openaccess-ai-collective_-_dpopenhermes-alpha-v0-4bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

dpopenhermes-alpha-v0 - bnb 4bits

• Model creator: https://huggingface.co/openaccess-ai-collective/
• Original model: https://huggingface.co/openaccess-ai-collective/dpopenhermes-alpha-v0/

Original model description:

license: apache-2.0 base_model: teknium/OpenHermes-2.5-Mistral-7B tags:

• generated_from_trainer model-index:
• name: qlora-out results: []

---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

DPOpenHermes

OpenHermes x NoRobots x Neural

This model is a RL fine-tuned version of teknium/OpenHermes-2.5-Mistral-7B on the Intel/orca_dpo_pairs and winglian/no_robots_rlhf datasets.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 20
• training_steps: 1348

Training results

Framework versions

• Transformers 4.35.2
• Pytorch 2.0.1+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0