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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 54556
    Prev
    Next
    INDEX
    Explanations

    references to military or defense-related topics and political statements

    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
    _EMIT
    -0.16
     Atatürk
    -0.14
    ç¡
    -0.14
    ÃĹ↵↵
    -0.14
    ì±
    -0.14
    ucha
    -0.13
     *č↵
    -0.13
     deltaX
    -0.13
    Americans
    -0.13
    æ°¸ä¹ħ
    -0.12
    POSITIVE LOGITS
     sources
    0.34
    sources
    0.25
     Sources
    0.24
     Prime
    0.22
     source
    0.22
     media
    0.22
     spokes
    0.22
     according
    0.21
     spokesman
    0.21
     spokesperson
    0.21
    Activations Density 0.192%

    No Known Activations