© 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. 56036
    Prev
    Next
    INDEX
    Explanations

    speed

    np_max-act · gemini-2.0-flash

    The neuron activates on terms that denote speed or efficiency—e.g. words like “fast,” “rapidly,” “efficient,” and even the “press” in product names like “Xpress.”

    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
    调用
    -0.06
     Craig
    -0.06
     prescriptions
    -0.06
    ereo
    -0.06
     ALOG
    -0.06
     لأن
    -0.06
     Voices
    -0.06
     freak
    -0.06
     hạ
    -0.06
    293
    -0.06
    POSITIVE LOGITS
    volatile
    0.07
    _BAND
    0.07
     vielen
    0.06
     губер
    0.06
     laughed
    0.06
    &_
    0.06
     Donne
    0.06
    (_
    0.06
    /business
    0.06
    _INITIALIZ
    0.06
    Activations Density 0.075%

    No Known Activations