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. 41822
    Prev
    Next
    INDEX
    Explanations

    theater

    np_max-act · gemini-2.0-flash

    The neuron activates on occurrences of the word “theater” (including the variant “theatre”).

    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
     Simpson
    -0.07
    _Close
    -0.07
    ерап
    -0.06
    _squared
    -0.06
     Discount
    -0.06
     disciple
    -0.06
     Source
    -0.06
    (limit
    -0.06
     diversified
    -0.06
    Mixin
    -0.06
    POSITIVE LOGITS
     Theatre
    0.13
     theater
    0.13
     theatre
    0.13
     Theater
    0.13
     теат
    0.09
     theaters
    0.09
    atre
    0.07
     theat
    0.07
     phía
    0.07
    ater
    0.07
    Activations Density 0.005%

    No Known Activations