© 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. 13380
    Prev
    Next
    INDEX
    Explanations

    comparative expressions related to quality and performance

    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
     Alonso
    -0.15
     nearer
    -0.15
    алÑĭ
    -0.15
    .bd
    -0.14
    itos
    -0.14
    ATEST
    -0.13
    utton
    -0.13
    umber
    -0.13
    ç¼ĺ
    -0.13
    iba
    -0.13
    POSITIVE LOGITS
     even
    0.26
     slightly
    0.25
     tháºŃm
    0.23
     actually
    0.23
     exceeded
    0.22
     possibly
    0.22
     maybe
    0.22
     sometimes
    0.22
    actually
    0.21
     surpassed
    0.21
    Activations Density 0.013%

    No Known Activations