| { | |
| "run_info": { | |
| "created_at": "2025-06-19T23:49:12+00:00", | |
| "total_time": 2210.184595478997, | |
| "experiment_name": "vblora/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": "VBLORA", | |
| "auto_mapping": null, | |
| "base_model_name_or_path": "meta-llama/Llama-3.2-3B", | |
| "revision": null, | |
| "inference_mode": false, | |
| "r": 4, | |
| "num_vectors": 256, | |
| "vector_length": 256, | |
| "topk": 2, | |
| "target_modules": [ | |
| "v_proj", | |
| "q_proj" | |
| ], | |
| "exclude_modules": null, | |
| "save_only_topk_weights": false, | |
| "vblora_dropout": 0.0, | |
| "fan_in_fan_out": false, | |
| "bias": "none", | |
| "modules_to_save": null, | |
| "init_vector_bank_bound": 0.02, | |
| "init_logits_std": 0.1, | |
| "layers_to_transform": null, | |
| "layers_pattern": null | |
| }, | |
| "error_msg": "" | |
| }, | |
| "train_info": { | |
| "accelerator_memory_reserved_avg": 11735344663, | |
| "accelerator_memory_max": 22181576704, | |
| "accelerator_memory_reserved_99th": 17635223797, | |
| "train_time": 1961.761054087001, | |
| "file_size": 4864912, | |
| "num_trainable_params": 1212416, | |
| "num_total_params": 3213962240, | |
| "status": "success", | |
| "metrics": [ | |
| { | |
| "step": 250, | |
| "valid accuracy": 0.0, | |
| "train loss": 1.308416832447052, | |
| "train samples": 1000, | |
| "train time": 40.12101128498034, | |
| "eval time": 12.847303112997906, | |
| "tokens / sec": 5277.010554299236, | |
| "mem allocated avg": 6798303909.888, | |
| "mem reserved avg": 11786547888.128, | |
| "elapsed time": 101.0704645309961 | |
| }, | |
| { | |
| "step": 500, | |
| "valid accuracy": 0.38, | |
| "train loss": 1.0353211228847503, | |
| "train samples": 2000, | |
| "train time": 39.44148263899842, | |
| "eval time": 12.918682472998626, | |
| "tokens / sec": 5273.508653408011, | |
| "mem allocated avg": 6790843463.68, | |
| "mem reserved avg": 11678368399.36, | |
| "elapsed time": 195.96383863499796 | |
| }, | |
| { | |
| "step": 750, | |
| "valid accuracy": 0.3, | |
| "train loss": 0.8149400608539581, | |
| "train samples": 3000, | |
| "train time": 39.73428049797076, | |
| "eval time": 12.816220013999555, | |
| "tokens / sec": 5395.869695210651, | |
| "mem allocated avg": 6801506754.56, | |
| "mem reserved avg": 11833448595.456, | |
| "elapsed time": 291.3562671039981 | |
| }, | |
| { | |
| "step": 1000, | |
| "valid accuracy": 0.34, | |
| "train loss": 0.766725031375885, | |
| "train samples": 4000, | |
| "train time": 39.66955411599338, | |
| "eval time": 12.954798815000686, | |
| "tokens / sec": 5251.785774824381, | |
| "mem allocated avg": 6791902521.344, | |
| "mem reserved avg": 11700715651.072, | |
| "elapsed time": 386.48246048299916 | |
| }, | |
| { | |
| "step": 1250, | |
| "valid accuracy": 0.4, | |
| "train loss": 0.7548577107191086, | |
| "train samples": 5000, | |
| "train time": 39.58740361595119, | |
| "eval time": 12.921318173001055, | |
| "tokens / sec": 5267.786744063522, | |
| "mem allocated avg": 6792241373.184, | |
| "mem reserved avg": 11698584944.64, | |
| "elapsed time": 481.6297270110008 | |
| }, | |
| { | |
| "step": 1500, | |
| "valid accuracy": 0.42, | |
| "train loss": 0.744083244919777, | |
| "train samples": 6000, | |
| "train time": 39.58100679998461, | |
| "eval time": 7.990361490003124, | |
| "tokens / sec": 5288.672950079719, | |
| "mem allocated avg": 6792975718.4, | |
| "mem reserved avg": 11729908006.912, | |
| "elapsed time": 571.7995823969977 | |
| }, | |
| { | |
| "step": 1750, | |
| "valid accuracy": 0.4, | |
| "train loss": 0.7353366105556488, | |
| "train samples": 7000, | |
| "train time": 39.89848869200068, | |
| "eval time": 9.013003496002057, | |
| "tokens / sec": 5247.191231129864, | |
| "mem allocated avg": 6795538806.784, | |
| "mem reserved avg": 11737281593.344, | |
| "elapsed time": 663.3418345429964 | |
| }, | |
| { | |
| "step": 2000, | |
| "valid accuracy": 0.36, | |
| "train loss": 0.735884799003601, | |
| "train samples": 8000, | |
| "train time": 39.573877232054656, | |
| "eval time": 13.005171182994673, | |
| "tokens / sec": 5248.3106161699825, | |
| "mem allocated avg": 6792414906.368, | |
| "mem reserved avg": 11683577724.928, | |
| "elapsed time": 758.2818646219966 | |
| }, | |
| { | |
| "step": 2250, | |
| "valid accuracy": 0.34, | |
| "train loss": 0.7294247032403945, | |
| "train samples": 9000, | |
| "train time": 40.16309046502283, | |
| "eval time": 12.827437616993848, | |
| "tokens / sec": 5351.878989172747, | |
