© 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. 74093
    Prev
    Next
    INDEX
    Explanations

    Complete sentences

    np_max-act · gemini-2.0-flash

    The neuron responds primarily to numeric tokens—especially multi‐digit sequences like years or other numbers.

    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
    環
    -0.08
    .players
    -0.06
    之前
    -0.06
     ها
    -0.06
    肌
    -0.06
    βε
    -0.06
    Playlist
    -0.06
    يدة
    -0.06
    Normally
    -0.06
    ona
    -0.06
    POSITIVE LOGITS
     endoth
    0.07
     []).
    0.06
     activist
    0.06
    >;↵↵
    0.06
     Platz
    0.06
    duct
    0.06
    -Cds
    0.06
    ngrx
    0.06
     discern
    0.06
    >`;↵
    0.06
    Activations Density 0.587%

    No Known Activations