© 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. 44682
    Prev
    Next
    INDEX
    Explanations

    topics related to availability and access to information or resources

    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
    ierge
    -0.13
    оÑģÑĢед
    -0.13
     dù
    -0.13
    ê°ĢìļĶ
    -0.13
    ÑĢÑĸй
    -0.12
    ordova
    -0.12
    SEX
    -0.12
    obil
    -0.12
    VERTISE
    -0.12
    iscard
    -0.12
    POSITIVE LOGITS
     so
    1.08
    so
    0.59
     så
    0.50
     So
    0.46
    So
    0.45
    æīĢ以
    0.42
    ãģ®ãģ§
    0.42
    å¦ĤæѤ
    0.41
     ÑĤак
    0.40
    "So
    0.39
    Activations Density 0.099%

    No Known Activations