© 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. 15824
    Prev
    Next
    INDEX
    Explanations

    occurrences of numerical values or financial figures

    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
     did
    -0.13
    uest
    -0.13
    uring
    -0.13
     since
    -0.13
    ailles
    -0.13
     seemed
    -0.13
    VV
    -0.13
    ening
    -0.13
     particular
    -0.13
     devis
    -0.13
    POSITIVE LOGITS
    alach
    0.16
    ovenant
    0.14
    'ÑıÑĤ
    0.14
    /renderer
    0.13
    475
    0.13
    nda
    0.13
    ansa
    0.13
    dependence
    0.12
    ãĥıãĤ¤
    0.12
    pee
    0.12
    Activations Density 0.033%

    No Known Activations