© 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. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 43362
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron activates on user‐interface call-to-action instructions, especially phrases telling readers to click buttons (e.g. “clicking the ‘Report Comment’ button”).

    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
    moved
    -0.08
     FIRE
    -0.07
     та
    -0.07
     �
    -0.07
    727
    -0.07
     Horny
    -0.07
     PATH
    -0.07
    449
    -0.07
     fv
    -0.06
    word
    -0.06
    POSITIVE LOGITS
    	mouse
    0.06
    =view
    0.06
    .relative
    0.06
     daddy
    0.06
     delle
    0.06
    _href
    0.06
    $array
    0.05
    (eval
    0.05
    (cor
    0.05
    )]);↵
    0.05
    Activations Density 0.007%

    No Known Activations