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

    Neuronpedia

    Jacobian LensNEW
    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. 43404
    Prev
    Next
    INDEX
    Explanations

    phrases related to economic hardship and its impact on vulnerable populations

    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
    cheon
    -0.15
    raud
    -0.15
    artner
    -0.15
    ucken
    -0.14
    еÑĤÑĮ
    -0.14
    ccione
    -0.14
    orque
    -0.14
    legate
    -0.14
     rau
    -0.14
     microtime
    -0.14
    POSITIVE LOGITS
     paying
    0.46
     pay
    0.42
     pays
    0.41
     bear
    0.39
     suffer
    0.39
     paid
    0.38
     bearing
    0.38
    pay
    0.37
     Pay
    0.34
     suffers
    0.34
    Activations Density 0.042%

    No Known Activations