© 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. 54985
    Prev
    Next
    INDEX
    Explanations

    references to human rights violations and associated individuals

    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
     Afghan
    -0.20
     Taliban
    -0.19
    ubble
    -0.17
     Kabul
    -0.17
     jih
    -0.17
    aille
    -0.16
    ÅĻÃŃd
    -0.16
    lect
    -0.16
     Ahmet
    -0.15
    olon
    -0.15
    POSITIVE LOGITS
     Bahrain
    0.42
    ahrain
    0.27
     Kingdom
    0.26
     kingdom
    0.25
     Gulf
    0.25
     pear
    0.24
     king
    0.23
     pearl
    0.23
     King
    0.23
     Pearl
    0.22
    Activations Density 0.020%

    No Known Activations