© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 25431
    Prev
    Next
    INDEX
    Explanations

    transfer-related discussions and player movement in sports contexts

    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
    ooke
    -0.18
    ovich
    -0.16
    ients
    -0.15
    Ymd
    -0.15
    nnen
    -0.15
    º«
    -0.14
    ạt
    -0.14
    .Virtual
    -0.14
     respir
    -0.14
    амеÑĤ
    -0.14
    POSITIVE LOGITS
     move
    0.41
    -move
    0.31
     switch
    0.29
     Move
    0.28
    move
    0.28
     transfer
    0.28
    Move
    0.26
     moves
    0.25
     MOVE
    0.23
     moved
    0.23
    Activations Density 0.031%

    No Known Activations