File size: 1,426 Bytes
22b0dcf |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 |
\# SNP-Universal-Embedding
\_A public-ready embedding model derived from the Substrate-Prism Neuron (SNP) framework.\_
\## Description
SNP-Universal-Embedding is a 6-dimensional embedding model built from emotional geometry principles
originating in the Substrate-Prism Neuron research.
It maps semantic and emotional relationships into compact geometric space — designed for
research in cognitive modeling, decision conflict analysis, and affective reasoning.
\## Key Info
\- \*\*Base model:\*\* BERT-base-uncased
\- \*\*Dimensions:\*\* 6
\- \*\*Purpose:\*\* Compact universal embeddings reflecting emotional \& relational context
\- \*\*Use cases:\*\*
- Compare semantic + affective similarity
- Feed into downstream emotional-reasoning models
- Multi-modal integration (text → cognitive vector space)
\## Example Usage (local)
```python
from transformers import AutoTokenizer, AutoModel
import torch
tokenizer = AutoTokenizer.from\_pretrained("./SNP-Universal-Embedding")
model = AutoModel.from\_pretrained("./SNP-Universal-Embedding")
inputs = tokenizer("A decision between love and duty.", return\_tensors="pt")
with torch.no\_grad():
output = model(\*\*inputs)
embedding = output.last\_hidden\_state.mean(dim=1)
print(embedding.shape) # torch.Size(\[1, 6])
|