© 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. 21846
    Prev
    Next
    INDEX
    Explanations

    references to scheduled television programming and events

    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
    rosso
    -0.15
     staged
    -0.14
     gent
    -0.13
    [rand
    -0.13
     last
    -0.13
    à¹Īาà¸ĩ
    -0.13
     rend
    -0.13
     attending
    -0.12
    urb
    -0.12
     generic
    -0.12
    POSITIVE LOGITS
     exclusively
    0.24
     Channel
    0.18
     on
    0.16
    Viewer
    0.15
     kênh
    0.15
    -exclusive
    0.15
    icamente
    0.15
     exclus
    0.15
    tere
    0.15
    -live
    0.15
    Activations Density 0.054%

    No Known Activations