© 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. 53962
    Prev
    Next
    INDEX
    Explanations

    components of a programming syntax, particularly related to function definitions and HTML manipulation in JavaScript

    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
    ãĥĥ
    -0.16
     w
    -0.15
     Vere
    -0.15
    znam
    -0.14
    vere
    -0.14
    Ïħ
    -0.14
     ""},↵
    -0.13
    aly
    -0.13
    .ws
    -0.13
     Mood
    -0.13
    POSITIVE LOGITS
     ,
    0.16
    elerik
    0.16
     addCriterion
    0.15
    Compose
    0.15
    ahoo
    0.14
    hest
    0.14
    urai
    0.14
    swire
    0.14
    IVITY
    0.14
    ìĶ
    0.14
    Activations Density 0.019%

    No Known Activations