nothashim/CodeSagE

CodeSaga Transformer

This model has been converted to Hugging Face format for easy use with the transformers library.

Usage

from transformers import T5ForConditionalGeneration, T5Tokenizer

model = T5ForConditionalGeneration.from_pretrained("nothashim/CodeSaga")
tokenizer = T5Tokenizer.from_pretrained("nothashim/CodeSaga")

# Generate text
inputs = tokenizer("Your input text", return_tensors="pt")
outputs = model.generate(**inputs, max_length=100)
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)

Model Info

• Architecture: T5-style encoder-decoder
• Vocabulary Size: 1000
• Hidden Size: 128
• Encoder Layers: 2
• Decoder Layers: 2
• Attention Heads: 4