Files
huggingface--peft/method_comparison/MetaMathQA/results/waveft--llama-3.2-3B-n_frequency-5000.json
wehub-resource-sync caf324b09d
Build documentation / build (push) Failing after 0s
Deploy "method_comparison" Gradio to Spaces / deploy (push) Has been cancelled
Deploy "PEFT shop" Gradio app to Spaces / deploy (push) Has been cancelled
tests on transformers main / tests (push) Has been cancelled
tests / check_code_quality (push) Has been cancelled
tests / tests (ubuntu-latest, 3.10) (push) Has been cancelled
tests / tests (ubuntu-latest, 3.11) (push) Has been cancelled
tests / tests (ubuntu-latest, 3.12) (push) Has been cancelled
tests / tests (ubuntu-latest, 3.13) (push) Has been cancelled
tests / tests (windows-latest, 3.10) (push) Has been cancelled
tests / tests (windows-latest, 3.11) (push) Has been cancelled
tests / tests (windows-latest, 3.12) (push) Has been cancelled
tests / tests (windows-latest, 3.13) (push) Has been cancelled
Secret Leaks / trufflehog (push) Has been cancelled
CI security linting / zizmor latest via Cargo (push) Has been cancelled
chore: import upstream snapshot with attribution
2026-07-13 13:24:42 +08:00

359 lines
13 KiB
JSON

{
"run_info": {
"created_at": "2026-01-10T10:48:59+00:00",
"total_time": 2531.969072496984,
"experiment_name": "waveft/llama-3.2-3B-n_frequency-5000",
"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": "CAUSAL_LM",
"peft_type": "WAVEFT",
"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": 5000,
"scaling": 25.0,
"wavelet_family": "db1",
"use_idwt": true,
"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": {},
"proportional_parameters": false,
"init_weights": true
},
"error_msg": ""
},
"train_info": {
"accelerator_memory_reserved_avg": 17216480005,
"accelerator_memory_max": 24895291392,
"accelerator_memory_reserved_99th": 23207084032,
"train_time": 2067.484345029923,
"file_size": 1127304,
"num_trainable_params": 280000,
"num_total_params": 3213029824,
"status": "success",
"metrics": [
{
"step": 250,
"valid accuracy": 0.34,
"train loss": 1.0574276540279388,
"train samples": 1000,
"train time": 64.88644388166722,
"eval time": 29.24885773198912,
"tokens / sec": 3262.915754577488,
"mem allocated avg": 6785503729.664,
"mem reserved avg": 17346156560.384,
"elapsed time": 119.97641712299082
},
{
"step": 500,
"valid accuracy": 0.36,
"train loss": 0.7510680348873139,
"train samples": 2000,
"train time": 64.65898592909798,
"eval time": 29.2429800920072,
"tokens / sec": 3216.79959886902,
"mem allocated avg": 6777740601.344,
"mem reserved avg": 16956279226.368,
"elapsed time": 217.2935731979669
},
{
"step": 750,
"valid accuracy": 0.36,
"train loss": 0.696870755314827,
"train samples": 3000,
"train time": 65.85057329281699,
"eval time": 22.638891860027798,
"tokens / sec": 3255.87142645859,
"mem allocated avg": 6787450871.808,
"mem reserved avg": 17161649127.424,
"elapsed time": 309.1054733619676
},
{
"step": 1000,
"valid accuracy": 0.38,
"train loss": 0.6762536237239838,
"train samples": 4000,
"train time": 64.82745491992682,
"eval time": 26.79089952702634,
"tokens / sec": 3213.700125314671,
"mem allocated avg": 6780707246.08,
"mem reserved avg": 17234697125.888,
"elapsed time": 404.1986583190155
},
{
"step": 1250,
"valid accuracy": 0.42,
"train loss": 0.6723555997610092,
"train samples": 5000,
"train time": 64.86200350278523,
"eval time": 29.28087447199505,
"tokens / sec": 3215.102660081186,
"mem allocated avg": 6779905163.264,
"mem reserved avg": 17208071684.096,
"elapsed time": 501.6932976419921
},
{
"step": 1500,
"valid accuracy": 0.44,
"train loss": 0.668335919380188,
"train samples": 6000,
"train time": 65.25852178613422,
"eval time": 29.221144348965026,
"tokens / sec": 3207.718996241155,
"mem allocated avg": 6780988528.64,
"mem reserved avg": 17179751743.488,
"elapsed time": 599.6254665869637
},
{
"step": 1750,
"valid accuracy": 0.52,
"train loss": 0.6587805707454681,
"train samples": 7000,
"train time": 65.32939753873507,
"eval time": 20.71052837802563,
"tokens / sec": 3204.606316411679,
"mem allocated avg": 6782190442.496,
"mem reserved avg": 17551023144.96,
"elapsed time": 689.0983284820104
},
{
"step": 2000,
"valid accuracy": 0.34,
"train loss": 0.6639124175310135,
