README.md

metadata

base_model:
  - NousResearch/Hermes-3-Llama-3.2-3B
  - EpistemeAI/ReasoningCore-Llama-3B-R1-aligned
library_name: transformers
tags:
  - mergekit
  - merge

merge1

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 using merge as a base.

Models Merged

The following models were included in the merge:

• NousResearch/Hermes-3-Llama-3.2-3B
• EpistemeAI/ReasoningCore-Llama-3B-R1-aligned

Configuration

The following YAML configuration was used to produce this model:

merge_method: passthrough
dtype: bfloat16
base_model: merge

slices:
  - sources:
      - model: EpistemeAI/ReasoningCore-Llama-3B-R1-aligned
        layer_range: [0, 12]
        parameters:
          weight: [0.6, 0.65, 0.7]  # Entropy-driven: List simulate
  - sources:
      - model: NousResearch/Hermes-3-Llama-3.2-3B
        layer_range: [8, 22]
        parameters:
          weight: 0.85  # Hyperscaled for transcendent boost
  - sources:
      - model: merge  # Reference the Step 1 output directory/model
        layer_range: [16, 28]
        tensors: ["self_attn"]
        parameters:
          weight: 0.75  # Optimized via simulated genius annealing
        tensors: ["mlp"]
        parameters:
          weight: 0.95
dtype: float16