© 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. 23832
    Prev
    Next
    INDEX
    Explanations

    instances of the word "silence" in various contexts

    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
    i
    -0.17
    UNE
    -0.15
    èĪĪ
    -0.15
    asser
    -0.14
    ascar
    -0.14
    essed
    -0.14
    Invariant
    -0.14
    ainment
    -0.14
    ello
    -0.14
    ño
    -0.14
    POSITIVE LOGITS
     Sil
    0.25
    ently
    0.24
    houette
    0.24
     sil
    0.23
    Sil
    0.21
    ences
    0.21
    vest
    0.21
    encing
    0.21
    sil
    0.20
    berman
    0.20
    Activations Density 0.004%

    No Known Activations