benchmark-graveyard / MetaMathQA /ptuning--llama-3.2-3B-vt30-MLP.json
BenjaminB's picture
BenjaminB HF Staff
Upload 8 files (#2)
c398987
Raw
History Blame Contribute Delete
14.4 kB
{
"run_info": {
"created_at": "2026-07-01T15:50:58+00:00",
"total_time": 1492.634669919993,
"experiment_name": "ptuning/llama-3.2-3B-vt30-MLP",
"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,
"use_gc": 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,
"init_kv_cache_prefix": null
},
"peft_config": {
"task_type": "CAUSAL_LM",
"peft_type": "P_TUNING",
"auto_mapping": null,
"peft_version": "0.19.2.dev0@UNKNOWN",
"base_model_name_or_path": "meta-llama/Llama-3.2-3B",
"revision": null,
"inference_mode": false,
"num_virtual_tokens": 30,
"token_dim": 3072,
"num_transformer_submodules": 1,
"num_attention_heads": 24,
"num_layers": 28,
"modules_to_save": null,
"encoder_reparameterization_type": "MLP",
"encoder_hidden_size": 3072,
"encoder_num_layers": 2,
"encoder_dropout": 0.0
},
"error_msg": ""
},
"train_info": {
"accelerator_memory_reserved_avg": 14323580875,
"accelerator_memory_max": 21458059264,
"accelerator_memory_reserved_99th": 19736297472,
"train_time": 1194.9091126141138,
"file_size": 368760,
"num_trainable_params": 28412928,
"num_total_params": 3241162752,
"status": "success",
"metrics": [
{
"step": 250,
"valid accuracy": 0.08,
"train loss": 0.9363302497863769,
"train samples": 1000,
"train time": 34.00575314508751,
"eval time": 9.754597692983225,
"tokens / sec": 6225.975913448782,
"mem allocated avg": 7281153935.36,
"mem reserved avg": 14393391710.208,
"elapsed time": 258.6250282860128
},
{
"step": 500,
"valid accuracy": 0.14,
"train loss": 0.7985411801338196,
"train samples": 2000,
"train time": 33.73720351737575,
"eval time": 9.62405165200471,
"tokens / sec": 6165.1523634102,
"mem allocated avg": 7274907719.68,
"mem reserved avg": 14096116219.904,
"elapsed time": 304.5419928780175
},
{
"step": 750,
"valid accuracy": 0.22,
"train loss": 0.7601689212322235,
"train samples": 3000,
"train time": 34.6107883961231,
"eval time": 9.648138158983784,
"tokens / sec": 6194.629187470805,
"mem allocated avg": 7283639179.264,
"mem reserved avg": 14317038600.192,
"elapsed time": 351.2987851499929
},
{
"step": 1000,
"valid accuracy": 0.28,
"train loss": 0.7293364794254303,
"train samples": 4000,
"train time": 34.106910182803404,
"eval time": 9.637816358997952,
"tokens / sec": 6108.322298425096,
"mem allocated avg": 7276656787.456,
"mem reserved avg": 14334067474.432,
"elapsed time": 397.75173320499016
},
{
"step": 1250,
"valid accuracy": 0.32,
"train loss": 0.7207725659608841,
"train samples": 5000,
"train time": 33.97287149247131,
"eval time": 9.646640760998707,
"tokens / sec": 6138.368375667445,
"mem allocated avg": 7275547160.576,
"mem reserved avg": 14281647063.04,
"elapsed time": 444.0086419559957
},
{
"step": 1500,
"valid accuracy": 0.38,
"train loss": 0.7102297298908233,
"train samples": 6000,
"train time": 34.26030106123653,
"eval time": 9.622574922977947,
"tokens / sec": 6110.016360505525,
"mem allocated avg": 7277322196.992,
"mem reserved avg": 14224101212.16,
"elapsed time": 490.5009809990006
},
{
"step": 1750,
"valid accuracy": 0.28,
"train loss": 0.7035963997840882,
