© 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
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 12595
    Prev
    Next
    INDEX
    Explanations

    olar

    np_max-act · gemini-2.0-flash

    The neuron detects occurrences of “depolarize/ depolarizing/ depolarization” (i.e. the membrane depolarization term) in the text.

    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
    -handler
    -0.07
     searchData
    -0.07
    	Set
    -0.06
    _at
    -0.06
     suitability
    -0.06
    >Password
    -0.06
    adera
    -0.06
     intersection
    -0.06
     Serum
    -0.06
    ardu
    -0.06
    POSITIVE LOGITS
     decentral
    0.06
    :"+
    0.06
    Report
    0.06
     postData
    0.06
     Más
    0.06
    mouth
    0.06
    香蕉
    0.06
    erap
    0.06
    àu
    0.06
     amd
    0.06
    Activations Density 0.001%

    No Known Activations