返回模型
说明文档
Llama-2 ONNX
本仓库包含 Llama-2 13B 的优化版本。
下载模型
你可以使用 huggingface_hub 下载本仓库。可以通过 Python 脚本和命令行两种方式完成。Python 示例请参考 HuggingFace Hub 文档。
使用命令行:
- 确保已安装最新版本的
huggingface_hub。
pip install -U huggingface_hub
- 下载仓库。
huggingface-cli download alpindale/Llama-2-13b-ONNX --repo-type model --cache-dir /path/to/custom/cache/directory --local-dir /path/to/download/dir --local-dir-use-symlinks False
--cache-dir 参数仅在你的默认缓存目录(~/.cache)没有足够磁盘空间容纳整个仓库时才需要使用。
使用模型
本仓库提供了运行模型的示例代码。
python llama2_onnx_inference.py --onnx_file FP16/LlamaV2_13B_float16.onnx --embedding_file embeddings.pth --tokenizer_path tokenizer.model --prompt \"What is the lightest element?\"
输出:
The lightest element is hydrogen. Hydrogen is the lightest element on the periodic table, with an atomic mass of 1.00794 u (unified atomic mass units).
聊天界面
或者,你可以使用 Gradio 聊天界面来运行模型。
首先,安装所需的包:
pip install -r requirements.txt
将 Python 路径设置为仓库的根目录(导入所需模块时需要):
export PYTHONPATH=$PYTHONPATH:$(pwd)
然后只需运行:
python ChatApp/app.py
然后你可以在浏览器中访问 http://localhost:7860 来使用该界面。
常见问题
为什么第一次推理会话很慢?
ONNX 运行时执行提供程序可能需要为底层硬件生成 JIT 二进制文件,通常该二进制文件会被缓存,并在后续运行中直接加载以减少开销。
为什么在我的设备上 FP16 比 FP32 慢?
你的设备可能不支持原生 FP16 运算,因此权重会在运行时转换为 FP32。使用 FP32 版本的模型可以避免转换开销。
如何优化推理?
建议将输入/输出放在目标设备上以避免昂贵的数据拷贝,详情请参考以下文档:
我应该使用什么生成参数来运行模型?
你可以使用提供的示例代码进行温度(temperature)和 top-p 采样。请参考 Meta 的示例 这里。
alpindale/Llama-2-13b-ONNX
作者 alpindale
text-generation
↓ 0
♥ 5
创建时间: 2023-09-21 20:28:10+00:00
更新时间: 2023-09-22 18:33:20+00:00
在 Hugging Face 上查看文件 (1491)
.gitattributes
ChatApp/app.py
ChatApp/app_modules/__pycache__/overwrites.cpython-39.pyc
ChatApp/app_modules/__pycache__/presets.cpython-39.pyc
ChatApp/app_modules/__pycache__/utils.cpython-39.pyc
ChatApp/app_modules/overwrites.py
ChatApp/app_modules/presets.py
ChatApp/app_modules/utils.py
ChatApp/assets/custom.css
ChatApp/assets/custom.js
ChatApp/interface/__pycache__/base_interface.cpython-39.pyc
ChatApp/interface/__pycache__/empty_stub_interface.cpython-39.pyc
ChatApp/interface/__pycache__/hddr_llama_onnx_interface.cpython-39.pyc
ChatApp/interface/base_interface.py
ChatApp/interface/empty_stub_interface.py
ChatApp/interface/hddr_llama_onnx_interface.py
ChatApp/requirements.txt
FP16-Chat/.gitattributes
FP16-Chat/.gitignore
FP16-Chat/Constant_69_attr__value
FP16-Chat/Constant_77_attr__value
FP16-Chat/LlamaV2_13B_FT_float16.onnx
ONNX
FP16-Chat/embeddings.data
FP16-Chat/onnx__MatMul_25830
FP16-Chat/onnx__MatMul_25831
FP16-Chat/onnx__MatMul_25832
FP16-Chat/onnx__MatMul_25852
FP16-Chat/onnx__MatMul_25853
FP16-Chat/onnx__MatMul_25854
FP16-Chat/onnx__MatMul_25855
FP16-Chat/onnx__MatMul_25856
FP16-Chat/onnx__MatMul_25857
FP16-Chat/onnx__MatMul_25858
FP16-Chat/onnx__MatMul_25878
FP16-Chat/onnx__MatMul_25879
FP16-Chat/onnx__MatMul_25880
FP16-Chat/onnx__MatMul_25881
FP16-Chat/onnx__MatMul_25882
FP16-Chat/onnx__MatMul_25883
FP16-Chat/onnx__MatMul_25884
FP16-Chat/onnx__MatMul_25904
FP16-Chat/onnx__MatMul_25905
FP16-Chat/onnx__MatMul_25906
FP16-Chat/onnx__MatMul_25907
FP16-Chat/onnx__MatMul_25908
FP16-Chat/onnx__MatMul_25909
FP16-Chat/onnx__MatMul_25910
FP16-Chat/onnx__MatMul_25930
FP16-Chat/onnx__MatMul_25931
FP16-Chat/onnx__MatMul_25932
FP16-Chat/onnx__MatMul_25933
FP16-Chat/onnx__MatMul_25934
FP16-Chat/onnx__MatMul_25935
FP16-Chat/onnx__MatMul_25936
FP16-Chat/onnx__MatMul_25956
FP16-Chat/onnx__MatMul_25957
FP16-Chat/onnx__MatMul_25958
FP16-Chat/onnx__MatMul_25959
FP16-Chat/onnx__MatMul_25960
FP16-Chat/onnx__MatMul_25961
FP16-Chat/onnx__MatMul_25962
FP16-Chat/onnx__MatMul_25982
FP16-Chat/onnx__MatMul_25983
FP16-Chat/onnx__MatMul_25984
FP16-Chat/onnx__MatMul_25985
FP16-Chat/onnx__MatMul_25986
FP16-Chat/onnx__MatMul_25987
FP16-Chat/onnx__MatMul_25988
FP16-Chat/onnx__MatMul_26008
FP16-Chat/onnx__MatMul_26009
FP16-Chat/onnx__MatMul_26010
FP16-Chat/onnx__MatMul_26011
FP16-Chat/onnx__MatMul_26012
FP16-Chat/onnx__MatMul_26013
FP16-Chat/onnx__MatMul_26014
FP16-Chat/onnx__MatMul_26034
FP16-Chat/onnx__MatMul_26035
FP16-Chat/onnx__MatMul_26036
FP16-Chat/onnx__MatMul_26037
FP16-Chat/onnx__MatMul_26038
FP16-Chat/onnx__MatMul_26039
FP16-Chat/onnx__MatMul_26040
FP16-Chat/onnx__MatMul_26060
FP16-Chat/onnx__MatMul_26061
FP16-Chat/onnx__MatMul_26062
FP16-Chat/onnx__MatMul_26063
FP16-Chat/onnx__MatMul_26064
FP16-Chat/onnx__MatMul_26065
FP16-Chat/onnx__MatMul_26066
FP16-Chat/onnx__MatMul_26086
FP16-Chat/onnx__MatMul_26087
FP16-Chat/onnx__MatMul_26088
FP16-Chat/onnx__MatMul_26089
FP16-Chat/onnx__MatMul_26090
FP16-Chat/onnx__MatMul_26091
FP16-Chat/onnx__MatMul_26092
FP16-Chat/onnx__MatMul_26112
FP16-Chat/onnx__MatMul_26113
FP16-Chat/onnx__MatMul_26114
FP16-Chat/onnx__MatMul_26115
FP16-Chat/onnx__MatMul_26116
FP16-Chat/onnx__MatMul_26117
FP16-Chat/onnx__MatMul_26118
FP16-Chat/onnx__MatMul_26138
FP16-Chat/onnx__MatMul_26139
FP16-Chat/onnx__MatMul_26140
FP16-Chat/onnx__MatMul_26141
FP16-Chat/onnx__MatMul_26142
FP16-Chat/onnx__MatMul_26143
FP16-Chat/onnx__MatMul_26144
FP16-Chat/onnx__MatMul_26164
FP16-Chat/onnx__MatMul_26165
FP16-Chat/onnx__MatMul_26166
FP16-Chat/onnx__MatMul_26167
FP16-Chat/onnx__MatMul_26168
FP16-Chat/onnx__MatMul_26169
FP16-Chat/onnx__MatMul_26170
FP16-Chat/onnx__MatMul_26190
FP16-Chat/onnx__MatMul_26191
FP16-Chat/onnx__MatMul_26192
FP16-Chat/onnx__MatMul_26193
FP16-Chat/onnx__MatMul_26194
FP16-Chat/onnx__MatMul_26195
FP16-Chat/onnx__MatMul_26196
FP16-Chat/onnx__MatMul_26216
FP16-Chat/onnx__MatMul_26217
FP16-Chat/onnx__MatMul_26218
FP16-Chat/onnx__MatMul_26219
FP16-Chat/onnx__MatMul_26220
FP16-Chat/onnx__MatMul_26221
FP16-Chat/onnx__MatMul_26222
FP16-Chat/onnx__MatMul_26242
FP16-Chat/onnx__MatMul_26243
FP16-Chat/onnx__MatMul_26244
FP16-Chat/onnx__MatMul_26245
FP16-Chat/onnx__MatMul_26246
FP16-Chat/onnx__MatMul_26247
FP16-Chat/onnx__MatMul_26248
FP16-Chat/onnx__MatMul_26268
FP16-Chat/onnx__MatMul_26269
FP16-Chat/onnx__MatMul_26270
FP16-Chat/onnx__MatMul_26271
FP16-Chat/onnx__MatMul_26272
FP16-Chat/onnx__MatMul_26273
FP16-Chat/onnx__MatMul_26274
FP16-Chat/onnx__MatMul_26294
FP16-Chat/onnx__MatMul_26295
FP16-Chat/onnx__MatMul_26296
FP16-Chat/onnx__MatMul_26297
FP16-Chat/onnx__MatMul_26298
FP16-Chat/onnx__MatMul_26299
FP16-Chat/onnx__MatMul_26300
FP16-Chat/onnx__MatMul_26320
