© 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. 125212
    Prev
    Next
    INDEX
    Explanations

    is

    np_max-act · gemini-2.0-flash

    This neuron detects markers of formal definitional or passive‐voice constructions, e.g. “is defined by.”

    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
     bicycle
    -0.07
     memorandum
    -0.07
     Popup
    -0.06
    _unused
    -0.06
    Connect
    -0.06
    levard
    -0.06
     Yak
    -0.06
     Diploma
    -0.06
    Safe
    -0.06
    Pixmap
    -0.06
    POSITIVE LOGITS
    lessness
    0.07
    ाव
    0.06
    -column
    0.06
     algún
    0.06
    rij
    0.06
    jn
    0.06
     gcc
    0.06
     ajust
    0.06
    个
    0.06
    bul
    0.06
    Activations Density 0.000%

    No Known Activations