from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = AutoModelForCausalLM.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")
model = PeftModel.from_pretrained(base, "your-username/TinyLlama-BobDylan-Lyric-Writer")
tokenizer = AutoTokenizer.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")

prompt = "Write a song about rain and regret"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=300)
print(tokenizer.decode(outputs[0]))

Training Details

Base Model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
LoRA Rank: 32
LoRA Alpha: 64
Training Data: Bob Dylan songs dataset (342 examples)
Format: Alpaca instruction-following

Results

Metric	Value
Final Validation Loss	3.91
Final Perplexity	49.89
Total Training Steps	158

Prompt Format

Write a song about [topic]

Example: "Write a song about loneliness and the open road"

culturebunker/TinyLlama-BobDylan-Lyric-Writer

TinyLlama Bob Dylan Lyric Writer (LoRA)

Usage

Training Details

Results

Prompt Format

culturebunker/
TinyLlama-BobDylan-Lyric-Writer