© 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. 22261
    Prev
    Next
    INDEX
    Explanations

    references to technical specifications or comparisons of products

    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
     wholes
    -0.21
    pherd
    -0.19
    ÑıÑĤелÑĮ
    -0.19
     ambassador
    -0.19
     policym
    -0.19
    ponent
    -0.18
     inventor
    -0.17
     newcomer
    -0.17
    /operator
    -0.17
     teammate
    -0.17
    POSITIVE LOGITS
    ators
    0.28
    ifiers
    0.26
    ainers
    0.25
     shooters
    0.25
    oders
    0.24
    aders
    0.24
    achers
    0.24
     writers
    0.23
    iators
    0.23
    ers
    0.23
    Activations Density 0.058%

    No Known Activations