| { | |
| "run_info": { | |
| "created_at": "2025-06-19T19:48:53+00:00", | |
| "total_time": 1918.2703526590012, | |
| "experiment_name": "ptuning/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": "CAUSAL_LM", | |
| "peft_type": "P_TUNING", | |
| "auto_mapping": null, | |
| "base_model_name_or_path": "meta-llama/Llama-3.2-3B", | |
| "revision": null, | |
| "inference_mode": false, | |
| "num_virtual_tokens": 20, | |
| "token_dim": 3072, | |
| "num_transformer_submodules": 1, | |
| "num_attention_heads": 24, | |
| "num_layers": 28, | |
| "encoder_reparameterization_type": "MLP", | |
| "encoder_hidden_size": 3072, | |
| "encoder_num_layers": 2, | |
| "encoder_dropout": 0.0 | |
| }, | |
| "error_msg": "" | |
| }, | |
| "train_info": { | |
| "accelerator_memory_reserved_avg": 11867101593, | |
| "accelerator_memory_max": 20937965568, | |
| "accelerator_memory_reserved_99th": 17215688540, | |
| "train_time": 1707.340225783013, | |
| "file_size": 245880, | |
| "num_trainable_params": 28382208, | |
| "num_total_params": 3241132032, | |
| "status": "success", | |
| "metrics": [ | |
| { | |
| "step": 250, | |
| "valid accuracy": 0.06, | |
| "train loss": 0.9461167964935303, | |
| "train samples": 1000, | |
| "train time": 29.476242057011405, | |
| "eval time": 11.075081511000462, | |
| "tokens / sec": 7182.699870305862, | |
| "mem allocated avg": 7263395393.536, | |
| "mem reserved avg": 11910330187.776, | |
| "elapsed time": 89.09710205499869 | |
| }, | |
| { | |
| "step": 500, | |
| "valid accuracy": 0.3, | |
| "train loss": 0.7913461194038391, | |
| "train samples": 2000, | |
| "train time": 28.956617519994325, | |
| "eval time": 11.047425028998987, | |
| "tokens / sec": 7182.986751003671, | |
| "mem allocated avg": 7255670497.28, | |
| "mem reserved avg": 11810254094.336, | |
| "elapsed time": 171.9022758780011 | |
| }, | |
| { | |
| "step": 750, | |
| "valid accuracy": 0.26, | |
| "train loss": 0.7562740923166275, | |
| "train samples": 3000, | |
| "train time": 29.73533859500094, | |
| "eval time": 11.056799476999004, | |
| "tokens / sec": 7210.309689765724, | |
| "mem allocated avg": 7266187038.72, | |
| "mem reserved avg": 11954009669.632, | |
| "elapsed time": 255.8485612350014 | |
| }, | |
| { | |
| "step": 1000, | |
| "valid accuracy": 0.3, | |
| "train loss": 0.7289484927654266, | |
| "train samples": 4000, | |
| "train time": 29.176458327034197, | |
| "eval time": 11.069810884997423, | |
| "tokens / sec": 7140.551387861937, | |
| "mem allocated avg": 7258589235.2, | |
| "mem reserved avg": 11838347542.528, | |
| "elapsed time": 338.5030210529985 | |
| }, | |
| { | |
| "step": 1250, | |
| "valid accuracy": 0.4, | |
| "train loss": 0.7231850942373276, | |
| "train samples": 5000, | |
| "train time": 29.15449026899296, | |
| "eval time": 11.055301014999714, | |
| "tokens / sec": 7152.860436794844, | |
| "mem allocated avg": 7257714087.936, | |
| "mem reserved avg": 11824925769.728, | |
| "elapsed time": 421.85765765199903 | |
| }, | |
| { | |
| "step": 1500, | |
| "valid accuracy": 0.38, | |
| "train loss": 0.711922277212143, | |
| "train samples": 6000, | |
| "train time": 29.099172437985544, | |
| "eval time": 11.07098460600173, | |
| "tokens / sec": 7193.709733364892, | |
| "mem allocated avg": 7259322730.496, | |
| "mem reserved avg": 11860233420.8, | |
| "elapsed time": 504.97678817400083 | |
| }, | |
| { | |
| "step": 1750, | |
| "valid accuracy": 0.44, | |
| "train loss": 0.7051182547807694, | |
| "train samples": 7000, | |
| "train time": 29.301267419017677, | |
| "eval time": 11.044947161997698, | |
| "tokens / sec": 7144.912778213831, | |
| "mem allocated avg": 7260392302.592, | |
| "mem reserved avg": 11872371736.576, | |
| "elapsed time": 588.3443257949984 | |
| }, | |
| { | |
| "step": 2000, | |
| "valid accuracy": 0.38, | |
| "train loss": 0.7055468891859055, | |
| "train samples": 8000, | |
| "train time": 29.128185330951965, | |
| "eval time": 11.045154800998716, | |
| "tokens / sec": 7130.413296955362, | |
| "mem allocated avg": 7257253203.968, | |
| "mem reserved avg": 11821100564.48, | |
| "elapsed time": 671.2971968860002 | |
| }, | |
| { | |
| "step": 2250, | |
| "valid accuracy": 0.3, | |
| "train loss": 0.699348534822464, | |
| "train samples": 9000, | |
| "train time": 29.44214156106318, | |
| "eval time": 11.039785496999684, | |
| "tokens / sec": 7300.691750095574, | |
| "mem allocated avg": 7268387997.696, | |
| "mem reserved avg": 11993788448.768, | |
