Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Qwen2.5-7B-IT
    4. Resid Post - 131k
    5. 15-RESID-POST-AA
    6. 20501
    Prev
    Next
    INDEX
    Explanations

    Narratives

    np_max-act · gemini-2.0-flash

    third-person narrative cues, especially pronouns and clause/sentence punctuation that structure story-like prose.

    oai_token-act-pair · gpt-5Triggered by @vetterc0

    third-person narrative prose about people, with strong emphasis on pronouns, character references, and sentence punctuation.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     might
    -0.08
    わけで
    -0.08
    /releases
    -0.07
     gu
    -0.07
    被认为
    -0.07
     vg
    -0.06
     כבר
    -0.06
     التي
    -0.06
    或者说
    -0.06
     entwick
    -0.06
    POSITIVE LOGITS
     Radar
    0.07
    esses
    0.07
     ordained
    0.07
    oled
    0.07
    ﳇ
    0.07
     Wolverine
    0.06
    caler
    0.06
    $total
    0.06
    éal
    0.06
    ernal
    0.06
    Activations Density 0.142%

    No Known Activations