© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 54273
    Prev
    Next
    INDEX
    Explanations

    the word "them" in various contexts

    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
     itself
    -0.25
    ibly
    -0.17
    اÙĨÙĩ
    -0.15
    uae
    -0.14
    quine
    -0.14
    gress
    -0.14
     its
    -0.14
    (es
    -0.14
    odge
    -0.14
    _________________↵↵
    -0.13
    POSITIVE LOGITS
    /us
    0.40
    atically
    0.39
    self
    0.29
    /her
    0.28
     selves
    0.27
    elves
    0.25
    /th
    0.24
    editary
    0.22
    SEL
    0.21
    aly
    0.21
    Activations Density 0.267%

    No Known Activations