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

    Neuronpedia

    Jacobian LensNEW
    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. 42996
    Prev
    Next
    INDEX
    Explanations

    references to specific times and moments in life

    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
     timed
    -0.19
    man
    -0.17
    ese
    -0.16
     timely
    -0.16
    name
    -0.15
    termin
    -0.15
    entimes
    -0.15
    matic
    -0.15
    scribe
    -0.15
    innen
    -0.14
    POSITIVE LOGITS
    punkt
    0.30
    piece
    0.28
     period
    0.26
    frames
    0.25
    pieces
    0.25
    /place
    0.25
    åĢĻ
    0.24
    frame
    0.24
     frame
    0.23
    keepers
    0.21
    Activations Density 0.097%

    No Known Activations