© 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. 9017
    Prev
    Next
    INDEX
    Explanations

    references to time or duration, particularly in the context of years and age

    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
     dần
    -0.16
    æ´
    -0.15
    isco
    -0.14
    INESS
    -0.14
     iParam
    -0.14
    etin
    -0.13
     ↵↵
    -0.13
    ushima
    -0.13
    ::$
    -0.13
    ONTAL
    -0.13
    POSITIVE LOGITS
     to
    0.20
    icari
    0.15
    ADA
    0.14
    Criterion
    0.14
    pcf
    0.14
    .HttpContext
    0.14
     olsa
    0.13
    ç´Ħ
    0.13
    onas
    0.13
    agli
    0.13
    Activations Density 0.059%

    No Known Activations