© 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. 29891
    Prev
    Next
    INDEX
    Explanations

    -year

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric duration phrases (e.g. “two-year,” “six-month”) specifying the length of contracts or loans.

    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
     上涨
    -0.07
    _pow
    -0.07
    	pt
    -0.07
    四
    -0.07
    ki
    -0.06
    けた
    -0.06
    すす
    -0.06
    ощи
    -0.06
     hot
    -0.06
    ायल
    -0.06
    POSITIVE LOGITS
     """↵↵
    0.06
     öldür
    0.06
     Ваш
    0.06
    نا
    0.06
    wil
    0.05
     الاح
    0.05
    ्वर
    0.05
    노출
    0.05
    reira
    0.05
     ड
    0.05
    Activations Density 0.004%

    No Known Activations