© 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. 42241
    Prev
    Next
    INDEX
    Explanations

    instances of defiance or resistance against authority

    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
    .setViewport
    -0.17
    .bd
    -0.17
    mando
    -0.16
    éŀ
    -0.15
     Undo
    -0.15
     bekl
    -0.14
     Singleton
    -0.14
    bd
    -0.14
    alley
    -0.14
    699
    -0.14
    POSITIVE LOGITS
     ref
    0.65
     Ref
    0.65
    ref
    0.62
    Ref
    0.59
    -ref
    0.55
     REF
    0.53
    REF
    0.51
    _ref
    0.50
    .ref
    0.45
    _REF
    0.45
    Activations Density 0.064%

    No Known Activations