LLMOX/FiscalOxLLM-stage012

FiscalOxLLM — Stage 0 + 1 + 2

Llama-3.1-8B-Instruct fine-tuned sequentially on Stage 0 + Stage 1 + Stage 2 datasets using QLoRA.

Details

Property	Value
Base Model	meta-llama/Meta-Llama-3.1-8B-Instruct
Fine-Tuning Stage	Stage 0 → 1 → 2
Method	QLoRA (r=16, alpha=32)
Precision	bfloat16

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("LLMOX/FiscalOxLLM-stage012")
tokenizer = AutoTokenizer.from_pretrained("LLMOX/FiscalOxLLM-stage012")

inputs = tokenizer("Your prompt here", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Part of Fiscal Ox LLM Research

This model is part of a multi-stage fine-tuning comparison study.