Quantization made by Richard Erkhov.

KoSOLAR-10.9B-v0.3 - GGUF

Model creator: https://huggingface.co/rrw-x2/
Original model: https://huggingface.co/rrw-x2/KoSOLAR-10.9B-v0.3/

Name	Quant method	Size
KoSOLAR-10.9B-v0.3.Q2_K.gguf	Q2_K	3.8GB
KoSOLAR-10.9B-v0.3.IQ3_XS.gguf	IQ3_XS	4.22GB
KoSOLAR-10.9B-v0.3.IQ3_S.gguf	IQ3_S	4.45GB
KoSOLAR-10.9B-v0.3.Q3_K_S.gguf	Q3_K_S	4.42GB
KoSOLAR-10.9B-v0.3.IQ3_M.gguf	IQ3_M	4.59GB
KoSOLAR-10.9B-v0.3.Q3_K.gguf	Q3_K	4.92GB
KoSOLAR-10.9B-v0.3.Q3_K_M.gguf	Q3_K_M	4.92GB
KoSOLAR-10.9B-v0.3.Q3_K_L.gguf	Q3_K_L	5.34GB
KoSOLAR-10.9B-v0.3.IQ4_XS.gguf	IQ4_XS	5.51GB
KoSOLAR-10.9B-v0.3.Q4_0.gguf	Q4_0	5.74GB
KoSOLAR-10.9B-v0.3.IQ4_NL.gguf	IQ4_NL	5.8GB
KoSOLAR-10.9B-v0.3.Q4_K_S.gguf	Q4_K_S	5.78GB
KoSOLAR-10.9B-v0.3.Q4_K.gguf	Q4_K	6.1GB
KoSOLAR-10.9B-v0.3.Q4_K_M.gguf	Q4_K_M	6.1GB
KoSOLAR-10.9B-v0.3.Q4_1.gguf	Q4_1	6.36GB
KoSOLAR-10.9B-v0.3.Q5_0.gguf	Q5_0	6.98GB
KoSOLAR-10.9B-v0.3.Q5_K_S.gguf	Q5_K_S	6.98GB
KoSOLAR-10.9B-v0.3.Q5_K.gguf	Q5_K	7.17GB
KoSOLAR-10.9B-v0.3.Q5_K_M.gguf	Q5_K_M	7.17GB
KoSOLAR-10.9B-v0.3.Q5_1.gguf	Q5_1	7.6GB
KoSOLAR-10.9B-v0.3.Q6_K.gguf	Q6_K	8.3GB
KoSOLAR-10.9B-v0.3.Q8_0.gguf	Q8_0	10.75GB

Original model description:

language: - ko base_model: - LDCC/LDCC-SOLAR-10.7B - hyeogi/SOLAR-10.7B-dpo-v1 tags: - mergekit - merge - LDCC/LDCC-SOLAR-10.7B - hyeogi/SOLAR-10.7B-dpo-v1 license: apache-2.0

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: LDCC/LDCC-SOLAR-10.7B

        layer_range: [0, 48]

      - model: hyeogi/SOLAR-10.7B-dpo-v1

        layer_range: [0, 48]

merge_method: slerp
tokenizer_source: base
base_model: LDCC/LDCC-SOLAR-10.7B

embed_slerp: true

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

Datasets

Finetuned using LoRA with kyujinpy/OpenOrca-KO