© 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. 44798
    Prev
    Next
    INDEX
    Explanations

    references to comparative statistics or contrasts

    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
    heimer
    -0.17
    дап
    -0.16
    à¸Ļาà¸Ķ
    -0.16
    createTime
    -0.15
     Obrázky
    -0.15
    à¸Ľà¸£à¸°à¸ģà¸Ńà¸ļ
    -0.15
    unders
    -0.14
    nett
    -0.14
    efeller
    -0.14
    igans
    -0.14
    POSITIVE LOGITS
     other
    0.51
    other
    0.43
     others
    0.34
    -other
    0.33
    åħ¶ä»ĸ
    0.30
     OTHER
    0.30
     autre
    0.30
     Other
    0.30
     andere
    0.29
    place
    0.29
    Activations Density 0.032%

    No Known Activations