© 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. 23041
    Prev
    Next
    INDEX
    Explanations

    instances of punctuation and special characters in the text

    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
     blown
    -0.24
     beaten
    -0.22
    thrown
    -0.21
     risen
    -0.20
     bitten
    -0.19
     venture
    -0.18
     ridden
    -0.17
     spoken
    -0.17
     ventures
    -0.17
     fallen
    -0.17
    POSITIVE LOGITS
     took
    0.37
     gave
    0.32
     drove
    0.29
     threw
    0.28
    took
    0.28
     went
    0.27
     drew
    0.27
     got
    0.26
     stole
    0.25
    ãģĹãģ¾ãģĹãģŁ
    0.25
    Activations Density 0.051%

    No Known Activations