Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 130451
    Prev
    Next
    INDEX
    Explanations

    Electrically conductive materials

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of electrical conductivity, especially words like “conduct,” “conductive,” or “conductivity” describing conductive materials.

    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
    Cs
    -0.07
     Kr
    -0.07
    	true
    -0.06
    输
    -0.06
     hardcoded
    -0.06
    言った
    -0.06
     نار
    -0.06
     meno
    -0.06
    Kr
    -0.06
     pups
    -0.06
    POSITIVE LOGITS
    .Scroll
    0.08
    ROUND
    0.08
    ieval
    0.07
     ERROR
    0.06
     kiếm
    0.06
    ,message
    0.06
    eventId
    0.06
     Oakland
    0.06
    cept
    0.06
     council
    0.06
    Activations Density 0.043%

    No Known Activations