| { |
| "run_info": { |
| "created_at": "2026-01-10T09:01:28+00:00", |
| "total_time": 1209.8416560290498, |
| "experiment_name": "road/llama-3.2-3B-lr_0.001", |
| "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": 0.001 |
| }, |
| "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": null, |
| "peft_type": "ROAD", |
| "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, |
| "variant": "road_2", |
| "group_size": 64, |
| "init_weights": true, |
| "target_modules": [ |
| "v_proj", |
| "q_proj" |
| ], |
| "modules_to_save": null |
| }, |
| "error_msg": "" |
| }, |
| "train_info": { |
| "accelerator_memory_reserved_avg": 14529478839, |
| "accelerator_memory_max": 22806528000, |
| "accelerator_memory_reserved_99th": 20495466496, |
| "train_time": 986.1049144260469, |
| "file_size": 931480, |
| "num_trainable_params": 229376, |
| "num_total_params": 3212979200, |
| "status": "success", |
| "metrics": [ |
| { |
| "step": 250, |
| "valid accuracy": 0.22, |
| "train loss": 1.1801395919322968, |
| "train samples": 1000, |
| "train time": 30.463671060162596, |
| "eval time": 12.11280647298554, |
| "tokens / sec": 6949.88465381854, |
| "mem allocated avg": 6785091975.168, |
| "mem reserved avg": 14651240742.912, |
| "elapsed time": 66.61390974500682 |
| }, |
| { |
| "step": 500, |
| "valid accuracy": 0.3, |
| "train loss": 0.8728977775573731, |
| "train samples": 2000, |
| "train time": 30.350360362324864, |
| "eval time": 12.122786874999292, |
| "tokens / sec": 6853.1311495791215, |
| "mem allocated avg": 6777472542.72, |
| "mem reserved avg": 14239108431.872, |
| "elapsed time": 112.51008765597362 |
| }, |
| { |
| "step": 750, |
| "valid accuracy": 0.42, |
| "train loss": 0.7738239982128143, |
| "train samples": 3000, |
| "train time": 31.060686238168273, |
| "eval time": 12.105645439994987, |
| "tokens / sec": 6902.648523474598, |
| "mem allocated avg": 6787427880.96, |
| "mem reserved avg": 14477940490.24, |
| "elapsed time": 159.04484326101374 |
| }, |
| { |
| "step": 1000, |
| "valid accuracy": 0.36, |
| "train loss": 0.7328966956138611, |
| "train samples": 4000, |
| "train time": 31.1341793507454, |
| "eval time": 12.271532059996389, |
| "tokens / sec": 6691.552639077739, |
| "mem allocated avg": 6779240853.504, |
| "mem reserved avg": 14547599491.072, |
| "elapsed time": 205.96491558500566 |
| }, |
| { |
| "step": 1250, |
| "valid accuracy": 0.32, |
| "train loss": 0.7226412436962127, |
| "train samples": 5000, |
| "train time": 30.673152873758227, |
| "eval time": 12.08470779901836, |
| "tokens / sec": 6798.714199948135, |
| "mem allocated avg": 6779142604.8, |
| "mem reserved avg": 14559075106.816, |
| "elapsed time": 252.09245199902216 |
| }, |
| { |
| "step": 1500, |
| "valid accuracy": 0.38, |
| "train loss": 0.7143599749803543, |
| "train samples": 6000, |
| "train time": 31.072990959743038, |
| "eval time": 12.123655039002188, |
| "tokens / sec": 6736.750905994249, |
| "mem allocated avg": 6781227952.128, |
| "mem reserved avg": 14496538034.176, |
| "elapsed time": 298.79225073900307 |
| }, |
| { |
| "step": 1750, |
| "valid accuracy": 0.38, |
| "train loss": 0.7038573758602142, |
| "train samples": 7000, |
| "train time": 30.712441952317022, |
| "eval time": 12.087091822992079, |
| "tokens / sec": 6816.618500249399, |
| "mem allocated avg": 6782587774.976, |
| "mem reserved avg": 14888504131.584, |
| "elapsed time": 345.02615644899197 |
| }, |
| { |
| "step": 2000, |
| "valid accuracy": 0.3, |
| "train loss": 0.7051438037157058, |
| "train samples": 8000, |
| "train time": 30.437189053802285, |
| "eval time": 12.269888009992428, |
| "tokens / sec": 6823.757595777529, |
| "mem allocated avg": 6778266038.272, |
| "mem reserved avg": 14524782477.312, |
| "elapsed time": 391.30238440597896 |
| }, |
| { |
| "step": 2250, |
| "valid accuracy": 0.4, |
| "train loss": 0.697867576956749, |
| "train samples": 9000, |
| "train time": 31.4548720899038, |
| "eval time": 12.078225647972431, |
| "tokens / sec": 6833.535974511012, |
| "mem allocated avg": 6789062723.584, |
| "mem reserved avg": 14819180675.072, |
| "elapsed time": 438.25679710297845 |
| }, |
| { |
| "step": 2500, |
| "valid accuracy": 0.38, |
