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

    Neuronpedia

    Jacobian LensNEW
    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. 57233
    Prev
    Next
    INDEX
    Explanations

    quotes

    np_max-act · gemini-2.0-flash

    The neuron activates on first‐person narrative phrasing—“I” plus introspective or action verbs (e.g. “glanced,” “noticed,” “finding”).

    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
     CLICK
    -0.08
    .TXT
    -0.07
     klein
    -0.07
     literary
    -0.06
    ازي
    -0.06
     queue
    -0.06
     mental
    -0.06
     revive
    -0.06
     mailed
    -0.06
    /button
    -0.06
    POSITIVE LOGITS
     �
    0.07
    ніч
    0.07
    _hot
    0.06
     withRouter
    0.06
     Miami
    0.06
    iant
    0.06
    UInteger
    0.06
    (newState
    0.06
    ously
    0.06
    	JLabel
    0.06
    Activations Density 0.177%

    No Known Activations