Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 26783
    Prev
    Next
    INDEX
    Explanations

    and

    np_max-act · gemini-2.0-flash

    This neuron detects key domain or category labels—specialized nouns that describe the type or field of something (e.g. CAD, festival, youth conference, kickboxing event).

    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
     Fakült
    -0.07
     Yog
    -0.06
     Hasan
    -0.06
     Lok
    -0.06
    Mahon
    -0.06
    Illustr
    -0.06
    ίο
    -0.06
    .)
    -0.06
     MouseEventArgs
    -0.06
     Ritual
    -0.06
    POSITIVE LOGITS
    inue
    0.07
    ीव
    0.07
    äl
    0.07
    _do
    0.06
     alimentos
    0.06
    Org
    0.06
    stub
    0.06
    vé
    0.06
     postData
    0.06
     debit
    0.06
    Activations Density 0.087%

    No Known Activations