© 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. 21577
    Prev
    Next
    INDEX
    Explanations

    research findings related to medical implications and advancements in technology

    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
    utherford
    -0.16
    uder
    -0.15
    allet
    -0.15
    utoff
    -0.14
    otu
    -0.14
    .fhir
    -0.14
    uze
    -0.14
    BuilderFactory
    -0.14
    atron
    -0.13
    loor
    -0.13
    POSITIVE LOGITS
     opening
    0.38
     opens
    0.36
     opened
    0.34
    opening
    0.33
    -opening
    0.32
    Opens
    0.30
    Opening
    0.30
     Opening
    0.30
     revolution
    0.30
    opens
    0.28
    Activations Density 0.055%

    No Known Activations