© 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. 15-RESID-POST-AA
    11. 6381
    Prev
    Next
    INDEX
    Explanations

    legal system

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of legal, courtroom, or judicial contexts (references to courts, trials, and the justice system).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Reliable
    -0.07
     SU
    -0.06
    adí
    -0.06
    (Chat
    -0.06
    BuildContext
    -0.06
    blocking
    -0.06
    Social
    -0.06
    Ai
    -0.06
    Le
    -0.06
     lm
    -0.06
    POSITIVE LOGITS
     nominate
    0.08
     marital
    0.06
     zakáz
    0.06
     CGSize
    0.06
     greatness
    0.06
    aria
    0.06
     retrieved
    0.06
     nackte
    0.06
    ーデ
    0.06
     Leave
    0.06
    Activations Density 0.126%

    No Known Activations