Commit
•
60eef1b
1
Parent(s):
cf406ac
Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,101 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
|
| 4 |
+
base_model: teknium/OpenHermes-2.5-Mistral-7B
|
| 5 |
+
tags:
|
| 6 |
+
- mistral
|
| 7 |
+
- instruct
|
| 8 |
+
- finetune
|
| 9 |
+
- chatml
|
| 10 |
+
model-index:
|
| 11 |
+
- name: OpenHermes-Emojitron-001
|
| 12 |
+
results: []
|
| 13 |
+
license: apache-2.0
|
| 14 |
+
language:
|
| 15 |
+
- en
|
| 16 |
+
|
| 17 |
+
---
|
| 18 |
+
|
| 19 |
+
[](https://hf.co/QuantFactory)
|
| 20 |
+
|
| 21 |
+
|
| 22 |
+
# QuantFactory/OpenHermes-Emojitron-001-GGUF
|
| 23 |
+
This is quantized version of [movaxbx/OpenHermes-Emojitron-001](https://huggingface.co/movaxbx/OpenHermes-Emojitron-001) created using llama.cpp
|
| 24 |
+
|
| 25 |
+
# Original Model Card
|
| 26 |
+
|
| 27 |
+
|
| 28 |
+
# OpenHermes-Emojitron-001
|
| 29 |
+
|
| 30 |
+
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
*🧑🎓📜🔄🌌➡️🗣️🌠🔮🖊️👀🤝🔄🧩🛠️👂📡🔐🌟*
|
| 34 |
+
|
| 35 |
+
## Model description
|
| 36 |
+
|
| 37 |
+
OpenHermes-Emojitron-001 is a state of the art Emoji Fine-tune, based on the OpenHermes 2.5 model.
|
| 38 |
+
|
| 39 |
+
It only speaks in emoji. Mostly. If using for chat, start it off with a nice 😊 to get going.
|
| 40 |
+
|
| 41 |
+
Huge thank you to Teknium and Mistral for a wonderful model.
|
| 42 |
+
|
| 43 |
+
Follow all my updates in ML and AI on Twitter: https://twitter.com/mov_axbx
|
| 44 |
+
|
| 45 |
+
|
| 46 |
+
# Table of Contents
|
| 47 |
+
1. [Example Outputs](#example-outputs)
|
| 48 |
+
- [Geography knowlege](#geography)
|
| 49 |
+
- [Chat about food](#food-chat)
|
| 50 |
+
- [Talk about movies](#movie-talk)
|
| 51 |
+
2. [Benchmark Results](#benchmark-results)
|
| 52 |
+
3. [Prompt Format](#prompt-format)
|
| 53 |
+
4. [Quantized Models](#quantized-models)
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
## Example Outputs
|
| 57 |
+
### Geography knowlege:
|
| 58 |
+
in what country is london
|
| 59 |
+
```
|
| 60 |
+
🇬🇧
|
| 61 |
+
```
|
| 62 |
+
|
| 63 |
+
### Chat about food:
|
| 64 |
+
what country has the best food
|
| 65 |
+
```
|
| 66 |
+
🤔🌎🍽️
|
| 67 |
+
|
| 68 |
+
1️⃣🇮🇳
|
| 69 |
+
|
| 70 |
+
2️⃣🇯🇵
|
| 71 |
+
|
| 72 |
+
3️⃣🇺🇸
|
| 73 |
+
|
| 74 |
+
4️⃣🇪🇸
|
| 75 |
+
|
| 76 |
+
5️⃣🇫🇷
|
| 77 |
+
```
|
| 78 |
+
|
| 79 |
+
### Talk about movies:
|
| 80 |
+
|
| 81 |
+
tell me a time travel story about a delorean
|
| 82 |
+
```
|
| 83 |
+
🛸🚗🕰️👨🔬🔍📅1985👴🚗💭🔜🌌😱💡💼🔢🔜📅1955🚗👨👩👧👦👀🔜🌉🕰️💥🌀🚗🕰️🔂📅1985👨🔬🎉
|
| 84 |
+
```
|
| 85 |
+
|
| 86 |
+
## Benchmark Results
|
| 87 |
+
|
| 88 |
+
There are no benchmarks for emoji models. Maybe someone can create one. EmojiBench 5K let's gooooooo
|
| 89 |
+
|
| 90 |
+
# Prompt Format
|
| 91 |
+
|
| 92 |
+
OpenHermes-Emojitron-001 uses ChatML as the prompt format, just like Open Hermes 2.5
|
| 93 |
+
|
| 94 |
+
It also appears to handle Mistral format great. Especially since I used that for the finetune (oops)
|
| 95 |
+
|
| 96 |
+
|
| 97 |
+
# Quantized Models:
|
| 98 |
+
|
| 99 |
+
Coming soon if TheBloke thinks this is worth his 🕰️
|
| 100 |
+
|
| 101 |
+
|