metadata

tags:
  - unsloth

kevin009/llama322

Model Description

This is a LoRA-tuned version of kevin009/llama322 using KTO (Kahneman-Tversky Optimization).

Training Parameters

Learning Rate: 5e-06
Batch Size: 1
Training Steps: 2043
LoRA Rank: 16
Training Date: 2024-12-29

Usage

from peft import AutoPeftModelForCausalLM
from transformers import AutoTokenizer

model = AutoPeftModelForCausalLM.from_pretrained("kevin009/llama322", token="YOUR_TOKEN")
tokenizer = AutoTokenizer.from_pretrained("kevin009/llama322")