{ "run_info": { "created_at": "2026-01-09T22:09:31+00:00", "total_time": 2139.03748531203, "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, "peft_version": "0.18.1.dev0@UNKNOWN", "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": 15762228825, "accelerator_memory_max": 23651680256, "accelerator_memory_reserved_99th": 21596471296, "train_time": 1726.894935239805, "file_size": 231416, "num_trainable_params": 56000, "num_total_params": 3212805824, "status": "success", "metrics": [ { "step": 250, "valid accuracy": 0.0, "train loss": 1.3261459288597106, "train samples": 1000, "train time": 54.26436190091772, "eval time": 20.152891070989426, "tokens / sec": 3901.62148016375, "mem allocated avg": 6780939091.968, "mem reserved avg": 15867958001.664, "elapsed time": 99.62701734097209 }, { "step": 500, "valid accuracy": 0.0, "train loss": 1.3400139141082763, "train samples": 2000, "train time": 53.535696760285646, "eval time": 20.12735749600688, "tokens / sec": 3885.16471414073, "mem allocated avg": 6773629601.792, "mem reserved avg": 15486259560.448, "elapsed time": 176.75905558397062 }, { "step": 750, "valid accuracy": 0.0, "train loss": 1.3044390759468079, "train samples": 3000, "train time": 53.646198650298174, "eval time": 20.156346999981906, "tokens / sec": 3996.573949211373, "mem allocated avg": 6783680942.08, "mem reserved avg": 15731836059.648, "elapsed time": 253.90234332997352 }, { "step": 1000, "valid accuracy": 0.0, "train loss": 1.311077395439148, "train samples": 4000, "train time": 52.92571030347608, "eval time": 20.228597007051576, "tokens / sec": 3936.3855261535678, "mem allocated avg": 6776640890.88, "mem reserved avg": 15724596690.944, "elapsed time": 330.55366935697384 }, { "step": 1250, "valid accuracy": 0.0, "train loss": 1.2998026401996612, "train samples": 5000, "train time": 53.52640636928845, "eval time": 20.14151435403619, "tokens / sec": 3895.9835741868837, "mem allocated avg": 6775686537.216, "mem reserved avg": 15805932634.112, "elapsed time": 407.61837184696924 }, { "step": 1500, "valid accuracy": 0.0, "train loss": 1.2868368306159974, "train samples": 6000, "train time": 53.88665995717747, "eval time": 20.14792396198027, "tokens / sec": 3884.6534590629794, "mem allocated avg": 6777282525.184, "mem reserved avg": 15735241834.496, "elapsed time": 485.17054163996363 }, { "step": 1750, "valid accuracy": 0.0, "train loss": 1.2805588779449464, "train samples": 7000, "train time": 53.43724203767488, "eval time": 20.139051661011763, "tokens / sec": 3917.773298487193, "mem allocated avg": 6777839691.776, "mem reserved avg": 16100456660.992, "elapsed time": 562.2017202629941 }, { "step": 2000, "valid accuracy": 0.0, "train loss": 1.2766080429553985, "train samples": 8000, "train time": 53.74037380289519, "eval time": 20.10542684403481, "tokens / sec": 3864.8037834975135, "mem allocated avg": 6774405505.024, "mem reserved avg": 15730795872.256, "elapsed time": 639.6445371389855 }, { "step": 2250, "valid accuracy": 0.0, "train loss": 1.2596942813396454, "train samples": 9000, "train time": 55.60716101707658, "eval time": 20.199656157987192, "tokens / sec": 3865.473368330941, "mem allocated avg": 6785559302.144, "mem reserved avg": 16046719238.144, "elapsed time": 718.9049943919526 }, { "step": 2500, "valid accuracy": 0.0, "train loss": 1.2683517651557923, "train samples": 10000, "train time": 53.73305493139196, "eval time": 20.08102407801198, "tokens / sec": 3833.152614586777, "mem allocated avg": 6771125086.208, "mem reserved avg": 15553687191.552, "elapsed time": 796.1963960899739 }, { "step": 2750, "valid accuracy": 0.0, "train loss": 1.254921770811081, "train samples": 11000, "train time": 55.10060942819109, "eval time": 20.137201737961732, "tokens / sec": 3845.347668543126, "mem allocated avg": 6781653977.088, "mem reserved avg": 15920974004.224, "elapsed time": 874.8866917559644 }, { "step": 3000, "valid accuracy": 0.0, "train loss": 1.2537723093032838, "train samples": 12000, "train time": 53.60143480350962, "eval time": 20.126665371004492, "tokens / sec": 3894.130833720389, "mem allocated avg": 6776695320.576, "mem reserved avg": 15812299587.584, "elapsed time": 952.1623955519754 }, { "step": 3250, "valid accuracy": 0.0, "train loss": 1.2485535402297974, "train