© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 110284
    Prev
    Next
    INDEX
    Explanations

    destroy

    np_max-act · gemini-2.0-flash

    The neuron activates on words built from the root “destroy” (e.g., “destroy,” “destroyer,” “destruction”).

    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
     lone
    -0.08
     climbed
    -0.08
     elevate
    -0.08
     page
    -0.07
     palms
    -0.07
     calibrated
    -0.07
     plausible
    -0.07
    Page
    -0.07
     stepping
    -0.07
     alta
    -0.07
    POSITIVE LOGITS
    destroy
    0.10
     destroyed
    0.10
     Destroy
    0.10
     destroy
    0.09
    ılmaktadır
    0.09
     Destruction
    0.09
     destruction
    0.09
     destructive
    0.08
    rypton
    0.08
    .DisplayStyle
    0.07
    Activations Density 0.013%

    No Known Activations