© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 33153
    Prev
    Next
    INDEX
    Explanations

    titles and roles associated with leadership or authority positions

    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
    kla
    -0.14
    anzeigen
    -0.13
     ÄĮech
    -0.13
     Fol
    -0.13
    [from
    -0.13
    //---------------------------------------------------------------------------↵↵
    -0.13
    clare
    -0.13
    PACE
    -0.12
     sobie
    -0.12
    iá»ĥu
    -0.12
    POSITIVE LOGITS
    -turned
    0.15
    -design
    0.14
    -elect
    0.14
    urtle
    0.13
    let
    0.13
    ä¹ĭä¸Ģ
    0.13
    (and
    0.13
    oux
    0.13
     extra
    0.13
    ager
    0.12
    Activations Density 0.207%

    No Known Activations