© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 29785
    Prev
    Next
    INDEX
    Explanations

    spatial references and directions related to positioning

    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
     frau
    -0.15
    adb
    -0.14
    itur
    -0.14
    .si
    -0.14
    iversit
    -0.14
    inke
    -0.13
    abal
    -0.13
     escorte
    -0.13
    amat
    -0.13
     suce
    -0.13
    POSITIVE LOGITS
    /right
    0.20
    /left
    0.18
    wards
    0.17
    -facing
    0.17
    CTest
    0.17
    ward
    0.16
     near
    0.16
     away
    0.15
    near
    0.15
     opposite
    0.15
    Activations Density 0.097%

    No Known Activations