FP16-Chat/onnx__MatMul_26321
FP16-Chat/onnx__MatMul_26322
FP16-Chat/onnx__MatMul_26323
FP16-Chat/onnx__MatMul_26324
FP16-Chat/onnx__MatMul_26325
FP16-Chat/onnx__MatMul_26326
FP16-Chat/onnx__MatMul_26346
FP16-Chat/onnx__MatMul_26347
FP16-Chat/onnx__MatMul_26348
FP16-Chat/onnx__MatMul_26349
FP16-Chat/onnx__MatMul_26350
FP16-Chat/onnx__MatMul_26351
FP16-Chat/onnx__MatMul_26352
FP16-Chat/onnx__MatMul_26372
FP16-Chat/onnx__MatMul_26373
FP16-Chat/onnx__MatMul_26374
FP16-Chat/onnx__MatMul_26375
FP16-Chat/onnx__MatMul_26376
FP16-Chat/onnx__MatMul_26377
FP16-Chat/onnx__MatMul_26378
FP16-Chat/onnx__MatMul_26398
FP16-Chat/onnx__MatMul_26399
FP16-Chat/onnx__MatMul_26400
FP16-Chat/onnx__MatMul_26401
FP16-Chat/onnx__MatMul_26402
FP16-Chat/onnx__MatMul_26403
FP16-Chat/onnx__MatMul_26404
FP16-Chat/onnx__MatMul_26424
FP16-Chat/onnx__MatMul_26425
FP16-Chat/onnx__MatMul_26426
FP16-Chat/onnx__MatMul_26427
FP16-Chat/onnx__MatMul_26428
FP16-Chat/onnx__MatMul_26429
FP16-Chat/onnx__MatMul_26430
FP16-Chat/onnx__MatMul_26450
FP16-Chat/onnx__MatMul_26451
FP16-Chat/onnx__MatMul_26452
FP16-Chat/onnx__MatMul_26453
FP16-Chat/onnx__MatMul_26454
FP16-Chat/onnx__MatMul_26455
FP16-Chat/onnx__MatMul_26456
FP16-Chat/onnx__MatMul_26476
FP16-Chat/onnx__MatMul_26477
FP16-Chat/onnx__MatMul_26478
FP16-Chat/onnx__MatMul_26479
FP16-Chat/onnx__MatMul_26480
FP16-Chat/onnx__MatMul_26481
FP16-Chat/onnx__MatMul_26482
FP16-Chat/onnx__MatMul_26502
FP16-Chat/onnx__MatMul_26503
FP16-Chat/onnx__MatMul_26504
FP16-Chat/onnx__MatMul_26505
FP16-Chat/onnx__MatMul_26506
FP16-Chat/onnx__MatMul_26507
FP16-Chat/onnx__MatMul_26508
FP16-Chat/onnx__MatMul_26528
FP16-Chat/onnx__MatMul_26529
FP16-Chat/onnx__MatMul_26530
FP16-Chat/onnx__MatMul_26531
FP16-Chat/onnx__MatMul_26532
FP16-Chat/onnx__MatMul_26533
FP16-Chat/onnx__MatMul_26534
FP16-Chat/onnx__MatMul_26554
FP16-Chat/onnx__MatMul_26555
FP16-Chat/onnx__MatMul_26556
FP16-Chat/onnx__MatMul_26557
FP16-Chat/onnx__MatMul_26558
FP16-Chat/onnx__MatMul_26559
FP16-Chat/onnx__MatMul_26560
FP16-Chat/onnx__MatMul_26580
FP16-Chat/onnx__MatMul_26581
FP16-Chat/onnx__MatMul_26582
FP16-Chat/onnx__MatMul_26583
FP16-Chat/onnx__MatMul_26584
FP16-Chat/onnx__MatMul_26585
FP16-Chat/onnx__MatMul_26586
FP16-Chat/onnx__MatMul_26606
FP16-Chat/onnx__MatMul_26607
FP16-Chat/onnx__MatMul_26608
FP16-Chat/onnx__MatMul_26609
FP16-Chat/onnx__MatMul_26610
FP16-Chat/onnx__MatMul_26611
FP16-Chat/onnx__MatMul_26612
FP16-Chat/onnx__MatMul_26632
FP16-Chat/onnx__MatMul_26633
FP16-Chat/onnx__MatMul_26634
FP16-Chat/onnx__MatMul_26635
FP16-Chat/onnx__MatMul_26636
FP16-Chat/onnx__MatMul_26637
FP16-Chat/onnx__MatMul_26638
FP16-Chat/onnx__MatMul_26658
FP16-Chat/onnx__MatMul_26659
FP16-Chat/onnx__MatMul_26660
FP16-Chat/onnx__MatMul_26661
FP16-Chat/onnx__MatMul_26662
FP16-Chat/onnx__MatMul_26663
FP16-Chat/onnx__MatMul_26664
FP16-Chat/onnx__MatMul_26684
FP16-Chat/onnx__MatMul_26685
FP16-Chat/onnx__MatMul_26686
FP16-Chat/onnx__MatMul_26687
FP16-Chat/onnx__MatMul_26688
FP16-Chat/onnx__MatMul_26689
FP16-Chat/onnx__MatMul_26690
FP16-Chat/onnx__MatMul_26710
FP16-Chat/onnx__MatMul_26711
FP16-Chat/onnx__MatMul_26712
FP16-Chat/onnx__MatMul_26713
FP16-Chat/onnx__MatMul_26714
FP16-Chat/onnx__MatMul_26715
FP16-Chat/onnx__MatMul_26716
FP16-Chat/onnx__MatMul_26736
FP16-Chat/onnx__MatMul_26737
FP16-Chat/onnx__MatMul_26738
FP16-Chat/onnx__MatMul_26739
FP16-Chat/onnx__MatMul_26740
FP16-Chat/onnx__MatMul_26741
FP16-Chat/onnx__MatMul_26742
FP16-Chat/onnx__MatMul_26762
FP16-Chat/onnx__MatMul_26763
FP16-Chat/onnx__MatMul_26764
FP16-Chat/onnx__MatMul_26765
FP16-Chat/onnx__MatMul_26766
FP16-Chat/onnx__MatMul_26767
FP16-Chat/onnx__MatMul_26768
FP16-Chat/onnx__MatMul_26788
FP16-Chat/onnx__MatMul_26789
FP16-Chat/onnx__MatMul_26790
FP16-Chat/onnx__MatMul_26791
FP16-Chat/onnx__MatMul_26792
FP16-Chat/onnx__MatMul_26793
FP16-Chat/onnx__MatMul_26794
FP16-Chat/onnx__MatMul_26814
FP16-Chat/onnx__MatMul_26815
FP16-Chat/onnx__MatMul_26816
FP16-Chat/onnx__MatMul_26817
FP16-Chat/onnx__MatMul_26818
FP16-Chat/onnx__MatMul_26819
FP16-Chat/onnx__MatMul_26820
FP16-Chat/onnx__MatMul_26840
FP16-Chat/onnx__MatMul_26841
FP16-Chat/onnx__MatMul_26842
FP16-Chat/onnx__MatMul_26843
FP16-Chat/onnx__MatMul_26844
FP16-Chat/onnx__MatMul_26845
FP16-Chat/onnx__MatMul_26846
FP16-Chat/onnx__MatMul_26866
FP16-Chat/onnx__MatMul_26867
FP16-Chat/onnx__MatMul_26868
FP16-Chat/onnx__MatMul_26869
FP16-Chat/onnx__MatMul_26870
FP16-Chat/transformer.block_list.0.attn_norm.weight
FP16-Chat/transformer.block_list.0.proj_norm.weight
FP16-Chat/transformer.block_list.1.attn_norm.weight
FP16-Chat/transformer.block_list.1.proj_norm.weight
FP16-Chat/transformer.block_list.10.attn_norm.weight
FP16-Chat/transformer.block_list.10.proj_norm.weight
FP16-Chat/transformer.block_list.11.attn_norm.weight
FP16-Chat/transformer.block_list.11.proj_norm.weight
FP16-Chat/transformer.block_list.12.attn_norm.weight
FP16-Chat/transformer.block_list.12.proj_norm.weight
FP16-Chat/transformer.block_list.13.attn_norm.weight
FP16-Chat/transformer.block_list.13.proj_norm.weight
FP16-Chat/transformer.block_list.14.attn_norm.weight
FP16-Chat/transformer.block_list.14.proj_norm.weight
FP16-Chat/transformer.block_list.15.attn_norm.weight
FP16-Chat/transformer.block_list.15.proj_norm.weight
FP16-Chat/transformer.block_list.16.attn_norm.weight
FP16-Chat/transformer.block_list.16.proj_norm.weight
FP16-Chat/transformer.block_list.17.attn_norm.weight
FP16-Chat/transformer.block_list.17.proj_norm.weight
FP16-Chat/transformer.block_list.18.attn_norm.weight
FP16-Chat/transformer.block_list.18.proj_norm.weight
FP16-Chat/transformer.block_list.19.attn_norm.weight
FP16-Chat/transformer.block_list.19.proj_norm.weight
FP16-Chat/transformer.block_list.2.attn_norm.weight
FP16-Chat/transformer.block_list.2.proj_norm.weight
FP16-Chat/transformer.block_list.20.attn_norm.weight
FP16-Chat/transformer.block_list.20.proj_norm.weight
FP16-Chat/transformer.block_list.21.attn_norm.weight
FP16-Chat/transformer.block_list.21.proj_norm.weight
FP16-Chat/transformer.block_list.22.attn_norm.weight
FP16-Chat/transformer.block_list.22.proj_norm.weight
FP16-Chat/transformer.block_list.23.attn_norm.weight
