© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 28890
    Prev
    Next
    INDEX
    Explanations

    references to familial relationships and emotional bonds

    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
     granddaughter
    -0.18
    ovah
    -0.17
    èn
    -0.16
    ÑĪÑĤ
    -0.15
     Mrs
    -0.15
    Mrs
    -0.15
    _Callback
    -0.14
    icari
    -0.14
    oplay
    -0.14
     Grandma
    -0.14
    POSITIVE LOGITS
     brother
    1.15
     brothers
    1.11
     sibling
    0.99
     siblings
    0.97
     Brother
    0.97
     Brothers
    0.93
     sister
    0.91
     sisters
    0.89
    bro
    0.77
     Sister
    0.71
    Activations Density 0.029%

    No Known Activations