experiment-A

DupletBoreas-7B-t0.0001

A merge of the following models using a custom NearSwap algorithm:

• fearlessdots/WizardLM-2-7B-abliterated
• Virt-io/Erebus-Holodeck-7B

With fearlessdots/WizardLM-2-7B-abliterated as the base model.

Thanks mradermacher for the quants!

• GGUF
• GGUF imatrix

https://huggingface.co/v000000/DupletBoreas-7B-t0.0001-Q8_0-GGUF

https://huggingface.co/v000000/DupletBoreas-7B-t0.0001-Q5_K_S-GGUF

#Fixed
def lerp(a, b, t):
    return a * (1 - t) + b * t

def nearswap(v0, v1, t):
    lweight = np.abs(v0 - v1)
    with np.errstate(divide='ignore', invalid='ignore'):
        lweight = np.where(lweight != 0, t / lweight, 1.0)
    lweight = np.nan_to_num(lweight, nan=1.0, posinf=1.0, neginf=1.0)
    np.clip(lweight, a_min=0.0, a_max=1.0, out=lweight)
    return lerp(v0, v1, lweight)

Credit Alchemonaut

WizardLM-2 adopts the prompt format from Vicuna and supports multi-turn conversation. The prompt should be as following:

A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, 
detailed, and polite answers to the user's questions. USER: Hi ASSISTANT: Hello.</s>
USER: Who are you? ASSISTANT: I am WizardLM.</s>......

Add Genre's in a list like this.

[Genre: <genre1>, <genre2>]