Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 23299
    Prev
    Next
    INDEX
    Explanations

    words indicating a growing trend or change 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
    geries
    -0.14
    ог
    -0.14
    ÎŃλ
    -0.14
    .navigateByUrl
    -0.14
     jadx
    -0.14
    UNET
    -0.14
    ERSIST
    -0.14
    èĹı
    -0.14
    ìį¨
    -0.14
    ÙĪÙĤ
    -0.14
    POSITIVE LOGITS
     becoming
    0.24
    ly
    0.20
     important
    0.19
     sophisticated
    0.18
     become
    0.18
     popular
    0.18
    bec
    0.16
    LY
    0.16
    ernet
    0.16
     became
    0.16
    Activations Density 0.023%

    No Known Activations