---
license: llama3.1
---
### Merged as below:


-----
slices:
  - sources:

  - 
      - model: athirdpath/Llama-3.1-Instruct_NSFW-pretrained_e1-plus_reddit
      - 
        layer_range: [0, 23]
        
  - sources:

  - 
      - model: athirdpath/Llama-3.1-Techne-RP-8b-v1
      - 
        layer_range: [9, 31]

merge_method: passthrough

dtype: float16

tokenizer_source: athirdpath/Llama-3.1-Techne-RP-8b-v1

-----

Then pretrained for 1 epoch on the Iambe dataset as an 11b model