© 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. 109469
    Prev
    Next
    INDEX
    Explanations

    verb "is"

    np_max-act · gemini-2.0-flash

    words related to medical and physiological concepts.

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

    This neuron responds to tokens about material properties and capabilities—most strongly “material” (and its plural) and related words like “enable” and “ability.”

    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
    (pattern
    -0.07
    ुभव
    -0.06
     สำ
    -0.06
     Açık
    -0.06
    ptom
    -0.06
    brid
    -0.06
    /aws
    -0.06
    xEB
    -0.06
    -slot
    -0.06
    vac
    -0.06
    POSITIVE LOGITS
     период
    0.07
    abilidad
    0.07
     lend
    0.07
     taco
    0.07
     inplace
    0.07
    .shopping
    0.07
     сид
    0.06
    intValue
    0.06
    оці
    0.06
    ону
    0.06
    Activations Density 0.086%

    No Known Activations