© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 63499
    Prev
    Next
    INDEX
    Explanations

    references to ascension to power or positions of authority

    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
    εÏį
    -0.17
     Tunnel
    -0.15
     Tarif
    -0.14
     complement
    -0.14
    oust
    -0.14
    tainment
    -0.14
    ace
    -0.14
    ynn
    -0.14
    unnel
    -0.14
     Toggle
    -0.14
    POSITIVE LOGITS
     throne
    0.54
     th
    0.50
    thr
    0.42
     Throne
    0.41
     thrown
    0.40
     Thrones
    0.36
     crown
    0.35
     пÑĢеÑģÑĤ
    0.35
    rones
    0.34
     THR
    0.34
    Activations Density 0.016%

    No Known Activations