Test98

DeepHermes passed through https://huggingface.co/spaces/Naphula/model_tools/blob/main/vocab_resizer_mistral24B.py

karcher still isnt merging assert max(tokenizer.vocab.values()) < vocab_size

trying SLERP with magidonia tokenizer instead

base_model: TheDrummer/Magidonia-24B-v4.2.0
architecture: MistralForCausalLM
merge_method: slerp
dtype: bfloat16
slices:
  - sources:
      - model: TheDrummer/Magidonia-24B-v4.2.0
        layer_range: [0, 40]
      - model: NousResearch/DeepHermes-3-Mistral-24B-Preview
        layer_range: [0, 40]
parameters:
  t: 0.5
tokenizer:
source: TheDrummer/Magidonia-24B-v4.2.0
chat_template: auto

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support