© 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. 13236
    Prev
    Next
    INDEX
    Explanations

    key elements related to the description of objects and entities

    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
     once
    -0.18
     themselves
    -0.17
     first
    -0.17
    DefaultValue
    -0.16
    åĢij
    -0.16
    iyel
    -0.16
     ones
    -0.16
    Angles
    -0.15
    ุล
    -0.14
    第ä¸Ģ
    -0.14
    POSITIVE LOGITS
     alone
    0.25
    alone
    0.24
     solitary
    0.23
    atat
    0.21
     lone
    0.21
    ìĶ©
    0.20
     Alone
    0.20
    ãĢģä¸Ģ
    0.18
     among
    0.18
     lonely
    0.18
    Activations Density 0.074%

    No Known Activations