© 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. 41628
    Prev
    Next
    INDEX
    Explanations

    references to political or social issues related to inequality

    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
    hes
    -0.15
    ìŀIJ
    -0.15
    λλά
    -0.15
    >>)
    -0.15
    оÑĥ
    -0.15
    URRED
    -0.14
    ourd
    -0.14
    åĿĬ
    -0.14
     TMPro
    -0.14
    ernote
    -0.14
    POSITIVE LOGITS
     meanwhile
    0.29
     still
    0.26
     continue
    0.25
     remain
    0.24
     STILL
    0.22
     wonder
    0.22
     continues
    0.22
     remained
    0.22
     remains
    0.21
     continued
    0.21
    Activations Density 0.232%

    No Known Activations