© 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. 24932
    Prev
    Next
    INDEX
    Explanations

    phrases related to change and development over time

    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
     Dud
    -0.16
    edback
    -0.15
    ÄĽtÅ¡
    -0.15
    urre
    -0.15
    å¢
    -0.14
    à¥įवव
    -0.14
     Surre
    -0.14
    cripcion
    -0.14
    ¦æĥħ
    -0.14
    insn
    -0.14
    POSITIVE LOGITS
     replace
    0.71
     replacing
    0.65
     sup
    0.64
     replaces
    0.64
     Replace
    0.62
     replacement
    0.60
     replaced
    0.59
    replace
    0.59
    Replace
    0.57
     REPLACE
    0.56
    Activations Density 0.061%

    No Known Activations