© 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. 42151
    Prev
    Next
    INDEX
    Explanations

    words related to public perception and approval ratings

    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
    261
    -0.18
    iggs
    -0.17
    /includes
    -0.16
    åŀ
    -0.15
    agit
    -0.14
    691
    -0.14
     trest
    -0.14
    ("(%
    -0.13
    IDX
    -0.13
     promo
    -0.13
    POSITIVE LOGITS
     các
    0.23
    ses
    0.21
    ä¹ĭä¸Ģ
    0.19
    maries
    0.18
    ests
    0.17
    Datum
    0.17
    Ps
    0.16
    izzes
    0.16
    sembles
    0.16
    åIJĦ
    0.16
    Activations Density 0.101%

    No Known Activations