Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

Kernels:
aws-neuron
/
vit-encoder-neuron-kernels

Kernel card Files Files and versions
xet
Community
3
vit-encoder-neuron-kernels / build /torch-neuron
270 kB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 11 commits
jburtoft's picture
jburtoft
v1.9: padding masks on the kernel path for NeuronXLMRobertaModel (per-image additive key bias in attention_block_l1_ln_post_kb); existing models bitwise identical to v1.8
954a874 6 days ago
  • nki_kernels
    v1.9: padding masks on the kernel path for NeuronXLMRobertaModel (per-image additive key bias in attention_block_l1_ln_post_kb); existing models bitwise identical to v1.8 6 days ago
  • __init__.py
    1.62 kB
    v1.8: post-LayerNorm encoders (NeuronXLMRobertaModel; post-LN epilogue outputs + no-RoPE switch) and LNC=2 fix for odd head-group counts 15 days ago
  • encoder.py
    14.5 kB
    v1.8: post-LayerNorm encoders (NeuronXLMRobertaModel; post-LN epilogue outputs + no-RoPE switch) and LNC=2 fix for odd head-group counts 15 days ago
  • layers.py
    17.9 kB
    v1.9: padding masks on the kernel path for NeuronXLMRobertaModel (per-image additive key bias in attention_block_l1_ln_post_kb); existing models bitwise identical to v1.8 6 days ago
  • metadata.json
    182 Bytes
    v1: NKI kernels for RoPE ViT encoders (DINOv3, V-JEPA2) on Trainium2 26 days ago