FP16-Chat/transformer.block_list.23.proj_norm.weight
FP16-Chat/transformer.block_list.24.attn_norm.weight
FP16-Chat/transformer.block_list.24.proj_norm.weight
FP16-Chat/transformer.block_list.25.attn_norm.weight
FP16-Chat/transformer.block_list.25.proj_norm.weight
FP16-Chat/transformer.block_list.26.attn_norm.weight
FP16-Chat/transformer.block_list.26.proj_norm.weight
FP16-Chat/transformer.block_list.27.attn_norm.weight
FP16-Chat/transformer.block_list.27.proj_norm.weight
FP16-Chat/transformer.block_list.28.attn_norm.weight
FP16-Chat/transformer.block_list.28.proj_norm.weight
FP16-Chat/transformer.block_list.29.attn_norm.weight
FP16-Chat/transformer.block_list.29.proj_norm.weight
FP16-Chat/transformer.block_list.3.attn_norm.weight
FP16-Chat/transformer.block_list.3.proj_norm.weight
FP16-Chat/transformer.block_list.30.attn_norm.weight
FP16-Chat/transformer.block_list.30.proj_norm.weight
FP16-Chat/transformer.block_list.31.attn_norm.weight
FP16-Chat/transformer.block_list.31.proj_norm.weight
FP16-Chat/transformer.block_list.32.attn_norm.weight
FP16-Chat/transformer.block_list.32.proj_norm.weight
FP16-Chat/transformer.block_list.33.attn_norm.weight
FP16-Chat/transformer.block_list.33.proj_norm.weight
FP16-Chat/transformer.block_list.34.attn_norm.weight
FP16-Chat/transformer.block_list.34.proj_norm.weight
FP16-Chat/transformer.block_list.35.attn_norm.weight
FP16-Chat/transformer.block_list.35.proj_norm.weight
FP16-Chat/transformer.block_list.36.attn_norm.weight
FP16-Chat/transformer.block_list.36.proj_norm.weight
FP16-Chat/transformer.block_list.37.attn_norm.weight
FP16-Chat/transformer.block_list.37.proj_norm.weight
FP16-Chat/transformer.block_list.38.attn_norm.weight
FP16-Chat/transformer.block_list.38.proj_norm.weight
FP16-Chat/transformer.block_list.39.attn_norm.weight
FP16-Chat/transformer.block_list.39.proj_norm.weight
FP16-Chat/transformer.block_list.4.attn_norm.weight
FP16-Chat/transformer.block_list.4.proj_norm.weight
FP16-Chat/transformer.block_list.5.attn_norm.weight
FP16-Chat/transformer.block_list.5.proj_norm.weight
FP16-Chat/transformer.block_list.6.attn_norm.weight
FP16-Chat/transformer.block_list.6.proj_norm.weight
FP16-Chat/transformer.block_list.7.attn_norm.weight
FP16-Chat/transformer.block_list.7.proj_norm.weight
FP16-Chat/transformer.block_list.8.attn_norm.weight
FP16-Chat/transformer.block_list.8.proj_norm.weight
FP16-Chat/transformer.block_list.9.attn_norm.weight
FP16-Chat/transformer.block_list.9.proj_norm.weight
FP16-Chat/transformer.layer_norm.weight
FP16/Constant_69_attr__value
FP16/Constant_77_attr__value
FP16/LlamaV2_13B_float16.onnx
ONNX
FP16/embeddings.data
FP16/onnx__MatMul_25830
FP16/onnx__MatMul_25831
FP16/onnx__MatMul_25832
FP16/onnx__MatMul_25852
FP16/onnx__MatMul_25853
FP16/onnx__MatMul_25854
FP16/onnx__MatMul_25855
FP16/onnx__MatMul_25856
FP16/onnx__MatMul_25857
FP16/onnx__MatMul_25858
FP16/onnx__MatMul_25878
FP16/onnx__MatMul_25879
FP16/onnx__MatMul_25880
FP16/onnx__MatMul_25881
FP16/onnx__MatMul_25882
FP16/onnx__MatMul_25883
FP16/onnx__MatMul_25884
FP16/onnx__MatMul_25904
FP16/onnx__MatMul_25905
FP16/onnx__MatMul_25906
FP16/onnx__MatMul_25907
FP16/onnx__MatMul_25908
FP16/onnx__MatMul_25909
FP16/onnx__MatMul_25910
FP16/onnx__MatMul_25930
FP16/onnx__MatMul_25931
FP16/onnx__MatMul_25932
FP16/onnx__MatMul_25933
FP16/onnx__MatMul_25934
FP16/onnx__MatMul_25935
FP16/onnx__MatMul_25936
FP16/onnx__MatMul_25956
FP16/onnx__MatMul_25957
FP16/onnx__MatMul_25958
FP16/onnx__MatMul_25959
FP16/onnx__MatMul_25960
FP16/onnx__MatMul_25961
FP16/onnx__MatMul_25962
FP16/onnx__MatMul_25982
FP16/onnx__MatMul_25983
FP16/onnx__MatMul_25984
FP16/onnx__MatMul_25985
FP16/onnx__MatMul_25986
FP16/onnx__MatMul_25987
FP16/onnx__MatMul_25988
FP16/onnx__MatMul_26008
FP16/onnx__MatMul_26009
FP16/onnx__MatMul_26010
FP16/onnx__MatMul_26011
FP16/onnx__MatMul_26012
FP16/onnx__MatMul_26013
FP16/onnx__MatMul_26014
FP16/onnx__MatMul_26034
FP16/onnx__MatMul_26035
FP16/onnx__MatMul_26036
FP16/onnx__MatMul_26037
FP16/onnx__MatMul_26038
FP16/onnx__MatMul_26039
FP16/onnx__MatMul_26040
FP16/onnx__MatMul_26060
FP16/onnx__MatMul_26061
FP16/onnx__MatMul_26062
FP16/onnx__MatMul_26063
FP16/onnx__MatMul_26064
FP16/onnx__MatMul_26065
FP16/onnx__MatMul_26066
FP16/onnx__MatMul_26086
FP16/onnx__MatMul_26087
FP16/onnx__MatMul_26088
FP16/onnx__MatMul_26089
FP16/onnx__MatMul_26090
FP16/onnx__MatMul_26091
FP16/onnx__MatMul_26092
FP16/onnx__MatMul_26112
FP16/onnx__MatMul_26113
FP16/onnx__MatMul_26114
FP16/onnx__MatMul_26115
FP16/onnx__MatMul_26116
FP16/onnx__MatMul_26117
FP16/onnx__MatMul_26118
FP16/onnx__MatMul_26138
FP16/onnx__MatMul_26139
FP16/onnx__MatMul_26140
FP16/onnx__MatMul_26141
FP16/onnx__MatMul_26142
FP16/onnx__MatMul_26143
FP16/onnx__MatMul_26144
FP16/onnx__MatMul_26164
FP16/onnx__MatMul_26165
FP16/onnx__MatMul_26166
FP16/onnx__MatMul_26167
FP16/onnx__MatMul_26168
FP16/onnx__MatMul_26169
FP16/onnx__MatMul_26170
FP16/onnx__MatMul_26190
FP16/onnx__MatMul_26191
FP16/onnx__MatMul_26192
FP16/onnx__MatMul_26193
FP16/onnx__MatMul_26194
FP16/onnx__MatMul_26195
FP16/onnx__MatMul_26196
FP16/onnx__MatMul_26216
FP16/onnx__MatMul_26217
FP16/onnx__MatMul_26218
FP16/onnx__MatMul_26219
FP16/onnx__MatMul_26220
FP16/onnx__MatMul_26221
FP16/onnx__MatMul_26222
FP16/onnx__MatMul_26242
FP16/onnx__MatMul_26243
FP16/onnx__MatMul_26244
FP16/onnx__MatMul_26245
FP16/onnx__MatMul_26246
FP16/onnx__MatMul_26247
FP16/onnx__MatMul_26248
FP16/onnx__MatMul_26268
FP16/onnx__MatMul_26269
FP16/onnx__MatMul_26270
FP16/onnx__MatMul_26271
FP16/onnx__MatMul_26272
FP16/onnx__MatMul_26273
FP16/onnx__MatMul_26274
FP16/onnx__MatMul_26294
FP16/onnx__MatMul_26295
FP16/onnx__MatMul_26296
FP16/onnx__MatMul_26297
FP16/onnx__MatMul_26298
FP16/onnx__MatMul_26299
FP16/onnx__MatMul_26300
FP16/onnx__MatMul_26320
FP16/onnx__MatMul_26321
FP16/onnx__MatMul_26322
FP16/onnx__MatMul_26323
FP16/onnx__MatMul_26324
FP16/onnx__MatMul_26325
FP16/onnx__MatMul_26326
FP16/onnx__MatMul_26346
FP16/onnx__MatMul_26347
FP16/onnx__MatMul_26348
FP16/onnx__MatMul_26349
FP16/onnx__MatMul_26350
FP16/onnx__MatMul_26351
FP16/onnx__MatMul_26352
FP16/onnx__MatMul_26372
