Downloads · 30 days
10
11% of all-time downloads
custom-deepseek-r1-4L
MollyHexapotato/custom-deepseek-r1-4L
custom-deepseek-r1-4L is a machine learning model from MollyHexapotato. 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.
⚠️ For Testing Purposes Only This is a modified version of DeepSeek-R1 with random weights, used for architecture experiments.
All-time downloads
91
Public
Parameters
15.1B
30.2 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors30.2 GB · 100%
At a glance
- License
- apache-2.0
- Model type
- deepseek_v3
- Access
- Public
- Created
- Jul 24, 2025
- Updated
- Jul 24, 2025
- SHA
- 7ef0babe
Model card
From the Hugging Face model README
Custom DeepSeek-R1 (4 Layers)
⚠️ For Testing Purposes Only
This is a modified version of DeepSeek-R1 with random weights, used for architecture experiments.
Key Modifications
- Reduced to 4 layers (original: 32+ layers)
- Contains:
- First 3 layers: MLA (Multi-head Latent Attention)
- Layer 4: MoE (Mixture of Experts)
- All weights randomly initialized (not performance-optimized)
Usage
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("MollyHexapotato/custom-deepseek-r1-4L")
- Type
- deepseek_v3
- License
- apache-2.0
- Languages
- en
- Created
- Jul 24, 2025
- Updated
- Jul 24, 2025