© 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. 41262
    Prev
    Next
    INDEX
    Explanations

    suggests

    np_max-act · gemini-2.0-flash

    The neuron detects hedge words and speculative phrases that indicate uncertainty or conjecture (e.g. “likely,” “suggests,” “perhaps,” “it has been proposed,” etc.).

    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
    _PB
    -0.07
    Ｃ
    -0.07
    оз
    -0.06
     inquiry
    -0.06
    تى
    -0.06
    تي
    -0.06
    omas
    -0.06
    晨
    -0.06
     alarms
    -0.06
    购买
    -0.06
    POSITIVE LOGITS
    ineTransform
    0.07
    �
    0.07
    shader
    0.07
    ливість
    0.07
     Leer
    0.07
    ��
    0.07
    吨
    0.06
     Ratings
    0.06
    ßer
    0.06
     laser
    0.06
    Activations Density 0.051%

    No Known Activations