{ "run_info": { "created_at": "2026-01-10T02:16:05+00:00", "total_time": 1280.0792806840036, "experiment_name": "lora/llama-3.2-3B-rank10-target-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, "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": "LORA", "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, "r": 10, "target_modules": [ "up_proj", "gate_proj", "down_proj" ], "exclude_modules": null, "lora_alpha": 20, "lora_dropout": 0.0, "fan_in_fan_out": false, "bias": "none", "use_rslora": false, "modules_to_save": null, "init_lora_weights": true, "layers_to_transform": null, "layers_pattern": null, "rank_pattern": {}, "alpha_pattern": {}, "megatron_config": null, "megatron_core": "megatron.core", "trainable_token_indices": null, "loftq_config": {}, "eva_config": null, "corda_config": null, "use_dora": false, "alora_invocation_tokens": null, "use_qalora": false, "qalora_group_size": 16, "layer_replication": null, "lora_bias": false, "target_parameters": null, "use_bdlora": null, "arrow_config": null, "ensure_weight_tying": false }, "error_msg": "" }, "train_info": { "accelerator_memory_reserved_avg": 15498352577, "accelerator_memory_max": 24693964800, "accelerator_memory_reserved_99th": 22066233344, "train_time": 1084.286043549073, "file_size": 37868984, "num_trainable_params": 9461760, "num_total_params": 3222211584, "status": "success", "metrics": [ { "step": 250, "valid accuracy": 0.28, "train loss": 0.956905131816864, "train samples": 1000, "train time": 34.45501961914124, "eval time": 8.187887417967431, "tokens / sec": 6144.794063108907, "mem allocated avg": 6931928963.072, "mem reserved avg": 15621727191.04, "elapsed time": 66.55439633503556 }, { "step": 500, "valid accuracy": 0.34, "train loss": 0.6960876834392548, "train samples": 2000, "train time": 34.311293395468965, "eval time": 12.610760554962326, "tokens / sec": 6061.998234886334, "mem allocated avg": 6925043042.304, "mem reserved avg": 15157191245.824, "elapsed time": 116.78655780700501 }, { "step": 750, "valid accuracy": 0.4, "train loss": 0.6628184849023819, "train samples": 3000, "train time": 34.70101865468314, "eval time": 7.451941114966758, "tokens / sec": 6178.5217930789795, "mem allocated avg": 6935014987.776, "mem reserved avg": 15434812227.584, "elapsed time": 162.2516898770118 }, { "step": 1000, "valid accuracy": 0.42, "train loss": 0.6416039798259735, "train samples": 4000, "train time": 34.48607983236434, "eval time": 6.998231331992429, "tokens / sec": 6041.162144630941, "mem allocated avg": 6926214045.696, "mem reserved avg": 15482434355.2, "elapsed time": 207.0800661740359 }, { "step": 1250, "valid accuracy": 0.38, "train loss": 0.6370971938371658, "train samples": 5000, "train time": 34.169997635355685, "eval time": 7.873161940020509, "tokens / sec": 6102.956231528264, "mem allocated avg": 6926345869.312, "mem reserved avg": 15533814579.2, "elapsed time": 252.415388560039 }, { "step": 1500, "valid accuracy": 0.5, "train loss": 0.6280006822347641, "train samples": 6000, "train time": 34.276999980502296, "eval time": 8.087436428992078, "tokens / sec": 6107.039709399109, "mem allocated avg": 6927777204.224, "mem reserved avg": 15449374851.072, "elapsed time": 298.10151039704215 }, { "step": 1750, "valid accuracy": 0.42, "train loss": 0.6198004070520401, "train samples": 7000, "train time": 34.87267060915474, "eval time": 12.61192423501052, "tokens / sec": 6003.411735981022, "mem allocated avg": 6928959924.224, "mem reserved avg": 15878183714.816, "elapsed time": 348.9116879070061 }, { "step": 2000, "valid accuracy": 0.36, "train loss": 0.62178213763237, "train samples": 8000, "train time": 34.02151767432224, "eval time": 7.202711585967336, "tokens / sec": 6104.842293874464, "mem allocated avg": 6924543111.168, "mem reserved avg": 15504924213.248, "elapsed time": 393.50626342202304 }, { "step": 2250, "valid accuracy": 0.52, "train