© 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. 53895
    Prev
    Next
    INDEX
    Explanations

    instances of the word "confirmed" suggesting verification or validation of information

    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
    É
    -0.18
    oe
    -0.17
    and
    -0.16
    ida
    -0.16
    oun
    -0.15
    ka
    -0.15
     Opinion
    -0.15
    UR
    -0.14
    /do
    -0.14
    ias
    -0.14
    POSITIVE LOGITS
     that
    0.25
     kepada
    0.20
     yesterday
    0.18
     rằng
    0.18
     exclusively
    0.18
     Monday
    0.17
    RTOS
    0.17
     bahwa
    0.16
     Tuesday
    0.16
     via
    0.16
    Activations Density 0.051%

    No Known Activations