© 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. 33359
    Prev
    Next
    INDEX
    Explanations

    phrases indicating contrast or comparison

    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
    .scalablytyped
    -0.17
    rani
    -0.16
     Kurd
    -0.15
     nuru
    -0.14
    senal
    -0.14
    ÐĬ
    -0.14
    ieten
    -0.14
    lbrace
    -0.14
    unsupported
    -0.13
    ickers
    -0.13
    POSITIVE LOGITS
     him
    0.17
    erras
    0.15
     someone
    0.15
     somebody
    0.15
     Guid
    0.15
    ep
    0.14
     Al
    0.14
    utter
    0.14
     Coun
    0.14
    两人
    0.14
    Activations Density 0.093%

    No Known Activations