© 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. 25790
    Prev
    Next
    INDEX
    Explanations

    references to destinations and the process of reaching them

    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
    odu
    -0.16
     Doub
    -0.15
     datings
    -0.14
    cket
    -0.14
     nackte
    -0.14
    ãĥ¬ãĥ³
    -0.14
    اجر
    -0.14
    ury
    -0.14
    uries
    -0.14
     Dob
    -0.14
    POSITIVE LOGITS
     destination
    0.96
     destinations
    0.82
     Destination
    0.78
    destination
    0.78
     Dest
    0.77
     destin
    0.74
     destino
    0.74
    dest
    0.74
     dest
    0.72
    Dest
    0.69
    Activations Density 0.021%

    No Known Activations