© 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. 25749
    Prev
    Next
    INDEX
    Explanations

    mentions of the United States and terms related to age or stages of 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
    ÑģÑĤанов
    -0.14
    ega
    -0.14
    orthand
    -0.14
    oup
    -0.14
    SocketAddress
    -0.13
    /******/
    -0.13
    ysi
    -0.13
    रल
    -0.13
    .VisualBasic
    -0.13
    ispiel
    -0.12
    POSITIVE LOGITS
     portion
    1.06
     portions
    0.97
     part
    0.93
     parts
    0.88
     parte
    0.83
    éĥ¨åĪĨ
    0.81
    portion
    0.78
     Portions
    0.70
     aspects
    0.70
     aspect
    0.67
    Activations Density 0.018%

    No Known Activations