README.md

aaa-2-sql

This is a finetuned version of Mistral-7B-Instruct-v0.3 using LoRA with LitGPT.

Training Details

• Base Model: mistralai/Mistral-7B-Instruct-v0.3
• Framework: LitGPT
• Finetuning Method: Low-Rank Adaptation (LoRA)
• LoRA Parameters:
  • Rank (r): 16
  • Alpha: 32
  • Dropout: 0.05
• Quantization: bnb.nf4
• Context Length: 4098 tokens
• Training Steps: 2000

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

# Load model and tokenizer
model = AutoModelForCausalLM.from_pretrained("exaler/aaa-2-sql")
tokenizer = AutoTokenizer.from_pretrained("exaler/aaa-2-sql")

# Create prompt
prompt = "Your prompt here"

# Generate text
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output = model.generate(**inputs, max_new_tokens=1024)
response = tokenizer.decode(output[0], skip_special_tokens=True)
print(response)