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. 88749
    Prev
    Next
    INDEX
    Explanations

    ac

    np_max-act · gemini-2.0-flash

    This neuron detects the prefix “Ac,” especially in capitalized words such as “Acid” or “Acoustic.”

    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
    46
    -0.09
     rủi
    -0.08
     MethodInvocation
    -0.07
     xảy
    -0.07
     následující
    -0.07
     佛
    -0.07
     망
    -0.07
    TableWidgetItem
    -0.07
    .Rule
    -0.07
    Let
    -0.07
    POSITIVE LOGITS
     ac
    0.15
     Ac
    0.15
     AC
    0.12
    AC
    0.11
    Ac
    0.11
    ac
    0.11
    -ac
    0.09
    IC
    0.09
    AK
    0.09
    _ac
    0.09
    Activations Density 0.032%

    No Known Activations