{ "run_info": { "created_at": "2025-06-20T04:48:22+00:00", "total_time": 2260.6744696069945, "experiment_name": "adaptionprompt/llama-3.2-3B-lr_0.0005", "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": 0.0005 }, "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": "CAUSAL_LM", "peft_type": "ADAPTION_PROMPT", "auto_mapping": null, "base_model_name_or_path": "meta-llama/Llama-3.2-3B", "revision": null, "inference_mode": false, "target_modules": "self_attn", "adapter_len": 100, "adapter_layers": 28 }, "error_msg": "" }, "train_info": { "accelerator_memory_reserved_avg": 11893757234, "accelerator_memory_max": 22410166272, "accelerator_memory_reserved_99th": 17907664814, "train_time": 1989.2834085189897, "file_size": 17210384, "num_trainable_params": 8601628, "num_total_params": 3221351452, "status": "success", "metrics": [ { "step": 250, "valid accuracy": 0.0, "train loss": 1.3201356165409088, "train samples": 1000, "train time": 36.18721537806414, "eval time": 13.46754032199533, "tokens / sec": 5850.657415556191, "mem allocated avg": 6848060076.032, "mem reserved avg": 11943163199.488, "elapsed time": 99.94861951399798 }, { "step": 500, "valid accuracy": 0.1, "train loss": 1.153662922859192, "train samples": 2000, "train time": 35.6493088029747, "eval time": 13.314302301005227, "tokens / sec": 5834.474972559473, "mem allocated avg": 6840933136.384, "mem reserved avg": 11833045942.272, "elapsed time": 193.4177081749949 }, { "step": 750, "valid accuracy": 0.22, "train loss": 0.9016587936878204, "train samples": 3000, "train time": 36.424757257977035, "eval time": 13.392894379001518, "tokens / sec": 5886.133941305707, "mem allocated avg": 6851972698.112, "mem reserved avg": 11989870968.832, "elapsed time": 288.2962625699947 }, { "step": 1000, "valid accuracy": 0.2, "train loss": 0.8571369113922119, "train samples": 4000, "train time": 35.59983186099271, "eval time": 13.363479856001504, "tokens / sec": 5852.1624712581015, "mem allocated avg": 6842572642.304, "mem reserved avg": 11863001661.44, "elapsed time": 381.66334240599826 }, { "step": 1250, "valid accuracy": 0.18, "train loss": 0.84929132604599, "train samples": 5000, "train time": 35.52914607799903, "eval time": 13.408120855005109, "tokens / sec": 5869.490911551474, "mem allocated avg": 6843078866.944, "mem reserved avg": 11855409971.2, "elapsed time": 475.2031378399988 }, { "step": 1500, "valid accuracy": 0.18, "train loss": 0.8379741818904877, "train samples": 6000, "train time": 35.84657208897261, "eval time": 13.451748254003178, "tokens / sec": 5839.637873335062, "mem allocated avg": 6844234328.064, "mem reserved avg": 11880013758.464, "elapsed time": 568.970056428996 }, { "step": 1750, "valid accuracy": 0.2, "train loss": 0.8320568509101868, "train samples": 7000, "train time": 36.04748217701126, "eval time": 13.354637482996623, "tokens / sec": 5807.756529900249, "mem allocated avg": 6845049858.048, "mem reserved avg": 11894333112.32, "elapsed time": 663.2131869919976 }, { "step": 2000, "valid accuracy": 0.2, "train loss": 0.83651398563385, "train samples": 8000, "train time": 35.70882848704787, "eval time": 13.407459709997056, "tokens / sec": 5816.376756110452, "mem allocated avg": 6842067818.496, "mem reserved avg": 11843724640.256, "elapsed time": 756.9679808469955 }, { "step": 2250, "valid accuracy": 0.18, "train loss": 0.8321560187339783, "train samples": 9000, "train time": 36.077689886013104, "eval time": 13.313609958000598, "tokens / sec": 5957.92027369615, "mem allocated avg": 6853360060.416, "mem reserved avg": 12025841319.936, "elapsed time": 