© 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. 21570
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the impact of policies on people's lives and well-being

    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
    .getElementsBy
    -0.15
    วà¸ĩศ
    -0.15
    azzo
    -0.15
    chalk
    -0.15
    ãĥ©ãĥĥãĤ¯
    -0.15
     Matching
    -0.14
    gable
    -0.14
    tlement
    -0.14
    oir
    -0.14
    á»ķi
    -0.14
    POSITIVE LOGITS
     protected
    0.23
     safe
    0.22
     winners
    0.22
     safer
    0.21
     Protected
    0.20
     better
    0.20
     worse
    0.19
     Safe
    0.18
     secure
    0.17
     richer
    0.17
    Activations Density 0.039%

    No Known Activations