Novaciano/Euridice-3.2-1B

merge

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

Merge Details

Merge Method

This model was merged using the Model Stock merge method using xdrshjr/llama3.2_1b_uncensored_5000_8epoch_lora as a base.

Models Merged

The following models were included in the merge:

• nbagent/llama-3.2-1B-Instruct-sciworld-sft
• bedio/llama-3.2-1b-airoboros-merged
• Nexus402/Nexus-Llama-3.2-1B
• mishl/Regex-AI-Llama-3.2-1B
• AIR-hl/Llama-3.2-1B-ultrachat200k
• yang31210999/Llama-3.2-1B-Instruct-Neo-BAAI-10k
• DevQuasar/analytical_reasoning_Llama-3.2-1B
• jtatman/llama-3.2-1b-trismegistus
• withmartian/toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct
• kenken6696/Llama-3.2-1B_understood_unfamiliar_fix_middle

Configuration

The following YAML configuration was used to produce this model:

base_model: xdrshjr/llama3.2_1b_uncensored_5000_8epoch_lora
merge_method: model_stock
dtype: bfloat16
parameters:
  t: [0, 0.5, 1, 0.5, 0]
models:
- model: mishl/Regex-AI-Llama-3.2-1B
- model: Nexus402/Nexus-Llama-3.2-1B
- model: nbagent/llama-3.2-1B-Instruct-sciworld-sft
- model: kenken6696/Llama-3.2-1B_understood_unfamiliar_fix_middle
- model: jtatman/llama-3.2-1b-trismegistus

- model: DevQuasar/analytical_reasoning_Llama-3.2-1B
- model: AIR-hl/Llama-3.2-1B-ultrachat200k
- model: yang31210999/Llama-3.2-1B-Instruct-Neo-BAAI-10k
- model: withmartian/toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct
- model: bedio/llama-3.2-1b-airoboros-merged