© 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 EvalsExportsAPI 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. 43299
    Prev
    Next
    INDEX
    Explanations

    actions related to creation, assembly, and collaboration in projects

    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
     further
    -0.22
    Others
    -0.21
     others
    -0.20
     anymore
    -0.20
     other
    -0.18
     Others
    -0.18
    others
    -0.18
     autres
    -0.17
     elsewhere
    -0.17
     altre
    -0.17
    POSITIVE LOGITS
     several
    0.32
     bunch
    0.28
     some
    0.28
    çļĦä¸Ģ个
    0.25
     sebuah
    0.24
    ä¸Ģ个
    0.24
    åĩłä¸ª
    0.22
     certain
    0.22
    ä¸ĢäºĽ
    0.22
     a
    0.22
    Activations Density 0.398%

    No Known Activations