© 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. 23918
    Prev
    Next
    INDEX
    Explanations

    instances of denial or deprivation related to social issues

    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
    antha
    -0.17
    antan
    -0.16
    chemas
    -0.16
    imore
    -0.15
    atown
    -0.15
    ecal
    -0.15
    rey
    -0.15
    rypton
    -0.14
    anc
    -0.14
    aks
    -0.14
    POSITIVE LOGITS
     opportunity
    0.37
     access
    0.34
     opportunities
    0.32
     even
    0.29
     chance
    0.28
     normal
    0.27
     basic
    0.27
     any
    0.26
     possibility
    0.26
     Opportunity
    0.25
    Activations Density 0.030%

    No Known Activations