Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 41506
    Prev
    Next
    INDEX
    Explanations

    references to inequality and poverty

    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
    blas
    -0.18
     deadliest
    -0.16
     Deadly
    -0.15
    ูà¸ģ
    -0.15
     deadly
    -0.15
     kø
    -0.15
    à¸Ĩ
    -0.14
    ÏĦικα
    -0.14
    chl
    -0.14
    xon
    -0.14
    POSITIVE LOGITS
     mis
    0.37
     poverty
    0.33
     Mis
    0.32
     suffer
    0.30
     misery
    0.29
    Mis
    0.29
     priv
    0.28
    mis
    0.28
     suffering
    0.27
     dest
    0.27
    Activations Density 0.089%

    No Known Activations