samples": 13000, "train time": 54.74126313335728, "eval time": 20.236808118002955, "tokens / sec": 3852.6878615536516, "mem allocated avg": 6778833948.672, "mem reserved avg": 15685539332.096, "elapsed time": 1030.4662453850033 }, { "step": 3500, "valid accuracy": 0.0, "train loss": 1.247687777519226, "train samples": 14000, "train time": 53.67239992291434, "eval time": 20.154386248032097, "tokens / sec": 3907.9676016210988, "mem allocated avg": 6776816490.496, "mem reserved avg": 15717592203.264, "elapsed time": 1107.7825586889521 }, { "step": 3750, "valid accuracy": 0.0, "train loss": 1.2286142303943635, "train samples": 15000, "train time": 55.43350543506676, "eval time": 20.094228011963423, "tokens / sec": 3909.2422227174457, "mem allocated avg": 6788307730.432, "mem reserved avg": 16017065508.864, "elapsed time": 1186.77993246197 }, { "step": 4000, "valid accuracy": 0.0, "train loss": 1.2608506062030793, "train samples": 16000, "train time": 53.14804647688288, "eval time": 20.161071109992918, "tokens / sec": 3845.353000676958, "mem allocated avg": 6769206956.032, "mem reserved avg": 15759719792.64, "elapsed time": 1263.6211953659658 }, { "step": 4250, "valid accuracy": 0.0, "train loss": 1.2272750456333161, "train samples": 17000, "train time": 54.87057439738419, "eval time": 20.140905109001324, "tokens / sec": 3852.502043610417, "mem allocated avg": 6780253456.384, "mem reserved avg": 15731970277.376, "elapsed time": 1341.9455141729559 }, { "step": 4500, "valid accuracy": 0.0, "train loss": 1.2504476809501648, "train samples": 18000, "train time": 53.83008128858637, "eval time": 20.128594986046664, "tokens / sec": 3860.629503527497, "mem allocated avg": 6775227097.088, "mem reserved avg": 15675707883.52, "elapsed time": 1419.423019112961 }, { "step": 4750, "valid accuracy": 0.0, "train loss": 1.234989553451538, "train samples": 19000, "train time": 53.29879065020941, "eval time": 20.26173467701301, "tokens / sec": 3938.9073830547627, "mem allocated avg": 6777659713.536, "mem reserved avg": 15754426580.992, "elapsed time": 1496.3555859409971 }, { "step": 5000, "valid accuracy": 0.0, "train loss": 1.2481540043354034, "train samples": 20000, "train time": 53.355687880772166, "eval time": 20.13233814504929, "tokens / sec": 3903.6138089985725, "mem allocated avg": 6774096742.4, "mem reserved avg": 15385797591.04, "elapsed time": 1573.2050500559853 }, { "step": 5000, "test accuracy": 0.000758150113722517, "train loss": 1.2481540043354034, "train samples": 20000, "train total tokens": 4198051, "forgetting": -0.012529373168945312 } ] }, "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": "cc7b047b6e5bb11b4f1af84efc572db110a51b3c", "created_at": "2022-04-12T10:22:10+00:00" } }, "package_info": { "transformers-version": "4.57.1", "transformers-commit-hash": null, "peft-version": "0.18.1.dev0", "peft-commit-hash": "8be1a16f5e06ca5e197d2af74bdfc5b3c8072d26", "datasets-version": "4.2.0", "datasets-commit-hash": null, "bitsandbytes-version": "0.46.0", "bitsandbytes-commit-hash": null, "torch-version": "2.9.0+cu128", "torch-commit-hash": null }, "system_info": { "system": "Linux", "release": "6.14.0-1016-aws", "version": "#16~24.04.1-Ubuntu SMP Tue Oct 14 02:15:09 UTC 2025", "machine": "x86_64", "processor": "x86_64", "accelerator": "NVIDIA L40S" }, "pytorch_info": "PyTorch built with:\n - GCC 13.3\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 12.8\n - NVCC architecture flags: -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;-gencode;arch=compute_100,code=sm_100;-gencode;arch=compute_120,code=sm_120\n - CuDNN 90.7.1\n - Built with CuDNN 90.8\n - Magma 2.6.1\n - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, COMMIT_SHA=0fabc3ba44823f257e70ce397d989c8de5e362c1, CUDA_VERSION=12.8, CUDNN_VERSION=9.8.0, CXX_COMPILER=/opt/rh/gcc-toolset-13/root/usr/bin/c++, CXX_FLAGS= -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 -DC10_NODEPRECATED -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 -faligned-new -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-dangling-reference -Wno-error=dangling-reference -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, TORCH_VERSION=2.9.0, 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, USE_XCCL=OFF, USE_XPU=OFF, \n" } }