© 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. 25289
    Prev
    Next
    INDEX
    Explanations

    numerical representations of various entities or events

    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
    2
    -0.23
    1
    -0.22
    guard
    -0.20
    3
    -0.20
    12
    -0.19
    10
    -0.19
    109
    -0.18
    202
    -0.18
    5
    -0.18
    list
    -0.18
    POSITIVE LOGITS
    âĢ³
    0.45
    âĢ²
    0.44
    s
    0.41
    ï¸ı
    0.40
    -го
    0.29
    -й
    0.25
    th
    0.24
    sheets
    0.24
    sak
    0.22
    ë²Ī
    0.22
    Activations Density 1.259%

    No Known Activations