1.2684449093341827, "train samples": 10000, "train time": 52.006629903029534, "eval time": 19.470633051998448, "tokens / sec": 3960.3989026791724, "mem allocated avg": 6771212331.008, "mem reserved avg": 12996118052.864, "elapsed time": 1142.5889472209965 }, { "step": 2750, "valid accuracy": 0.0, "train loss": 1.2548872971534728, "train samples": 11000, "train time": 53.403087337108445, "eval time": 19.463876378998975, "tokens / sec": 3967.579601952513, "mem allocated avg": 6781916252.16, "mem reserved avg": 13168084516.864, "elapsed time": 1257.0122518049902 }, { "step": 3000, "valid accuracy": 0.0, "train loss": 1.253697858095169, "train samples": 12000, "train time": 53.20096563108382, "eval time": 19.472515105997445, "tokens / sec": 3923.443823321214, "mem allocated avg": 6777045135.36, "mem reserved avg": 13084844359.68, "elapsed time": 1370.94780872899 }, { "step": 3250, "valid accuracy": 0.0, "train loss": 1.248513156414032, "train samples": 13000, "train time": 52.962746563891415, "eval time": 19.54665829600708, "tokens / sec": 3982.06312328573, "mem allocated avg": 6779038627.84, "mem reserved avg": 13110345728.0, "elapsed time": 1484.7621198889974 }, { "step": 3500, "valid accuracy": 0.0, "train loss": 1.2477959940433503, "train samples": 14000, "train time": 52.93443578510778, "eval time": 19.444701158994576, "tokens / sec": 3962.4489595298505, "mem allocated avg": 6776803573.76, "mem reserved avg": 13097142059.008, "elapsed time": 1598.8772237269877 }, { "step": 3750, "valid accuracy": 0.0, "train loss": 1.228544222354889, "train samples": 15000, "train time": 53.31031796212483, "eval time": 19.472959079008433, "tokens / sec": 4064.9354249577, "mem allocated avg": 6788200585.216, "mem reserved avg": 13268999471.104, "elapsed time": 1713.6814467679942 }, { "step": 4000, "valid accuracy": 0.0, "train loss": 1.2609001460075377, "train samples": 16000, "train time": 51.9827769130934, "eval time": 19.473652824002784, "tokens / sec": 3931.552182017475, "mem allocated avg": 6770180233.216, "mem reserved avg": 12983610638.336, "elapsed time": 1826.5604049959948 }, { "step": 4250, "valid accuracy": 0.0, "train loss": 1.227214762210846, "train samples": 17000, "train time": 53.09942602888623, "eval time": 19.547112297004787, "tokens / sec": 3981.0034836347163, "mem allocated avg": 6779591426.048, "mem reserved avg": 13132760088.576, "elapsed time": 1940.5098487799987 }, { "step": 4500, "valid accuracy": 0.0, "train loss": 1.2504195840358734, "train samples": 18000, "train time": 52.23909889203787, "eval time": 19.522137050997117, "tokens / sec": 3978.207978462565, "mem allocated avg": 6775933241.344, "mem reserved avg": 13056079822.848, "elapsed time": 2053.2267840139975 }, { "step": 4750, "valid accuracy": 0.0, "train loss": 1.2349513354301453, "train samples": 19000, "train time": 53.36620609794045, "eval time": 19.541859832999762, "tokens / sec": 3933.931514912433, "mem allocated avg": 6777532579.84, "mem reserved avg": 13101604798.464, "elapsed time": 2167.8329333979927 }, { "step": 5000, "valid accuracy": 0.0, "train loss": 1.2480293517112733, "train samples": 20000, "train time": 52.46977503092785, "eval time": 19.44991449599911, "tokens / sec": 3969.5234042309344, "mem allocated avg": 6773533165.568, "mem reserved avg": 13049645760.512, "elapsed time": 2281.220151823989 }, { "step": 5000, "test accuracy": 0.000758150113722517, "train loss": 1.2480293517112733, "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" } }