Text Generation
Transformers
Safetensors
English
llama
conversational
text-generation-inference
Inference Endpoints
Tulu 2.5 banner image

Model Card for Tulu V2.5 DPO 13B - UltraFeedback Mean

Tulu is a series of language models that are trained to act as helpful assistants. Tulu V2.5 is a series of models trained using DPO and PPO starting from the Tulu 2 suite. This model is trained on UltraFeedback, using the average of the finegrained scores to determine chosen and rejected.

For more details, read the paper: Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback.

.Model description

  • Model type: One model belonging to a suite of RLHF tuned chat models on a mix of publicly available, synthetic and human-created datasets.
  • Language(s) (NLP): English
  • License: Apache 2.0.
  • Finetuned from model: meta-llama/Llama-2-13b-hf

Model Sources

Input Format

The model is trained to use the following format (note the newlines):

<|user|>
Your message here!
<|assistant|>

For best results, format all inputs in this manner. Make sure to include a newline after <|assistant|>, this can affect generation quality quite a bit. We have included a chat template in the tokenizer implementing this template.

Intended uses & limitations

The model was initially fine-tuned on a filtered and preprocessed of the Tulu V2 mix dataset, which contains a diverse range of human created instructions and synthetic dialogues generated primarily by other LLMs. We then further aligned the model with a Jax DPO trainer built on EasyLM on the dataset mentioned above.

Bias, Risks, and Limitations

The Tulu models have not been aligned to generate safe completions within the RLHF phase or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so). It is also unknown what the size and composition of the corpus was used to train the base Llama 2 models, however it is likely to have included a mix of Web data and technical sources like books and code. See the Falcon 180B model card for an example of this.

Training hyperparameters

The following hyperparameters were used during DPO training:

  • learning_rate: 5e-07
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3.0

Citation

If you find Tulu 2.5 is useful in your work, please cite it with:

@misc{ivison2024unpacking,
      title={{Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback}}, 
      author={{Hamish Ivison and Yizhong Wang and Jiacheng Liu and Ellen Wu and Valentina Pyatkin and Nathan Lambert and Yejin Choi and Noah A. Smith and Hannaneh Hajishirzi}}
      year={2024},
      eprint={2406.09279},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}
Downloads last month
63
Safetensors
Model size
13B params
Tensor type
BF16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for allenai/tulu-v2.5-dpo-13b-uf-mean

Finetuned
allenai/tulu-2-13b
Finetuned
(33)
this model
Quantizations
1 model

Datasets used to train allenai/tulu-v2.5-dpo-13b-uf-mean

Collection including allenai/tulu-v2.5-dpo-13b-uf-mean