851.5264306229947 }, { "step": 2500, "valid accuracy": 0.22, "train loss": 0.830465945482254, "train samples": 10000, "train time": 35.51607862501987, "eval time": 13.570960901000944, "tokens / sec": 5799.260728488849, "mem allocated avg": 6838232895.488, "mem reserved avg": 11785499312.128, "elapsed time": 945.1205676109967 }, { "step": 2750, "valid accuracy": 0.2, "train loss": 0.8323929319381714, "train samples": 11000, "train time": 36.33290277811466, "eval time": 13.340032396001334, "tokens / sec": 5831.6562619276265, "mem allocated avg": 6849506107.392, "mem reserved avg": 11957667102.72, "elapsed time": 1039.698461469001 }, { "step": 3000, "valid accuracy": 0.22, "train loss": 0.8273163681030273, "train samples": 12000, "train time": 36.133581758025684, "eval time": 13.486512909999874, "tokens / sec": 5776.648476140576, "mem allocated avg": 6844330549.248, "mem reserved avg": 11874754101.248, "elapsed time": 1134.0729920019949 }, { "step": 3250, "valid accuracy": 0.18, "train loss": 0.8321007430553437, "train samples": 13000, "train time": 35.81564853595046, "eval time": 13.383609317002993, "tokens / sec": 5888.515456820645, "mem allocated avg": 6845503963.136, "mem reserved avg": 11903065653.248, "elapsed time": 1228.1345331240009 }, { "step": 3500, "valid accuracy": 0.18, "train loss": 0.8267617487907409, "train samples": 14000, "train time": 35.759473790014454, "eval time": 13.568141147006827, "tokens / sec": 5865.578482269809, "mem allocated avg": 6844375582.72, "mem reserved avg": 11893385199.616, "elapsed time": 1322.3741278140005 }, { "step": 3750, "valid accuracy": 0.18, "train loss": 0.822540352344513, "train samples": 15000, "train time": 36.6447854490616, "eval time": 13.383382205000089, "tokens / sec": 5913.610827418539, "mem allocated avg": 6855454945.28, "mem reserved avg": 12064244367.36, "elapsed time": 1417.8726171529997 }, { "step": 4000, "valid accuracy": 0.22, "train loss": 0.842738341331482, "train samples": 16000, "train time": 35.83419257100468, "eval time": 13.484180120998644, "tokens / sec": 5703.295800373884, "mem allocated avg": 6837201041.408, "mem reserved avg": 11769015697.408, "elapsed time": 1511.8286734409994 }, { "step": 4250, "valid accuracy": 0.24, "train loss": 0.8195172207355499, "train samples": 17000, "train time": 36.032976000991766, "eval time": 13.43221827600064, "tokens / sec": 5866.542913196561, "mem allocated avg": 6847173238.784, "mem reserved avg": 11924070727.68, "elapsed time": 1606.2413196950001 }, { "step": 4500, "valid accuracy": 0.22, "train loss": 0.8333091423511505, "train samples": 18000, "train time": 35.92476197002543, "eval time": 13.364069708994066, "tokens / sec": 5784.812163081199, "mem allocated avg": 6842308513.792, "mem reserved avg": 11840637632.512, "elapsed time": 1700.1633438569988 }, { "step": 4750, "valid accuracy": 0.24, "train loss": 0.8247289218902588, "train samples": 19000, "train time": 36.319470202004595, "eval time": 13.367499373998726, "tokens / sec": 5780.343128144329, "mem allocated avg": 6845010323.456, "mem reserved avg": 11893443919.872, "elapsed time": 1795.0117048679967 }, { "step": 5000, "valid accuracy": 0.24, "train loss": 0.8317011270523071, "train samples": 20000, "train time": 35.778475134953624, "eval time": 13.382634160996531, "tokens / sec": 5821.377216731123, "mem allocated avg": 6841479706.624, "mem reserved avg": 11840956399.616, "elapsed time": 1888.9356832179983 }, { "step": 5000, "test accuracy": 0.22062168309325247, "train loss": 0.8317011270523071, "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" } }