Quantization made by Richard Erkhov.

which8_mistral - GGUF

Model creator: https://huggingface.co/s1ghhh/
Original model: https://huggingface.co/s1ghhh/which8_mistral/

Name	Quant method	Size
which8_mistral.Q2_K.gguf	Q2_K	2.53GB
which8_mistral.IQ3_XS.gguf	IQ3_XS	2.81GB
which8_mistral.IQ3_S.gguf	IQ3_S	2.96GB
which8_mistral.Q3_K_S.gguf	Q3_K_S	2.95GB
which8_mistral.IQ3_M.gguf	IQ3_M	3.06GB
which8_mistral.Q3_K.gguf	Q3_K	3.28GB
which8_mistral.Q3_K_M.gguf	Q3_K_M	3.28GB
which8_mistral.Q3_K_L.gguf	Q3_K_L	3.56GB
which8_mistral.IQ4_XS.gguf	IQ4_XS	3.67GB
which8_mistral.Q4_0.gguf	Q4_0	3.83GB
which8_mistral.IQ4_NL.gguf	IQ4_NL	3.87GB
which8_mistral.Q4_K_S.gguf	Q4_K_S	3.86GB
which8_mistral.Q4_K.gguf	Q4_K	4.07GB
which8_mistral.Q4_K_M.gguf	Q4_K_M	4.07GB
which8_mistral.Q4_1.gguf	Q4_1	4.24GB
which8_mistral.Q5_0.gguf	Q5_0	4.65GB
which8_mistral.Q5_K_S.gguf	Q5_K_S	4.65GB
which8_mistral.Q5_K.gguf	Q5_K	4.78GB
which8_mistral.Q5_K_M.gguf	Q5_K_M	4.78GB
which8_mistral.Q5_1.gguf	Q5_1	5.07GB
which8_mistral.Q6_K.gguf	Q6_K	5.53GB
which8_mistral.Q8_0.gguf	Q8_0	7.17GB

Original model description:

base_model:

HuggingFaceH4/zephyr-7b-beta library_name: transformers tags:
mergekit
merge

round_6

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

Merge Details

Merge Method

This model was merged using the linear merge method.

Models Merged

The following models were included in the merge:

HuggingFaceH4/zephyr-7b-beta
/workspace/0814_nips_db_rebuttal/heur/old_codebase/avg/round_4

Configuration

The following YAML configuration was used to produce this model:

dtype: bfloat16
merge_method: linear
models:
- model: /workspace/0814_nips_db_rebuttal/heur/old_codebase/avg/round_4
  parameters:
    weight: 0.75
- model: HuggingFaceH4/zephyr-7b-beta
  parameters:
    weight: 0.25