The BigWeave models aim to experimentally identify merge settings for increasing model performance. The version number merely tracks various attempts and is not a quality indicator. Only results demonstrating good performance are retained and shared.

Prompting Format

Chatml, Mistral, Vicuna.

Merge process

This is a self-merge of 152334H/miqu-1-70b-sf. Layers are repeated in groups of 4 with a 2 layer overlap. The first and last 8/9 layers are not repeated.

Merge configuration:

slices:
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [0,11]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [9,13]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [11,15]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [13,17]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [15,19]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [17,21]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [19,23]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [21,25]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [23,27]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [25,29]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [27,31]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [29,33]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [31,35]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [33,37]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [35,39]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [37,41]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [39,43]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [41,45]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [43,47]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [45,49]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [47,51]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [49,53]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [51,55]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [53,57]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [55,59]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [57,61]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [59,63]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [61,65]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [63,67]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [65,69]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [67,71]
  - sources:
    - model: 152334H/miqu-1-70b-sf
      layer_range: [69,80]
merge_method: passthrough
dtype: float16

llmixer/BigWeave-v29-122b

BigWeave v29 122b

Prompting Format

Merge process

llmixer/
BigWeave-v29-122b