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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 128309
    Prev
    Next
    INDEX
    Explanations

    handling choices and problems

    np_max-act · gemini-2.0-flash

    The neuron fires on vivid, visceral disgust actions—especially mentions of forcing oneself (or others) to gag or vomit (e.g. “stick their fingers down their throats”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _TODO
    -0.06
     WALL
    -0.06
    StateToProps
    -0.06
    ası
    -0.06
     Trap
    -0.06
     Dem
    -0.06
    	This
    -0.06
     remarkable
    -0.06
     APC
    -0.06
    _b
    -0.06
    POSITIVE LOGITS
    ็กชาย
    0.06
    홍
    0.06
    英語
    0.06
     specifically
    0.06
    提高
    0.06
     CSC
    0.06
    —
    0.06
     새로운
    0.06
     permitting
    0.06
     Мож
    0.06
    Activations Density 0.122%

    No Known Activations