prototype-0.4x241

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 /workspace/prototype-0.4x239 as a base.

Models Merged

The following models were included in the merge:

  • /workspace/prototype-0.4x231
  • /workspace/prototype-0.4x232
  • /workspace/prototype-0.4x233

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: /workspace/prototype-0.4x233
    parameters:
      weight: 0.25
      density: [0.65, 0.7, 0.75, 0.8]
  - model: /workspace/prototype-0.4x232
    parameters:
      weight: 0.25
      density: [0.7, 0.6, 0.8, 0.7]
  - model: /workspace/prototype-0.4x231
    parameters:
      weight: 0.25
      density: [0.7, 0.8, 0.6, 0.7]
  - model: /workspace/prototype-0.4x239
    parameters:
      weight: 0.25
      density: 0.4
merge_method: dare_ties
base_model: /workspace/prototype-0.4x239
parameters:
  normalize: false
  lambda: [0.9, 1.1, 0.95, 1.2, 0.85, 1.1]
chat_template: llama3
pad_to_multiple_of: 8
int8_mask: true
tokenizer:
 source: base
dtype: bfloat16
Downloads last month
-
Safetensors
Model size
71B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Paper for bruhzair/prototype-0.4x241