© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 33306
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "exchange" in various contexts

    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
    ewolf
    -0.16
    rev
    -0.15
    ly
    -0.14
    çek
    -0.14
    lier
    -0.14
    ÑĻ
    -0.14
    mes
    -0.14
    rent
    -0.14
    /exec
    -0.14
    rame
    -0.14
    POSITIVE LOGITS
    able
    0.34
    ability
    0.25
    ables
    0.23
    ably
    0.21
    ABLE
    0.20
    Rate
    0.20
    -rate
    0.20
     rate
    0.19
     Rate
    0.19
     rates
    0.17
    Activations Density 0.013%

    No Known Activations