© 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. 63563
    Prev
    Next
    INDEX
    Explanations

    unique and individualized project descriptions or characteristics

    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
    ัย
    -0.18
    ä¸ĢåĪĩ
    -0.17
    avail
    -0.17
    одав
    -0.16
     everything
    -0.14
     ÙĥØ«ÙĬر
    -0.14
    erin
    -0.14
     suá»ijt
    -0.14
    ahr
    -0.14
    PARATOR
    -0.13
    POSITIVE LOGITS
     unique
    0.35
    unique
    0.29
     distinct
    0.28
     differently
    0.27
     separately
    0.26
     Unique
    0.26
     its
    0.26
     respective
    0.25
    çį¨
    0.25
    _unique
    0.25
    Activations Density 0.074%

    No Known Activations