© 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. 125104
    Prev
    Next
    INDEX
    Explanations

    <?

    np_max-act · gemini-2.0-flash

    The neuron responds to Java generic wildcard type parameters that use an “extends” bound (i.e. occurrences of “<? extends …>”).

    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
    _metadata
    -0.07
    ＿
    -0.07
    .br
    -0.07
    (`
    -0.07
     Men
    -0.06
    -ind
    -0.06
    ();↵↵↵↵
    -0.06
     KUR
    -0.06
     DbSet
    -0.06
     UNION
    -0.06
    POSITIVE LOGITS
    cce
    0.06
     lắng
    0.06
    违
    0.06
     downt
    0.06
    ，並
    0.06
     fades
    0.06
     темп
    0.06
     informational
    0.06
     ye
    0.06
     actresses
    0.06
    Activations Density 0.000%

    No Known Activations