© 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
    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. 110201
    Prev
    Next
    INDEX
    Explanations

    legal cases with defendants

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of criminal parties, particularly “defendant” and its variants (e.g. “codefendant,” “co‐defendant”).

    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
     Chan
    -0.06
     Duty
    -0.06
     demonstrating
    -0.06
    yy
    -0.06
    ाहर
    -0.06
    prox
    -0.06
    .spin
    -0.06
     DOE
    -0.06
     Hawk
    -0.06
    slice
    -0.06
    POSITIVE LOGITS
     tant
    0.07
    ระด
    0.06
    .Mar
    0.06
    Scroll
    0.06
    _CAN
    0.06
    ีพ
    0.06
    еси
    0.06
    .An
    0.06
    .How
    0.06
     rehabilitation
    0.06
    Activations Density 0.012%

    No Known Activations