© 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. 29722
    Prev
    Next
    INDEX
    Explanations

    structures related to professional roles and the creative process

    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
     accr
    -0.15
    linger
    -0.14
    ÑĥÑģÑĤа
    -0.14
    Å¡ti
    -0.14
    оба
    -0.14
     segreg
    -0.13
    aira
    -0.13
    mite
    -0.13
    omb
    -0.13
    еÑģÑı
    -0.13
    POSITIVE LOGITS
    mitted
    0.18
    ted
    0.18
    ded
    0.17
    ged
    0.17
    ified
    0.17
    zed
    0.17
    lated
    0.16
    ced
    0.16
    iced
    0.16
    ided
    0.16
    Activations Density 0.065%

    No Known Activations