© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 41568
    Prev
    Next
    INDEX
    Explanations

    explosion or fire

    np_max-act · gemini-2.0-flash

    The neuron detects hazard terms related to flammability and explosive risks.

    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
     dealt
    -0.07
    .Buffer
    -0.06
    .robot
    -0.06
     attitudes
    -0.06
     pond
    -0.06
    Head
    -0.06
     Paint
    -0.06
    Paint
    -0.06
    cky
    -0.06
     Behaviour
    -0.06
    POSITIVE LOGITS
     nově
    0.07
     Devin
    0.07
    _LANGUAGE
    0.06
     exploration
    0.06
     erotisch
    0.06
    ถนน
    0.06
    탄
    0.06
     terse
    0.06
     HinderedRotor
    0.06
    fight
    0.06
    Activations Density 0.009%

    No Known Activations