Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 23364
    Prev
    Next
    INDEX
    Explanations

    terms related to change, communication, and safety

    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
    .↵↵
    -0.33
    .↵
    -0.26
    ;↵↵
    -0.24
    ãĢĤ↵↵
    -0.23
    ).↵↵
    -0.22
    ↵↵
    -0.21
    ();↵↵
    -0.21
    :↵↵
    -0.21
    ()↵↵
    -0.20
    /.↵↵
    -0.20
    POSITIVE LOGITS
    malink
    0.20
     seedu
    0.17
    $LANG
    0.17
    aeper
    0.17
    ]]>
    0.16
    _mD
    0.16
     =================================================================
    0.16
    ``↵
    0.15
     ```↵
    0.15
    .Endpoint
    0.15
    Activations Density 0.000%

    No Known Activations