FP16/onnx__MatMul_26373
FP16/onnx__MatMul_26374
FP16/onnx__MatMul_26375
FP16/onnx__MatMul_26376
FP16/onnx__MatMul_26377
FP16/onnx__MatMul_26378
FP16/onnx__MatMul_26398
FP16/onnx__MatMul_26399
FP16/onnx__MatMul_26400
FP16/onnx__MatMul_26401
FP16/onnx__MatMul_26402
FP16/onnx__MatMul_26403
FP16/onnx__MatMul_26404
FP16/onnx__MatMul_26424
FP16/onnx__MatMul_26425
FP16/onnx__MatMul_26426
FP16/onnx__MatMul_26427
FP16/onnx__MatMul_26428
FP16/onnx__MatMul_26429
FP16/onnx__MatMul_26430
FP16/onnx__MatMul_26450
FP16/onnx__MatMul_26451
FP16/onnx__MatMul_26452
FP16/onnx__MatMul_26453
FP16/onnx__MatMul_26454
FP16/onnx__MatMul_26455
FP16/onnx__MatMul_26456
FP16/onnx__MatMul_26476
FP16/onnx__MatMul_26477
FP16/onnx__MatMul_26478
FP16/onnx__MatMul_26479
FP16/onnx__MatMul_26480
FP16/onnx__MatMul_26481
FP16/onnx__MatMul_26482
FP16/onnx__MatMul_26502
FP16/onnx__MatMul_26503
FP16/onnx__MatMul_26504
FP16/onnx__MatMul_26505
FP16/onnx__MatMul_26506
FP16/onnx__MatMul_26507
FP16/onnx__MatMul_26508
FP16/onnx__MatMul_26528
FP16/onnx__MatMul_26529
FP16/onnx__MatMul_26530
FP16/onnx__MatMul_26531
FP16/onnx__MatMul_26532
FP16/onnx__MatMul_26533
FP16/onnx__MatMul_26534
FP16/onnx__MatMul_26554
FP16/onnx__MatMul_26555
FP16/onnx__MatMul_26556
FP16/onnx__MatMul_26557
FP16/onnx__MatMul_26558
FP16/onnx__MatMul_26559
FP16/onnx__MatMul_26560
FP16/onnx__MatMul_26580
FP16/onnx__MatMul_26581
FP16/onnx__MatMul_26582
FP16/onnx__MatMul_26583
FP16/onnx__MatMul_26584
FP16/onnx__MatMul_26585
FP16/onnx__MatMul_26586
FP16/onnx__MatMul_26606
FP16/onnx__MatMul_26607
FP16/onnx__MatMul_26608
FP16/onnx__MatMul_26609
FP16/onnx__MatMul_26610
FP16/onnx__MatMul_26611
FP16/onnx__MatMul_26612
FP16/onnx__MatMul_26632
FP16/onnx__MatMul_26633
FP16/onnx__MatMul_26634
FP16/onnx__MatMul_26635
FP16/onnx__MatMul_26636
FP16/onnx__MatMul_26637
FP16/onnx__MatMul_26638
FP16/onnx__MatMul_26658
FP16/onnx__MatMul_26659
FP16/onnx__MatMul_26660
FP16/onnx__MatMul_26661
FP16/onnx__MatMul_26662
FP16/onnx__MatMul_26663
FP16/onnx__MatMul_26664
FP16/onnx__MatMul_26684
FP16/onnx__MatMul_26685
FP16/onnx__MatMul_26686
FP16/onnx__MatMul_26687
FP16/onnx__MatMul_26688
FP16/onnx__MatMul_26689
FP16/onnx__MatMul_26690
FP16/onnx__MatMul_26710
FP16/onnx__MatMul_26711
FP16/onnx__MatMul_26712
FP16/onnx__MatMul_26713
FP16/onnx__MatMul_26714
FP16/onnx__MatMul_26715
FP16/onnx__MatMul_26716
FP16/onnx__MatMul_26736
FP16/onnx__MatMul_26737
FP16/onnx__MatMul_26738
FP16/onnx__MatMul_26739
FP16/onnx__MatMul_26740
FP16/onnx__MatMul_26741
FP16/onnx__MatMul_26742
FP16/onnx__MatMul_26762
FP16/onnx__MatMul_26763
FP16/onnx__MatMul_26764
FP16/onnx__MatMul_26765
FP16/onnx__MatMul_26766
FP16/onnx__MatMul_26767
FP16/onnx__MatMul_26768
FP16/onnx__MatMul_26788
FP16/onnx__MatMul_26789
FP16/onnx__MatMul_26790
FP16/onnx__MatMul_26791
FP16/onnx__MatMul_26792
FP16/onnx__MatMul_26793
FP16/onnx__MatMul_26794
FP16/onnx__MatMul_26814
FP16/onnx__MatMul_26815
FP16/onnx__MatMul_26816
FP16/onnx__MatMul_26817
FP16/onnx__MatMul_26818
FP16/onnx__MatMul_26819
FP16/onnx__MatMul_26820
FP16/onnx__MatMul_26840
FP16/onnx__MatMul_26841
FP16/onnx__MatMul_26842
FP16/onnx__MatMul_26843
FP16/onnx__MatMul_26844
FP16/onnx__MatMul_26845
FP16/onnx__MatMul_26846
FP16/onnx__MatMul_26866
FP16/onnx__MatMul_26867
FP16/onnx__MatMul_26868
FP16/onnx__MatMul_26869
FP16/onnx__MatMul_26870
FP16/transformer.block_list.0.attn_norm.weight
FP16/transformer.block_list.0.proj_norm.weight
FP16/transformer.block_list.1.attn_norm.weight
FP16/transformer.block_list.1.proj_norm.weight
FP16/transformer.block_list.10.attn_norm.weight
FP16/transformer.block_list.10.proj_norm.weight
FP16/transformer.block_list.11.attn_norm.weight
FP16/transformer.block_list.11.proj_norm.weight
FP16/transformer.block_list.12.attn_norm.weight
FP16/transformer.block_list.12.proj_norm.weight
FP16/transformer.block_list.13.attn_norm.weight
FP16/transformer.block_list.13.proj_norm.weight
FP16/transformer.block_list.14.attn_norm.weight
FP16/transformer.block_list.14.proj_norm.weight
FP16/transformer.block_list.15.attn_norm.weight
FP16/transformer.block_list.15.proj_norm.weight
FP16/transformer.block_list.16.attn_norm.weight
FP16/transformer.block_list.16.proj_norm.weight
FP16/transformer.block_list.17.attn_norm.weight
FP16/transformer.block_list.17.proj_norm.weight
FP16/transformer.block_list.18.attn_norm.weight
FP16/transformer.block_list.18.proj_norm.weight
FP16/transformer.block_list.19.attn_norm.weight
FP16/transformer.block_list.19.proj_norm.weight
FP16/transformer.block_list.2.attn_norm.weight
FP16/transformer.block_list.2.proj_norm.weight
FP16/transformer.block_list.20.attn_norm.weight
FP16/transformer.block_list.20.proj_norm.weight
FP16/transformer.block_list.21.attn_norm.weight
FP16/transformer.block_list.21.proj_norm.weight
FP16/transformer.block_list.22.attn_norm.weight
FP16/transformer.block_list.22.proj_norm.weight
FP16/transformer.block_list.23.attn_norm.weight
FP16/transformer.block_list.23.proj_norm.weight
FP16/transformer.block_list.24.attn_norm.weight
FP16/transformer.block_list.24.proj_norm.weight
FP16/transformer.block_list.25.attn_norm.weight
FP16/transformer.block_list.25.proj_norm.weight
FP16/transformer.block_list.26.attn_norm.weight
FP16/transformer.block_list.26.proj_norm.weight
FP16/transformer.block_list.27.attn_norm.weight
FP16/transformer.block_list.27.proj_norm.weight
FP16/transformer.block_list.28.attn_norm.weight
FP16/transformer.block_list.28.proj_norm.weight
FP16/transformer.block_list.29.attn_norm.weight
FP16/transformer.block_list.29.proj_norm.weight
FP16/transformer.block_list.3.attn_norm.weight
FP16/transformer.block_list.3.proj_norm.weight
FP16/transformer.block_list.30.attn_norm.weight
FP16/transformer.block_list.30.proj_norm.weight
FP16/transformer.block_list.31.attn_norm.weight
FP16/transformer.block_list.31.proj_norm.weight
FP16/transformer.block_list.32.attn_norm.weight
FP16/transformer.block_list.32.proj_norm.weight
FP16/transformer.block_list.33.attn_norm.weight
FP16/transformer.block_list.33.proj_norm.weight
FP16/transformer.block_list.34.attn_norm.weight
FP16/transformer.block_list.34.proj_norm.weight
FP16/transformer.block_list.35.attn_norm.weight
