© 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 EvalsExportsAPI 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. 43363
    Prev
    Next
    INDEX
    Explanations

    repetitive phrases or expressions that emphasize particular subjects or concepts

    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
    ondo
    -0.14
    λÎŃον
    -0.14
     siguientes
    -0.13
     Incontri
    -0.12
    een
    -0.12
    à¥ľ
    -0.12
    awe
    -0.12
     CGRectGet
    -0.12
    shint
    -0.11
    rest
    -0.11
    POSITIVE LOGITS
     other
    1.07
    other
    0.89
     Other
    0.78
    Other
    0.75
     OTHER
    0.74
    -other
    0.71
     andere
    0.70
    _other
    0.69
    åħ¶ä»ĸ
    0.68
    (other
    0.67
    Activations Density 0.045%

    No Known Activations