© 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. 55089
    Prev
    Next
    INDEX
    Explanations

    references to judicial and executive powers in relation to legal cases

    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
    ovu
    -0.18
    ocha
    -0.17
    adius
    -0.16
    åħħ
    -0.16
    ube
    -0.15
    TRS
    -0.15
    asz
    -0.15
    umed
    -0.14
    ADIUS
    -0.14
    etime
    -0.14
    POSITIVE LOGITS
    eng
    0.21
    enos
    0.20
    dale
    0.18
    low
    0.17
    berg
    0.17
    led
    0.17
    ault
    0.17
    pires
    0.16
    dma
    0.16
    loss
    0.16
    Activations Density 0.004%

    No Known Activations