© 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. 42589
    Prev
    Next
    INDEX
    Explanations

    group statements

    np_max-act · gemini-2.0-flash

    This neuron activates on deictic and possessive pronouns (e.g. “this,” “our,” “their”) that signal a company’s own actions, products, or responsibilities.

    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
     увид
    -0.08
    drag
    -0.07
    Because
    -0.06
    dale
    -0.06
    され
    -0.06
    (mapping
    -0.06
     فال
    -0.06
    ред
    -0.06
    untos
    -0.06
    Nevertheless
    -0.06
    POSITIVE LOGITS
    ivia
    0.07
     cleric
    0.07
     ['$
    0.06
    _pix
    0.06
     fundament
    0.06
     anonymity
    0.06
    _TextChanged
    0.06
     hurricanes
    0.06
    	record
    0.06
    gba
    0.06
    Activations Density 0.071%

    No Known Activations