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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 44501
    Prev
    Next
    INDEX
    Explanations

    references to financial motivations or transactions involving individuals

    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
     entire
    -0.16
     altogether
    -0.15
     both
    -0.14
    AREST
    -0.14
     --
    -0.13
    ldr
    -0.13
    :async
    -0.13
     entirely
    -0.13
     very
    -0.13
     Neo
    -0.12
    POSITIVE LOGITS
     etc
    0.24
    etc
    0.21
    /etc
    0.17
     ÙħØ«ÙĦا
    0.16
     ÑĤоÑīо
    0.16
    íĨłíĨł
    0.15
    æŁIJ
    0.15
    agrams
    0.14
    ฯ
    0.14
    ãģªãģ©
    0.14
    Activations Density 0.355%

    No Known Activations