© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 63307
    Prev
    Next
    INDEX
    Explanations

    references to auditory perception or listening

    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
    iling
    -0.17
    ongan
    -0.17
    loat
    -0.16
    acer
    -0.16
    astle
    -0.15
    ement
    -0.15
    grounds
    -0.15
    gli
    -0.15
    ulent
    -0.15
    екаÑĢ
    -0.15
    POSITIVE LOGITS
    /view
    0.22
    AndServe
    0.22
    ening
    0.21
    eners
    0.21
     ears
    0.21
    /watch
    0.20
    erner
    0.19
    ners
    0.18
     comprehension
    0.17
    ardo
    0.17
    Activations Density 0.005%

    No Known Activations