metadata

base_model: google/gemma-3-270m-it
library_name: transformers
model_name: Philosopher
tags:
  - generated_from_trainer
  - sft
  - trl
licence: license

Model Card for Philosopher

This model is a fine-tuned version of google/gemma-3-270m-it. It has been trained using TRL.

Quick start

from transformers import pipeline

# Load text-generation pipeline
generator = pipeline(
    "text-generation",
    model="TanishkB/RandomNumberGenerator",
    device=-1   # use 0 if you have GPU
)

print("Chat with it — type 'exit' to quit.")

while True:
    user_input = input(">> ").strip()
    if user_input.lower() in ("exit", "quit"):
        break

    # Build single-turn prompt (no history)
    prompt = f"User: {user_input}\nAssistant:"

    # Generate reply
    response = generator(
        prompt,
        max_new_tokens=64,
        return_full_text=False
    )[0]["generated_text"]

    # Clean up model output (remove repeated labels if any)
    reply = response.strip()
    if reply.lower().startswith("assistant:"):
        reply = reply[len("assistant:"):].strip()

    print(reply)

Training procedure

This model was trained with SFT.

Framework versions

TRL: 0.21.0
Transformers: 4.55.1
Pytorch: 2.6.0+cu124
Datasets: 4.0.0
Tokenizers: 0.21.4

Citations

Cite TRL as:

@misc{vonwerra2022trl,
    title        = {{TRL: Transformer Reinforcement Learning}},
    author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
    year         = 2020,
    journal      = {GitHub repository},
    publisher    = {GitHub},
    howpublished = {\url{https://github.com/huggingface/trl}}
}