Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 9246
    Prev
    Next
    INDEX
    Explanations

    entities associated with governmental or regulatory bodies

    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
    _mD
    -0.18
    ICODE
    -0.16
    eguard
    -0.16
    kla
    -0.16
    slaught
    -0.16
    .Dom
    -0.16
    Titles
    -0.15
    ERSHEY
    -0.15
    ovit
    -0.15
    icaret
    -0.15
    POSITIVE LOGITS
     officials
    0.33
     sources
    0.31
     authorities
    0.27
     President
    0.24
     records
    0.24
     efforts
    0.24
     Chief
    0.21
     spokes
    0.20
     spokeswoman
    0.20
     representatives
    0.19
    Activations Density 0.106%

    No Known Activations