© 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. 43781
    Prev
    Next
    INDEX
    Explanations

    modal verbs, specifically variations of the word "can."

    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
    ÂĿ
    -0.17
     �
    -0.15
    $h
    -0.14
    anch
    -0.14
    lette
    -0.14
     Abort
    -0.14
    ÌĨ
    -0.14
    ĵ¨
    -0.14
    edom
    -0.14
    eters
    -0.14
    POSITIVE LOGITS
    't
    0.57
    ’t
    0.53
    ä¸įäºĨ
    0.25
     ikke
    0.25
    'T
    0.24
    ’te
    0.24
    ÃŃt
    0.23
    ’T
    0.23
     not
    0.23
     neither
    0.22
    Activations Density 0.080%

    No Known Activations