© 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. 22961
    Prev
    Next
    INDEX
    Explanations

    words associated with escalating situations 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
    ated
    -0.52
    ized
    -0.50
    äºĨ
    -0.44
    ened
    -0.43
    ified
    -0.42
    ged
    -0.39
    IFIED
    -0.26
    IZED
    -0.24
    ished
    -0.23
    ured
    -0.23
    POSITIVE LOGITS
    äºĨä¸Ģ
    0.26
    atedRoute
    0.23
    ised
    0.23
    ating
    0.20
    izedName
    0.16
    yth
    0.16
    ates
    0.15
    глÑıд
    0.14
     TED
    0.14
     Barker
    0.14
    Activations Density 0.048%

    No Known Activations