© 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. 22814
    Prev
    Next
    INDEX
    Explanations

    references to financial transactions and investments

    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
    ngr
    -0.14
     кÑĥлÑĮÑĤÑĥ
    -0.14
    orta
    -0.14
     eskort
    -0.14
    mpr
    -0.14
    izmet
    -0.14
     Sesso
    -0.13
    à¤Ĥपर
    -0.13
    éĻµ
    -0.13
    odon
    -0.13
    POSITIVE LOGITS
     minority
    0.41
     ownership
    0.38
     majority
    0.37
     equity
    0.37
     share
    0.36
     shares
    0.35
     stake
    0.35
     stakes
    0.32
     Minority
    0.31
     Majority
    0.29
    Activations Density 0.049%

    No Known Activations