"train samples": 8000,
"train time": 64.99063060310436,
"eval time": 19.205011637997814,
"tokens / sec": 3195.783731787319,
"mem allocated avg": 6778450325.504,
"mem reserved avg": 17222315540.48,
"elapsed time": 776.8548289329628
},
{
"step": 2250,
"valid accuracy": 0.42,
"train loss": 0.6567040872573853,
"train samples": 9000,
"train time": 65.94732432649471,
"eval time": 18.143796648015268,
"tokens / sec": 3259.389250363315,
"mem allocated avg": 6790019246.08,
"mem reserved avg": 17537441988.608,
"elapsed time": 864.3628261939739
},
{
"step": 2500,
"valid accuracy": 0.5,
"train loss": 0.6553765784502029,
"train samples": 10000,
"train time": 64.10474358743522,
"eval time": 26.53317212499678,
"tokens / sec": 3212.975958933097,
"mem allocated avg": 6775553220.608,
"mem reserved avg": 16962285469.696,
"elapsed time": 958.4253370200167
},
{
"step": 2750,
"valid accuracy": 0.4,
"train loss": 0.6487860177755356,
"train samples": 11000,
"train time": 65.63589023012901,
"eval time": 21.08229285199195,
"tokens / sec": 3228.1271611783477,
"mem allocated avg": 6786175535.104,
"mem reserved avg": 17393409589.248,
"elapsed time": 1048.5621143350145
},
{
"step": 3000,
"valid accuracy": 0.42,
"train loss": 0.6410797102451324,
"train samples": 12000,
"train time": 64.99408569413936,
"eval time": 29.25125474098604,
"tokens / sec": 3211.5383695415485,
"mem allocated avg": 6781518624.768,
"mem reserved avg": 17271095296.0,
"elapsed time": 1146.3423907670076
},
{
"step": 3250,
"valid accuracy": 0.4,
"train loss": 0.6509387036561965,
"train samples": 13000,
"train time": 65.00038516824134,
"eval time": 18.81445889500901,
"tokens / sec": 3244.611542748896,
"mem allocated avg": 6782426165.248,
"mem reserved avg": 17123514515.456,
"elapsed time": 1233.4203302449896
},
{
"step": 3500,
"valid accuracy": 0.44,
"train loss": 0.6376135617494583,
"train samples": 14000,
"train time": 64.8011383570265,
"eval time": 18.342268240987323,
"tokens / sec": 3236.8258539590365,
"mem allocated avg": 6781259872.256,
"mem reserved avg": 17207123771.392,
"elapsed time": 1320.0369267839706
},
{
"step": 3750,
"valid accuracy": 0.4,
"train loss": 0.635018677353859,
"train samples": 15000,
"train time": 66.17575967306038,
"eval time": 19.382110312988516,
"tokens / sec": 3274.6582898423158,
"mem allocated avg": 6792251662.336,
"mem reserved avg": 17455057469.44,
"elapsed time": 1409.0309616610175
},
{
"step": 4000,
"valid accuracy": 0.42,
"train loss": 0.6503389171361923,
"train samples": 16000,
"train time": 64.66550021909643,
"eval time": 17.806172358978074,
"tokens / sec": 3160.464224471373,
"mem allocated avg": 6773483292.672,
"mem reserved avg": 17174861185.024,
"elapsed time": 1494.9754901370034
},
{
"step": 4250,
"valid accuracy": 0.4,
"train loss": 0.6333471950292587,
"train samples": 17000,
"train time": 64.98916661646217,
"eval time": 20.801376049988903,
"tokens / sec": 3252.680577480337,
"mem allocated avg": 6783976222.72,
"mem reserved avg": 17191504183.296,
"elapsed time": 1583.9985767480102
},
{
"step": 4500,
"valid accuracy": 0.46,
"train loss": 0.6411651639938355,
"train samples": 18000,
"train time": 64.85606565064518,
"eval time": 20.16082675504731,
"tokens / sec": 3204.295510607074,
"mem allocated avg": 6779666743.296,
"mem reserved avg": 17107123175.424,
"elapsed time": 1672.4678985319915
},
{
"step": 4750,
"valid accuracy": 0.42,
"train loss": 0.6345895240306855,
"train samples": 19000,
"train time": 65.05685202759923,
"eval time": 18.8133876810316,
"tokens / sec": 3227.0082774822404,
"mem allocated avg": 6781692170.24,
"mem reserved avg": 17216258965.504,
"elapsed time": 1759.7040043339948
},
{
"step": 5000,
"valid accuracy": 0.44,
"train loss": 0.63977869784832,
"train samples": 20000,
"train time": 64.65817459212849,
"eval time": 19.78138606599532,
"tokens / sec": 3221.24775890837,
"mem allocated avg": 6778381412.352,
"mem reserved avg": 16829980344.32,
"elapsed time": 1847.474526909995
},
{
"step": 5000,
"test accuracy": 0.4200151630022744,
"train loss": 0.63977869784832,
"train samples": 20000,
"train total tokens": 4198051,
"forgetting": -0.13996505737304688
}
]
},
"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"
}
}