© 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. 24197
    Prev
    Next
    INDEX
    Explanations

    elements that indicate items, lists, or notable updates within a context

    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
     mouseClicked
    -0.14
    worm
    -0.14
    ãĢĶ
    -0.14
    çĺ
    -0.14
    inh
    -0.13
    _REUSE
    -0.13
    posted
    -0.13
    ulo
    -0.13
    /LICENSE
    -0.13
    .layouts
    -0.13
    POSITIVE LOGITS
     host
    0.79
    host
    0.56
     hosts
    0.55
     Host
    0.51
    Host
    0.50
    _host
    0.45
     HOST
    0.44
    .host
    0.43
    (host
    0.43
    -host
    0.40
    Activations Density 0.123%

    No Known Activations