| { | |
| "run_info": { | |
| "created_at": "2025-06-19T21:59:33+00:00", | |
| "total_time": 2004.8640038169979, | |
| "experiment_name": "ia3/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": null, | |
| "peft_type": "IA3", | |
| "auto_mapping": null, | |
| "base_model_name_or_path": "meta-llama/Llama-3.2-3B", | |
| "revision": null, | |
| "inference_mode": false, | |
| "target_modules": [ | |
| "down_proj", | |
| "v_proj", | |
| "k_proj" | |
| ], | |
| "exclude_modules": null, | |
| "feedforward_modules": [ | |
| "down_proj" | |
| ], | |
| "fan_in_fan_out": false, | |
| "modules_to_save": null, | |
| "init_ia3_weights": true | |
| }, | |
| "error_msg": "" | |
| }, | |
| "train_info": { | |
| "accelerator_memory_reserved_avg": 12023227429, | |
| "accelerator_memory_max": 23137878016, | |
| "accelerator_memory_reserved_99th": 18398566154, | |
| "train_time": 1782.9318781230104, | |
| "file_size": 1157064, | |
| "num_trainable_params": 286720, | |
| "num_total_params": 3213036544, | |
| "status": "success", | |
| "metrics": [ | |
| { | |
| "step": 250, | |
| "valid accuracy": 0.0, | |
| "train loss": 1.3155810165405273, | |
| "train samples": 1000, | |
| "train time": 30.56459548201383, | |
| "eval time": 10.972947114001727, | |
| "tokens / sec": 6926.936105684404, | |
| "mem allocated avg": 6780994971.648, | |
| "mem reserved avg": 12076433014.784, | |
| "elapsed time": 90.53726772200025 | |
| }, | |
| { | |
| "step": 500, | |
| "valid accuracy": 0.0, | |
| "train loss": 1.205229633808136, | |
| "train samples": 2000, | |
| "train time": 30.221456803970796, | |
| "eval time": 10.954313254995213, | |
| "tokens / sec": 6882.361805029583, | |
| "mem allocated avg": 6773721065.472, | |
| "mem reserved avg": 11963673346.048, | |
| "elapsed time": 175.07058417100052 | |
| }, | |
| { | |
| "step": 750, | |
| "valid accuracy": 0.1, | |
| "train loss": 1.0194582087993622, | |
| "train samples": 3000, | |
| "train time": 30.774312397006724, | |
| "eval time": 10.944943730006344, | |
| "tokens / sec": 6966.881899231445, | |
| "mem allocated avg": 6784231882.752, | |
| "mem reserved avg": 12126680776.704, | |
| "elapsed time": 260.540154495 | |
| }, | |
| { | |
| "step": 1000, | |
| "valid accuracy": 0.24, | |
| "train loss": 0.9196457831859589, | |
| "train samples": 4000, | |
| "train time": 30.61534244806535, | |
| "eval time": 10.960088267995161, | |
| "tokens / sec": 6804.95409624808, | |
| "mem allocated avg": 6775492155.392, | |
| "mem reserved avg": 11986893012.992, | |
| "elapsed time": 345.30987053900026 | |
| }, | |
| { | |
| "step": 1250, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.8685842225551605, | |
| "train samples": 5000, | |
| "train time": 29.97266351111466, | |
| "eval time": 10.924794500999269, | |
| "tokens / sec": 6957.606551138459, | |
| "mem allocated avg": 6775089207.296, | |
| "mem reserved avg": 11983428517.888, | |
| "elapsed time": 429.5542291879974 | |
| }, | |
| { | |
| "step": 1500, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.8332846148014068, | |
| "train samples": 6000, | |
| "train time": 29.98314001694962, | |
| "eval time": 10.942266878999362, | |
| "tokens / sec": 6981.6236685572, | |
| "mem allocated avg": 6776724867.072, | |
| "mem reserved avg": 12008594341.888, | |
| "elapsed time": 513.8152235820016 | |
| }, | |
| { | |
| "step": 1750, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.8169269208908081, | |
| "train samples": 7000, | |
| "train time": 30.245623568014707, | |
| "eval time": 10.940915298000618, | |
| "tokens / sec": 6921.8278647558345, | |
| "mem allocated avg": 6777912934.4, | |
| "mem reserved avg": 12032065667.072, | |
| "elapsed time": 598.2868188970024 | |
| }, | |
| { | |
| "step": 2000, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.8072074156999588, | |
| "train samples": 8000, | |
| "train time": 30.292844633964705, | |
| "eval time": 10.95617212200159, | |
| "tokens / sec": 6856.272578875894, | |
| "mem allocated avg": 6775099170.816, | |
| "mem reserved avg": 11967473385.472, | |
| "elapsed time": 682.7948923380027 | |
| }, | |
| { | |
| "step": 2250, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.7952859619855881, | |
| "train samples": 9000, | |
| "train time": 31.20892413101683, | |
| "eval time": 10.942549917002907, | |
| "tokens / sec": 6887.388975590319, | |
| "mem allocated avg": 6786161477.632, | |
| "mem reserved avg": 12167709458.432, | |
