README.md

metadata

library_name: transformers
datasets:
  - eswardivi/telugu_instruction_dataset
  - Telugu-LLM-Labs/yahma_alpaca_cleaned_telugu_filtered_and_romanized
  - >-
    Telugu-LLM-Labs/teknium_GPTeacher_general_instruct_telugu_filtered_and_romanized
language:
  - te

Model card for mistral7b_telugu (romanized)

mistralai/Mistral-7B-Instruct-v0.2 is finetuned on tranliterated telugu sentences.

usage

from transformers import pipeline

pipe = pipeline(
    "text-generation", 
    model="eswardivi/mistral7b_telugu", 
    device_map="auto"
)

def create_prompt(instruction: str, input: str = "") -> str:
    prompt = f"""
### Instruction:
{instruction}

### Response:
"""
    return prompt

prompt = create_prompt("Naku python Program 1 to 10 count cheyadaniki ivvu")

out = pipe(
    prompt, 
    num_return_sequences=1, 
    max_new_tokens=1024, 
    temperature=0.7, 
    top_p=0.9, 
    do_sample=True
)

print(out[0]['generated_text'])