© 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 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. 28215
    Prev
    Next
    INDEX
    Explanations

    phrases related to perceptions and attitudes towards others

    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
     how
    -0.16
    how
    -0.15
    oret
    -0.14
     two
    -0.14
    gw
    -0.14
    doch
    -0.13
    uit
    -0.13
    zero
    -0.13
    illo
    -0.13
     a
    -0.13
    POSITIVE LOGITS
     themselves
    0.52
     thems
    0.31
     herself
    0.30
     himself
    0.28
     oneself
    0.23
    èĩªå·±
    0.22
     siÄĻ
    0.22
     itself
    0.21
     mình
    0.20
    à¸ķà¸Ļ
    0.20
    Activations Density 0.109%

    No Known Activations