© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 24493
    Prev
    Next
    INDEX
    Explanations

    phrases related to surveillance and data privacy issues

    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
    uv
    -0.16
     latest
    -0.14
    atomic
    -0.14
     atomic
    -0.14
     Latest
    -0.14
    oid
    -0.14
     Atomic
    -0.14
    eral
    -0.14
     Ann
    -0.13
    _defs
    -0.13
    POSITIVE LOGITS
    Bulk
    0.24
     Bulk
    0.24
     communications
    0.23
    communications
    0.22
     metadata
    0.22
    bulk
    0.22
     bulk
    0.21
    -metadata
    0.20
     content
    0.19
    metadata
    0.18
    Activations Density 0.040%

    No Known Activations