© 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. 41340
    Prev
    Next
    INDEX
    Explanations

    instances of the word "talk."

    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
    icari
    -0.17
    arily
    -0.16
    wu
    -0.16
    ordin
    -0.16
    roys
    -0.15
    uning
    -0.15
    ez
    -0.15
    rib
    -0.14
    licted
    -0.14
    halb
    -0.14
    POSITIVE LOGITS
    ative
    0.32
    SPORT
    0.23
    er
    0.23
    ATIVE
    0.21
    nowled
    0.19
    athon
    0.18
    shop
    0.18
    sgiving
    0.17
    ings
    0.17
    ingham
    0.17
    Activations Density 0.016%

    No Known Activations