© 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. 44406
    Prev
    Next
    INDEX
    Explanations

    discussions about political shifts and personal identity struggles

    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
    yk
    -0.15
    .getComponent
    -0.15
    emann
    -0.14
    uner
    -0.14
     defaultMessage
    -0.14
    .newBuilder
    -0.14
    idebar
    -0.13
    าà¸ķร
    -0.13
    oleon
    -0.13
    ØŃÙĨ
    -0.13
    POSITIVE LOGITS
     began
    0.44
     stopped
    0.41
     stop
    0.39
     became
    0.36
     started
    0.36
     begun
    0.35
     stops
    0.34
     become
    0.34
    stop
    0.34
    Stop
    0.33
    Activations Density 0.150%

    No Known Activations