{ "run_info": { "created_at": "2025-06-20T07:20:24+00:00", "total_time": 2457.3893872150074, "experiment_name": "randlora/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": "RANDLORA", "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" ], "projection_prng_key": 0, "save_projection": true, "sparse": false, "very_sparse": false, "randlora_dropout": 0.0, "fan_in_fan_out": false, "randlora_alpha": 640, "bias": "none", "modules_to_save": null, "init_weights": true, "layers_to_transform": null, "layers_pattern": null }, "error_msg": "" }, "train_info": { "accelerator_memory_reserved_avg": 12743670025, "accelerator_memory_max": 22798139392, "accelerator_memory_reserved_99th": 18436063232, "train_time": 2213.072415724004, "file_size": 2211281240, "num_trainable_params": 9289728, "num_total_params": 3222039552, "status": "success", "metrics": [ { "step": 250, "valid accuracy": 0.38, "train loss": 0.9159075767993927, "train samples": 1000, "train time": 50.62416129904159, "eval time": 13.32173753400275, "tokens / sec": 4182.172989481373, "mem allocated avg": 6983776778.24, "mem reserved avg": 12791771561.984, "elapsed time": 114.85611474100733 }, { "step": 500, "valid accuracy": 0.34, "train loss": 0.7009325810670852, "train samples": 2000, "train time": 49.47734279213182, "eval time": 13.318595108998124, "tokens / sec": 4203.843380875268, "mem allocated avg": 6975756310.528, "mem reserved avg": 12690437177.344, "elapsed time": 222.717683150011 }, { "step": 750, "valid accuracy": 0.38, "train loss": 0.6809726172685623, "train samples": 3000, "train time": 50.701564677088754, "eval time": 6.592474952994962, "tokens / sec": 4228.6860645325305, "mem allocated avg": 6985956540.416, "mem reserved avg": 12840031223.808, "elapsed time": 325.2694208340108 }, { "step": 1000, "valid accuracy": 0.32, "train loss": 0.6661903276443482, "train samples": 4000, "train time": 49.452677299879724, "eval time": 13.326040301006287, "tokens / sec": 4212.835611238114, "mem allocated avg": 6977344550.912, "mem reserved avg": 12711484194.816, "elapsed time": 432.82023598300293 }, { "step": 1250, "valid accuracy": 0.44, "train loss": 0.665697453379631, "train samples": 5000, "train time": 49.56871296803001, "eval time": 6.698036557994783, "tokens / sec": 4207.0489127789, "mem allocated avg": 6977509738.496, "mem reserved avg": 12708397187.072, "elapsed time": 534.2725243740133 }, { "step": 1500, "valid accuracy": 0.44, "train loss": 0.658678293466568, "train samples": 6000, "train time": 49.71162069692218, "eval time": 13.42559558400535, "tokens / sec": 4210.906767176883, "mem allocated avg": 6978434217.984, "mem reserved avg": 12733680451.584, "elapsed time": 642.8949007330084 }, { "step": 1750, "valid accuracy": 0.44, "train loss": 0.6513392345905304, "train samples": 7000, "train time": 49.957065908936784, "eval time": 8.692238900999655, "tokens / sec": 4190.698476600257, "mem allocated avg": 6980155148.288, "mem reserved avg": 12746875731.968, "elapsed time": 746.9297674360132 }, { "step": 2000, "valid accuracy": 0.36, "train loss": 0.6511732361316681, "train samples": 8000, "train time": 49.75638979690848, "eval time": 13.350251003997982, "tokens / sec": 4174.257835983607, "mem allocated avg": 6976487055.36, "mem reserved avg": 12692744044.544, "elapsed time": 855.1831161730079 }, { "step": 2250, "valid accuracy": 0.38, "train loss": 0.6382467728853226, "train samples": 9000, "train time": 51.20128064297023, "eval time": 13.277926524999202, "tokens / sec": 4198.098119827237, "mem allocated avg": 6988260448.256, "mem reserved avg": 12868644765.696, "elapsed