© 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. 33643
    Prev
    Next
    INDEX
    Explanations

    terms related to fan and user engagement in various contexts

    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
    º
    -0.15
     Burl
    -0.14
    íĤ
    -0.14
    nis
    -0.13
    (Array
    -0.13
    chemy
    -0.13
    351
    -0.13
     gall
    -0.13
    æ´²
    -0.13
    /moment
    -0.13
    POSITIVE LOGITS
     base
    0.90
    base
    0.83
     Base
    0.71
    -base
    0.67
    Base
    0.65
    _base
    0.59
    (base
    0.56
    .base
    0.54
     bases
    0.49
    BASE
    0.48
    Activations Density 0.038%

    No Known Activations