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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 54231
    Prev
    Next
    INDEX
    Explanations

    phrases indicating disagreement or a critique of common beliefs about divinity or existence

    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
    apon
    -0.17
     normal
    -0.16
     Hey
    -0.15
     sf
    -0.14
    RLF
    -0.14
    okes
    -0.14
     subt
    -0.14
     not
    -0.14
     Fare
    -0.14
    ogi
    -0.14
    POSITIVE LOGITS
     merely
    0.26
     mere
    0.21
     simply
    0.20
    mere
    0.20
     пÑĢоÑģÑĤо
    0.17
    .getRaw
    0.17
    åıªæĺ¯
    0.17
     simplement
    0.16
     static
    0.16
     seperate
    0.16
    Activations Density 0.049%

    No Known Activations