© 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. 91995
    Prev
    Next
    INDEX
    Explanations

    ict

    np_max-act · gemini-2.0-flash

    The neuron fires on terms containing the “ict” root—especially seizure‐related jargon like “ictal,” “epileptic,” and “EEG.”

    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
    \Desktop
    -0.06
    chemes
    -0.06
     recur
    -0.06
    	line
    -0.06
    enie
    -0.06
     الت
    -0.06
    memberOf
    -0.06
    -being
    -0.06
    vary
    -0.06
     Instead
    -0.05
    POSITIVE LOGITS
     RTS
    0.07
    Fetch
    0.06
     oč
    0.06
    컴
    0.06
    čit
    0.06
    τίου
    0.06
    気に入
    0.06
    ICS
    0.06
     صفحه
    0.06
    inputs
    0.06
    Activations Density 0.002%

    No Known Activations