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. 29289
    Prev
    Next
    INDEX
    Explanations

    terms related to naming conventions and structure in code development frameworks

    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
    .code
    -0.16
    365
    -0.15
    ึà¸ģ
    -0.15
    æŁĦ
    -0.15
    eyh
    -0.14
     code
    -0.14
    arak
    -0.14
    SocketAddress
    -0.14
    ynos
    -0.14
    arch
    -0.14
    POSITIVE LOGITS
     convention
    0.27
     conventions
    0.24
     suff
    0.23
     camel
    0.22
     snake
    0.22
     Camel
    0.21
    Convention
    0.21
     Suff
    0.20
     Convention
    0.20
    camel
    0.20
    Activations Density 0.034%

    No Known Activations