Quantization made by Richard Erkhov.

ECE-TW3-JRGL-V1 - GGUF

Model creator: https://huggingface.co/paloalma/
Original model: https://huggingface.co/paloalma/ECE-TW3-JRGL-V1/

Name	Quant method	Size
ECE-TW3-JRGL-V1.Q2_K.gguf	Q2_K	23.71GB
ECE-TW3-JRGL-V1.Q3_K_S.gguf	Q3_K_S	27.86GB
ECE-TW3-JRGL-V1.Q3_K.gguf	Q3_K	30.99GB
ECE-TW3-JRGL-V1.Q3_K_M.gguf	Q3_K_M	30.99GB
ECE-TW3-JRGL-V1.Q3_K_L.gguf	Q3_K_L	33.67GB
ECE-TW3-JRGL-V1.IQ4_XS.gguf	IQ4_XS	34.64GB
ECE-TW3-JRGL-V1.Q4_0.gguf	Q4_0	36.2GB
ECE-TW3-JRGL-V1.IQ4_NL.gguf	IQ4_NL	36.55GB
ECE-TW3-JRGL-V1.Q4_K_S.gguf	Q4_K_S	36.55GB
ECE-TW3-JRGL-V1.Q4_K.gguf	Q4_K	38.58GB
ECE-TW3-JRGL-V1.Q4_K_M.gguf	Q4_K_M	38.58GB
ECE-TW3-JRGL-V1.Q4_1.gguf	Q4_1	40.2GB
ECE-TW3-JRGL-V1.Q5_0.gguf	Q5_0	44.2GB
ECE-TW3-JRGL-V1.Q5_K_S.gguf	Q5_K_S	44.2GB
ECE-TW3-JRGL-V1.Q5_K.gguf	Q5_K	45.41GB
ECE-TW3-JRGL-V1.Q5_K_M.gguf	Q5_K_M	45.41GB
ECE-TW3-JRGL-V1.Q5_1.gguf	Q5_1	48.2GB
ECE-TW3-JRGL-V1.Q6_K.gguf	Q6_K	52.7GB
ECE-TW3-JRGL-V1.Q8_0.gguf	Q8_0	68.26GB

Original model description:

license: apache-2.0 tags: - merge - mergekit - ShinojiResearch/Senku-70B-Full - 152334H/miqu-1-70b-sf

ECE-TW3-JRGL-V1

This model has been produced by :

Louis Garcia, engineering student at French Engineering School ECE
Matthieu Jollard, engineering student at French Engineering School ECE

Under the supervision of :

Andre-Louis Rochet, Lecturer at ECE & Co-Founder of TW3 Partners
Paul Lemaistre, CTO of TW3 Partners

With the contribution of :

ECE engineering school as sponsor and financial contributor
RunPod as financial contributor

About ECE

ECE, a multi-program, multi-campus, and multi-sector engineering school specializing in digital engineering, trains engineers and technology experts for the 21st century, capable of meeting the challenges of the dual digital and sustainable development revolutions. French Engineering School ECE

Description

ECE-TW3-JRGL-V1 is a merge of the following models using mergekit:

slices:
  - sources:
      - model: ShinojiResearch/Senku-70B-Full
        layer_range: [0, 80]
      - model: 152334H/miqu-1-70b-sf
        layer_range: [0, 80]
merge_method: slerp
base_model: 152334H/miqu-1-70b-sf
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: float16

Results

ECE-TW3-JRGL-v1 scores 83.07 on EQ-Bench V2

@misc{paech2023eqbench, title={EQ-Bench: An Emotional Intelligence Benchmark for Large Language Models}, author={Samuel J. Paech}, year={2023}, eprint={2312.06281}, archivePrefix={arXiv}, primaryClass={cs.CL} }