FP16/transformer.block_list.35.proj_norm.weight
FP16/transformer.block_list.36.attn_norm.weight
FP16/transformer.block_list.36.proj_norm.weight
FP16/transformer.block_list.37.attn_norm.weight
FP16/transformer.block_list.37.proj_norm.weight
FP16/transformer.block_list.38.attn_norm.weight
FP16/transformer.block_list.38.proj_norm.weight
FP16/transformer.block_list.39.attn_norm.weight
FP16/transformer.block_list.39.proj_norm.weight
FP16/transformer.block_list.4.attn_norm.weight
FP16/transformer.block_list.4.proj_norm.weight
FP16/transformer.block_list.5.attn_norm.weight
FP16/transformer.block_list.5.proj_norm.weight
FP16/transformer.block_list.6.attn_norm.weight
FP16/transformer.block_list.6.proj_norm.weight
FP16/transformer.block_list.7.attn_norm.weight
FP16/transformer.block_list.7.proj_norm.weight
FP16/transformer.block_list.8.attn_norm.weight
FP16/transformer.block_list.8.proj_norm.weight
FP16/transformer.block_list.9.attn_norm.weight
FP16/transformer.block_list.9.proj_norm.weight
FP16/transformer.layer_norm.weight
FP32-Chat/.gitattributes
FP32-Chat/.gitignore
FP32-Chat/Constant_68_attr__value
FP32-Chat/Constant_76_attr__value
FP32-Chat/LlamaV2_13B_FT_float32.onnx
ONNX
FP32-Chat/embeddings.data
FP32-Chat/onnx__MatMul_25749
FP32-Chat/onnx__MatMul_25750
FP32-Chat/onnx__MatMul_25751
FP32-Chat/onnx__MatMul_25771
FP32-Chat/onnx__MatMul_25772
FP32-Chat/onnx__MatMul_25773
FP32-Chat/onnx__MatMul_25774
FP32-Chat/onnx__MatMul_25775
FP32-Chat/onnx__MatMul_25776
FP32-Chat/onnx__MatMul_25777
FP32-Chat/onnx__MatMul_25797
FP32-Chat/onnx__MatMul_25798
FP32-Chat/onnx__MatMul_25799
FP32-Chat/onnx__MatMul_25800
FP32-Chat/onnx__MatMul_25801
FP32-Chat/onnx__MatMul_25802
FP32-Chat/onnx__MatMul_25803
FP32-Chat/onnx__MatMul_25823
FP32-Chat/onnx__MatMul_25824
FP32-Chat/onnx__MatMul_25825
FP32-Chat/onnx__MatMul_25826
FP32-Chat/onnx__MatMul_25827
FP32-Chat/onnx__MatMul_25828
FP32-Chat/onnx__MatMul_25829
FP32-Chat/onnx__MatMul_25849
FP32-Chat/onnx__MatMul_25850
FP32-Chat/onnx__MatMul_25851
FP32-Chat/onnx__MatMul_25852
FP32-Chat/onnx__MatMul_25853
FP32-Chat/onnx__MatMul_25854
FP32-Chat/onnx__MatMul_25855
FP32-Chat/onnx__MatMul_25875
FP32-Chat/onnx__MatMul_25876
FP32-Chat/onnx__MatMul_25877
FP32-Chat/onnx__MatMul_25878
FP32-Chat/onnx__MatMul_25879
FP32-Chat/onnx__MatMul_25880
FP32-Chat/onnx__MatMul_25881
FP32-Chat/onnx__MatMul_25901
FP32-Chat/onnx__MatMul_25902
FP32-Chat/onnx__MatMul_25903
FP32-Chat/onnx__MatMul_25904
FP32-Chat/onnx__MatMul_25905
FP32-Chat/onnx__MatMul_25906
FP32-Chat/onnx__MatMul_25907
FP32-Chat/onnx__MatMul_25927
FP32-Chat/onnx__MatMul_25928
FP32-Chat/onnx__MatMul_25929
FP32-Chat/onnx__MatMul_25930
FP32-Chat/onnx__MatMul_25931
FP32-Chat/onnx__MatMul_25932
FP32-Chat/onnx__MatMul_25933
FP32-Chat/onnx__MatMul_25953
FP32-Chat/onnx__MatMul_25954
FP32-Chat/onnx__MatMul_25955
FP32-Chat/onnx__MatMul_25956
FP32-Chat/onnx__MatMul_25957
FP32-Chat/onnx__MatMul_25958
FP32-Chat/onnx__MatMul_25959
FP32-Chat/onnx__MatMul_25979
FP32-Chat/onnx__MatMul_25980
FP32-Chat/onnx__MatMul_25981
FP32-Chat/onnx__MatMul_25982
FP32-Chat/onnx__MatMul_25983
FP32-Chat/onnx__MatMul_25984
FP32-Chat/onnx__MatMul_25985
FP32-Chat/onnx__MatMul_26005
FP32-Chat/onnx__MatMul_26006
FP32-Chat/onnx__MatMul_26007
FP32-Chat/onnx__MatMul_26008
FP32-Chat/onnx__MatMul_26009
FP32-Chat/onnx__MatMul_26010
FP32-Chat/onnx__MatMul_26011
FP32-Chat/onnx__MatMul_26031
FP32-Chat/onnx__MatMul_26032
FP32-Chat/onnx__MatMul_26033
FP32-Chat/onnx__MatMul_26034
FP32-Chat/onnx__MatMul_26035
FP32-Chat/onnx__MatMul_26036
FP32-Chat/onnx__MatMul_26037
FP32-Chat/onnx__MatMul_26057
FP32-Chat/onnx__MatMul_26058
FP32-Chat/onnx__MatMul_26059
FP32-Chat/onnx__MatMul_26060
FP32-Chat/onnx__MatMul_26061
FP32-Chat/onnx__MatMul_26062
FP32-Chat/onnx__MatMul_26063
FP32-Chat/onnx__MatMul_26083
FP32-Chat/onnx__MatMul_26084
FP32-Chat/onnx__MatMul_26085
FP32-Chat/onnx__MatMul_26086
FP32-Chat/onnx__MatMul_26087
FP32-Chat/onnx__MatMul_26088
FP32-Chat/onnx__MatMul_26089
FP32-Chat/onnx__MatMul_26109
FP32-Chat/onnx__MatMul_26110
FP32-Chat/onnx__MatMul_26111
FP32-Chat/onnx__MatMul_26112
FP32-Chat/onnx__MatMul_26113
FP32-Chat/onnx__MatMul_26114
FP32-Chat/onnx__MatMul_26115
FP32-Chat/onnx__MatMul_26135
FP32-Chat/onnx__MatMul_26136
FP32-Chat/onnx__MatMul_26137
FP32-Chat/onnx__MatMul_26138
FP32-Chat/onnx__MatMul_26139
FP32-Chat/onnx__MatMul_26140
FP32-Chat/onnx__MatMul_26141
FP32-Chat/onnx__MatMul_26161
FP32-Chat/onnx__MatMul_26162
FP32-Chat/onnx__MatMul_26163
FP32-Chat/onnx__MatMul_26164
FP32-Chat/onnx__MatMul_26165
FP32-Chat/onnx__MatMul_26166
FP32-Chat/onnx__MatMul_26167
FP32-Chat/onnx__MatMul_26187
FP32-Chat/onnx__MatMul_26188
FP32-Chat/onnx__MatMul_26189
FP32-Chat/onnx__MatMul_26190
FP32-Chat/onnx__MatMul_26191
FP32-Chat/onnx__MatMul_26192
FP32-Chat/onnx__MatMul_26193
FP32-Chat/onnx__MatMul_26213
FP32-Chat/onnx__MatMul_26214
FP32-Chat/onnx__MatMul_26215
FP32-Chat/onnx__MatMul_26216
FP32-Chat/onnx__MatMul_26217
FP32-Chat/onnx__MatMul_26218
FP32-Chat/onnx__MatMul_26219
FP32-Chat/onnx__MatMul_26239
FP32-Chat/onnx__MatMul_26240
FP32-Chat/onnx__MatMul_26241
FP32-Chat/onnx__MatMul_26242
FP32-Chat/onnx__MatMul_26243
FP32-Chat/onnx__MatMul_26244
FP32-Chat/onnx__MatMul_26245
FP32-Chat/onnx__MatMul_26265
FP32-Chat/onnx__MatMul_26266
FP32-Chat/onnx__MatMul_26267
FP32-Chat/onnx__MatMul_26268
FP32-Chat/onnx__MatMul_26269
FP32-Chat/onnx__MatMul_26270
FP32-Chat/onnx__MatMul_26271
FP32-Chat/onnx__MatMul_26291
FP32-Chat/onnx__MatMul_26292
FP32-Chat/onnx__MatMul_26293
FP32-Chat/onnx__MatMul_26294
FP32-Chat/onnx__MatMul_26295
FP32-Chat/onnx__MatMul_26296
FP32-Chat/onnx__MatMul_26297
FP32-Chat/onnx__MatMul_26317
FP32-Chat/onnx__MatMul_26318
FP32-Chat/onnx__MatMul_26319
FP32-Chat/onnx__MatMul_26320
FP32-Chat/onnx__MatMul_26321
FP32-Chat/onnx__MatMul_26322
FP32-Chat/onnx__MatMul_26323
FP32-Chat/onnx__MatMul_26343
FP32-Chat/onnx__MatMul_26344
FP32-Chat/onnx__MatMul_26345
FP32-Chat/onnx__MatMul_26346
FP32-Chat/onnx__MatMul_26347
FP32-Chat/onnx__MatMul_26348
FP32-Chat/onnx__MatMul_26349
FP32-Chat/onnx__MatMul_26369
FP32-Chat/onnx__MatMul_26370
FP32-Chat/onnx__MatMul_26371
FP32-Chat/onnx__MatMul_26372
FP32-Chat/onnx__MatMul_26373
