© 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. 73539
    Prev
    Next
    INDEX
    Explanations

    causing or preventing

    np_max-act · gemini-2.0-flash

    descriptions of inventions and technical innovations related to locking mechanisms.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on verbs that signal safety or hazard actions—especially words like “prevent,” “secure,” or “cause.”

    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
    249
    -0.06
     dür
    -0.06
    agger
    -0.06
     apoptosis
    -0.06
    esktop
    -0.06
    audi
    -0.06
    851
    -0.06
    订
    -0.06
    انس
    -0.06
    	ST
    -0.06
    POSITIVE LOGITS
     sexist
    0.07
     Lindsay
    0.07
    %");↵
    0.06
    жень
    0.06
     обрат
    0.06
    -match
    0.06
    :^{↵
    0.06
     생성
    0.06
     директор
    0.06
     ціл
    0.06
    Activations Density 0.179%

    No Known Activations