Downloads · 30 days
7
3% of all-time downloads
kkkkkkatherine/llama-3.2-1b-finetome-1000steps-gguf
llama-3.2-1b-finetome-1000steps-gguf is a machine learning model from kkkkkkatherine. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
This is the best performing model from systematic optimization experiments for ID2223 Scalable Machine Learning Lab 2.
Downloads · 30 days
7
3% of all-time downloads
All-time downloads
201
Public
Repo size
808 MB
Likes
0
Public
Click a slice to open those files.
.gguf808 MB · 100%
From the Hugging Face model README
This is the best performing model from systematic optimization experiments for ID2223 Scalable Machine Learning Lab 2.
Achieved 23% loss reduction compared to baseline (60 steps):
This model was selected as the best from 8 systematic experiments testing:
See experiment report for detailed analysis.
from llama_cpp import Llama
llm = Llama.from_pretrained(
repo_id="kkkkkkatherine/llama-3.2-1b-finetome-1000steps-gguf",
filename="model.gguf",
n_ctx=2048,
)
response = llm("User: What is machine learning?\nAssistant:", max_tokens=100)
print(response['choices'][0]['text'])
./llama-cli -m model.gguf -p "User: Hello!\nAssistant:"
@misc{llama32-finetome-1000steps,
author = {Katherine},
title = {Llama 3.2 1B Fine-tuned on FineTome-100k},
year = {2025},
publisher = {HuggingFace},
howpublished = {\url{https://huggingface.co/kkkkkkatherine/llama-3.2-1b-finetome-1000steps-gguf}}
}
Apache 2.0 (same as base model)