Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 86050
    Prev
    Next
    INDEX
    Explanations

    comma

    np_max-act · gemini-2.0-flash

    information related to geographic locations and community statistics.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron selectively activates on tokens that are decimal numbers (i.e. numeric values containing a “.”), flagging floating-point numerals in the text.

    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
    _bd
    -0.07
    _STATES
    -0.07
     //↵↵
    -0.06
    cháze
    -0.06
     grosse
    -0.06
     contrad
    -0.06
    //:
    -0.06
     merak
    -0.06
     должны
    -0.06
    Say
    -0.06
    POSITIVE LOGITS
    UILTIN
    0.07
    `
    0.06
     tho
    0.06
    <const
    0.06
    zza
    0.06
    _todo
    0.06
     суду
    0.06
    Cİ
    0.06
    溫
    0.06
    aptops
    0.06
    Activations Density 0.026%

    No Known Activations