© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 12188
    Prev
    Next
    INDEX
    Explanations

    phrases expressing warnings and consequences

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    ัมà¸ŀ
    -0.15
     prostituer
    -0.14
    akit
    -0.14
    ilig
    -0.14
    stdafx
    -0.14
    à¹Ģà¸
    -0.14
     Washer
    -0.14
    853
    -0.14
    illard
    -0.13
    -alist
    -0.13
    POSITIVE LOGITS
     mess
    0.28
    cross
    0.27
     crossed
    0.27
     cross
    0.26
     crossing
    0.26
     challenge
    0.26
     crosses
    0.26
    æķ¢
    0.25
     Cross
    0.24
     CROSS
    0.23
    Activations Density 0.055%

    No Known Activations