fpgaminer

`B` is the batch size, `T` is the sequence length, and `C` is the dimensionality of the embedding (`n_embd`). At the first layer, if your batch size were 16, `n_embd=768`,...

The model and training code has seen a number of updates, some of which fixed bugs that improved training, but looking at the history for `play_math` it hasn't been updated...

You can use a docker-compose container with docker-credential-gcr added to it, ala: https://hub.docker.com/r/cryptopants/docker-compose-gcr It's a drop-in replacement for the containerized docker-compose suggested by COS docs, and can pull from private...

No worries. Regarding the failing automated test: looks like it's because of an unused argument which occurs when the parallel feature is off. I can fix that once the exact...

I ran into this bug and have been doing some investigation. I'm using a Yard Stick 1, sdcc 3.5.0. If I patch the firmware code to pass repeat correctly: `buf[0]...

My modified Dockerfile for testing: ``` FROM nvidia/cuda:11.7.0-devel-ubuntu22.04 as builder RUN apt-get update && \ apt-get install --no-install-recommends -y git vim build-essential python3-dev python3-venv && \ rm -rf /var/lib/apt/lists/* RUN...

> This should already work. How so? I don't see any support for GPTQ-triton on `main`?

> Closing this in favor of #1229 That pull request doesn't have anything to do with GPTQ-triton? Are you perhaps confusing GPTQ-triton with GPTQ-for-LLaMa? I'm not affiliated with the latter.

> What's the difference (advantage) of GPTQ-triton over GPTQ-for-LLaMa's triton branch? Most of GPTQ-for-LLaMa's triton branch is copied from my code in GPTQ-triton, so it's always going to lag my...

I'm not sure what you mean by "Hardware miner."