© 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 (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 128893
    Prev
    Next
    INDEX
    Explanations

    Named places or titles

    np_max-act · gemini-2.0-flash

    This neuron detects honorific epithets or nicknames—phrases like “the Albanian Homer,” “Friendly City,” or “Queen of Gypsys” that appear as titles or sobriquets.

    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
    採
    -0.07
    尔
    -0.07
    爾
    -0.06
     podpor
    -0.06
    忽
    -0.06
     honoured
    -0.06
    CNT
    -0.06
    andez
    -0.06
    لعاب
    -0.06
     イ
    -0.06
    POSITIVE LOGITS
    mdir
    0.07
    "){
    0.07
    ufacturer
    0.07
    ً
    0.06
    }$/
    0.06
     frozen
    0.06
     Cat
    0.06
     sebe
    0.06
    GetInt
    0.06
     },{
    0.06
    Activations Density 0.116%

    No Known Activations