Downloads · 30 days
196
19% of all-time downloads
alea-institute/kl3m-006-170m-checkpoint-37000
kl3m-006-170m-checkpoint-37000 is a text generation model from alea-institute. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
A 170M parameter language model trained on legal agreements using the Muon optimizer with spectral clamping.
Downloads · 30 days
196
19% of all-time downloads
All-time downloads
1K
Public
Parameters
182M
739 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors727 MB · 98%
From the Hugging Face model README
A 170M parameter language model trained on legal agreements using the Muon optimizer with spectral clamping.
alea-institute/kl3m-data-sample-002-shuffledAnalysis of weight matrix conditioning shows excellent manifold quality:
Key Finding: Many attention projection layers (q_proj, o_proj) are actively hitting the spectral clamp ceiling of 2000, indicating the regularization is working as intended to prevent ill-conditioning.
Generates coherent, fluent legal text with no repetition issues. Suitable for legal/contractual content generation and analysis.
from transformers import pipeline
# Create text generation pipeline
generator = pipeline(
"text-generation",
model="alea-institute/kl3m-006-170m-checkpoint-37000",
torch_dtype="auto",
device_map="auto"
)
# Generate text
outputs = generator(
"This Agreement is entered into as of",
max_new_tokens=100,
temperature=0.7,
top_p=0.95,
repetition_penalty=1.15
)
print(outputs[0]['generated_text'])
Alea Institute
For technical details, see the paper: https://arxiv.org/abs/2504.07854
@misc{kl3m2025,
title={KL3M: Knowledge-Guided Language Model Training},
author={Alea Institute},
year={2025},
url={https://arxiv.org/abs/2504.07854},
note={Trained with Muon optimizer and spectral clamping}
}
Apache 2.0