© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 12225
    Prev
    Next
    INDEX
    Explanations

    instances of the word "in" indicating a location or context

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
     sense
    -0.18
     hindsight
    -0.15
    sense
    -0.15
    viewer
    -0.14
     glare
    -0.14
    .opens
    -0.14
    verbs
    -0.14
    indsight
    -0.14
     Sense
    -0.14
    γά
    -0.14
    POSITIVE LOGITS
    co
    0.18
    iami
    0.17
    _PORTS
    0.17
     surprise
    0.17
     greeting
    0.17
     excitement
    0.17
     approval
    0.16
    ane
    0.16
     delight
    0.16
    assen
    0.15
    Activations Density 0.034%

    No Known Activations