| "mem allocated avg": 6803159742.464, | |
| "mem reserved avg": 11872145244.16, | |
| "elapsed time": 854.2901024749954 | |
| }, | |
| { | |
| "step": 2500, | |
| "valid accuracy": 0.36, | |
| "train loss": 0.7273153622150421, | |
| "train samples": 10000, | |
| "train time": 39.40831322706072, | |
| "eval time": 12.578817460002028, | |
| "tokens / sec": 5226.48606686793, | |
| "mem allocated avg": 6788682082.304, | |
| "mem reserved avg": 11624815525.888, | |
| "elapsed time": 948.6046375669976 | |
| }, | |
| { | |
| "step": 2750, | |
| "valid accuracy": 0.3, | |
| "train loss": 0.7221734907627105, | |
| "train samples": 11000, | |
| "train time": 39.99460277392063, | |
| "eval time": 12.831681943003787, | |
| "tokens / sec": 5297.739827488966, | |
| "mem allocated avg": 6798795204.608, | |
| "mem reserved avg": 11800045158.4, | |
| "elapsed time": 1044.2291650330008 | |
| }, | |
| { | |
| "step": 3000, | |
| "valid accuracy": 0.44, | |
| "train loss": 0.7163265677690506, | |
| "train samples": 12000, | |
| "train time": 39.72457089692762, | |
| "eval time": 9.721318816998973, | |
| "tokens / sec": 5254.455750864856, | |
| "mem allocated avg": 6794274019.328, | |
| "mem reserved avg": 11717786468.352, | |
| "elapsed time": 1136.276137433997 | |
| }, | |
| { | |
| "step": 3250, | |
| "valid accuracy": 0.24, | |
| "train loss": 0.7239821909666061, | |
| "train samples": 13000, | |
| "train time": 39.57894092098286, | |
| "eval time": 12.939295003001462, | |
| "tokens / sec": 5328.616559524724, | |
| "mem allocated avg": 6796102031.36, | |
| "mem reserved avg": 11749923225.6, | |
| "elapsed time": 1231.5789504099957 | |
| }, | |
| { | |
| "step": 3500, | |
| "valid accuracy": 0.3, | |
| "train loss": 0.7123430745601654, | |
| "train samples": 14000, | |
| "train time": 39.774808847985696, | |
| "eval time": 12.81972120499995, | |
| "tokens / sec": 5273.438291096208, | |
| "mem allocated avg": 6794877718.528, | |
| "mem reserved avg": 11727257206.784, | |
| "elapsed time": 1327.2042175199967 | |
| }, | |
| { | |
| "step": 3750, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.7080619329214096, | |
| "train samples": 15000, | |
| "train time": 40.429172475058294, | |
| "eval time": 12.810948685997573, | |
| "tokens / sec": 5360.065188910042, | |
| "mem allocated avg": 6804612114.432, | |
| "mem reserved avg": 11907847159.808, | |
| "elapsed time": 1424.0900874009967 | |
| }, | |
| { | |
| "step": 4000, | |
| "valid accuracy": 0.42, | |
| "train loss": 0.7257569855451584, | |
| "train samples": 16000, | |
| "train time": 39.64596449997771, | |
| "eval time": 12.844434396996803, | |
| "tokens / sec": 5154.950890401844, | |
| "mem allocated avg": 6787030419.456, | |
| "mem reserved avg": 11605689499.648, | |
| "elapsed time": 1519.344677588997 | |
| }, | |
| { | |
| "step": 4250, | |
| "valid accuracy": 0.38, | |
| "train loss": 0.7041294666528701, | |
| "train samples": 17000, | |
| "train time": 39.938396073041076, | |
| "eval time": 12.83599700799823, | |
| "tokens / sec": 5292.876549508964, | |
| "mem allocated avg": 6797280624.64, | |
| "mem reserved avg": 11765333098.496, | |
| "elapsed time": 1614.829351510998 | |
| }, | |
| { | |
| "step": 4500, | |
| "valid accuracy": 0.38, | |
| "train loss": 0.7148806138038635, | |
| "train samples": 18000, | |
| "train time": 39.55479707601626, | |
| "eval time": 12.901207727001747, | |
| "tokens / sec": 5253.926586972907, | |
| "mem allocated avg": 6791958038.528, | |
| "mem reserved avg": 11679299534.848, | |
| "elapsed time": 1710.0793527279966 | |
| }, | |
| { | |
| "step": 4750, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.7083848255872727, | |
| "train samples": 19000, | |
| "train time": 39.88160159892141, | |
| "eval time": 12.8585010780007, | |
| "tokens / sec": 5264.056396513368, | |
| "mem allocated avg": 6794248144.896, | |
| "mem reserved avg": 11730923028.48, | |
| "elapsed time": 1806.240128286001 | |
| }, | |
| { | |
| "step": 5000, | |
| "valid accuracy": 0.36, | |
| "train loss": 0.7142883945703506, | |
| "train samples": 20000, | |
| "train time": 39.631631882970396, | |
| "eval time": 12.818420095005422, | |
| "tokens / sec": 5255.398026885119, | |
| "mem allocated avg": 6791235981.312, | |
| "mem reserved avg": 11677395320.832, | |
| "elapsed time": 1901.3912653839943 | |
| }, | |
| { | |
| "step": 5000, | |
| "test accuracy": 0.36997725549658833, | |
| "train loss": 0.7142883945703506, | |
| "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" | |
| } | |
| } | |