© 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 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. 28298
    Prev
    Next
    INDEX
    Explanations

    mentions of time-related phrases indicating age or duration

    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
    DEST
    -0.15
    _hierarchy
    -0.14
     sooner
    -0.14
    £¨
    -0.13
     tslib
    -0.13
    ocking
    -0.13
    ÙĪÙĩ
    -0.13
     EAR
    -0.13
    ÑĢоÑģÑĤ
    -0.12
    lav
    -0.12
    POSITIVE LOGITS
     old
    0.58
     older
    0.52
    old
    0.46
    -old
    0.46
    older
    0.42
     OLD
    0.41
     Older
    0.38
    _old
    0.38
    èĢģ
    0.38
     Old
    0.36
    Activations Density 0.026%

    No Known Activations