© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109305
    Prev
    Next
    INDEX
    Explanations

    programming languages

    np_max-act · gemini-2.0-flash

    The neuron activates on CamelCase identifiers—especially class or type names in code.

    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
    apos
    -0.08
    urement
    -0.07
     drama
    -0.07
     waterfront
    -0.06
     slavery
    -0.06
    able
    -0.06
    -turned
    -0.06
     CASE
    -0.06
     Jersey
    -0.06
    _receive
    -0.06
    POSITIVE LOGITS
     Wouldn
    0.07
    ','%
    0.06
     نب
    0.06
    (()=>{↵
    0.06
     Famil
    0.06
    bew
    0.06
     Challenger
    0.06
    ้ช
    0.06
     kendisine
    0.06
    conut
    0.06
    Activations Density 0.013%

    No Known Activations