FP32-Chat/onnx__MatMul_26374
FP32-Chat/onnx__MatMul_26375
FP32-Chat/onnx__MatMul_26395
FP32-Chat/onnx__MatMul_26396
FP32-Chat/onnx__MatMul_26397
FP32-Chat/onnx__MatMul_26398
FP32-Chat/onnx__MatMul_26399
FP32-Chat/onnx__MatMul_26400
FP32-Chat/onnx__MatMul_26401
FP32-Chat/onnx__MatMul_26421
FP32-Chat/onnx__MatMul_26422
FP32-Chat/onnx__MatMul_26423
FP32-Chat/onnx__MatMul_26424
FP32-Chat/onnx__MatMul_26425
FP32-Chat/onnx__MatMul_26426
FP32-Chat/onnx__MatMul_26427
FP32-Chat/onnx__MatMul_26447
FP32-Chat/onnx__MatMul_26448
FP32-Chat/onnx__MatMul_26449
FP32-Chat/onnx__MatMul_26450
FP32-Chat/onnx__MatMul_26451
FP32-Chat/onnx__MatMul_26452
FP32-Chat/onnx__MatMul_26453
FP32-Chat/onnx__MatMul_26473
FP32-Chat/onnx__MatMul_26474
FP32-Chat/onnx__MatMul_26475
FP32-Chat/onnx__MatMul_26476
FP32-Chat/onnx__MatMul_26477
FP32-Chat/onnx__MatMul_26478
FP32-Chat/onnx__MatMul_26479
FP32-Chat/onnx__MatMul_26499
FP32-Chat/onnx__MatMul_26500
FP32-Chat/onnx__MatMul_26501
FP32-Chat/onnx__MatMul_26502
FP32-Chat/onnx__MatMul_26503
FP32-Chat/onnx__MatMul_26504
FP32-Chat/onnx__MatMul_26505
FP32-Chat/onnx__MatMul_26525
FP32-Chat/onnx__MatMul_26526
FP32-Chat/onnx__MatMul_26527
FP32-Chat/onnx__MatMul_26528
FP32-Chat/onnx__MatMul_26529
FP32-Chat/onnx__MatMul_26530
FP32-Chat/onnx__MatMul_26531
FP32-Chat/onnx__MatMul_26551
FP32-Chat/onnx__MatMul_26552
FP32-Chat/onnx__MatMul_26553
FP32-Chat/onnx__MatMul_26554
FP32-Chat/onnx__MatMul_26555
FP32-Chat/onnx__MatMul_26556
FP32-Chat/onnx__MatMul_26557
FP32-Chat/onnx__MatMul_26577
FP32-Chat/onnx__MatMul_26578
FP32-Chat/onnx__MatMul_26579
FP32-Chat/onnx__MatMul_26580
FP32-Chat/onnx__MatMul_26581
FP32-Chat/onnx__MatMul_26582
FP32-Chat/onnx__MatMul_26583
FP32-Chat/onnx__MatMul_26603
FP32-Chat/onnx__MatMul_26604
FP32-Chat/onnx__MatMul_26605
FP32-Chat/onnx__MatMul_26606
FP32-Chat/onnx__MatMul_26607
FP32-Chat/onnx__MatMul_26608
FP32-Chat/onnx__MatMul_26609
FP32-Chat/onnx__MatMul_26629
FP32-Chat/onnx__MatMul_26630
FP32-Chat/onnx__MatMul_26631
FP32-Chat/onnx__MatMul_26632
FP32-Chat/onnx__MatMul_26633
FP32-Chat/onnx__MatMul_26634
FP32-Chat/onnx__MatMul_26635
FP32-Chat/onnx__MatMul_26655
FP32-Chat/onnx__MatMul_26656
FP32-Chat/onnx__MatMul_26657
FP32-Chat/onnx__MatMul_26658
FP32-Chat/onnx__MatMul_26659
FP32-Chat/onnx__MatMul_26660
FP32-Chat/onnx__MatMul_26661
FP32-Chat/onnx__MatMul_26681
FP32-Chat/onnx__MatMul_26682
FP32-Chat/onnx__MatMul_26683
FP32-Chat/onnx__MatMul_26684
FP32-Chat/onnx__MatMul_26685
FP32-Chat/onnx__MatMul_26686
FP32-Chat/onnx__MatMul_26687
FP32-Chat/onnx__MatMul_26707
FP32-Chat/onnx__MatMul_26708
FP32-Chat/onnx__MatMul_26709
FP32-Chat/onnx__MatMul_26710
FP32-Chat/onnx__MatMul_26711
FP32-Chat/onnx__MatMul_26712
FP32-Chat/onnx__MatMul_26713
FP32-Chat/onnx__MatMul_26733
FP32-Chat/onnx__MatMul_26734
FP32-Chat/onnx__MatMul_26735
FP32-Chat/onnx__MatMul_26736
FP32-Chat/onnx__MatMul_26737
FP32-Chat/onnx__MatMul_26738
FP32-Chat/onnx__MatMul_26739
FP32-Chat/onnx__MatMul_26759
FP32-Chat/onnx__MatMul_26760
FP32-Chat/onnx__MatMul_26761
FP32-Chat/onnx__MatMul_26762
FP32-Chat/onnx__MatMul_26763
FP32-Chat/onnx__MatMul_26764
FP32-Chat/onnx__MatMul_26765
FP32-Chat/onnx__MatMul_26785
FP32-Chat/onnx__MatMul_26786
FP32-Chat/onnx__MatMul_26787
FP32-Chat/onnx__MatMul_26788
FP32-Chat/onnx__MatMul_26789
FP32-Chat/transformer.block_list.0.attn_norm.weight
FP32-Chat/transformer.block_list.0.proj_norm.weight
FP32-Chat/transformer.block_list.1.attn_norm.weight
FP32-Chat/transformer.block_list.1.proj_norm.weight
FP32-Chat/transformer.block_list.10.attn_norm.weight
FP32-Chat/transformer.block_list.10.proj_norm.weight
FP32-Chat/transformer.block_list.11.attn_norm.weight
FP32-Chat/transformer.block_list.11.proj_norm.weight
FP32-Chat/transformer.block_list.12.attn_norm.weight
FP32-Chat/transformer.block_list.12.proj_norm.weight
FP32-Chat/transformer.block_list.13.attn_norm.weight
FP32-Chat/transformer.block_list.13.proj_norm.weight
FP32-Chat/transformer.block_list.14.attn_norm.weight
FP32-Chat/transformer.block_list.14.proj_norm.weight
FP32-Chat/transformer.block_list.15.attn_norm.weight
FP32-Chat/transformer.block_list.15.proj_norm.weight
FP32-Chat/transformer.block_list.16.attn_norm.weight
FP32-Chat/transformer.block_list.16.proj_norm.weight
FP32-Chat/transformer.block_list.17.attn_norm.weight
FP32-Chat/transformer.block_list.17.proj_norm.weight
FP32-Chat/transformer.block_list.18.attn_norm.weight
FP32-Chat/transformer.block_list.18.proj_norm.weight
FP32-Chat/transformer.block_list.19.attn_norm.weight
FP32-Chat/transformer.block_list.19.proj_norm.weight
FP32-Chat/transformer.block_list.2.attn_norm.weight
FP32-Chat/transformer.block_list.2.proj_norm.weight
FP32-Chat/transformer.block_list.20.attn_norm.weight
FP32-Chat/transformer.block_list.20.proj_norm.weight
FP32-Chat/transformer.block_list.21.attn_norm.weight
FP32-Chat/transformer.block_list.21.proj_norm.weight
FP32-Chat/transformer.block_list.22.attn_norm.weight
FP32-Chat/transformer.block_list.22.proj_norm.weight
FP32-Chat/transformer.block_list.23.attn_norm.weight
FP32-Chat/transformer.block_list.23.proj_norm.weight
FP32-Chat/transformer.block_list.24.attn_norm.weight
FP32-Chat/transformer.block_list.24.proj_norm.weight
FP32-Chat/transformer.block_list.25.attn_norm.weight
FP32-Chat/transformer.block_list.25.proj_norm.weight
FP32-Chat/transformer.block_list.26.attn_norm.weight
FP32-Chat/transformer.block_list.26.proj_norm.weight
FP32-Chat/transformer.block_list.27.attn_norm.weight
FP32-Chat/transformer.block_list.27.proj_norm.weight
FP32-Chat/transformer.block_list.28.attn_norm.weight
FP32-Chat/transformer.block_list.28.proj_norm.weight
FP32-Chat/transformer.block_list.29.attn_norm.weight
FP32-Chat/transformer.block_list.29.proj_norm.weight
FP32-Chat/transformer.block_list.3.attn_norm.weight
FP32-Chat/transformer.block_list.3.proj_norm.weight
FP32-Chat/transformer.block_list.30.attn_norm.weight
FP32-Chat/transformer.block_list.30.proj_norm.weight
FP32-Chat/transformer.block_list.31.attn_norm.weight
FP32-Chat/transformer.block_list.31.proj_norm.weight
FP32-Chat/transformer.block_list.32.attn_norm.weight
FP32-Chat/transformer.block_list.32.proj_norm.weight
