© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 29237
    Prev
    Next
    INDEX
    Explanations

    expressions of doubt or controversy regarding significant events or situations

    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
    illo
    -0.16
    brero
    -0.14
    ilo
    -0.14
    rias
    -0.14
    ogl
    -0.14
    è³Ģ
    -0.13
    ذر
    -0.13
    ãĥ¼ãĥģ
    -0.13
    енÑĮ
    -0.13
     Nej
    -0.13
    POSITIVE LOGITS
     big
    0.73
    big
    0.62
     Big
    0.58
     BIG
    0.55
    -big
    0.53
    Big
    0.52
    BIG
    0.48
    _big
    0.48
    .big
    0.44
    /big
    0.44
    Activations Density 0.011%

    No Known Activations