© 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. 42450
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and transitional phrases in the text

    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
     esos
    -0.15
    å®ĥ们
    -0.14
    them
    -0.14
     esas
    -0.14
    à¹Ģหล
    -0.13
    “These
    -0.13
    ifu
    -0.13
    以ä¸Ĭ
    -0.12
     them
    -0.12
    "These
    -0.12
    POSITIVE LOGITS
     this
    0.94
    this
    0.81
     This
    0.74
    This
    0.69
     THIS
    0.65
    -this
    0.64
    (this
    0.64
     nÃły
    0.63
    ,this
    0.62
    	this
    0.60
    Activations Density 0.214%

    No Known Activations