FP32-Chat/transformer.block_list.33.attn_norm.weight
FP32-Chat/transformer.block_list.33.proj_norm.weight
FP32-Chat/transformer.block_list.34.attn_norm.weight
FP32-Chat/transformer.block_list.34.proj_norm.weight
FP32-Chat/transformer.block_list.35.attn_norm.weight
FP32-Chat/transformer.block_list.35.proj_norm.weight
FP32-Chat/transformer.block_list.36.attn_norm.weight
FP32-Chat/transformer.block_list.36.proj_norm.weight
FP32-Chat/transformer.block_list.37.attn_norm.weight
FP32-Chat/transformer.block_list.37.proj_norm.weight
FP32-Chat/transformer.block_list.38.attn_norm.weight
FP32-Chat/transformer.block_list.38.proj_norm.weight
FP32-Chat/transformer.block_list.39.attn_norm.weight
FP32-Chat/transformer.block_list.39.proj_norm.weight
FP32-Chat/transformer.block_list.4.attn_norm.weight
FP32-Chat/transformer.block_list.4.proj_norm.weight
FP32-Chat/transformer.block_list.5.attn_norm.weight
FP32-Chat/transformer.block_list.5.proj_norm.weight
FP32-Chat/transformer.block_list.6.attn_norm.weight
FP32-Chat/transformer.block_list.6.proj_norm.weight
FP32-Chat/transformer.block_list.7.attn_norm.weight
FP32-Chat/transformer.block_list.7.proj_norm.weight
FP32-Chat/transformer.block_list.8.attn_norm.weight
FP32-Chat/transformer.block_list.8.proj_norm.weight
FP32-Chat/transformer.block_list.9.attn_norm.weight
FP32-Chat/transformer.block_list.9.proj_norm.weight
FP32-Chat/transformer.layer_norm.weight
FP32/Constant_68_attr__value
FP32/Constant_76_attr__value
FP32/LlamaV2_13B_float32.onnx
ONNX
FP32/embeddings.data
FP32/onnx__MatMul_25749
FP32/onnx__MatMul_25750
FP32/onnx__MatMul_25751
FP32/onnx__MatMul_25771
FP32/onnx__MatMul_25772
FP32/onnx__MatMul_25773
FP32/onnx__MatMul_25774
FP32/onnx__MatMul_25775
FP32/onnx__MatMul_25776
FP32/onnx__MatMul_25777
FP32/onnx__MatMul_25797
FP32/onnx__MatMul_25798
FP32/onnx__MatMul_25799
FP32/onnx__MatMul_25800
FP32/onnx__MatMul_25801
FP32/onnx__MatMul_25802
FP32/onnx__MatMul_25803
FP32/onnx__MatMul_25823
FP32/onnx__MatMul_25824
FP32/onnx__MatMul_25825
FP32/onnx__MatMul_25826
FP32/onnx__MatMul_25827
FP32/onnx__MatMul_25828
FP32/onnx__MatMul_25829
FP32/onnx__MatMul_25849
FP32/onnx__MatMul_25850
FP32/onnx__MatMul_25851
FP32/onnx__MatMul_25852
FP32/onnx__MatMul_25853
FP32/onnx__MatMul_25854
FP32/onnx__MatMul_25855
FP32/onnx__MatMul_25875
FP32/onnx__MatMul_25876
FP32/onnx__MatMul_25877
FP32/onnx__MatMul_25878
FP32/onnx__MatMul_25879
FP32/onnx__MatMul_25880
FP32/onnx__MatMul_25881
FP32/onnx__MatMul_25901
FP32/onnx__MatMul_25902
FP32/onnx__MatMul_25903
FP32/onnx__MatMul_25904
FP32/onnx__MatMul_25905
FP32/onnx__MatMul_25906
FP32/onnx__MatMul_25907
FP32/onnx__MatMul_25927
FP32/onnx__MatMul_25928
FP32/onnx__MatMul_25929
FP32/onnx__MatMul_25930
FP32/onnx__MatMul_25931
FP32/onnx__MatMul_25932
FP32/onnx__MatMul_25933
FP32/onnx__MatMul_25953
FP32/onnx__MatMul_25954
FP32/onnx__MatMul_25955
FP32/onnx__MatMul_25956
FP32/onnx__MatMul_25957
FP32/onnx__MatMul_25958
FP32/onnx__MatMul_25959
FP32/onnx__MatMul_25979
FP32/onnx__MatMul_25980
FP32/onnx__MatMul_25981
FP32/onnx__MatMul_25982
FP32/onnx__MatMul_25983
FP32/onnx__MatMul_25984
FP32/onnx__MatMul_25985
FP32/onnx__MatMul_26005
FP32/onnx__MatMul_26006
FP32/onnx__MatMul_26007
FP32/onnx__MatMul_26008
FP32/onnx__MatMul_26009
FP32/onnx__MatMul_26010
FP32/onnx__MatMul_26011
FP32/onnx__MatMul_26031
FP32/onnx__MatMul_26032
FP32/onnx__MatMul_26033
FP32/onnx__MatMul_26034
FP32/onnx__MatMul_26035
FP32/onnx__MatMul_26036
FP32/onnx__MatMul_26037
FP32/onnx__MatMul_26057
FP32/onnx__MatMul_26058
FP32/onnx__MatMul_26059
FP32/onnx__MatMul_26060
FP32/onnx__MatMul_26061
FP32/onnx__MatMul_26062
FP32/onnx__MatMul_26063
FP32/onnx__MatMul_26083
FP32/onnx__MatMul_26084
FP32/onnx__MatMul_26085
FP32/onnx__MatMul_26086
FP32/onnx__MatMul_26087
FP32/onnx__MatMul_26088
FP32/onnx__MatMul_26089
FP32/onnx__MatMul_26109
FP32/onnx__MatMul_26110
FP32/onnx__MatMul_26111
FP32/onnx__MatMul_26112
FP32/onnx__MatMul_26113
FP32/onnx__MatMul_26114
FP32/onnx__MatMul_26115
FP32/onnx__MatMul_26135
FP32/onnx__MatMul_26136
FP32/onnx__MatMul_26137
FP32/onnx__MatMul_26138
FP32/onnx__MatMul_26139
FP32/onnx__MatMul_26140
FP32/onnx__MatMul_26141
FP32/onnx__MatMul_26161
FP32/onnx__MatMul_26162
FP32/onnx__MatMul_26163
FP32/onnx__MatMul_26164
FP32/onnx__MatMul_26165
FP32/onnx__MatMul_26166
FP32/onnx__MatMul_26167
FP32/onnx__MatMul_26187
FP32/onnx__MatMul_26188
FP32/onnx__MatMul_26189
FP32/onnx__MatMul_26190
FP32/onnx__MatMul_26191
FP32/onnx__MatMul_26192
FP32/onnx__MatMul_26193
FP32/onnx__MatMul_26213
FP32/onnx__MatMul_26214
FP32/onnx__MatMul_26215
FP32/onnx__MatMul_26216
FP32/onnx__MatMul_26217
FP32/onnx__MatMul_26218
FP32/onnx__MatMul_26219
FP32/onnx__MatMul_26239
FP32/onnx__MatMul_26240
FP32/onnx__MatMul_26241
FP32/onnx__MatMul_26242
FP32/onnx__MatMul_26243
FP32/onnx__MatMul_26244
FP32/onnx__MatMul_26245
FP32/onnx__MatMul_26265
FP32/onnx__MatMul_26266
FP32/onnx__MatMul_26267
FP32/onnx__MatMul_26268
FP32/onnx__MatMul_26269
FP32/onnx__MatMul_26270
FP32/onnx__MatMul_26271
FP32/onnx__MatMul_26291
FP32/onnx__MatMul_26292
FP32/onnx__MatMul_26293
FP32/onnx__MatMul_26294
FP32/onnx__MatMul_26295
FP32/onnx__MatMul_26296
FP32/onnx__MatMul_26297
FP32/onnx__MatMul_26317
FP32/onnx__MatMul_26318
FP32/onnx__MatMul_26319
FP32/onnx__MatMul_26320
FP32/onnx__MatMul_26321
FP32/onnx__MatMul_26322
FP32/onnx__MatMul_26323
FP32/onnx__MatMul_26343
FP32/onnx__MatMul_26344
FP32/onnx__MatMul_26345
FP32/onnx__MatMul_26346
FP32/onnx__MatMul_26347
FP32/onnx__MatMul_26348
FP32/onnx__MatMul_26349
FP32/onnx__MatMul_26369
FP32/onnx__MatMul_26370
FP32/onnx__MatMul_26371
FP32/onnx__MatMul_26372
FP32/onnx__MatMul_26373
FP32/onnx__MatMul_26374
FP32/onnx__MatMul_26375
FP32/onnx__MatMul_26395
FP32/onnx__MatMul_26396
FP32/onnx__MatMul_26397
FP32/onnx__MatMul_26398
FP32/onnx__MatMul_26399
FP32/onnx__MatMul_26400
FP32/onnx__MatMul_26401
FP32/onnx__MatMul_26421
FP32/onnx__MatMul_26422
FP32/onnx__MatMul_26423
FP32/onnx__MatMul_26424
FP32/onnx__MatMul_26425
FP32/onnx__MatMul_26426
FP32/onnx__MatMul_26427
FP32/onnx__MatMul_26447
FP32/onnx__MatMul_26448
FP32/onnx__MatMul_26449
FP32/onnx__MatMul_26450
FP32/onnx__MatMul_26451
FP32/onnx__MatMul_26452
