© 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. 8726
    Prev
    Next
    INDEX
    Explanations

    instances of the word "then" indicating sequential or conditional actions

    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
     then
    -0.25
     now
    -0.21
    then
    -0.21
     thus
    -0.20
     entonces
    -0.20
     Then
    -0.20
     alors
    -0.19
    	then
    -0.19
    Then
    -0.18
     dabei
    -0.17
    POSITIVE LOGITS
     latter
    0.22
    umber
    0.20
     later
    0.19
    after
    0.19
    ew
    0.18
     proced
    0.17
    abouts
    0.17
    -current
    0.16
     proceed
    0.16
    -President
    0.16
    Activations Density 0.107%

    No Known Activations