RichardErkhov/rbgo_-_Super-phi-2-dpo-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Super-phi-2-dpo - GGUF

• Model creator: https://huggingface.co/rbgo/
• Original model: https://huggingface.co/rbgo/Super-phi-2-dpo/

Name	Quant method	Size
Super-phi-2-dpo.Q2K.gguf	Q2K	1.01GB
Super-phi-2-dpo.Q3KS.gguf	Q3K_S	1.16GB
Super-phi-2-dpo.Q3K.gguf	Q3K	1.38GB
Super-phi-2-dpo.Q3KM.gguf	Q3K_M	1.38GB
Super-phi-2-dpo.Q3KL.gguf	Q3K_L	1.49GB
Super-phi-2-dpo.IQ4XS.gguf	IQ4XS	1.43GB
Super-phi-2-dpo.Q40.gguf	Q40	1.49GB
Super-phi-2-dpo.IQ4NL.gguf	IQ4NL	1.5GB
Super-phi-2-dpo.Q4KS.gguf	Q4K_S	1.5GB
Super-phi-2-dpo.Q4K.gguf	Q4K	1.67GB
Super-phi-2-dpo.Q4KM.gguf	Q4K_M	1.67GB
Super-phi-2-dpo.Q41.gguf	Q41	1.65GB
Super-phi-2-dpo.Q50.gguf	Q50	1.8GB
Super-phi-2-dpo.Q5KS.gguf	Q5K_S	1.8GB
Super-phi-2-dpo.Q5K.gguf	Q5K	1.93GB
Super-phi-2-dpo.Q5KM.gguf	Q5K_M	1.93GB
Super-phi-2-dpo.Q51.gguf	Q51	1.95GB
Super-phi-2-dpo.Q6K.gguf	Q6K	2.13GB
Super-phi-2-dpo.Q80.gguf	Q80	2.75GB

Original model description:

---

base_model: microsoft/phi-2
inference: false
language:

• en

license: mit
model-index:

• name: phi-2

results: []
model_creator: microsoft
model_name: phi-2
model_type: phi
prompt_template: |
<|im_start|>system
{systemmessage}<|imend|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
finetuned_by: Inferless
tags:

• finetune
• rl
• dpo
• phi
• nlp

pipeline_tag: text-generation
datasets:

• argilla/distilabel-intel-orca-dpo-pairs

---

---

#

• Model creator: microsoft
• Original model: phi-2

Description

This repo contains DPO Finetuned model files for Microsoft Phi-2.