© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 91641
    Prev
    Next
    INDEX
    Explanations

    4

    np_max-act · gemini-2.0-flash

    This neuron picks out the numeric token “4” (especially in the context of “4.0” or section numbering, as in “Industry 4.0”).

    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
    nim
    -0.06
    rine
    -0.06
    paint
    -0.06
    _walk
    -0.06
    増
    -0.06
     Stap
    -0.06
    _ten
    -0.06
    _des
    -0.06
     Swim
    -0.06
    tip
    -0.05
    POSITIVE LOGITS
    είται
    0.07
     руковод
    0.07
     availability
    0.07
     rejuven
    0.07
    政治
    0.07
    方法
    0.06
    uck
    0.06
     inverted
    0.06
     variables
    0.06
    保护
    0.06
    Activations Density 0.002%

    No Known Activations