© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 54848
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     eskort
    -0.18
     RuntimeObject
    -0.17
     RuntimeMethod
    -0.16
    .btnClose
    -0.15
     destinationViewController
    -0.15
     addCriterion
    -0.14
    .Companion
    -0.14
    jejer
    -0.14
    gii
    -0.13
    ieme
    -0.13
    POSITIVE LOGITS
     "...
    0.16
     “â̦
    0.15
     whilst
    0.14
    .
    0.13
     that
    0.13
     (“
    0.13
     “
    0.13
     (...
    0.13
     ...
    0.12
     this
    0.12
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.