© 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. 25933
    Prev
    Next
    INDEX
    Explanations

    topics related to financial transactions and payments

    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
     Alto
    -0.15
    quam
    -0.15
    dech
    -0.14
    thic
    -0.14
    rome
    -0.14
    ingham
    -0.14
    versible
    -0.14
    quist
    -0.13
    ome
    -0.13
    usable
    -0.13
    POSITIVE LOGITS
     full
    0.32
    -full
    0.25
     what
    0.24
     twice
    0.24
    full
    0.23
     FULL
    0.23
     Full
    0.20
    FULL
    0.20
     gì
    0.20
    (full
    0.19
    Activations Density 0.057%

    No Known Activations