Quantization made by Richard Erkhov.

DolphinLake-7B - GGUF

Model creator: https://huggingface.co/Noodlz/
Original model: https://huggingface.co/Noodlz/DolphinLake-7B/

Name	Quant method	Size
DolphinLake-7B.Q2_K.gguf	Q2_K	2.53GB
DolphinLake-7B.IQ3_XS.gguf	IQ3_XS	2.81GB
DolphinLake-7B.IQ3_S.gguf	IQ3_S	2.96GB
DolphinLake-7B.Q3_K_S.gguf	Q3_K_S	2.95GB
DolphinLake-7B.IQ3_M.gguf	IQ3_M	3.06GB
DolphinLake-7B.Q3_K.gguf	Q3_K	3.28GB
DolphinLake-7B.Q3_K_M.gguf	Q3_K_M	3.28GB
DolphinLake-7B.Q3_K_L.gguf	Q3_K_L	3.56GB
DolphinLake-7B.IQ4_XS.gguf	IQ4_XS	3.67GB
DolphinLake-7B.Q4_0.gguf	Q4_0	3.83GB
DolphinLake-7B.IQ4_NL.gguf	IQ4_NL	3.87GB
DolphinLake-7B.Q4_K_S.gguf	Q4_K_S	3.86GB
DolphinLake-7B.Q4_K.gguf	Q4_K	4.07GB
DolphinLake-7B.Q4_K_M.gguf	Q4_K_M	4.07GB
DolphinLake-7B.Q4_1.gguf	Q4_1	4.24GB
DolphinLake-7B.Q5_0.gguf	Q5_0	4.65GB
DolphinLake-7B.Q5_K_S.gguf	Q5_K_S	4.65GB
DolphinLake-7B.Q5_K.gguf	Q5_K	4.78GB
DolphinLake-7B.Q5_K_M.gguf	Q5_K_M	4.78GB
DolphinLake-7B.Q5_1.gguf	Q5_1	5.07GB
DolphinLake-7B.Q6_K.gguf	Q6_K	5.53GB
DolphinLake-7B.Q8_0.gguf	Q8_0	7.17GB

Original model description:

license: apache-2.0

My first successful Dare-Ties merge. Because of the tokenizer difference of the model types (also bf16 vs f16), Had to use Slerp as well.

Seems to perform well! Did a local lm-eval and HellaSWAG gives me around 84.5, which seems decent. will be submitting this for eval on the openLLM leaderboard as well.

Preset for this should be ChatML, but standard default presets should work ok too.

base_model:

senseable/WestLake-7B-v2
cognitivecomputations/dolphin-2.8-mistral-7b-v02 library_name: transformers tags:
mergekit
merge

Noodlz_DolphinLake-DARE_TIE_SLERP-tokenwest

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using cognitivecomputations/dolphin-2.8-mistral-7b-v02 as a base.

Models Merged

The following models were included in the merge:

senseable/WestLake-7B-v2

Configuration

The following YAML configuration was used to produce this model:

merge_method: dare_ties

parameters:
  int8_mask: true
  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 # fallback for rest of tensors
  embed_slerp: true

models:
  - model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
    # No parameters necessary for base model
  - model: senseable/WestLake-7B-v2
    parameters:
      density: 0.58
      weight: 0.8

base_model: cognitivecomputations/dolphin-2.8-mistral-7b-v02

tokenizer_source: model:senseable/WestLake-7B-v2

dtype: bfloat16