© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 15777
    Prev
    Next
    INDEX
    Explanations

    references to language and its implications in societal contexts

    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
    mers
    -0.17
    ucas
    -0.15
    ickle
    -0.14
     ==============================================================
    -0.14
    ÑĪ
    -0.13
    веÑĢж
    -0.13
    اÙĨÙĩ
    -0.13
    -await
    -0.13
    oling
    -0.13
    Å¡il
    -0.13
    POSITIVE LOGITS
     usage
    0.23
     used
    0.22
    usage
    0.21
    ology
    0.21
     Used
    0.21
     barrier
    0.19
     surrounding
    0.18
    Used
    0.18
     USED
    0.18
     use
    0.17
    Activations Density 0.016%

    No Known Activations