© 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. 41810
    Prev
    Next
    INDEX
    Explanations

    descriptions of loud noises and sounds in the context of an event

    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
     DM
    -0.15
    itura
    -0.13
    olas
    -0.13
    ä¼ı
    -0.13
    ente
    -0.13
    ле
    -0.13
     (!!
    -0.13
    aeda
    -0.13
     responseBody
    -0.13
    apture
    -0.13
    POSITIVE LOGITS
     noise
    0.33
     noises
    0.31
     sound
    0.31
    noise
    0.29
     sounds
    0.26
    sound
    0.23
     clicking
    0.23
     Noise
    0.22
    Sound
    0.22
     Sounds
    0.22
    Activations Density 0.040%

    No Known Activations