© 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. 8567
    Prev
    Next
    INDEX
    Explanations

    phrases referring to time periods or durations

    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
    ouch
    -0.15
    маз
    -0.15
    uner
    -0.14
    -hours
    -0.14
    inder
    -0.14
    elia
    -0.13
    eric
    -0.13
    maz
    -0.13
    Times
    -0.12
     thing
    -0.12
    POSITIVE LOGITS
     years
    0.29
     decades
    0.24
     decade
    0.23
     Jahren
    0.22
    years
    0.21
     year
    0.21
     jaar
    0.19
     года
    0.18
     ÑĢокÑĸв
    0.18
     fiscal
    0.18
    Activations Density 0.060%

    No Known Activations