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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 24481
    Prev
    Next
    INDEX
    Explanations

    statements emphasizing the importance or necessity of certain actions or conditions

    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
    ContentLoaded
    -0.16
    hte
    -0.14
    thinkable
    -0.14
    PTY
    -0.13
    _translate
    -0.13
    uale
    -0.13
    ouver
    -0.13
     discre
    -0.13
    evin
    -0.13
    _unused
    -0.13
    POSITIVE LOGITS
     important
    0.77
    important
    0.64
     Important
    0.64
     importance
    0.63
    Important
    0.57
    -important
    0.55
    éĩįè¦ģ
    0.55
     importante
    0.55
     Importance
    0.53
     importantes
    0.52
    Activations Density 0.083%

    No Known Activations