crestf411's picture
Update README.md
4c89912 verified
|
raw
history blame
1.02 kB
metadata
language:
  - en
tags:
  - not-for-all-audiences

Daybreak-Mixtral-8x7b v24.02-7

An experimental model trained on a (currently) private ERP dataset of highly curated niche content (crestfall/daybreak as of 2024-02-10).

Not suitable for any audience.

Model was fine tuned on top of mistralai/Mistral-7B-Instruct-v0.2, and follows that model's instruction format.

Prompt format:

The model uses the Mixtral-8x7b-instruct format:

text = "<s>[INST] What is your favourite condiment? [/INST]"
"Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!</s> "
"[INST] Do you have mayonnaise recipes? [/INST]"

Training details:

The model was trained for 1.83 epochs (eval minima based on 1% of dataset) using Axolotl.

See axolotl.yml for details.