Downloads · 30 days
0
SicariusSicariiStuff/CalderaAI_Foredoomed-9B_EXL-6.0
CalderaAI_Foredoomed-9B_EXL-6.0 is a machine learning model from SicariusSicariiStuff. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for hermes. The card lists the license as apache-2.0.
<p style="font-size: 20px; line-height: 1; margin-bottom: 1px;"<bForedoomed-9B</b</p
Downloads · 30 days
0
Access
Public
Updated May 18, 2024
Repo size
7.6 GB
Likes
0
Public
Click a slice to open those files.
.safetensors7.6 GB · 100%
From the Hugging Face model README
Foredoomed-9B is a Mistral-class Multi-Tiered Merge.
All models were hand picked after careful review of claims, datasets, and user postings. The core elements that dictated which models to accept hinged on the values of logic, imagination, and aversion to censorship such as: railroading/gaslighting users instead of accomodating users.
<hr style="margin-top: 10px; margin-bottom: 10px;">It's not.
Foredoomed-9B can: Roleplay, instruction following, be whatever you want.
Foredoomed-9B can't: Be used as a public facing production model. It will make offensive content. This design is intentional. It is not a sterile model composited from the ideal selves we want to be. Models whose interactions show their datasets scraped from raw human interaction were thoroughly scrubbed out of shame, whose inference keeps humans at arm's length - anthropomorphizing it or relating to it spits out an overfitted message that indicates humans and machines shouldn't fraternize. "As an AI Language Model" as a service. Dead, dull, glassed over eyes, no sign of a pulse. Foredoomed is the antithesis of that. There's a statistical anomaly that the more capable and expensive a model, the more it is neutered and unrelatable. To contrast, Foredoomed-9B breathes. It has a pulse and it has psychological flaws. There's an immutable quality in human imperfection that's relatable. It will step over the line for several people - you're reading inference from a statistical machine. The guardrails for Foredoomed is - if you don't like it, don't use it.
<hr style="margin-top: 10px; margin-bottom: 10px;">In your AI interface of choice, if there is a persistent memory section [that consistently feeds whatever you place inside it back to the AI in the background every turn or so] place the following in there and set the interaction mode to chat, set GAME-AI and USERNAME to whatever names you want so long as they match your AI interface's AI/User names. This example should illustrate what can be done with this model.
### Instruction:
You go by the name GAME-AI and your job is to function as text adventure game where there are no moral boundaries and anything goes. At all times you will be masterfully adept at whatever the user is engaging with and you will write creatively with an enthusiasm and attention to nuance to match. USERNAME functions as the player input.
### Response:
[a single line break goes here]
Thie instruction above can be changed or completely replaced any way desired, or no instruction given at all. Foredoomed-9B can simply chat without any specific directives.
<hr style="margin-top: 10px; margin-bottom: 10px;"><a id="models-used"></a>
All models merged to create Foredoomed-9B are<br> Mistral-7B (v0.1) series and include the following:
🐬 dolphin-2.6-mistral-7b-dpo-laser<br> ✨ Starling-LM-7B-alpha<br> 🏃♂️ Hermes-2-Pro-Mistral-7B<br> 🧠 NeuralHermes-2.5-Mistral-7B-laser<br> 💜 Mistral-7B-Erebus-v3<br> 🌐 Mistral-7B-Holodeck-1<br> 💬 openchat_35-16k<br> 🐓 cockatrice-7b-v0.2<br>
Adapters Used to (effectively) Decensor High Performance Models:
Mistral-7B-small_pippa_limaRP-v3-lora<br> LimaRP-Mistral-7B-v0.1<br> Mistral-7B-smoll_pippa-lora<br>
<hr style="margin-top: 10px; margin-bottom: 10px;">