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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 23457
    Prev
    Next
    INDEX
    Explanations

    references to armed conflict and wars

    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
    æĿ¥çļĦ
    -0.16
    ำ
    -0.15
    ombat
    -0.15
    bih
    -0.15
    å¹ħ
    -0.15
    atak
    -0.15
    ogh
    -0.15
    usk
    -0.15
    بÙĬØ©
    -0.14
    slu
    -0.14
    POSITIVE LOGITS
     break
    0.60
     broke
    0.60
     breakout
    0.57
     breaking
    0.57
     breaks
    0.55
     Break
    0.53
    break
    0.48
    Break
    0.46
    -break
    0.46
    _break
    0.45
    Activations Density 0.028%

    No Known Activations