© 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. 28945
    Prev
    Next
    INDEX
    Explanations

    references to catastrophic events or conditions

    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
     пÑĢави
    -0.16
    ane
    -0.16
    que
    -0.15
    elial
    -0.14
    ampo
    -0.14
    lie
    -0.14
    ellular
    -0.14
    ie
    -0.14
    oser
    -0.14
    ka
    -0.14
    POSITIVE LOGITS
    cly
    0.24
    comb
    0.19
    estro
    0.18
     cata
    0.18
     freaking
    0.16
    æĬľ
    0.16
    emey
    0.16
    umont
    0.16
    oom
    0.16
     cats
    0.15
    Activations Density 0.001%

    No Known Activations