"train samples": 7000,
"train time": 34.326304641144816,
"eval time": 9.675930661993334,
"tokens / sec": 6098.967022190298,
"mem allocated avg": 7277080172.544,
"mem reserved avg": 14590775656.448,
"elapsed time": 537.2490275430027
},
{
"step": 2000,
"valid accuracy": 0.36,
"train loss": 0.7042187093496323,
"train samples": 8000,
"train time": 33.91284731691121,
"eval time": 7.696966222982155,
"tokens / sec": 6124.404655825785,
"mem allocated avg": 7274753998.848,
"mem reserved avg": 14271337463.808,
"elapsed time": 581.5433360229945
},
{
"step": 2250,
"valid accuracy": 0.38,
"train loss": 0.6965467965602875,
"train samples": 9000,
"train time": 35.00853349498357,
"eval time": 9.622937389998697,
"tokens / sec": 6139.874440350387,
"mem allocated avg": 7287149615.104,
"mem reserved avg": 14514305105.92,
"elapsed time": 628.8494034899923
},
{
"step": 2500,
"valid accuracy": 0.34,
"train loss": 0.6948007898330688,
"train samples": 10000,
"train time": 33.72064900404075,
"eval time": 5.387909529992612,
"tokens / sec": 6108.037836855362,
"mem allocated avg": 7272138412.032,
"mem reserved avg": 14066789646.336,
"elapsed time": 670.5936161220015
},
{
"step": 2750,
"valid accuracy": 0.32,
"train loss": 0.6874945278167724,
"train samples": 11000,
"train time": 34.540225339180324,
"eval time": 9.61607060598908,
"tokens / sec": 6134.325932137308,
"mem allocated avg": 7281916778.496,
"mem reserved avg": 14505924886.528,
"elapsed time": 717.2573533540126
},
{
"step": 3000,
"valid accuracy": 0.3,
"train loss": 0.6796259649991989,
"train samples": 12000,
"train time": 34.26134033527342,
"eval time": 9.50968838101835,
"tokens / sec": 6092.318571235321,
"mem allocated avg": 7276503754.752,
"mem reserved avg": 14410059874.304,
"elapsed time": 763.75420076499
},
{
"step": 3250,
"valid accuracy": 0.32,
"train loss": 0.68837134039402,
"train samples": 13000,
"train time": 34.255930323008215,
"eval time": 9.612164622027194,
"tokens / sec": 6156.627422211534,
"mem allocated avg": 7278948605.952,
"mem reserved avg": 14338605711.36,
"elapsed time": 810.1472433689923
},
{
"step": 3500,
"valid accuracy": 0.34,
"train loss": 0.6742574932575226,
"train samples": 14000,
"train time": 34.28129725190229,
"eval time": 6.6055597880040295,
"tokens / sec": 6118.49658018297,
"mem allocated avg": 7278363557.888,
"mem reserved avg": 14407107084.288,
"elapsed time": 853.6943467550154
},
{
"step": 3750,
"valid accuracy": 0.4,
"train loss": 0.6714792121648788,
"train samples": 15000,
"train time": 35.124471633374924,
"eval time": 8.83637132399599,
"tokens / sec": 6169.573232643049,
"mem allocated avg": 7288971911.168,
"mem reserved avg": 14592495321.088,
"elapsed time": 900.199428027001
},
{
"step": 4000,
"valid accuracy": 0.32,
"train loss": 0.6868616794347763,
"train samples": 16000,
"train time": 33.620701181003824,
"eval time": 9.615348125982564,
"tokens / sec": 6078.784582740162,
"mem allocated avg": 7270337929.216,
"mem reserved avg": 14266983776.256,
"elapsed time": 946.0162565729988
},
{
"step": 4250,
"valid accuracy": 0.38,
"train loss": 0.6681802560091019,
"train samples": 17000,
"train time": 34.270348728779936,
"eval time": 6.986196043988457,
"tokens / sec": 6168.276887783093,
"mem allocated avg": 7279699822.592,
