© 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. 42808
    Prev
    Next
    INDEX
    Explanations

    phrases indicating conjunctions or associations, particularly focusing on the word "with."

    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
    ains
    -0.14
    914
    -0.14
    ãģ¾ãģŁ
    -0.14
    aten
    -0.14
     Ñģк
    -0.13
    ÙĨدا
    -0.13
    į
    -0.13
    åѤ
    -0.13
    ÄĽr
    -0.13
    ç´Ħ
    -0.13
    POSITIVE LOGITS
     minimal
    0.49
    minimal
    0.40
     ease
    0.39
     Minimal
    0.37
    Minimal
    0.35
     Ease
    0.32
     minimum
    0.32
     little
    0.29
     just
    0.28
     only
    0.28
    Activations Density 0.054%

    No Known Activations