© 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. 30539
    Prev
    Next
    INDEX
    Explanations

    expressions of sentiment and emotional connections

    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
    edom
    -0.14
    uced
    -0.13
    LTR
    -0.12
    ÃŃÅ¡
    -0.12
    /Sub
    -0.12
    âu
    -0.12
    slu
    -0.12
    _EVT
    -0.12
    @nate
    -0.12
    avar
    -0.12
    POSITIVE LOGITS
     special
    1.48
     Special
    1.26
    special
    1.22
    Special
    1.18
     SPECIAL
    1.13
    _special
    1.02
    çī¹æ®Ĭ
    1.01
    -special
    1.00
     specials
    1.00
    pecial
    0.94
    Activations Density 0.008%

    No Known Activations