kromcomp/L3.1-Charmedv1-8B

charmed

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

Merge Details

Merge Method

This model was merged using the NearSwap merge method using hf-100/Llama-3.1-8b-Spellbound-NaturalWriter-instruct-0.1-chkpt-608-16-bit as a base.

Models Merged

The following models were included in the merge:

• hf-100/Llama-3.1-8b-Spellbound-NaturalWriter-instruct-0.1-chkpt-304-16-bit

Configuration

The following YAML configuration was used to produce this model:

base_model: hf-100/Llama-3.1-8b-Spellbound-NaturalWriter-instruct-0.1-chkpt-608-16-bit
chat_template: llama3
dtype: float32
merge_method: nearswap
parameters:
  t: 0.0001
slices:
- sources:
  - layer_range: [0, 32]
    model: hf-100/Llama-3.1-8b-Spellbound-NaturalWriter-instruct-0.1-chkpt-304-16-bit
  - layer_range: [0, 32]
    model: hf-100/Llama-3.1-8b-Spellbound-NaturalWriter-instruct-0.1-chkpt-608-16-bit
tokenizer:
  pad_to_multiple_of: 4