© 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. 42403
    Prev
    Next
    INDEX
    Explanations

    phrases related to achieving or accomplishing something

    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
     Out
    -0.18
    zb
    -0.17
    _Out
    -0.16
     Outlet
    -0.15
    eren
    -0.15
    Specifier
    -0.15
    zial
    -0.15
    UPS
    -0.14
    ups
    -0.14
    agua
    -0.13
    POSITIVE LOGITS
     off
    0.90
    -off
    0.72
    off
    0.63
     Off
    0.59
    Off
    0.58
    _off
    0.54
    .off
    0.52
    -offs
    0.47
     OFF
    0.47
     offs
    0.47
    Activations Density 0.032%

    No Known Activations