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. 19
    Prev
    Next
    INDEX
    Explanations

    articles and determiners related to political discussions

    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
    ons
    -0.18
    ups
    -0.17
    ess
    -0.17
    åŃ
    -0.15
     progress
    -0.15
    arks
    -0.15
    643
    -0.15
    bid
    -0.14
    anno
    -0.14
    983
    -0.14
    POSITIVE LOGITS
    ozem
    0.19
    conomy
    0.16
     functioning
    0.15
    erdale
    0.15
    .easing
    0.15
    distributed
    0.15
    fdc
    0.14
     hung
    0.14
    виÑĩ
    0.14
    oningen
    0.14
    Activations Density 0.055%

    No Known Activations