LoneStriker/BoreanGale-70B-2.4bpw-h6-exl2

BoreanGale-70B

A merge using a custom algorithm (NearSwap) of:

• 152334H/miqu-1-70b-sf
• Sao10K/WinterGoddess-1.4x-70B-L2

Quants

Several quants are available thanks to community efforts.

Type	Misc	Author
GGUF	iMat Q3	Nexesenex
GGUF	iMat	mradermacher
GGUF	Q8_0	mradermacher

NearSwap Algorithm

NearSwap retains most of the weights of the base model (Miqu), but when a weight is similar between the two, it is interpolated to the secondary model (WinterGoddess) value. A parameter t specifies the sameness threshold. When the distance between two values is below t, the weight from the secondary model (WinterGoddess) is used.

This version of the model uses t = 0.001. At this t, about 10% of weights are fully switched to WinterGoddess. Model quality rapidly degrades above t = 0.0025:

• t = 0.0001 (~0.8% full swap): QuartetAnemoi-70B-t0.0001
• t = 0.0003 (~2% full swap)
• t = 0.001 (~10% full swap): This model
• t = 0.0025 (~18% full swap): Generates one paragraph okay, but then reverts to garbage
• t = 0.005 (~35% full swap): Garbage; semi-related word lists
• t = 0.01 (~55% full swap): Garbage; pseudorandom tokens output

NearSwap implementation:

    t: Union[float, np.ndarray],
    v0: Union[np.ndarray, torch.Tensor],
    v1: Union[np.ndarray, torch.Tensor],
...
    lweight = numpy.absolute(v0-v1)
    lweight = t / lweight
    lweight = numpy.nan_to_num(lweight, nan=1.0, posinf=1.0, neginf=1.0)
    numpy.clip(lweight, a_min=0.0, a_max=1.0, out=lweight)
    res = lerp(lweight,v0,v1)

License and Use

Since the ultimate origin of Miqu is at this time unknown beyond speculation, this model is for noncommercial research use only.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	76.48
AI2 Reasoning Challenge (25-Shot)	73.89
HellaSwag (10-Shot)	89.37
MMLU (5-Shot)	75.19
TruthfulQA (0-shot)	68.6
Winogrande (5-shot)	84.53
GSM8k (5-shot)	67.32