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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 28461
    Prev
    Next
    INDEX
    Explanations

    phrases indicating goals or actions related to personal transformation or social change

    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
    lds
    -0.15
    uje
    -0.15
    landı
    -0.14
    PPER
    -0.13
    ãģ«ãģ¯
    -0.13
     hearing
    -0.13
    xmm
    -0.13
    ¤íĶĦ
    -0.13
    vido
    -0.12
    ovÄĽ
    -0.12
    POSITIVE LOGITS
     being
    0.65
     be
    0.62
     Being
    0.59
    being
    0.59
    Being
    0.54
     бÑĭÑĤÑĮ
    0.54
     být
    0.52
     essere
    0.49
     être
    0.48
    被
    0.47
    Activations Density 0.053%

    No Known Activations