"mem reserved avg": 14257932468.224,
"elapsed time": 989.7307399639976
},
{
"step": 4500,
"valid accuracy": 0.4,
"train loss": 0.6755580611228943,
"train samples": 18000,
"train time": 34.087618150311755,
"eval time": 8.943682793003973,
"tokens / sec": 6096.583195798893,
"mem allocated avg": 7274260011.008,
"mem reserved avg": 14269450027.008,
"elapsed time": 1035.466415124014
},
{
"step": 4750,
"valid accuracy": 0.38,
"train loss": 0.6677044332027435,
"train samples": 19000,
"train time": 34.23474362070556,
"eval time": 9.606772648985498,
"tokens / sec": 6132.337438421082,
"mem allocated avg": 7277436657.664,
"mem reserved avg": 14318162673.664,
"elapsed time": 1081.8005726340052
},
{
"step": 5000,
"valid accuracy": 0.44,
"train loss": 0.673769939661026,
"train samples": 20000,
"train time": 33.93519900739193,
"eval time": 9.618838055990636,
"tokens / sec": 6137.580037607307,
"mem allocated avg": 7274845607.936,
"mem reserved avg": 14015325536.256,
"elapsed time": 1127.8809662260173
},
{
"step": 5000,
"test accuracy": 0.3419257012888552,
"train loss": 0.673769939661026,
"train samples": 20000,
"train total tokens": 4198051,
"forgetting": 1.8561630249023438
}
]
},
"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": "740312add88f781978c0658806c59bc2815b9866",
"created_at": "2022-04-12T10:22:10+00:00"
}
},
"package_info": {
"transformers-version": "5.12.1",
"transformers-commit-hash": null,
"peft-version": "0.19.2.dev0",
"peft-commit-hash": "0fba0c34a9900c5bb11ce7c6964f9d14cf7a5c4b",
"datasets-version": "5.0.0",
"datasets-commit-hash": null,
"bitsandbytes-version": "0.49.2",
"bitsandbytes-commit-hash": null,
"torch-version": "2.12.0+cu130",
"torch-commit-hash": null
},
"system_info": {
"system": "Linux",
"release": "5.15.0-185-generic",
"version": "#195-Ubuntu SMP Fri Jun 19 17:11:50 UTC 2026",
"machine": "x86_64",
"processor": "x86_64",
"accelerator": "NVIDIA GeForce RTX 4090"
},
"pytorch_info": "PyTorch built with:\n - GCC 13.3\n - C++ Version: 202002\n - Intel(R) oneAPI Math Kernel Library Version 2024.2-Product Build 20240605 for Intel(R) 64 architecture applications\n - Intel(R) MKL-DNN v3.11.2 (Git Hash 03c022d3ffdcee958cfacbe720048e725fdf644c)\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 13.0\n - NVCC architecture flags: -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 92.0 (built against CUDA 13.2)\n - Magma 2.6.1\n - Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, COMMIT_SHA=7661cd9c6b841b62b7f411aa52ec51f05457263b, CUDA_FLAGS= -DLIBCUDACXX_ENABLE_SIMPLIFIED_COMPLEX_OPERATIONS -Xfatbin -compress-all -DONNX_NAMESPACE=onnx_torch -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 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -Xfatbin -compress-all --threads 2 -compress-mode=size -Wno-deprecated-gpu-targets --expt-extended-lambda -DCUB_WRAPPED_NAMESPACE=at_cuda_detail -DDISABLE_CUSPARSE_DEPRECATED -DCUDA_HAS_FP16=1 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -DC10_NODEPRECATED, CUDA_VERSION=13.0, CUDNN_VERSION=9.20.0, 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_MSLK -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=2.12.0, 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"
}
}