© 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 EvalsExportsinterp-engineNEWAPI 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. 24770
    Prev
    Next
    INDEX
    Explanations

    phrases related to issues or concerns regarding transparency and integrity in systems

    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
    lings
    -0.15
    ipur
    -0.15
    (es
    -0.14
    .Env
    -0.14
    _OW
    -0.13
    /domain
    -0.13
    ruta
    -0.13
    entimes
    -0.13
    /maps
    -0.13
    -0.13
    POSITIVE LOGITS
    atically
    0.25
    atic
    0.22
    atical
    0.21
     with
    0.20
    atik
    0.19
    -solving
    0.18
    atics
    0.18
     solver
    0.18
    /question
    0.18
    /issues
    0.18
    Activations Density 0.046%

    No Known Activations