Code

Support for Phi-3 MLP layer#84

Code

Merged

alessandropalla merged 26 commits intointel:mainintel/intel-npu-acceleration-library:mainfrom SarahByrneIntel:sarah/feature/phi3MLP_layerSarahByrneIntel/intel-npu-acceleration-library:sarah/feature/phi3MLP_layerCopy head branch name to clipboardJul 19, 2024

+421-64

Commits

Commits on Jul 1, 2024

• Add support for phi-3 MLP layer

  

  SarahByrneIntel

  committed
  22c627c

  Copy full SHA for 22c627c

Commits on Jul 2, 2024

• Updating support for Phi-3 MLP

  

  SarahByrneIntel

  committed
  ea4b27a

  Copy full SHA for ea4b27a

Commits on Jul 3, 2024

• Update for Phi-3 MLP testing

  

  SarahByrneIntel

  committed
  39c070c

  Copy full SHA for 39c070c

Commits on Jul 5, 2024

• Merge branch 'main' into sarah/feature/phi3MLP_layer

  

  SarahByrneIntel

  committed
  2042fab

  Copy full SHA for 2042fab

• Merge branch 'intel:main' into sarah/feature/phi3MLP_layer

  SarahByrneIntel

  authored
  5660cc3

  Copy full SHA for 5660cc3

Commits on Jul 8, 2024

• Update for phi-3 mlp layer

  

  SarahByrneIntel

  committed
  727454e

  Copy full SHA for 727454e

• Merge branch 'sarah/feature/phi3MLP_layer' of https://github.com/SarahByrneIntel/intel-npu-acceleration-library into sarah/feature/phi3MLP_layer

  

  SarahByrneIntel

  committed
  00a64f0

  Copy full SHA for 00a64f0

• Merge branch 'intel:main' into sarah/feature/phi3MLP_layer

  SarahByrneIntel

  authored
  100fe88

  Copy full SHA for 100fe88

• Remove old code for phi-3 mlp layer

  

  SarahByrneIntel

  committed
  ea4ea19

  Copy full SHA for ea4ea19

• Merge branch 'sarah/feature/phi3MLP_layer' of https://github.com/SarahByrneIntel/intel-npu-acceleration-library into sarah/feature/phi3MLP_layer

  

  SarahByrneIntel

  committed
  53c7b0d

  Copy full SHA for 53c7b0d

Commits on Jul 12, 2024

• Add type tensor op and quantisation support

  

  SarahByrneIntel

  committed
  1fef8a4

  Copy full SHA for 1fef8a4

Commits on Jul 15, 2024

• add support for model quantisation and code clean up

  

  SarahByrneIntel

  committed
  cc5d373

  Copy full SHA for cc5d373

• Merge branch 'main' into sarah/feature/phi3MLP_layer

  SarahByrneIntel

  authored
  ff47c1d

  Copy full SHA for ff47c1d

• Fix for model quantization

  

  SarahByrneIntel

  committed
  d2fe9fe

  Copy full SHA for d2fe9fe

Commits on Jul 16, 2024

• Add testing for phi-3 mlp quantisation

  

  SarahByrneIntel

  committed
  b7825e7

  Copy full SHA for b7825e7

Commits on Jul 17, 2024

• Add phi-3 mlp test and enable model profiling toggling

  

  SarahByrneIntel

  committed
  c652859

  Copy full SHA for c652859

• Update for model profiling toggle

  

  SarahByrneIntel

  committed
  786c663

  Copy full SHA for 786c663

Commits on Jul 18, 2024

• Add compile config feature

  

  SarahByrneIntel

  committed
  003d639

  Copy full SHA for 003d639

• Fix test for compile config and remove old code

  

  SarahByrneIntel

  committed
  c63c223

  Copy full SHA for c63c223

• Fix tests with compile config

  

  SarahByrneIntel

  committed
  e652eaa

  Copy full SHA for e652eaa

• Fix for compiler, updates for tests and examples, doc update

  

  SarahByrneIntel

  committed
  7f2faf9

  Copy full SHA for 7f2faf9

• Update for model examples and remove test code

  

  SarahByrneIntel

  committed
  4b5f857

  Copy full SHA for 4b5f857

Commits on Jul 19, 2024

• Merge branch 'main' into sarah/feature/phi3MLP_layer

  alessandropalla

  authored
  2718e13

  Copy full SHA for 2718e13

• Fix for quantization and remove unused code

  

  SarahByrneIntel

  committed
  ae1fd61

  Copy full SHA for ae1fd61

• Merge branch 'sarah/feature/phi3MLP_layer' of https://github.com/SarahByrneIntel/intel-npu-acceleration-library into sarah/feature/phi3MLP_layer

  

  SarahByrneIntel

  committed
  5d578a1

  Copy full SHA for 5d578a1

• Update for quantization of a model

  

  SarahByrneIntel

  committed
  2890299

  Copy full SHA for 2890299