© 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. 22597
    Prev
    Next
    INDEX
    Explanations

    terms related to personal identification and capabilities

    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
    onga
    -0.18
    ucer
    -0.16
    assin
    -0.15
    utar
    -0.15
    het
    -0.14
    utable
    -0.14
    ungeon
    -0.14
     عÙĨÙĩ
    -0.14
    ekim
    -0.14
    ↵↵
    -0.13
    POSITIVE LOGITS
     of
    0.69
     của
    0.54
    à¸Ĥà¸Ńà¸ĩà¸ľ
    0.50
    à¸Ĥà¸Ńà¸ĩ
    0.40
     ÏĦÏīν
    0.38
     à¸Ĥà¸Ńà¸ĩ
    0.36
    าà¸Ĥà¸Ńà¸ĩ
    0.33
    à¸Ĥà¸Ńà¸ĩà¸Ħ
    0.32
    à¹Įà¸Ĥà¸Ńà¸ĩ
    0.30
    ủa
    0.30
    Activations Density 0.117%

    No Known Activations