time": 965.4057929810078 }, { "step": 2500, "valid accuracy": 0.42, "train loss": 0.6324679807424546, "train samples": 10000, "train time": 47.79617171884456, "eval time": 13.268003197008511, "tokens / sec": 4309.278182603765, "mem allocated avg": 6973276801.024, "mem reserved avg": 12640810172.416, "elapsed time": 1071.8066454930085 }, { "step": 2750, "valid accuracy": 0.42, "train loss": 0.6214727911949157, "train samples": 11000, "train time": 49.63376283789694, "eval time": 8.245235234993743, "tokens / sec": 4268.888512281446, "mem allocated avg": 6983764305.92, "mem reserved avg": 12802987130.88, "elapsed time": 1175.2128759590123 }, { "step": 3000, "valid accuracy": 0.46, "train loss": 0.6079807863235474, "train samples": 12000, "train time": 49.776777152961586, "eval time": 13.29031453501375, "tokens / sec": 4193.340990289104, "mem allocated avg": 6978680711.168, "mem reserved avg": 12727900700.672, "elapsed time": 1283.6379908250092 }, { "step": 3250, "valid accuracy": 0.5, "train loss": 0.6133705099821091, "train samples": 13000, "train time": 50.014745363077964, "eval time": 7.092912267995416, "tokens / sec": 4216.77644200688, "mem allocated avg": 6980747913.216, "mem reserved avg": 12754257707.008, "elapsed time": 1386.1836155580095 }, { "step": 3500, "valid accuracy": 0.52, "train loss": 0.5912622555494308, "train samples": 14000, "train time": 49.560089439110016, "eval time": 13.321606318990234, "tokens / sec": 4232.236107194697, "mem allocated avg": 6979099045.888, "mem reserved avg": 12738579398.656, "elapsed time": 1494.848658177012 }, { "step": 3750, "valid accuracy": 0.48, "train loss": 0.5849999967813492, "train samples": 15000, "train time": 51.10861245104752, "eval time": 13.350408840997261, "tokens / sec": 4240.048586870968, "mem allocated avg": 6990205292.544, "mem reserved avg": 12906016014.336, "elapsed time": 1605.1250539940083 }, { "step": 4000, "valid accuracy": 0.52, "train loss": 0.5914600425958634, "train samples": 16000, "train time": 48.92153307204717, "eval time": 13.309209176004515, "tokens / sec": 4177.567364845621, "mem allocated avg": 6971749750.784, "mem reserved avg": 12621868695.552, "elapsed time": 1712.7276146870136 }, { "step": 4250, "valid accuracy": 0.54, "train loss": 0.575433883190155, "train samples": 17000, "train time": 50.056106529867975, "eval time": 13.322275185011677, "tokens / sec": 4223.041196259767, "mem allocated avg": 6981706383.36, "mem reserved avg": 12772226105.344, "elapsed time": 1821.8377219670074 }, { "step": 4500, "valid accuracy": 0.48, "train loss": 0.5807004086971282, "train samples": 18000, "train time": 49.559131018977496, "eval time": 13.371259606996318, "tokens / sec": 4193.334219690434, "mem allocated avg": 6976847835.136, "mem reserved avg": 12694061056.0, "elapsed time": 1929.8977656790084 }, { "step": 4750, "valid accuracy": 0.52, "train loss": 0.5704656873941422, "train samples": 19000, "train time": 49.80019182183605, "eval time": 13.346957685993402, "tokens / sec": 4215.62633234572, "mem allocated avg": 6979789905.92, "mem reserved avg": 12742303940.608, "elapsed time": 2038.7162974260136 }, { "step": 5000, "valid accuracy": 0.52, "train loss": 0.5784689987897873, "train samples": 20000, "train time": 49.38921916205436, "eval time": 13.307282750000013, "tokens / sec": 4217.114656471855, "mem allocated avg": 6976297842.688, "mem reserved avg": 12688323248.128, "elapsed time": 2146.6737093550037 }, { "step": 5000, "test accuracy": 0.5072024260803639, "train loss": 0.5784689987897873, "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" } }