© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 2152
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    sentence-ending punctuation, especially closing quotes or final periods that mark emphatic statements or questions.

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

    sentence- and clause-ending punctuation (especially closing quotes and question marks) that mark emphatic or rhetorical statements.

    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
     constructors
    -0.08
    薪酬
    -0.07
    と思います
    -0.07
     toolkit
    -0.07
    吗
    -0.07
    orney
    -0.07
     International
    -0.07
     heavyweight
    -0.07
    enario
    -0.07
    (PRO
    -0.07
    POSITIVE LOGITS
     economies
    0.07
    气血
    0.07
     ACT
    0.07
     Di
    0.06
    tà
    0.06
    ◑
    0.06
    !<
    0.06
     gi
    0.06
     massaggi
    0.06
    ✚
    0.06
    Activations Density 0.176%

    No Known Activations