Edens-Gate_-_Qwen1.5B-Instruct-Upscale-3.5B-4bits

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Qwen1.5B-Instruct-Upscale-3.5B - bnb 4bits

• Model creator: https://huggingface.co/Edens-Gate/
• Original model: https://huggingface.co/Edens-Gate/Qwen1.5B-Instruct-Upscale-3.5B/

Original model description:

base_model:

• Qwen/Qwen2-1.5B-Instruct library_name: transformers tags:
• mergekit
• merge

---

merge

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

Merge Details

Merge Method

This model was merged using the passthrough merge method.

Models Merged

The following models were included in the merge:

• Qwen/Qwen2-1.5B-Instruct

Configuration

The following YAML configuration was used to produce this model:

slices:
  - sources:
      - model: Qwen/Qwen2-1.5B-Instruct
        layer_range: [0, 24]
  - sources: # add middle layers with residuals scaled to zero
      - model: Qwen/Qwen2-1.5B-Instruct
        layer_range: [8, 24]
        parameters:
          scale:
            - filter: o_proj
              value: 0.0
            - filter: down_proj
              value: 0.0
            - value: 1.0
  - sources: # add middle layers with residuals scaled to zero
      - model: Qwen/Qwen2-1.5B-Instruct
        layer_range: [8, 28]
        parameters:
          scale:
            - filter: o_proj
              value: 0.0
            - filter: down_proj
              value: 0.0
            - value: 1.0
  - sources:
      - model: Qwen/Qwen2-1.5B-Instruct
        layer_range: [24, 28]
merge_method: passthrough
dtype: bfloat16