© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 13375
    Prev
    Next
    INDEX
    Explanations

    expressions related to interpersonal connections and familiarity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ugi
    -0.15
    ickle
    -0.14
    à¸ļรร
    -0.14
     thing
    -0.14
    ãĢħ
    -0.13
     VÅ¡ech
    -0.13
    ometimes
    -0.13
    unist
    -0.13
    getApplication
    -0.13
    parator
    -0.13
    POSITIVE LOGITS
     someone
    1.39
    someone
    1.20
     somebody
    1.16
     Someone
    1.12
    Someone
    1.09
     alguien
    0.91
     Somebody
    0.83
     jemand
    0.82
    æľī人
    0.63
     nÄĽkdo
    0.59
    Activations Density 0.092%

    No Known Activations