| "train loss": 0.6966045496463775, |
| "train samples": 10000, |
| "train time": 30.316960251017008, |
| "eval time": 12.1344860860263, |
| "tokens / sec": 6793.78797526677, |
| "mem allocated avg": 6775575060.48, |
| "mem reserved avg": 14285396770.816, |
| "elapsed time": 484.1519402990234 |
| }, |
| { |
| "step": 2750, |
| "valid accuracy": 0.4, |
| "train loss": 0.6900997126102447, |
| "train samples": 11000, |
| "train time": 31.133579084766097, |
| "eval time": 12.133857114997227, |
| "tokens / sec": 6805.545852056406, |
| "mem allocated avg": 6784651790.336, |
| "mem reserved avg": 14683520106.496, |
| "elapsed time": 530.7991969200084 |
| }, |
| { |
| "step": 3000, |
| "valid accuracy": 0.38, |
| "train loss": 0.6820800434350968, |
| "train samples": 12000, |
| "train time": 31.004261121852323, |
| "eval time": 6.647833172988612, |
| "tokens / sec": 6732.332668069386, |
| "mem allocated avg": 6781542193.152, |
| "mem reserved avg": 14604935626.752, |
| "elapsed time": 571.967426163028 |
| }, |
| { |
| "step": 3250, |
| "valid accuracy": 0.4, |
| "train loss": 0.6917668293714523, |
| "train samples": 13000, |
| "train time": 30.61097677750513, |
| "eval time": 8.696813674003351, |
| "tokens / sec": 6889.718075085513, |
| "mem allocated avg": 6783328645.12, |
| "mem reserved avg": 14447456288.768, |
| "elapsed time": 614.5629294660175 |
| }, |
| { |
| "step": 3500, |
| "valid accuracy": 0.44, |
| "train loss": 0.6792756502628327, |
| "train samples": 14000, |
| "train time": 30.631839248526376, |
| "eval time": 8.344476633996237, |
| "tokens / sec": 6847.450402773009, |
| "mem allocated avg": 6780746895.36, |
| "mem reserved avg": 14468931125.248, |
| "elapsed time": 657.0185567580047 |
| }, |
| { |
| "step": 3750, |
| "valid accuracy": 0.4, |
| "train loss": 0.6768034971952438, |
| "train samples": 15000, |
| "train time": 31.22799357509939, |
| "eval time": 7.472214682027698, |
| "tokens / sec": 6939.3827521725525, |
| "mem allocated avg": 6791758675.968, |
| "mem reserved avg": 14830622736.384, |
| "elapsed time": 699.1621996440226 |
| }, |
| { |
| "step": 4000, |
| "valid accuracy": 0.4, |
| "train loss": 0.693776785492897, |
| "train samples": 16000, |
| "train time": 30.5784999235766, |
| "eval time": 7.540951641974971, |
| "tokens / sec": 6683.552185711522, |
| "mem allocated avg": 6772682035.2, |
| "mem reserved avg": 14486186491.904, |
| "elapsed time": 740.7859192459728 |
| }, |
| { |
| "step": 4250, |
| "valid accuracy": 0.42, |
| "train loss": 0.674255707025528, |
| "train samples": 17000, |
| "train time": 31.011587454471737, |
| "eval time": 12.230024265998509, |
| "tokens / sec": 6816.452086186856, |
| "mem allocated avg": 6784054867.968, |
| "mem reserved avg": 14463939903.488, |
| "elapsed time": 787.3081514210207 |
| }, |
| { |
| "step": 4500, |
| "valid accuracy": 0.44, |
| "train loss": 0.6836280490159988, |
| "train samples": 18000, |
| "train time": 30.32553677726537, |
| "eval time": 12.078615510021336, |
| "tokens / sec": 6852.904254469726, |
| "mem allocated avg": 6778766051.328, |
| "mem reserved avg": 14435452190.72, |
| "elapsed time": 833.2261762300041 |
| }, |
| { |
| "step": 4750, |
| "valid accuracy": 0.4, |
| "train loss": 0.676575911641121, |
| "train samples": 19000, |
| "train time": 30.607126664719544, |
| "eval time": 12.086251140979584, |
| "tokens / sec": 6859.154153859013, |
| "mem allocated avg": 6781519785.984, |
| "mem reserved avg": 14521225707.52, |
| "elapsed time": 879.2816601090017 |
| }, |
| { |
| "step": 5000, |
| "valid accuracy": 0.42, |
| "train loss": 0.6835739253759384, |
| "train samples": 20000, |
| "train time": 30.988149417680688, |
| "eval time": 12.111999842978548, |
| "tokens / sec": 6721.279066802329, |
| "mem allocated avg": 6777674483.712, |
| "mem reserved avg": 14157940260.864, |
| "elapsed time": 925.7869706979836 |
| }, |
| { |
| "step": 5000, |
| "test accuracy": 0.3866565579984837, |
| "train loss": 0.6835739253759384, |
| "train samples": 20000, |
| "train total tokens": 4198051, |
| "forgetting": 0.1559743881225586 |
| } |
| ] |
| }, |
| "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" |
| } |
| } |
