© 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. 29602
    Prev
    Next
    INDEX
    Explanations

    references to branding and brand management concepts

    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
    даÑĤ
    -0.16
    lej
    -0.15
    outil
    -0.14
    bic
    -0.14
    leston
    -0.14
    iture
    -0.14
    abilit
    -0.14
    GRES
    -0.14
    .biz
    -0.14
    omba
    -0.13
    POSITIVE LOGITS
     brand
    0.92
     brands
    0.84
    brand
    0.76
     Brand
    0.76
     Brands
    0.72
     BRAND
    0.71
    Brand
    0.71
    -brand
    0.69
    åĵģçīĮ
    0.64
    brands
    0.64
    Activations Density 0.035%

    No Known Activations