© 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. 12365
    Prev
    Next
    INDEX
    Explanations

    instances or phrases that indicate contradiction or transition in narratives

    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
    rowable
    -0.14
    izione
    -0.14
    raz
    -0.14
    orre
    -0.14
     Rebel
    -0.14
    ุย
    -0.13
    atat
    -0.12
    plode
    -0.12
    rió
    -0.12
    INAL
    -0.12
    POSITIVE LOGITS
     return
    1.10
    return
    0.94
     returning
    0.94
     returns
    0.93
     returned
    0.92
     Return
    0.89
    -return
    0.88
    Return
    0.83
     RETURN
    0.83
    è¿ĶåĽŀ
    0.81
    Activations Density 0.045%

    No Known Activations