Skip to content
View n3ural-h4x's full-sized avatar
1 follower · 0 following

Block or report n3ural-h4x

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse

Popular repositories Loading

  1. Flash-attn-v1and-v2-fwd Flash-attn-v1and-v2-fwd Public

    This Project shows the implementation of the Flash attention v1 and v2 both version and also benchmarks it against the Naive and Pytorch SDPA implementation.*Support any dtype**Tested on t4 gpu*

    Jupyter Notebook

  2. Flash_attn_v1_and_v2_bwd Flash_attn_v1_and_v2_bwd Public

    Backward pass for the backward pass of flash attention v1 and v2 backward.It specifies on the correct gradient accuracy rather than speed.Works only on float tested on p100

    Jupyter Notebook

  3. ArcForge ArcForge Public

    ArcForge is a high-performance multimodal vision-language model built around a scalable Mixture-of-Experts (MoE) architecture. It combines a ResNet-150 visual backbone with rotary position embeddin…

    Jupyter Notebook