You can deploy this model and start earning money today!

RichardErkhov/
alchemonaut_-_BoreanGale-70B-gguf

Name: alchemonaut_-_BoreanGale-70B-gguf
Availability: InStock

endpoints_compatible

gguf

template

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

BoreanGale-70B - GGUF

Model creator: https://huggingface.co/alchemonaut/
Original model: https://huggingface.co/alchemonaut/BoreanGale-70B/

Name	Quant method	Size
BoreanGale-70B.Q2_K.gguf	Q2_K	15.36GB
BoreanGale-70B.IQ3_XS.gguf	IQ3_XS	2.64GB
BoreanGale-70B.IQ3_S.gguf	IQ3_S	27.86GB
BoreanGale-70B.Q3_K_S.gguf	Q3_K_S	27.86GB
BoreanGale-70B.IQ3_M.gguf	IQ3_M	28.82GB
BoreanGale-70B.Q3_K.gguf	Q3_K	30.99GB
BoreanGale-70B.Q3_K_M.gguf	Q3_K_M	30.99GB
BoreanGale-70B.Q3_K_L.gguf	Q3_K_L	29.86GB
BoreanGale-70B.IQ4_XS.gguf	IQ4_XS	17.81GB
BoreanGale-70B.Q4_0.gguf	Q4_0	36.2GB
BoreanGale-70B.IQ4_NL.gguf	IQ4_NL	36.55GB
BoreanGale-70B.Q4_K_S.gguf	Q4_K_S	36.55GB
BoreanGale-70B.Q4_K.gguf	Q4_K	38.58GB
BoreanGale-70B.Q4_K_M.gguf	Q4_K_M	38.58GB
BoreanGale-70B.Q4_1.gguf	Q4_1	40.2GB
BoreanGale-70B.Q5_0.gguf	Q5_0	44.2GB
BoreanGale-70B.Q5_K_S.gguf	Q5_K_S	44.2GB
BoreanGale-70B.Q5_K.gguf	Q5_K	45.41GB
BoreanGale-70B.Q5_K_M.gguf	Q5_K_M	45.41GB
BoreanGale-70B.Q5_1.gguf	Q5_1	48.2GB
BoreanGale-70B.Q6_K.gguf	Q6_K	52.7GB
BoreanGale-70B.Q8_0.gguf	Q8_0	68.26GB

Original model description:

tags:

merge license: other model-index:
name: BoreanGale-70B results:
- task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics:
  - type: acc_norm value: 73.89 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=alchemonaut/BoreanGale-70B name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics:
  - type: acc_norm value: 89.37 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=alchemonaut/BoreanGale-70B name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics:
  - type: acc value: 75.19 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=alchemonaut/BoreanGale-70B name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics:
  - type: mc2 value: 68.6 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=alchemonaut/BoreanGale-70B name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics:
  - type: acc value: 84.53 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=alchemonaut/BoreanGale-70B name: Open LLM Leaderboard
- task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics:
  - type: acc value: 67.32 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=alchemonaut/BoreanGale-70B name: Open LLM Leaderboard

BoreanGale-70B

A merge using a custom algorithm (NearSwap) of:

Quants

Several quants are available thanks to community efforts.

Type	Misc	Author
GGUF	iMat Q3	Nexesenex
GGUF	iMat	mradermacher
GGUF	Full Set	mradermacher
GGUF	Misc	LoneStriker
exl2	2.4 bpw	LoneStriker
exl2	3.5 bpw	LoneStriker
exl2	4.0 bpw	LoneStriker
exl2	4.65 bpw	LoneStriker

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

RichardErkhov/alchemonaut_-_BoreanGale-70B-gguf