loss": 0.6131408748626709, "train samples": 9000, "train time": 34.81766810180852, "eval time": 10.382788593007717, "tokens / sec": 6173.532339141203, "mem allocated avg": 6936508715.008, "mem reserved avg": 15821845823.488, "elapsed time": 441.9828996449942 }, { "step": 2500, "valid accuracy": 0.48, "train loss": 0.6092700402736664, "train samples": 10000, "train time": 33.81983969331486, "eval time": 12.668855660012923, "tokens / sec": 6090.123485733533, "mem allocated avg": 6922254147.584, "mem reserved avg": 15216221880.32, "elapsed time": 491.83312584704254 }, { "step": 2750, "valid accuracy": 0.52, "train loss": 0.599810087800026, "train samples": 11000, "train time": 34.98453600978246, "eval time": 12.589807912008837, "tokens / sec": 6056.418754296279, "mem allocated avg": 6932940275.712, "mem reserved avg": 15688190132.224, "elapsed time": 542.6733375560143 }, { "step": 3000, "valid accuracy": 0.38, "train loss": 0.5911425656080246, "train samples": 12000, "train time": 34.17171706445515, "eval time": 10.20739641803084, "tokens / sec": 6108.297092776719, "mem allocated avg": 6927596806.144, "mem reserved avg": 15592476114.944, "elapsed time": 590.4343665740453 }, { "step": 3250, "valid accuracy": 0.48, "train loss": 0.5990802057981491, "train samples": 13000, "train time": 34.13041735702427, "eval time": 9.451952209987212, "tokens / sec": 6179.268123030882, "mem allocated avg": 6930062843.904, "mem reserved avg": 15432622800.896, "elapsed time": 637.2517090940382 }, { "step": 3500, "valid accuracy": 0.54, "train loss": 0.5822520101070404, "train samples": 14000, "train time": 34.94589354854543, "eval time": 12.59894744300982, "tokens / sec": 6002.135836321477, "mem allocated avg": 6928036900.864, "mem reserved avg": 15451773992.96, "elapsed time": 688.1364196360228 }, { "step": 3750, "valid accuracy": 0.62, "train loss": 0.5812049109935761, "train samples": 15000, "train time": 35.20199201494688, "eval time": 12.605403171968646, "tokens / sec": 6155.986851766434, "mem allocated avg": 6939365515.264, "mem reserved avg": 15827197755.392, "elapsed time": 739.258546972007 }, { "step": 4000, "valid accuracy": 0.48, "train loss": 0.5900991085767746, "train samples": 16000, "train time": 33.90330129902577, "eval time": 8.236152897006832, "tokens / sec": 6028.115026245918, "mem allocated avg": 6919923267.584, "mem reserved avg": 15453300719.616, "elapsed time": 784.7485570590361 }, { "step": 4250, "valid accuracy": 0.52, "train loss": 0.5774198567867279, "train samples": 17000, "train time": 34.68033379589906, "eval time": 7.178619690006599, "tokens / sec": 6095.356556948615, "mem allocated avg": 6930677893.12, "mem reserved avg": 15433386164.224, "elapsed time": 829.8555172120105 }, { "step": 4500, "valid accuracy": 0.5, "train loss": 0.5839589636325836, "train samples": 18000, "train time": 33.97633226803737, "eval time": 7.210191433026921, "tokens / sec": 6116.551909150627, "mem allocated avg": 6925906124.8, "mem reserved avg": 15420786475.008, "elapsed time": 874.4303943190025 }, { "step": 4750, "valid accuracy": 0.52, "train loss": 0.575189305305481, "train samples": 19000, "train time": 34.245364292524755, "eval time": 10.291229791007936, "tokens / sec": 6130.435588498806, "mem allocated avg": 6928215662.592, "mem reserved avg": 15476939816.96, "elapsed time": 922.2494909689995 }, { "step": 5000, "valid accuracy": 0.5, "train loss": 0.5811728711128235, "train samples": 20000, "train time": 34.228812442102935, "eval time": 8.269593706994783, "tokens / sec": 6084.932112450577, "mem allocated avg": 6926067277.824, "mem reserved avg": 15089847500.8, "elapsed time": 968.0756093530217 }, { "step": 5000, "test accuracy": 0.5109931766489765, "train loss": 0.5811728711128235, "train samples": 20000, "train total tokens": 4198051, "forgetting": 0.4244680404663086 } ] }, "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" } }