© 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. 13850
    Prev
    Next
    INDEX
    Explanations

    if/whether

    np_max-act · gemini-2.0-flash

    This neuron detects expressions of uncertainty or lack of clarity, such as “if,” “whether,” “unclear,” and similar hedging language.

    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
    报道
    -0.07
     dimensions
    -0.06
    _ratio
    -0.06
    На
    -0.06
    ﾝ
    -0.06
     ├
    -0.06
     workforce
    -0.06
    _tid
    -0.06
    _ordered
    -0.06
     theatrical
    -0.06
    POSITIVE LOGITS
     لغ
    0.06
    ğ
    0.06
    lahoma
    0.06
    ющее
    0.06
     Sử
    0.06
     strav
    0.06
    -%
    0.06
    ानम
    0.06
    ”).
    0.06
    	cs
    0.06
    Activations Density 0.034%

    No Known Activations