© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 109730
    Prev
    Next
    INDEX
    Explanations

    company

    np_max-act · gemini-2.0-flash

    This neuron activates on corporate and organizational terms—words like “company,” “policies,” “employees,” and similar business-context vocabulary.

    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
    urray
    -0.07
     Murray
    -0.07
    (recipe
    -0.07
    .chk
    -0.07
     BLE
    -0.06
    God
    -0.06
     Denied
    -0.06
     IMG
    -0.06
     等
    -0.06
     همسر
    -0.06
    POSITIVE LOGITS
    !↵
    0.06
    Clinical
    0.06
    /";↵
    0.06
     <>↵
    0.06
     interfere
    0.06
    реп
    0.05
    _DEFINE
    0.05
    '}}>↵
    0.05
    -с
    0.05
    нез
    0.05
    Activations Density 0.042%

    No Known Activations