Arsenic-Shahrazad-12B-v4.4

Continued from the 4.3 lineage with Gemma staying securely in her role as judge rather than rewriting directly.

600+ steps RLVR GRPO in spicy roleplaying first-turns, with Gemma 4 31b as judge paying attention to writing craft; followed by DPO including self-rewrites of about half of the particularly bad trajectories from RLVR (until Gemma rated them well in comparison, and particular issues like a godmoded POV were fixed). Ran 5 seeds at low batch size to explore and merged via Karcher mean.

A mix of approaches including utilizing and not-utilizing the <feelings> tags were tried in RLVR. She's trying some interesting things with the tokens, including compound-word emotions that split the difference between word and phrase, but the ratings aren't notably better with the tags than without them - so I wouldn't consider her a reasoning model yet, and it's not in the chat template.

More Arsenic self-portraits:

Amethyst:

Aegis:

)

arsenic-v4.4-merged

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

Merge Details

Merge Method

This model was merged using the Karcher Mean merge method.

Models Merged

The following models were included in the merge:

• /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed1
• /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed2
• /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed5
• /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed4
• /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed3

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed1
  - model: /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed2
  - model: /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed3
  - model: /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed4
  - model: /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed5
merge_method: karcher
parameters:
  normalize: true
  int8_mask: true
tokenizer_source: /home/ann/Documents/Projects/training_models/outputs/baked_v44_craft_seed1
dtype: bfloat16