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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 21465
    Prev
    Next
    INDEX
    Explanations

    references to gaming and character development

    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
    ¹Ħ
    -0.14
    ynchronize
    -0.13
     ucwords
    -0.13
    eton
    -0.13
    ymm
    -0.13
    arcer
    -0.13
    oldemort
    -0.13
    lili
    -0.13
     INCIDENTAL
    -0.13
    enberg
    -0.13
    POSITIVE LOGITS
     again
    1.67
    again
    1.49
     Again
    1.41
    Again
    1.34
     AGAIN
    1.09
    AGAIN
    1.03
     wieder
    1.02
     novamente
    0.99
    _again
    0.96
    åĨį
    0.85
    Activations Density 0.091%

    No Known Activations