© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 28376
    Prev
    Next
    INDEX
    Explanations

    phrases related to changes and modifications within a context, particularly in relation to people and roles

    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
     çķ
    -0.15
    ç»Īäºİ
    -0.15
     Snowden
    -0.15
    .Tables
    -0.14
     ousted
    -0.14
    акон
    -0.14
    _DIRS
    -0.13
     breakout
    -0.13
     Woj
    -0.13
    apore
    -0.13
    POSITIVE LOGITS
     replaced
    0.36
     replacements
    0.26
     replacement
    0.26
     replace
    0.25
     yerine
    0.23
     Replace
    0.23
     never
    0.22
     Replacement
    0.22
    replacement
    0.21
    replace
    0.21
    Activations Density 0.048%

    No Known Activations