© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 63256
    Prev
    Next
    INDEX
    Explanations

    words and phrases that indicate legal or regulatory 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
    /*č↵
    -0.14
    mens
    -0.14
    _POINTER
    -0.13
     aby
    -0.13
     frau
    -0.12
     einz
    -0.12
    اÙĪÙĨ
    -0.12
    defs
    -0.12
    elia
    -0.12
     åĦ
    -0.12
    POSITIVE LOGITS
     to
    0.39
     kepada
    0.29
    åΰ
    0.27
    äºİ
    0.27
    æĸ¼
    0.27
     unto
    0.25
     Ø¥ÙĦÙī
    0.23
    ToBounds
    0.23
    äºİæĺ¯
    0.21
     zum
    0.21
    Activations Density 0.184%

    No Known Activations