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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 41150
    Prev
    Next
    INDEX
    Explanations

    phrases related to participation and activity in events or gatherings

    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
    oppel
    -0.14
     all
    -0.14
    VERBOSE
    -0.14
    areth
    -0.14
    åIJĮæĻĤ
    -0.14
    @student
    -0.14
    å°Ī
    -0.13
     вÑģеÑħ
    -0.13
     nearly
    -0.13
     تÙħاÙħ
    -0.13
    POSITIVE LOGITS
     occasional
    0.60
     occasionally
    0.51
     Occasionally
    0.38
     spor
    0.37
     limited
    0.35
     SOME
    0.34
     small
    0.34
     few
    0.33
     ocas
    0.33
    occ
    0.32
    Activations Density 0.167%

    No Known Activations