© 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. 25801
    Prev
    Next
    INDEX
    Explanations

    phrases related to aggregating services or information into a single platform

    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
    uns
    -0.15
    uyu
    -0.14
    enga
    -0.14
     uns
    -0.13
    /copyleft
    -0.13
    سÙħ
    -0.13
     slova
    -0.13
     subtype
    -0.13
    олом
    -0.13
     Leaf
    -0.12
    POSITIVE LOGITS
     central
    0.37
     centralized
    0.33
    central
    0.32
    -central
    0.29
     single
    0.28
     place
    0.28
    single
    0.26
     ONE
    0.25
     one
    0.25
    Central
    0.24
    Activations Density 0.025%

    No Known Activations