© 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. 54397
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "through."

    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
     μην
    -0.15
    ves
    -0.14
    .wp
    -0.14
    disposing
    -0.14
    elf
    -0.14
    ised
    -0.14
    fortawesome
    -0.14
    ulong
    -0.14
    Ħĸ
    -0.14
    bstract
    -0.13
    POSITIVE LOGITS
    puts
    0.30
    ought
    0.30
    s
    0.28
    put
    0.26
    -out
    0.25
    out
    0.25
    ogh
    0.24
    OUT
    0.24
    wards
    0.22
    ly
    0.21
    Activations Density 0.063%

    No Known Activations