© 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. 29741
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate intent or desire

    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
    this
    -0.19
     this
    -0.19
    è¿Ļ个
    -0.18
     This
    -0.17
    ÑıÑī
    -0.15
    [this
    -0.15
     questa
    -0.14
    RelativeTo
    -0.14
     codec
    -0.14
    itchens
    -0.14
    POSITIVE LOGITS
    ypress
    0.14
    imbus
    0.14
    agini
    0.13
    окол
    0.13
    onis
    0.13
    isten
    0.13
     gourmet
    0.13
    иÑĩниÑħ
    0.13
    иÑĩеÑģкие
    0.13
     traditional
    0.13
    Activations Density 0.003%

    No Known Activations