bennmann

Whole project may work with python 3.9; dag.py could be made compatable with python 3.9 by changing: from fractions import gcd To from math import gcd Do not install cynetworkx...

Code instructions with a single Input coding questions and a Target pair wrapped in JSON

1

https://huggingface.co/datasets/bigscience/xP3/blob/main/code/merged_code.jsonl The XP3 dataset includes a section on Code instructions with a single Input coding questions and a Target pair wrapped in JSON. This task has millions of lines of...

Stanford Human Preferences Dataset (SHP)

1

https://huggingface.co/datasets/stanfordnlp/SHP This may best be one of the best first datasets in the training of a model

include "timeit" module in benchmark.py

2

I'm not exactly sure the best way to implement this, however it would be good to use "timeit" python module in the todo loop at the end of benchmark.py somehow...

AMD Ubuntu 22.04 GPU - guide

7

- The below bash history assembled from my local on premise 6900 XT machine - Inference directions, untested for training - not containerized, raw bare metal, no conda or docker...

``` ~/llama$ pip3 install -e . __init__.py", line 2685, in parsed_version raise packaging.version.InvalidVersion(f"{str(ex)} {info}") from None pkg_resources.extern.packaging.version.InvalidVersion: Invalid version: '1.1build1' (package: distro-info) ``` i have all dependencies installed. 12400F CPU,...

docker-compose method could not load rwkv model, in spite of proper folder structure

9

hello, i tried the docker-compose method outlined in the README and here is the output: `$ docker-compose up -d --build /usr/lib/python3/dist-packages/paramiko/transport.py:236: CryptographyDeprecationWarning: Blowfish has been deprecated "class": algorithms.Blowfish, Building api...

docker-compose rwkv method folder structure

5

looks like some filename variables may not be gracefully accounting for the example/rwkv folder structure now after #255 ? this is closer though (using the docker-compose method) ``` $ docker-compose...

Replace Chain of Thought with Thread of Thought

2

https://arxiv.org/abs/2311.08734 is Thread of Thought new SOTA over Chain of Thought?

Hello, The docker run "--gpus" flag fails to limit the number of GPUs in a single machine to a specific container. If I wanted to run 2 instances of rocm-docker...