© 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. 28716
    Prev
    Next
    INDEX
    Explanations

    mentions of financial instruments and fundraising efforts

    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
    flare
    -0.15
    HeaderValue
    -0.14
    stalk
    -0.14
    ampie
    -0.14
    uggy
    -0.13
    éĥ
    -0.13
    wner
    -0.13
    _FIFO
    -0.13
    лад
    -0.13
    ẩm
    -0.13
    POSITIVE LOGITS
     raised
    1.11
     raise
    1.10
     raising
    1.09
     raises
    0.99
     Raised
    0.98
    raised
    0.97
    raise
    0.96
     Raise
    0.94
    raising
    0.91
    Raise
    0.86
    Activations Density 0.040%

    No Known Activations