Downloads · 30 days
11
4% of all-time downloads
Kanjdoes/apollo-ministral-3b
apollo-ministral-3b is a machine learning model from Kanjdoes. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Hello! This is a simple finetuning experiment, no less. Feel free to use these.
Downloads · 30 days
11
4% of all-time downloads
All-time downloads
253
Public
Parameters
3.8B
16.3 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors6.9 GB · 100%
From the Hugging Face model README
Hello! This is a simple finetuning experiment, no less. Feel free to use these.
Trained on 300 examples, train_loss of ~1.61, runtime of about half an hour on Colab T4 Runtime. Model frankensteined into Mistral instead of Ministral for tool support.
Quants are done with Imatrix data based on a subset of training data. (except Q8 and FP16) See quants over at https://hf.co/Kanjdoes/apollo-ministral-3b-GGUF
This model was trained on Apollo, which is a dataset that contains both SFW and NSFW content.