| "elapsed time": 768.9645714229991 | |
| }, | |
| { | |
| "step": 2500, | |
| "valid accuracy": 0.28, | |
| "train loss": 0.7890167078971863, | |
| "train samples": 10000, | |
| "train time": 30.187670495011844, | |
| "eval time": 10.954304017002869, | |
| "tokens / sec": 6822.884860692832, | |
| "mem allocated avg": 6771082014.72, | |
| "mem reserved avg": 11910984499.2, | |
| "elapsed time": 853.427360558002 | |
| }, | |
| { | |
| "step": 2750, | |
| "valid accuracy": 0.3, | |
| "train loss": 0.7823473591804504, | |
| "train samples": 11000, | |
| "train time": 30.410061570059042, | |
| "eval time": 10.93302121299348, | |
| "tokens / sec": 6967.4636965751015, | |
| "mem allocated avg": 6782254225.408, | |
| "mem reserved avg": 12090903363.584, | |
| "elapsed time": 938.3584665200033 | |
| }, | |
| { | |
| "step": 3000, | |
| "valid accuracy": 0.24, | |
| "train loss": 0.7709820411205291, | |
| "train samples": 12000, | |
| "train time": 30.02989622000314, | |
| "eval time": 10.940404225999373, | |
| "tokens / sec": 6950.773271769175, | |
| "mem allocated avg": 6776725577.728, | |
| "mem reserved avg": 12003133358.08, | |
| "elapsed time": 1022.4627897890023 | |
| }, | |
| { | |
| "step": 3250, | |
| "valid accuracy": 0.3, | |
| "train loss": 0.7755767168998718, | |
| "train samples": 13000, | |
| "train time": 30.172652364024543, | |
| "eval time": 10.940153044000908, | |
| "tokens / sec": 6989.806446431653, | |
| "mem allocated avg": 6778589339.648, | |
| "mem reserved avg": 12038298402.816, | |
| "elapsed time": 1107.0076802080002 | |
| }, | |
| { | |
| "step": 3500, | |
| "valid accuracy": 0.34, | |
| "train loss": 0.7658302361965179, | |
| "train samples": 14000, | |
| "train time": 30.384311634006735, | |
| "eval time": 10.941136569999799, | |
| "tokens / sec": 6903.233567590308, | |
| "mem allocated avg": 6777534660.608, | |
| "mem reserved avg": 12020623605.76, | |
| "elapsed time": 1191.893303306002 | |
| }, | |
| { | |
| "step": 3750, | |
| "valid accuracy": 0.34, | |
| "train loss": 0.7585167481899261, | |
| "train samples": 15000, | |
| "train time": 31.250990667955193, | |
| "eval time": 10.924158087997057, | |
| "tokens / sec": 6934.276173913666, | |
| "mem allocated avg": 6788426940.416, | |
| "mem reserved avg": 12209652498.432, | |
| "elapsed time": 1278.4574160839984 | |
| }, | |
| { | |
| "step": 4000, | |
| "valid accuracy": 0.26, | |
| "train loss": 0.7766438691616059, | |
| "train samples": 16000, | |
| "train time": 30.222231689898763, | |
| "eval time": 10.98030305699649, | |
| "tokens / sec": 6762.339793335249, | |
| "mem allocated avg": 6769563977.728, | |
| "mem reserved avg": 11885533462.528, | |
| "elapsed time": 1362.9405450319973 | |
| }, | |
| { | |
| "step": 4250, | |
| "valid accuracy": 0.34, | |
| "train loss": 0.7542061095237732, | |
| "train samples": 17000, | |
| "train time": 30.273203028933494, | |
| "eval time": 10.948997009996674, | |
| "tokens / sec": 6982.710081849145, | |
| "mem allocated avg": 6780103426.048, | |
| "mem reserved avg": 12047483928.576, | |
| "elapsed time": 1447.661586811002 | |
| }, | |
| { | |
| "step": 4500, | |
| "valid accuracy": 0.32, | |
| "train loss": 0.7659628703594208, | |
| "train samples": 18000, | |
| "train time": 29.84466753601737, | |
| "eval time": 10.942651322002348, | |
| "tokens / sec": 6963.320993581165, | |
| "mem allocated avg": 6775043430.4, | |
| "mem reserved avg": 11968387743.744, | |
| "elapsed time": 1531.5572027719973 | |
| }, | |
| { | |
| "step": 4750, | |
| "valid accuracy": 0.28, | |
| "train loss": 0.7580052223205567, | |
| "train samples": 19000, | |
| "train time": 30.03731635398435, | |
| "eval time": 10.927273799999966, | |
| "tokens / sec": 6989.272860661278, | |
| "mem allocated avg": 6776962899.968, | |
| "mem reserved avg": 12017695981.568, | |
| "elapsed time": 1615.9832882379997 | |
| }, | |
| { | |
| "step": 5000, | |
| "valid accuracy": 0.36, | |
| "train loss": 0.7657463653087616, | |
| "train samples": 20000, | |
| "train time": 30.07570726004633, | |
| "eval time": 10.953207714999735, | |
| "tokens / sec": 6925.19042691597, | |
| "mem allocated avg": 6774270615.552, | |
| "mem reserved avg": 11958900228.096, | |
| "elapsed time": 1700.4354192270039 | |
| }, | |
| { | |
| "step": 5000, | |
| "test accuracy": 0.34495830174374525, | |
| "train loss": 0.7657463653087616, | |
| "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" | |
| } | |
| } | |