© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 8703
    Prev
    Next
    INDEX
    Explanations

    references to increases or upward movements

    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
    rens
    -0.20
    hire
    -0.16
    ifton
    -0.16
    ipa
    -0.15
    AsyncResult
    -0.15
     whip
    -0.14
    imo
    -0.14
     rel
    -0.14
    unas
    -0.14
    едж
    -0.14
    POSITIVE LOGITS
    ledge
    0.20
    imentary
    0.20
    elage
    0.17
    inely
    0.17
    olph
    0.16
    enstein
    0.16
    ination
    0.15
    ghest
    0.15
    .GPIO
    0.14
    iform
    0.14
    Activations Density 0.014%

    No Known Activations