© 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. 13091
    Prev
    Next
    INDEX
    Explanations

    removing

    np_max-act · gemini-2.0-flash

    The neuron is detecting terms related to content removal or takedown (e.g., “remove,” “removing,” “takedown”).

    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
    ลอง
    -0.06
    uali
    -0.06
    ola
    -0.06
    meno
    -0.06
     حين
    -0.06
     moderately
    -0.06
    ớm
    -0.06
    LEGRO
    -0.06
    ound
    -0.06
     Heights
    -0.06
    POSITIVE LOGITS
     하면
    0.08
    .transfer
    0.07
    _ax
    0.07
     ايران
    0.07
     Accred
    0.07
    ,:),
    0.07
     ''),
    0.07
     expr
    0.06
    -gun
    0.06
     eventdata
    0.06
    Activations Density 0.010%

    No Known Activations