© 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. 14110
    Prev
    Next
    INDEX
    Explanations

    phrases related to recovery or restoration in a financial or organizational context

    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
    etÃŃ
    -0.16
    wahl
    -0.14
     뼰
    -0.14
    utely
    -0.14
    æ´ŀ
    -0.14
    æľĢé«ĺ
    -0.13
    adÃŃ
    -0.13
     пÑĢиÑĤ
    -0.13
    ugin
    -0.13
    plen
    -0.13
    POSITIVE LOGITS
     track
    0.43
     path
    0.38
    track
    0.37
     Track
    0.36
    Track
    0.33
     tracks
    0.33
    -track
    0.32
    _track
    0.31
    path
    0.30
     firm
    0.29
    Activations Density 0.032%

    No Known Activations