© 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. 22780
    Prev
    Next
    INDEX
    Explanations

    statements indicating awareness or knowledge about a situation

    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
     instead
    -0.21
     Instead
    -0.19
    Instead
    -0.18
    instead
    -0.18
    ess
    -0.16
     вмеÑģÑĤ
    -0.16
     Seems
    -0.15
     seeming
    -0.14
    Ess
    -0.14
    MBOL
    -0.14
    POSITIVE LOGITS
     otherwise
    0.66
    otherwise
    0.57
     Otherwise
    0.56
    Otherwise
    0.50
     OTHERWISE
    0.47
     else
    0.44
    åIJ¦
    0.38
     sonst
    0.38
    	else
    0.33
    наÑĩе
    0.30
    Activations Density 0.013%

    No Known Activations