© 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. 29561
    Prev
    Next
    INDEX
    Explanations

    phrases related to promotional offers and account services

    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
    .scalablytyped
    -0.15
    iedo
    -0.15
    ifton
    -0.13
    pic
    -0.13
    phia
    -0.13
    ãĥ³ãĤº
    -0.13
    पत
    -0.13
    orrow
    -0.12
    roker
    -0.12
    _reporting
    -0.12
    POSITIVE LOGITS
     pre
    1.11
     Pre
    1.03
    pre
    0.98
    Pre
    0.93
    -pre
    0.86
     PRE
    0.81
    .pre
    0.81
    _pre
    0.78
    	pre
    0.77
     pré
    0.77
    Activations Density 0.006%

    No Known Activations