© 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. 54325
    Prev
    Next
    INDEX
    Explanations

    words related to music, legal documents, and self-referential behavior

    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
    f
    -0.21
      ↵  ↵
    -0.21
    à¸Ħ
    -0.20
    ร
    -0.20
    face
    -0.18
    flix
    -0.18
    ت
    -0.18
    à¸ľ
    -0.17
    .future
    -0.17
    l
    -0.17
    POSITIVE LOGITS
    teenth
    0.26
    entimes
    0.21
    iciency
    0.20
    erring
    0.20
    icient
    0.18
    entially
    0.18
    ossil
    0.18
    etimes
    0.18
    teen
    0.18
    shire
    0.17
    Activations Density 0.060%

    No Known Activations