© 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. 28991
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal obligations and rights

    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
    POCH
    -0.18
    UNUSED
    -0.16
    ercul
    -0.16
    aeda
    -0.16
    stro
    -0.15
    Ðĩ
    -0.15
    egl
    -0.15
    eref
    -0.15
    éľ
    -0.15
    hwnd
    -0.15
    POSITIVE LOGITS
     required
    0.40
     entitled
    0.36
     subject
    0.34
     bound
    0.29
    required
    0.29
    Required
    0.28
     Required
    0.28
     allowed
    0.27
     Bound
    0.24
    subject
    0.24
    Activations Density 0.144%

    No Known Activations