© 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. 43109
    Prev
    Next
    INDEX
    Explanations

    advice articles

    np_max-act · gemini-2.0-flash

    This neuron fires on the infinitival construction “to find … in …,” identifying user queries that ask where to find something in a particular location.

    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
    NX
    -0.08
    767
    -0.07
     mega
    -0.07
    偏
    -0.06
    》的
    -0.06
    metadata
    -0.06
     qos
    -0.06
     hoch
    -0.06
    Rated
    -0.06
    奶
    -0.06
    POSITIVE LOGITS
    .states
    0.07
    .intersection
    0.06
    ,G
    0.06
     ));↵↵
    0.06
     ogs
    0.06
     Tanner
    0.06
    .mixin
    0.06
     deste
    0.06
     tors
    0.06
    .logout
    0.06
    Activations Density 0.099%

    No Known Activations