© 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. 43271
    Prev
    Next
    INDEX
    Explanations

    instances where a majority or significant amount is referenced in relation to a topic

    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
    ÑĤаж
    -0.17
    ile
    -0.15
    //{{
    -0.14
    clud
    -0.14
    elier
    -0.14
    ovit
    -0.14
    iner
    -0.13
    iez
    -0.13
    ilim
    -0.13
    andWhere
    -0.13
    POSITIVE LOGITS
     people
    0.40
     times
    0.33
    people
    0.29
     folks
    0.28
    人ãģ¯
    0.26
    times
    0.24
    人ãģĮ
    0.23
     ppl
    0.23
    -times
    0.22
     PEOPLE
    0.22
    Activations Density 0.384%

    No Known Activations