đ LLAMA2-13B-Psyfighter2
Psyfighter is a merged model created by the KoboldAI community, aiming to supplement fictional abilities with medical data. However, due to the data ratio, it's not suitable for medical advice.
đ Quick Start
Psyfighter is a merged model created by the KoboldAI community members Jeb Carter and TwistedShadows, made possible by the KoboldAI merge request service. The intention was to incorporate medical data to enhance the model's fictional capabilities with more details on anatomy and mental states.
⨠Features
- Medical Data Integration: Attempts to add medical data to the model.
- Broad Instruct Format Support: Should support a wide range of instruct formats or plain chat mode.
- Adventure Game Compatibility: Contains a LoRA trained on the same adventure dataset as the KoboldAI Skein model.
đĻ Installation
No installation steps are provided in the original document, so this section is skipped.
đģ Usage Examples
Story Writing
Regular story writing in the traditional way is supported. Simply copy - paste your story and continue writing. Optionally, use an instruction in memory or an author's note to guide the story's direction.
# Generate a story on demand
"Write a novel about X, use chapters and dialogue"
Chatbots and Persona's
You can put an instruction in memory in regular chat mode, for example:
### Instruction:
Generate a conversation between Alice and Jeb where they discuss language models.
In this conversation Henk is excited to teach Alice about Psyfighter.
### Response:
Instruct Prompting
When testing the model, Alpaca format was used. You can try different formats as well. In some cases, adventure data may leak. Using >
as the prefix for a user command can be experimented with, but it may lead to a stronger fiction bias.
Adventuring and Adventure Games
Best done with a small introduction to the world and your objective, using the >
prefix for a user command (KoboldAI's adventure mode). If the model doesn't engage correctly at first, manually correct the output and continue.
đ Documentation
This model contains the following ingredients from their upstream models for as far as we can track them:
- KoboldAI/LLaMA2-13B-Tiefighter
- Undi95/Xwin-MLewd-13B-V0.2
- Undi95/ReMM-S-Light
- Undi95/CreativeEngine
- Brouz/Slerpeno
- elinas/chronos-13b-v2
- jondurbin/airoboros-l2-13b-2.1
- NousResearch/Nous-Hermes-Llama2-13b+nRuaif/Kimiko-v2
- CalderaAI/13B-Legerdemain-L2+lemonilia/limarp-llama2-v2
- KoboldAI/LLAMA2-13B-Holodeck-1
- NousResearch/Nous-Hermes-13b
- OpenAssistant/llama2-13b-orca-8k-3319
- ehartford/WizardLM-1.0-Uncensored-Llama2-13b
- Henk717/spring-dragon
- The-Face-Of-Goonery/Huginn-v3-13b (Contains undisclosed model versions, those we assumed where possible)
- SuperCOT (Undisclosed version)
- elinas/chronos-13b-v2 (Version assumed)
- NousResearch/Nous-Hermes-Llama2-13b
- stabilityai/StableBeluga-13B (Version assumed)
- zattio770/120-Days-of-LORA-v2-13B
- PygmalionAI/pygmalion-2-13b
- Undi95/Storytelling-v1-13B-lora
- TokenBender/sakhi_13B_roleplayer_NSFW_chat_adapter
- nRuaif/Kimiko-v2-13B
- The-Face-Of-Goonery/Huginn-13b-FP16
- "a lot of different models, like hermes, beluga, airoboros, chronos.. limarp"
- lemonilia/LimaRP-Llama2-13B-v3-EXPERIMENT
- Xwin-LM/Xwin-LM-13B-V0.2
- PocketDoc/Dans-RetroRodeo-13b
- Blackroot/Llama-2-13B-Storywriter-LORA
- Doctor-Shotgun/cat-v1.0-13b
- Doctor-Shotgun/llama-2-13b-chat-limarp-v2-merged
- meta-llama/Llama-2-13b-chat-hf
- lemonilia/limarp-llama2-v2
The following mergekit recipe was used:
merge_method: task_arithmetic
base_model: TheBloke/Llama-2-13B-fp16
models:
- model: TheBloke/Llama-2-13B-fp16
- model: KoboldAI/LLaMA2-13B-Tiefighter
parameters:
weight: 1.0
- model: Doctor-Shotgun/cat-v1.0-13b
parameters:
weight: 0.01
- model: Doctor-Shotgun/llama-2-13b-chat-limarp-v2-merged
parameters:
weight: 0.02
dtype: float16
đ§ Technical Details
The model is a merge of multiple models, aiming to combine different capabilities. The focus was on enhancing fictional abilities with some medical data, but it has a high chance of pulling in fictional data due to the data ratio.
đ License
This model is licensed under Llama2.
â ī¸ Important Note
This model should not be used for medical advice or therapy because of its high chance of pulling in fictional data. The information provided by the model can be made up as the focus was on fiction.
đĄ Usage Tip
When using the model, if you want to generate content, less detailed input may lead to better results as the model is good at improvising. For adventuring, use a small introduction to the world and your objective with the >
prefix for a user command. If the model doesn't behave as expected, manually correct the output.
If you discover something cool and want to engage with us, join our community at https://koboldai.org/discord ! We can also provide assistance in making your own merges.