schonsense/SOGvorio-s_gguf

VARcrumb

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

Merge Details

Merge Method

This model was merged using the Model Breadcrumbs merge method using Steelskull/L3.3-MS-Nevoria-70b as a base.

Models Merged

The following models were included in the merge:

• Jolly-Q/SOG_10k_70B
• Delta-Vector/Plesio-70B
• schonsense/ll3_3_70B_r128_VAR2

Configuration

The following YAML configuration was used to produce this model:

merge_method: breadcrumbs

models:
   
  - model: schonsense/ll3_3_70B_r128_VAR2
    parameters:
      gamma: 0.05
      density: .5
      weight: 0.33
  - model: Delta-Vector/Plesio-70B
    parameters:
      gamma: 0.05
      density: .5
      weight: 0.33
  - model: Jolly-Q/SOG_10k_70B
    parameters:
      gamma: 0.05
      density: .5
      weight: 0.33
  - model: Steelskull/L3.3-MS-Nevoria-70b

base_model: Steelskull/L3.3-MS-Nevoria-70b
chat_template: llama3
tokenizer_source: union
parameters:
  normalize: false
  int8_mask: true
  lambda: 1.01

dtype: float32
out_dtype: bfloat16

merge_method: breadcrumbs

models:
   
  - model: schonsense/ll3_3_70B_r128_VAR2
    parameters:
      gamma: 0.01
      density: .5
      weight: 0.6
  - model: Delta-Vector/Plesio-70B
    parameters:
      gamma: 0.01
      density: .5
      weight: 0.2
  - model: Jolly-Q/SOG_10k_70B
    parameters:
      gamma: 0.01
      density: .5
      weight: 0.2
  - model: Steelskull/L3.3-MS-Nevoria-70b

base_model: Steelskull/L3.3-MS-Nevoria-70b
chat_template: llama3
tokenizer_source: union
parameters:
  normalize: false
  int8_mask: true
  lambda: 1.01

dtype: float32
out_dtype: bfloat16

merge_method: breadcrumbs

models:
   
  - model: schonsense/ll3_3_70B_r128_VAR2
    parameters:
      gamma: 0.02
      density: .8
      weight: 0.6
  - model: Delta-Vector/Plesio-70B
    parameters:
      gamma: 0.02
      density: .4
      weight: 0.2
  - model: Jolly-Q/SOG_10k_70B
    parameters:
      gamma: 0.02
      density: .8
      weight: 0.2
  - model: Steelskull/L3.3-MS-Nevoria-70b

base_model: Steelskull/L3.3-MS-Nevoria-70b
chat_template: llama3
tokenizer_source: union
parameters:
  normalize: false
  int8_mask: true
  lambda: 1.0

dtype: float32
out_dtype: bfloat16