© 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. 6992
    Prev
    Next
    INDEX
    Explanations

    strange behavior

    np_max-act · gemini-2.0-flash

    This neuron responds to descriptions of uncanny or unsettling events—particularly phrases like “strange things happening.”

    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
    (Field
    -0.06
     legitimate
    -0.06
     tenth
    -0.06
    иш
    -0.06
    ще
    -0.06
     Achievement
    -0.06
     robbed
    -0.06
    _ball
    -0.06
    只要
    -0.06
    ieves
    -0.06
    POSITIVE LOGITS
     progressive
    0.07
     chlorine
    0.07
    EXTERN
    0.07
    (store
    0.07
    uitka
    0.07
     Einstein
    0.07
    .XRLabel
    0.07
    ceil
    0.06
     dorsal
    0.06
    _when
    0.06
    Activations Density 0.032%

    No Known Activations