© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 14943
    Prev
    Next
    INDEX
    Explanations

    emotional responses to tragic events, particularly related to loss

    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
     embarrassing
    -0.15
    rons
    -0.15
    uintptr
    -0.15
     kaç
    -0.15
    apult
    -0.15
    erais
    -0.15
     suicidal
    -0.15
    uper
    -0.15
    ihar
    -0.15
     embarrass
    -0.14
    POSITIVE LOGITS
     sense
    0.41
    sense
    0.31
     Sense
    0.31
    Sense
    0.28
     loss
    0.25
     young
    0.21
    loss
    0.21
    trag
    0.20
     innocent
    0.20
     tragedy
    0.20
    Activations Density 0.138%

    No Known Activations