© 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. 28064
    Prev
    Next
    INDEX
    Explanations

    contractions indicating negation or absence

    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
     chuck
    -0.16
    lander
    -0.16
     reconstruct
    -0.15
    ãĥ¡ãĥ©
    -0.15
    ä¹ĥ
    -0.15
     KNOW
    -0.15
    yu
    -0.14
     accr
    -0.14
    bote
    -0.14
    lant
    -0.14
    POSITIVE LOGITS
     yet
    0.51
     Yet
    0.39
    yet
    0.37
    Yet
    0.37
     been
    0.34
     Been
    0.26
    Been
    0.25
     BEEN
    0.24
    been
    0.24
     lately
    0.18
    Activations Density 0.074%

    No Known Activations