© 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. 23118
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "it" and related functional phrases

    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
    chten
    -0.16
    dden
    -0.15
    .struts
    -0.14
    ť
    -0.14
    iggs
    -0.14
    aul
    -0.14
     Many
    -0.14
    ishly
    -0.14
     MANY
    -0.14
    otte
    -0.14
    POSITIVE LOGITS
     further
    0.47
     enough
    0.40
     more
    0.38
     Further
    0.32
    æĽ´å¤ļ
    0.31
    Further
    0.31
     much
    0.31
     Enough
    0.27
     weiter
    0.27
     weitere
    0.26
    Activations Density 0.069%

    No Known Activations