d_model: 3584
extraction:
  injection_scale: 150.0
extraction_layer_index: 20
kind: nla_model
prompt_templates:
  ar: 'Summary of the following text: <text>{explanation}</text> <summary>'
  av: 'You are a meticulous AI researcher conducting an important investigation into
    activation vectors from a language model. Your overall task is to describe the
    semantic content of that activation vector.


    We will pass the vector enclosed in <concept> tags into your context. You must
    then produce an explanation for the vector, enclosed within <explanation> tags.
    The explanation consists of 2-3 text snippets describing that vector.


    Here is the vector:


    <concept>{injection_char}</concept>


    Please provide an explanation.


    <explanation>'
role: av
schema_version: 2
stage: rl
tokens:
  injection_char: "\u320E"
  injection_token_id: 149705
training:
  ar_critic: output/nla-qwen25-7b-L20-ar-twin
  base_av: output/nla-qwen25-7b-L20-av-twin-clean
  best_mean_reward: 0.08581196278333664
  chat_template: true
  epochs: 6
  group_size: 4
  injection_mode: normalize
  lr: 5.0e-06
  method: grpo
  samples_per_epoch: 100