© 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. 63868
    Prev
    Next
    INDEX
    Explanations

    references to violence, state actions, and oppression

    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
    leting
    -0.13
    ÅĻev
    -0.13
    bette
    -0.13
    .grad
    -0.13
    ivel
    -0.13
    ħn
    -0.13
    antine
    -0.13
    िà¤ķà¤Ł
    -0.12
    sono
    -0.12
    Ïģον
    -0.12
    POSITIVE LOGITS
     stop
    1.10
     stops
    0.96
     Stop
    0.96
    stop
    0.92
    -stop
    0.92
     stopped
    0.92
     stopping
    0.90
    Stop
    0.89
     STOP
    0.88
    .stop
    0.84
    Activations Density 0.029%

    No Known Activations