© 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. 8695
    Prev
    Next
    INDEX
    Explanations

    concepts related to organizational efforts, systems, and evaluations

    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
    .↵
    -0.14
    irit
    -0.13
    ácil
    -0.13
    lech
    -0.12
    大人
    -0.12
     zel
    -0.12
    ÑĤаб
    -0.12
    leaflet
    -0.12
    ifu
    -0.12
     plá
    -0.12
    POSITIVE LOGITS
     to
    0.27
     Äijá»ĥ
    0.23
     ÑĩÑĤобÑĭ
    0.20
     called
    0.20
     длÑı
    0.20
     aimed
    0.19
     for
    0.19
    called
    0.18
     Ñīоб
    0.18
    ãĢįãĤĴ
    0.17
    Activations Density 0.591%

    No Known Activations