© 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. 30204
    Prev
    Next
    INDEX
    Explanations

    parts of speech related to programming concepts and module declarations

    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
    luv
    -0.14
    åIJIJ
    -0.14
     Rich
    -0.14
    avers
    -0.14
    PEND
    -0.14
    illery
    -0.14
    ìĿ´íĬ¸
    -0.13
    iversary
    -0.13
     Delegate
    -0.13
    Rich
    -0.13
    POSITIVE LOGITS
     variables
    0.34
     variable
    0.31
     Variables
    0.29
     declaration
    0.26
     Variable
    0.26
    variables
    0.26
     declarations
    0.25
     global
    0.24
     Declare
    0.24
     Declaration
    0.24
    Activations Density 0.030%

    No Known Activations