© 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. 89455
    Prev
    Next
    INDEX
    Explanations

    Numbers and code

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric tokens that contain decimal points (e.g., parts of DOIs, version numbers, or other floating-point values).

    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
     klient
    -0.07
    보았다
    -0.07
    ()/
    -0.07
    ANGES
    -0.07
     hard
    -0.06
    ICC
    -0.06
     средства
    -0.06
     самой
    -0.06
    ائيل
    -0.06
    PEnd
    -0.06
    POSITIVE LOGITS
     nh
    0.06
    ìm
    0.06
     dib
    0.06
    ashington
    0.06
    افق
    0.06
     прог
    0.06
     벽
    0.06
    赢
    0.06
    .erb
    0.06
    (Messages
    0.06
    Activations Density 0.001%

    No Known Activations