© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 25750
    Prev
    Next
    INDEX
    Explanations

    phrases related to technology and devices being turned on or activated

    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
    .SDK
    -0.15
     uninstall
    -0.14
    sko
    -0.14
    ulls
    -0.14
    ounge
    -0.14
    ãi
    -0.14
    bsite
    -0.14
    eway
    -0.13
    nero
    -0.13
    rors
    -0.13
    POSITIVE LOGITS
     turned
    0.51
     turn
    0.51
     Turn
    0.47
     TURN
    0.42
    Turn
    0.41
    turn
    0.41
    -turn
    0.40
     turning
    0.40
     turns
    0.40
    turned
    0.39
    Activations Density 0.028%

    No Known Activations