Kobi কবি Bhab
Collection
I am developing the world's first Bengali poet large language model (LLM) by training a custom tokenizer for Gemma 2. #AISprint
•
4 items
•
Updated
•
1
prompt = """You are tasked with writing a poem in the style of the poet mentioned below. The poem should fit the specified category and adhere to the poet's distinctive tone, language, and themes.
### Poet:
{}
### Category:
{}
### Poem:
{}"""
inputs = tokenizer(
[
prompt.format(
"সুনীল গঙ্গোপাধ্যায়",
"প্রেমমূলক",
"",
)
], return_tensors = "pt").to("cuda")
outputs = model.generate(**inputs, max_new_tokens = 256, use_cache = True)
tokenizer.batch_decode(outputs)
Base model
unsloth/gemma-2-9b-bnb-4bit