© 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. 29599
    Prev
    Next
    INDEX
    Explanations

    the phrase "the elephant in the room."

    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
    xima
    -0.18
    κÏģα
    -0.15
    _itr
    -0.14
    ÃŃm
    -0.14
    èĻ
    -0.14
    &m
    -0.14
     surfaces
    -0.14
    Millis
    -0.14
    meric
    -0.13
    æ±Ĺ
    -0.13
    POSITIVE LOGITS
     otherwise
    0.16
     Jar
    0.15
     jar
    0.15
     headlights
    0.15
    isas
    0.15
    opoulos
    0.14
    ased
    0.14
     Otherwise
    0.14
    uel
    0.14
    iqu
    0.14
    Activations Density 0.031%

    No Known Activations