Quantization made by Richard Erkhov.

SahabatAI-Rebase-8B-Test - GGUF

Model creator: https://huggingface.co/gmonsoon/
Original model: https://huggingface.co/gmonsoon/SahabatAI-Rebase-8B-Test/

Name	Quant method	Size
SahabatAI-Rebase-8B-Test.Q2_K.gguf	Q2_K	2.96GB
SahabatAI-Rebase-8B-Test.IQ3_XS.gguf	IQ3_XS	0.41GB
SahabatAI-Rebase-8B-Test.IQ3_S.gguf	IQ3_S	3.43GB
SahabatAI-Rebase-8B-Test.Q3_K_S.gguf	Q3_K_S	3.41GB
SahabatAI-Rebase-8B-Test.IQ3_M.gguf	IQ3_M	3.52GB
SahabatAI-Rebase-8B-Test.Q3_K.gguf	Q3_K	3.74GB
SahabatAI-Rebase-8B-Test.Q3_K_M.gguf	Q3_K_M	3.74GB
SahabatAI-Rebase-8B-Test.Q3_K_L.gguf	Q3_K_L	4.03GB
SahabatAI-Rebase-8B-Test.IQ4_XS.gguf	IQ4_XS	4.18GB
SahabatAI-Rebase-8B-Test.Q4_0.gguf	Q4_0	4.34GB
SahabatAI-Rebase-8B-Test.IQ4_NL.gguf	IQ4_NL	4.38GB
SahabatAI-Rebase-8B-Test.Q4_K_S.gguf	Q4_K_S	4.37GB
SahabatAI-Rebase-8B-Test.Q4_K.gguf	Q4_K	4.58GB
SahabatAI-Rebase-8B-Test.Q4_K_M.gguf	Q4_K_M	4.58GB
SahabatAI-Rebase-8B-Test.Q4_1.gguf	Q4_1	4.78GB
SahabatAI-Rebase-8B-Test.Q5_0.gguf	Q5_0	5.21GB
SahabatAI-Rebase-8B-Test.Q5_K_S.gguf	Q5_K_S	5.21GB
SahabatAI-Rebase-8B-Test.Q5_K.gguf	Q5_K	5.34GB
SahabatAI-Rebase-8B-Test.Q5_K_M.gguf	Q5_K_M	5.34GB
SahabatAI-Rebase-8B-Test.Q5_1.gguf	Q5_1	5.65GB
SahabatAI-Rebase-8B-Test.Q6_K.gguf	Q6_K	6.14GB
SahabatAI-Rebase-8B-Test.Q8_0.gguf	Q8_0	7.95GB

Original model description:

base_model:

GoToCompany/llama3-8b-cpt-sahabatai-v1-instruct
mlabonne/Daredevil-8B library_name: transformers tags:
mergekit
merge

merge

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

slices:
- sources:
  - model: GoToCompany/llama3-8b-cpt-sahabatai-v1-instruct
    layer_range:
    - 0
    - 32
  - model: mlabonne/Daredevil-8B
    layer_range:
    - 0
    - 32
merge_method: slerp
base_model: GoToCompany/llama3-8b-cpt-sahabatai-v1-instruct
parameters:
  t:
  - filter: self_attn
    value:
    - 0
    - 0.5
    - 0.3
    - 0.7
    - 1
  - filter: mlp
    value:
    - 1
    - 0.5
    - 0.7
    - 0.3
    - 0
  - value: 0.5
dtype: bfloat16