© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 8957
    Prev
    Next
    INDEX
    Explanations

    references to the Catholic Church and teaching responsibilities

    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
    onga
    -0.17
     Tune
    -0.16
     Tong
    -0.16
    ÑĪев
    -0.15
    TXT
    -0.15
    txt
    -0.15
    Typed
    -0.15
    din
    -0.15
     Titan
    -0.14
    TTY
    -0.14
    POSITIVE LOGITS
     teaching
    0.75
     teach
    0.72
     taught
    0.68
     teaches
    0.64
     Teach
    0.59
     Teaching
    0.57
     teachings
    0.56
    æķĻ
    0.56
     ense
    0.53
    -te
    0.46
    Activations Density 0.028%

    No Known Activations