© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 41283
    Prev
    Next
    INDEX
    Explanations

    references to identifiable contributions or traces left by individuals in various 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
    azen
    -0.16
    cept
    -0.14
     perpetual
    -0.14
     polar
    -0.14
    .epam
    -0.14
     Anything
    -0.14
    ìĻ¸
    -0.14
    illet
    -0.13
    ssi
    -0.13
    None
    -0.13
    POSITIVE LOGITS
     all
    0.55
     everywhere
    0.53
     throughout
    0.46
     ALL
    0.37
     Throughout
    0.35
    .all
    0.34
    =all
    0.34
    	all
    0.33
     across
    0.33
     Everywhere
    0.32
    Activations Density 0.053%

    No Known Activations