© 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. 55173
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "run" and its variations used in different contexts

    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
    andom
    -0.17
    igen
    -0.17
    antage
    -0.16
    ANDOM
    -0.16
    lash
    -0.15
    utom
    -0.15
    ih
    -0.14
    igo
    -0.14
    uter
    -0.14
    robe
    -0.14
    POSITIVE LOGITS
     af
    0.40
     rough
    0.28
     ag
    0.27
     Af
    0.25
     counter
    0.25
     foul
    0.24
    .af
    0.22
     am
    0.22
     interference
    0.21
     circles
    0.21
    Activations Density 0.022%

    No Known Activations