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. 80787
    Prev
    Next
    INDEX
    Explanations

    growing

    np_max-act · gemini-2.0-flash

    The neuron responds to quantitative data in text, activating on numbers, percentages, growth rates, and other measurement expressions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    statistics and data regarding market trends and economic analysis.

    oai_token-act-pair · gpt-4o-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
    oucí
    -0.07
    .magic
    -0.07
    ersed
    -0.07
     उच
    -0.06
     Crack
    -0.06
     butt
    -0.06
     شخص
    -0.06
    年龄
    -0.06
    _ENCODING
    -0.06
     JNI
    -0.06
    POSITIVE LOGITS
    groupBox
    0.06
    е
    0.06
     dönemde
    0.06
     spanning
    0.06
    Discuss
    0.06
    实施
    0.06
    Holder
    0.06
     Кон
    0.05
    .nextToken
    0.05
    Contr
    0.05
    Activations Density 0.078%

    No Known Activations