© 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. 42086
    Prev
    Next
    INDEX
    Explanations

    phrases indicating conditions or contingencies

    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
    adol
    -0.18
    uhl
    -0.17
    olist
    -0.14
    ÑĢап
    -0.14
    :///
    -0.14
    uzzi
    -0.14
     hã
    -0.13
    ÙĤÙī
    -0.13
    stad
    -0.13
    cntl
    -0.13
    POSITIVE LOGITS
     long
    1.21
    long
    0.93
    .long
    0.71
    -long
    0.71
     LONG
    0.69
    éķ¿
    0.67
    (long
    0.66
    éķ·
    0.63
    	long
    0.63
    _long
    0.61
    Activations Density 0.021%

    No Known Activations