© 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. 25052
    Prev
    Next
    INDEX
    Explanations

    instances of communication and authorities' involvement in situations

    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
    usan
    -0.15
    alin
    -0.15
    igen
    -0.15
    ajo
    -0.15
    .libs
    -0.14
    addtogroup
    -0.14
    vise
    -0.14
    Ïĥί
    -0.14
    оÑĩки
    -0.14
     prisoners
    -0.14
    POSITIVE LOGITS
     authorities
    0.36
     police
    0.30
     Authorities
    0.29
    Authorities
    0.24
    èŃ¦å¯Ł
    0.23
     поли
    0.22
     authority
    0.22
     cops
    0.21
    Police
    0.21
     Police
    0.21
    Activations Density 0.054%

    No Known Activations