Model Type | |
Use Cases |
Limitations: | The model may produce incorrect, nonsensical, or irrelevant responses. |
|
|
Training Details |
Data Sources: | HuggingFaceH4/ultrafeedback_binarized, Intel/orca_dpo_pairs, argilla/distilabel-math-preference-dpo, Open-Orca/OpenOrca, OpenAssistant/oasst2, HuggingFaceH4/ultrachat_200k, meta-math/MetaMathQA |
|
Context Length: | |
Model Architecture: | MistralForCausalLM with Llama 2 modifications |
|
|
Responsible Ai Considerations |
Fairness: | This model may contain biased, racist, offensive, or inappropriate content. |
|
|
Input Output |
Accepted Modalities: | |
Output Format: | |
Performance Tips: | Apply the HF Tokenizer chat template for better results. |
|
|