Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 20
    Prev
    Next
    INDEX
    Explanations

    terms related to war, famine, and conflict

    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
     Panic
    -0.16
    IRR
    -0.16
     zug
    -0.15
    auses
    -0.15
    uben
    -0.14
    erti
    -0.14
    anut
    -0.14
    aus
    -0.14
    ipay
    -0.14
    itre
    -0.14
    POSITIVE LOGITS
     rav
    0.37
     bes
    0.35
     mar
    0.31
     hit
    0.30
    -hit
    0.29
     str
    0.29
    bes
    0.28
     dog
    0.27
    hit
    0.27
     affected
    0.27
    Activations Density 0.052%

    No Known Activations