© 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. 109126
    Prev
    Next
    INDEX
    Explanations

    Book series and dates

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric tokens denoting publication years, volume numbers, and similar bibliographic indices.

    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
     jaké
    -0.07
    -------------</
    -0.07
     datasets
    -0.07
     Consultants
    -0.06
     підпис
    -0.06
    _Box
    -0.06
     thing
    -0.06
    早
    -0.06
    .accept
    -0.06
    Same
    -0.06
    POSITIVE LOGITS
     grub
    0.07
    .conf
    0.07
     pozem
    0.06
     ];
    0.06
     quảng
    0.06
     kde
    0.06
    jít
    0.06
     gluten
    0.06
    rodu
    0.06
     kval
    0.06
    Activations Density 0.008%

    No Known Activations