© 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. 92639
    Prev
    Next
    INDEX
    Explanations

    would

    np_max-act · gemini-2.0-flash

    The neuron predominantly activates on the word “would.”

    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
    StateToProps
    -0.07
    explained
    -0.07
    ières
    -0.06
     breeds
    -0.06
    .dx
    -0.06
     Marks
    -0.06
     anus
    -0.06
    bps
    -0.06
     asymmetric
    -0.06
    repair
    -0.06
    POSITIVE LOGITS
    "</
    0.07
     Adult
    0.07
     will
    0.07
    이가
    0.06
     would
    0.06
     renting
    0.06
    oen
    0.06
     poder
    0.06
    ":{"
    0.06
     what
    0.06
    Activations Density 0.012%

    No Known Activations