© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 12240
    Prev
    Next
    INDEX
    Explanations

    conditional phrases expressing potential outcomes or hypothetical situations

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
     appears
    -0.15
    HttpException
    -0.15
    æĪIJäºĨ
    -0.14
    enson
    -0.14
     overwhelming
    -0.14
    $__
    -0.13
    illac
    -0.13
     guint
    -0.13
    žil
    -0.13
    unic
    -0.13
    POSITIVE LOGITS
     be
    0.22
    hev
    0.15
    nt
    0.15
     mean
    0.15
     make
    0.15
    mean
    0.15
    MAKE
    0.14
    acci
    0.14
     depend
    0.14
    antt
    0.14
    Activations Density 0.075%

    No Known Activations