© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109250
    Prev
    Next
    INDEX
    Explanations

    introductions

    np_max-act · gemini-2.0-flash

    The neuron fires on first- and second-person pronouns and related tokens (“me”, “you”, “looking”, etc.), i.e. direct personal/addressing language.

    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
     therapists
    -0.07
     controlled
    -0.07
    _unsigned
    -0.07
     Tcl
    -0.06
     OT
    -0.06
     articulated
    -0.06
    579
    -0.06
    厂
    -0.06
     filed
    -0.06
    نم
    -0.06
    POSITIVE LOGITS
    ดา
    0.07
    ної
    0.07
     Stations
    0.06
    .dm
    0.06
    ουλίου
    0.06
     essays
    0.06
    ivní
    0.06
    avec
    0.06
     Quản
    0.06
    цять
    0.06
    Activations Density 0.044%

    No Known Activations