© 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. 44144
    Prev
    Next
    INDEX
    Explanations

    phrases related to familial and spiritual relationships

    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
    ÏĨοÏģ
    -0.17
     earthly
    -0.17
    ê°ģ
    -0.15
     redistribution
    -0.15
    วย
    -0.15
    assel
    -0.14
    782
    -0.14
    å®Ļ
    -0.14
    aney
    -0.14
    ellan
    -0.14
    POSITIVE LOGITS
     God
    0.32
     Christ
    0.29
     Him
    0.28
    God
    0.24
     Jesus
    0.23
    Christ
    0.19
     His
    0.18
     god
    0.18
     ç¥ŀ
    0.18
     GOD
    0.18
    Activations Density 0.049%

    No Known Activations