| "elapsed time": 755.1838785660002 | |
| }, | |
| { | |
| "step": 2500, | |
| "valid accuracy": 0.4, | |
| "train loss": 0.6970288401842117, | |
| "train samples": 10000, | |
| "train time": 28.56064905500898, | |
| "eval time": 11.062792377000733, | |
| "tokens / sec": 7211.565801718971, | |
| "mem allocated avg": 7253500915.712, | |
| "mem reserved avg": 11774535401.472, | |
| "elapsed time": 837.4507786270005 | |
| }, | |
| { | |
| "step": 2750, | |
| "valid accuracy": 0.38, | |
| "train loss": 0.6885807738304138, | |
| "train samples": 11000, | |
| "train time": 29.626391561985656, | |
| "eval time": 11.040969151999889, | |
| "tokens / sec": 7151.765329121947, | |
| "mem allocated avg": 7264164755.456, | |
| "mem reserved avg": 11929330384.896, | |
| "elapsed time": 921.4017121549987 | |
| }, | |
| { | |
| "step": 3000, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.6827223267555237, | |
| "train samples": 12000, | |
| "train time": 29.296160228008375, | |
| "eval time": 11.056816091997462, | |
| "tokens / sec": 7124.85862909926, | |
| "mem allocated avg": 7259324233.728, | |
| "mem reserved avg": 11842046918.656, | |
| "elapsed time": 1004.5840267519998 | |
| }, | |
| { | |
| "step": 3250, | |
| "valid accuracy": 0.5, | |
| "train loss": 0.6894591153860092, | |
| "train samples": 13000, | |
| "train time": 29.611147850035195, | |
| "eval time": 11.049655115999485, | |
| "tokens / sec": 7122.351388338677, | |
| "mem allocated avg": 7259635709.952, | |
| "mem reserved avg": 11876809310.208, | |
| "elapsed time": 1088.4846693049985 | |
| }, | |
| { | |
| "step": 3500, | |
| "valid accuracy": 0.42, | |
| "train loss": 0.6757243422269821, | |
| "train samples": 14000, | |
| "train time": 28.982272775025194, | |
| "eval time": 8.037888349997957, | |
| "tokens / sec": 7237.182591861713, | |
| "mem allocated avg": 7260029884.416, | |
| "mem reserved avg": 11864100569.088, | |
| "elapsed time": 1168.5907526180017 | |
| }, | |
| { | |
| "step": 3750, | |
| "valid accuracy": 0.44, | |
| "train loss": 0.6726652181148529, | |
| "train samples": 15000, | |
| "train time": 29.461453213014465, | |
| "eval time": 11.036738884999068, | |
| "tokens / sec": 7355.475591552708, | |
| "mem allocated avg": 7270358327.296, | |
| "mem reserved avg": 12018115411.968, | |
| "elapsed time": 1252.6760096750004 | |
| }, | |
| { | |
| "step": 4000, | |
| "valid accuracy": 0.44, | |
| "train loss": 0.6872537672519684, | |
| "train samples": 16000, | |
| "train time": 28.49340438899526, | |
| "eval time": 11.04012111100019, | |
| "tokens / sec": 7172.642384527876, | |
| "mem allocated avg": 7252451676.16, | |
| "mem reserved avg": 11753454829.568, | |
| "elapsed time": 1334.9961819890013 | |
| }, | |
| { | |
| "step": 4250, | |
| "valid accuracy": 0.46, | |
| "train loss": 0.6691881531476974, | |
| "train samples": 17000, | |
| "train time": 29.36704957404436, | |
| "eval time": 11.048986494999554, | |
| "tokens / sec": 7198.169481309866, | |
| "mem allocated avg": 7262467567.616, | |
| "mem reserved avg": 11896405098.496, | |
| "elapsed time": 1418.9507249929993 | |
| }, | |
| { | |
| "step": 4500, | |
| "valid accuracy": 0.5, | |
| "train loss": 0.6769082483053207, | |
| "train samples": 18000, | |
| "train time": 29.086171291994106, | |
| "eval time": 8.132250926999404, | |
| "tokens / sec": 7144.907382746569, | |
| "mem allocated avg": 7257195100.16, | |
| "mem reserved avg": 11816553938.944, | |
| "elapsed time": 1499.0322536989988 | |
| }, | |
| { | |
| "step": 4750, | |
| "valid accuracy": 0.46, | |
| "train loss": 0.6686601461172104, | |
| "train samples": 19000, | |
| "train time": 29.45103387799827, | |
| "eval time": 7.564945229998557, | |
| "tokens / sec": 7128.408492200246, | |
| "mem allocated avg": 7260019183.616, | |
| "mem reserved avg": 11863848910.848, | |
| "elapsed time": 1579.1494789060016 | |
| }, | |
| { | |
| "step": 5000, | |
| "valid accuracy": 0.48, | |
| "train loss": 0.6739867876768112, | |
| "train samples": 20000, | |
| "train time": 29.24236888399173, | |
| "eval time": 6.952750485001161, | |
| "tokens / sec": 7122.541980995923, | |
| "mem allocated avg": 7256318291.968, | |
| "mem reserved avg": 11821469663.232, | |
| "elapsed time": 1658.0220765080012 | |
| }, | |
| { | |
| "step": 5000, | |
| "test accuracy": 0.3707354056103108, | |
| "train loss": 0.6739867876768112, | |
| "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" | |
| } | |
| } | |