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. 4
    Prev
    Next
    INDEX
    Explanations

    references to government institutions and officials

    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
    å¡
    -0.14
    araoh
    -0.13
    ấn
    -0.13
    ç½²
    -0.13
    .inverse
    -0.13
    iba
    -0.13
    iance
    -0.12
    ización
    -0.12
    _mB
    -0.12
    ãĥ¶
    -0.12
    POSITIVE LOGITS
     sources
    0.33
     officials
    0.29
     source
    0.28
     official
    0.27
     insiders
    0.26
     expert
    0.26
     insider
    0.24
     experts
    0.23
    -watch
    0.23
     circles
    0.22
    Activations Density 0.091%

    No Known Activations