© 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. 42976
    Prev
    Next
    INDEX
    Explanations

    nouns that describe specific physical entities or attributes

    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
    ytut
    -0.17
    olley
    -0.17
    quets
    -0.16
    hea
    -0.16
    oldemort
    -0.15
    unci
    -0.15
    zcze
    -0.15
    .toObject
    -0.15
    arnation
    -0.15
    Ñĩенко
    -0.14
    POSITIVE LOGITS
    izer
    0.32
    ides
    0.30
    izing
    0.30
    ized
    0.28
    ate
    0.27
    ite
    0.27
    ine
    0.26
    ake
    0.26
    ide
    0.26
    izers
    0.24
    Activations Density 0.100%

    No Known Activations