ybelkada's picture
ybelkada HF staff
Update README.md
337f685
metadata
license: apache-2.0
tags:
  - trl
  - transformers
  - reinforcement-learning

TRL Model

This is a TRL language model that has been fine-tuned with reinforcement learning to guide the model outputs according to a value, function, or human feedback. The model can be used for text generation.

Training logs

Training logs can be found here

Usage

To use this model for inference, first install the TRL library:

python -m pip install trl

You can then generate text as follows:

from transformers import pipeline

generator = pipeline("text-generation", model="ybelkada//var/tmp/tmpxljq8o84/ybelkada/gpt-neo-125m-detoxified-long-context")
outputs = generator("Hello, my llama is cute")

If you want to use the model for training or to obtain the outputs from the value head, load the model as follows:

from transformers import AutoTokenizer
from trl import AutoModelForCausalLMWithValueHead

tokenizer = AutoTokenizer.from_pretrained("ybelkada//var/tmp/tmpxljq8o84/ybelkada/gpt-neo-125m-detoxified-long-context")
model = AutoModelForCausalLMWithValueHead.from_pretrained("ybelkada//var/tmp/tmpxljq8o84/ybelkada/gpt-neo-125m-detoxified-long-context")

inputs = tokenizer("Hello, my llama is cute", return_tensors="pt")
outputs = model(**inputs, labels=inputs["input_ids"])