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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 15774
    Prev
    Next
    INDEX
    Explanations

    phrases relating to uncertainty or the status of actions yet to be completed

    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
     Already
    -0.26
     already
    -0.26
    already
    -0.23
    Already
    -0.22
    _already
    -0.20
     déjÃł
    -0.18
     continu
    -0.16
    å·²ç»ı
    -0.16
    gems
    -0.16
     już
    -0.16
    POSITIVE LOGITS
     fully
    0.40
     officially
    0.28
     Fully
    0.28
     official
    0.26
     full
    0.26
     entirely
    0.24
     completely
    0.24
    official
    0.23
     FULL
    0.23
    Fully
    0.23
    Activations Density 0.102%

    No Known Activations