© 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. 28526
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly periods, indicating the end of sentences

    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
    [
    -0.15
    *
    -0.15
    .eval
    -0.15
    a
    -0.14
    it
    -0.14
    'n
    -0.14
    ll
    -0.14
    av
    -0.14
    anj
    -0.13
    ër
    -0.13
    POSITIVE LOGITS
    00
    0.22
    000
    0.18
    istrovstvÃŃ
    0.16
    yonel
    0.16
    styleType
    0.16
    rais
    0.15
    /Branch
    0.14
    0
    0.14
    rippling
    0.14
    olean
    0.14
    Activations Density 0.051%

    No Known Activations