Downloads · 30 days
19
6% of all-time downloads
Custom-Activations-GPT-KAN
AISE-TUDelft/Custom-Activations-GPT-KAN
Custom-Activations-GPT-KAN is a text generation model from AISE-TUDelft. Use it when you need the model to write or continue text. It is set up for transformers.
All-time downloads
319
Public
Repo size
47.3 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin47.3 MB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- Model type
- activations_gpt_neo
- Access
- Public
- Created
- Jun 25, 2024
- Updated
- Jun 25, 2024
- SHA
- 998cd1c2
Try a prompt
Datasets
Model card
From the Hugging Face model README
Basemodel: GPT-Neo
Configs: Vocab size: 10,000 Hidden size: 512 Max position embeddings: 512 Number of layers: 2 Number of heads: 4 Window size: 256 Intermediate-size: 256
Results:
- Task: glue Score: 55.15 Confidence Interval: [52.54, 56.73]
- Task: blimp Score: 55.38 Confidence Interval: [53.68, 56.47]
- Task
- Text Generation
- Library
- transformers
- Type
- activations_gpt_neo
- Created
- Jun 25, 2024
- Updated
- Jun 25, 2024