© 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. 54778
    Prev
    Next
    INDEX
    Explanations

    mentions of notable individuals, particularly in a speaking or event context

    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
    abar
    -0.16
    è®Ń
    -0.15
    eso
    -0.15
     Ages
    -0.15
    bgcolor
    -0.15
    εÏģÏĮ
    -0.14
     Halloween
    -0.14
     createAction
    -0.14
    resh
    -0.14
    Å¡il
    -0.14
    POSITIVE LOGITS
     Grad
    0.56
     graduation
    0.56
     gradu
    0.51
    Grad
    0.50
    grad
    0.50
     graduates
    0.47
     graduating
    0.44
     grad
    0.42
     grads
    0.41
     commencement
    0.40
    Activations Density 0.022%

    No Known Activations