© 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. 44691
    Prev
    Next
    INDEX
    Explanations

    expressions related to societal value and influence

    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
     even
    -0.22
    even
    -0.21
     EVEN
    -0.19
     zwar
    -0.18
     Even
    -0.17
     accordingly
    -0.17
    Even
    -0.17
    çĶļèĩ³
    -0.16
     además
    -0.16
     tháºŃm
    -0.16
    POSITIVE LOGITS
     nonetheless
    0.58
     nevertheless
    0.55
    Nevertheless
    0.48
     Nonetheless
    0.43
     Nevertheless
    0.42
     certainly
    0.31
     toch
    0.30
     still
    0.30
    è¿ĺæĺ¯
    0.30
    Anyway
    0.29
    Activations Density 0.154%

    No Known Activations