© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 24356
    Prev
    Next
    INDEX
    Explanations

    references to historical and cultural connections, particularly concerning Cuba

    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
    raud
    -0.14
    itrust
    -0.14
    )ï¼Į
    -0.14
    γοÏħ
    -0.13
    ï¼Į”
    -0.13
     unless
    -0.13
    allen
    -0.13
    _customize
    -0.13
     `}↵
    -0.13
     Whereas
    -0.12
    POSITIVE LOGITS
    ï¼Į以åıĬ
    0.28
    以åıĬ
    0.25
     ê·¸ë¦¬ê³ł
    0.24
     plus
    0.22
    åıĬåħ¶
    0.21
     oraz
    0.21
     nhé
    0.19
     PLUS
    0.18
    plus
    0.18
    :↵
    0.18
    Activations Density 0.043%

    No Known Activations