© 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. 43895
    Prev
    Next
    INDEX
    Explanations

    terms related to social class and economic conditions

    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
    owards
    -0.16
    彦
    -0.14
    าà¸ĵ
    -0.13
    arehouse
    -0.13
    tÄĽÅ¾
    -0.13
    luž
    -0.13
    vÃŃce
    -0.13
    lei
    -0.13
    едини
    -0.13
    eyse
    -0.13
    POSITIVE LOGITS
     precisely
    0.19
     or
    0.17
     indeed
    0.15
     (“
    0.15
     ('
    0.15
    -,
    0.14
     ("
    0.14
    ļ
    0.14
     --
    0.14
     prest
    0.13
    Activations Density 0.493%

    No Known Activations