© 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. 41654
    Prev
    Next
    INDEX
    Explanations

    the power and effectiveness of cards in a game context

    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
    ystick
    -0.15
    auc
    -0.14
    ukkan
    -0.14
    orra
    -0.14
    erb
    -0.14
    untu
    -0.14
    iden
    -0.14
    ç½²
    -0.14
    hir
    -0.13
    Å¡ka
    -0.13
    POSITIVE LOGITS
     situ
    0.26
     powerful
    0.24
     OP
    0.23
     Powerful
    0.21
     broken
    0.19
    .gg
    0.19
    ãĥªãĥ¼
    0.18
     meta
    0.18
     strong
    0.18
    (OP
    0.18
    Activations Density 0.086%

    No Known Activations