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. 85561
    Prev
    Next
    INDEX
    Explanations

    qualitative research interviews

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric expressions in study methods—especially participant counts, interview numbers, and other sample‐size figures.

    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
     spurred
    -0.07
    welcome
    -0.07
    olin
    -0.06
     holistic
    -0.06
     (^)(
    -0.06
    <>();↵
    -0.06
    ????
    -0.06
     repr
    -0.06
    icits
    -0.06
     namely
    -0.06
    POSITIVE LOGITS
     reflect
    0.07
    .lab
    0.07
     τά
    0.07
    .images
    0.07
    _LS
    0.07
     ConnectionState
    0.06
     userAgent
    0.06
     vinc
    0.06
    .method
    0.06
    (WIN
    0.06
    Activations Density 0.013%

    No Known Activations