© 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. 91600
    Prev
    Next
    INDEX
    Explanations

    audio equipment

    np_max-act · gemini-2.0-flash

    The neuron predominantly activates on numeric tokens and technical model or specification identifiers (e.g. “3.4,” “118,” “GTX-DR”).

    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
     Boards
    -0.07
    luet
    -0.07
    unt
    -0.06
     Fast
    -0.06
     Зап
    -0.06
     Finn
    -0.06
    :mm
    -0.06
    _known
    -0.06
    Fight
    -0.06
    _board
    -0.06
    POSITIVE LOGITS
     %+
    0.07
    _EXIST
    0.06
     Fuß
    0.06
     Virt
    0.06
    .sz
    0.06
     سود
    0.06
    нят
    0.06
     člov
    0.06
    .medium
    0.06
     heterosexual
    0.06
    Activations Density 0.010%

    No Known Activations