© 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. 63644
    Prev
    Next
    INDEX
    Explanations

    descriptors related to contrasting qualities or characteristics

    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
    hots
    -0.14
    )const
    -0.13
    rk
    -0.13
    juries
    -0.13
    ãĥ¬ãĤ¹
    -0.13
    orz
    -0.12
    onas
    -0.12
     <<<
    -0.12
    aug
    -0.12
    quests
    -0.12
    POSITIVE LOGITS
    ness
    0.27
    -looking
    0.17
    NESS
    0.17
     and
    0.16
     ness
    0.16
    lest
    0.15
    -as
    0.15
    ä¸Ķ
    0.15
    ãĢģé«ĺ
    0.14
    est
    0.14
    Activations Density 0.351%

    No Known Activations