© 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. 14446
    Prev
    Next
    INDEX
    Explanations

    the suffix "ish" in words

    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
    asse
    -0.20
    eh
    -0.19
    eel
    -0.18
    hti
    -0.18
    eeee
    -0.17
    eam
    -0.17
    adius
    -0.16
    zig
    -0.16
    zin
    -0.16
    iangle
    -0.15
    POSITIVE LOGITS
    s
    0.37
    Ùĩ
    0.26
    (es
    0.24
    mallow
    0.22
    hh
    0.22
    y
    0.22
    ers
    0.21
    ness
    0.21
    awn
    0.20
    rooms
    0.20
    Activations Density 0.028%

    No Known Activations