© 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. 14820
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to political ideologies and affiliations

    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
    verts
    -0.13
    ÃľM
    -0.13
     whether
    -0.13
    roud
    -0.13
    až
    -0.13
    wr
    -0.13
     mim
    -0.13
    CS
    -0.12
    ãĥ©ãĤ¤ãĥĪ
    -0.12
     assist
    -0.12
    POSITIVE LOGITS
     nature
    0.22
    nature
    0.21
     confines
    0.17
    ìŀIJìĿ¸
    0.17
    Nature
    0.16
    olla
    0.16
    ouver
    0.15
    liest
    0.15
    illez
    0.15
    usb
    0.14
    Activations Density 0.090%

    No Known Activations