FP32/onnx__MatMul_26453
FP32/onnx__MatMul_26473
FP32/onnx__MatMul_26474
FP32/onnx__MatMul_26475
FP32/onnx__MatMul_26476
FP32/onnx__MatMul_26477
FP32/onnx__MatMul_26478
FP32/onnx__MatMul_26479
FP32/onnx__MatMul_26499
FP32/onnx__MatMul_26500
FP32/onnx__MatMul_26501
FP32/onnx__MatMul_26502
FP32/onnx__MatMul_26503
FP32/onnx__MatMul_26504
FP32/onnx__MatMul_26505
FP32/onnx__MatMul_26525
FP32/onnx__MatMul_26526
FP32/onnx__MatMul_26527
FP32/onnx__MatMul_26528
FP32/onnx__MatMul_26529
FP32/onnx__MatMul_26530
FP32/onnx__MatMul_26531
FP32/onnx__MatMul_26551
FP32/onnx__MatMul_26552
FP32/onnx__MatMul_26553
FP32/onnx__MatMul_26554
FP32/onnx__MatMul_26555
FP32/onnx__MatMul_26556
FP32/onnx__MatMul_26557
FP32/onnx__MatMul_26577
FP32/onnx__MatMul_26578
FP32/onnx__MatMul_26579
FP32/onnx__MatMul_26580
FP32/onnx__MatMul_26581
FP32/onnx__MatMul_26582
FP32/onnx__MatMul_26583
FP32/onnx__MatMul_26603
FP32/onnx__MatMul_26604
FP32/onnx__MatMul_26605
FP32/onnx__MatMul_26606
FP32/onnx__MatMul_26607
FP32/onnx__MatMul_26608
FP32/onnx__MatMul_26609
FP32/onnx__MatMul_26629
FP32/onnx__MatMul_26630
FP32/onnx__MatMul_26631
FP32/onnx__MatMul_26632
FP32/onnx__MatMul_26633
FP32/onnx__MatMul_26634
FP32/onnx__MatMul_26635
FP32/onnx__MatMul_26655
FP32/onnx__MatMul_26656
FP32/onnx__MatMul_26657
FP32/onnx__MatMul_26658
FP32/onnx__MatMul_26659
FP32/onnx__MatMul_26660
FP32/onnx__MatMul_26661
FP32/onnx__MatMul_26681
FP32/onnx__MatMul_26682
FP32/onnx__MatMul_26683
FP32/onnx__MatMul_26684
FP32/onnx__MatMul_26685
FP32/onnx__MatMul_26686
FP32/onnx__MatMul_26687
FP32/onnx__MatMul_26707
FP32/onnx__MatMul_26708
FP32/onnx__MatMul_26709
FP32/onnx__MatMul_26710
FP32/onnx__MatMul_26711
FP32/onnx__MatMul_26712
FP32/onnx__MatMul_26713
FP32/onnx__MatMul_26733
FP32/onnx__MatMul_26734
FP32/onnx__MatMul_26735
FP32/onnx__MatMul_26736
FP32/onnx__MatMul_26737
FP32/onnx__MatMul_26738
FP32/onnx__MatMul_26739
FP32/onnx__MatMul_26759
FP32/onnx__MatMul_26760
FP32/onnx__MatMul_26761
FP32/onnx__MatMul_26762
FP32/onnx__MatMul_26763
FP32/onnx__MatMul_26764
FP32/onnx__MatMul_26765
FP32/onnx__MatMul_26785
FP32/onnx__MatMul_26786
FP32/onnx__MatMul_26787
FP32/onnx__MatMul_26788
FP32/onnx__MatMul_26789
FP32/transformer.block_list.0.attn_norm.weight
FP32/transformer.block_list.0.proj_norm.weight
FP32/transformer.block_list.1.attn_norm.weight
FP32/transformer.block_list.1.proj_norm.weight
FP32/transformer.block_list.10.attn_norm.weight
FP32/transformer.block_list.10.proj_norm.weight
FP32/transformer.block_list.11.attn_norm.weight
FP32/transformer.block_list.11.proj_norm.weight
FP32/transformer.block_list.12.attn_norm.weight
FP32/transformer.block_list.12.proj_norm.weight
FP32/transformer.block_list.13.attn_norm.weight
FP32/transformer.block_list.13.proj_norm.weight
FP32/transformer.block_list.14.attn_norm.weight
FP32/transformer.block_list.14.proj_norm.weight
FP32/transformer.block_list.15.attn_norm.weight
FP32/transformer.block_list.15.proj_norm.weight
FP32/transformer.block_list.16.attn_norm.weight
FP32/transformer.block_list.16.proj_norm.weight
FP32/transformer.block_list.17.attn_norm.weight
FP32/transformer.block_list.17.proj_norm.weight
FP32/transformer.block_list.18.attn_norm.weight
FP32/transformer.block_list.18.proj_norm.weight
FP32/transformer.block_list.19.attn_norm.weight
FP32/transformer.block_list.19.proj_norm.weight
FP32/transformer.block_list.2.attn_norm.weight
FP32/transformer.block_list.2.proj_norm.weight
FP32/transformer.block_list.20.attn_norm.weight
FP32/transformer.block_list.20.proj_norm.weight
FP32/transformer.block_list.21.attn_norm.weight
FP32/transformer.block_list.21.proj_norm.weight
FP32/transformer.block_list.22.attn_norm.weight
FP32/transformer.block_list.22.proj_norm.weight
FP32/transformer.block_list.23.attn_norm.weight
FP32/transformer.block_list.23.proj_norm.weight
FP32/transformer.block_list.24.attn_norm.weight
FP32/transformer.block_list.24.proj_norm.weight
FP32/transformer.block_list.25.attn_norm.weight
FP32/transformer.block_list.25.proj_norm.weight
FP32/transformer.block_list.26.attn_norm.weight
FP32/transformer.block_list.26.proj_norm.weight
FP32/transformer.block_list.27.attn_norm.weight
FP32/transformer.block_list.27.proj_norm.weight
FP32/transformer.block_list.28.attn_norm.weight
FP32/transformer.block_list.28.proj_norm.weight
FP32/transformer.block_list.29.attn_norm.weight
FP32/transformer.block_list.29.proj_norm.weight
FP32/transformer.block_list.3.attn_norm.weight
FP32/transformer.block_list.3.proj_norm.weight
FP32/transformer.block_list.30.attn_norm.weight
FP32/transformer.block_list.30.proj_norm.weight
FP32/transformer.block_list.31.attn_norm.weight
FP32/transformer.block_list.31.proj_norm.weight
FP32/transformer.block_list.32.attn_norm.weight
FP32/transformer.block_list.32.proj_norm.weight
FP32/transformer.block_list.33.attn_norm.weight
FP32/transformer.block_list.33.proj_norm.weight
FP32/transformer.block_list.34.attn_norm.weight
FP32/transformer.block_list.34.proj_norm.weight
FP32/transformer.block_list.35.attn_norm.weight
FP32/transformer.block_list.35.proj_norm.weight
FP32/transformer.block_list.36.attn_norm.weight
FP32/transformer.block_list.36.proj_norm.weight
FP32/transformer.block_list.37.attn_norm.weight
FP32/transformer.block_list.37.proj_norm.weight
FP32/transformer.block_list.38.attn_norm.weight
FP32/transformer.block_list.38.proj_norm.weight
FP32/transformer.block_list.39.attn_norm.weight
FP32/transformer.block_list.39.proj_norm.weight
FP32/transformer.block_list.4.attn_norm.weight
FP32/transformer.block_list.4.proj_norm.weight
FP32/transformer.block_list.5.attn_norm.weight
FP32/transformer.block_list.5.proj_norm.weight
FP32/transformer.block_list.6.attn_norm.weight
FP32/transformer.block_list.6.proj_norm.weight
FP32/transformer.block_list.7.attn_norm.weight
FP32/transformer.block_list.7.proj_norm.weight
FP32/transformer.block_list.8.attn_norm.weight
FP32/transformer.block_list.8.proj_norm.weight
FP32/transformer.block_list.9.attn_norm.weight
FP32/transformer.block_list.9.proj_norm.weight
FP32/transformer.layer_norm.weight
README.md
embeddings-chat.pth
embeddings.pth
llama2_onnx_inference.py
requirements.txt
tokenizer.model