© 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
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 6950
    Prev
    Next
    INDEX
    Explanations

    Android Activity/Fragment classes

    np_max-act · gemini-2.0-flash

    The neuron activates on Java class‐declaration lines (e.g. "public class X extends Y"), especially where an Android component is extended.

    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
     BET
    -0.07
    equiv
    -0.07
     grinding
    -0.07
    209
    -0.06
    allery
    -0.06
     oscill
    -0.06
    hape
    -0.06
    (alpha
    -0.06
    (&
    -0.06
     vznik
    -0.06
    POSITIVE LOGITS
    .selection
    0.06
    odie
    0.06
    lid
    0.06
    Drive
    0.06
     readings
    0.06
    Boot
    0.06
     grub
    0.06
    měr
    0.06
     "`
    0.06
    сии
    0.06
    Activations Density 0.015%

    No Known Activations