© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 63493
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to hopes, expectations, and thought processes

    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
    515
    -0.16
     Garr
    -0.14
    ardon
    -0.14
    -caret
    -0.14
    CONDITION
    -0.14
    .Require
    -0.14
    кÑĥл
    -0.13
    .SizeType
    -0.13
    514
    -0.13
    upp
    -0.13
    POSITIVE LOGITS
    orman
    0.15
    gency
    0.14
     for
    0.14
     between
    0.14
     surrounding
    0.14
    arsers
    0.14
     besar
    0.14
    dfa
    0.14
    etus
    0.13
    arb
    0.13
    Activations Density 0.035%

    No Known Activations