© 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. 41161
    Prev
    Next
    INDEX
    Explanations

    the word "just" 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
    #
    -0.18
    ijken
    -0.16
    ìŀ¥ìĿĢ
    -0.15
    essler
    -0.14
    ìŀ¥ìĿĦ
    -0.14
    âĸį
    -0.14
    à¥Įल
    -0.14
    .CopyTo
    -0.14
    .Invariant
    -0.13
    chyb
    -0.13
    POSITIVE LOGITS
    ifi
    0.34
    ifies
    0.30
    ifiable
    0.29
    ifying
    0.27
    ifications
    0.25
     recently
    0.23
     finished
    0.23
    åĪļ
    0.22
     days
    0.21
    IFI
    0.20
    Activations Density 0.058%

    No Known Activations