© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 11-RESID-POST-AA
    11. 14079
    Prev
    Next
    INDEX
    Explanations

    Meaning/Implication

    np_max-act · gemini-2.0-flash

    The neuron is sensitive to causal‐link words or phrases (e.g. “causes,” “because,” “means,” “caused by”) that mark one event as leading to another.

    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
    security
    -0.07
    PRESS
    -0.07
     Nah
    -0.06
    Unmount
    -0.06
     наслед
    -0.06
    operations
    -0.06
    DESC
    -0.06
     Implement
    -0.06
     glory
    -0.06
    .enterprise
    -0.06
    POSITIVE LOGITS
    ็ว
    0.07
    ,’”
    0.06
    스의
    0.06
     मई
    0.06
     전용면적
    0.06
    ör
    0.06
    'ın
    0.06
    .Category
    0.06
    ekim
    0.06
    ρέπει
    0.06
    Activations Density 0.001%

    No Known Activations