© 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. 24647
    Prev
    Next
    INDEX
    Explanations

    instances of the word "started" and its variations, indicating beginnings or initiations

    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
     another
    -0.24
     next
    -0.21
     upcoming
    -0.20
     tomorrow
    -0.19
    another
    -0.19
     several
    -0.18
     future
    -0.18
     a
    -0.17
     very
    -0.17
     soon
    -0.17
    POSITIVE LOGITS
    æľĢåĪĿ
    0.17
     initially
    0.16
    以æĿ¥
    0.15
     originally
    0.15
     beri
    0.14
    avez
    0.14
    ÑĢок
    0.14
    ãĥĥãĤ°
    0.14
    è¿Ļ个
    0.13
    .nodeType
    0.13
    Activations Density 0.075%

    No Known Activations