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

    Neuronpedia

    Jacobian LensNEW
    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. 56586
    Prev
    Next
    INDEX
    Explanations

    query

    np_max-act · gemini-2.0-flash

    This neuron detects Chinese-language tokens (especially those naming API elements like “基站信息查询” and related parameter terms).

    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
     better
    -0.06
     vrouw
    -0.06
     proti
    -0.06
    avenport
    -0.06
     implants
    -0.06
    ',{
    -0.06
     unde
    -0.06
    ottle
    -0.06
     minors
    -0.06
     gent
    -0.06
    POSITIVE LOGITS
     клас
    0.07
     ι
    0.07
     FontWeight
    0.07
     xử
    0.07
    .flush
    0.06
    _CRE
    0.06
     �
    0.06
    .FloatTensor
    0.06
    .credit
    0.06
     aktiv
    0.06
    Activations Density 0.002%

    No Known Activations