© 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. 44309
    Prev
    Next
    INDEX
    Explanations

    phrases related to the construction or formation of something, particularly in a collaborative context

    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
    actus
    -0.16
    acco
    -0.15
    enberg
    -0.14
    meer
    -0.14
    street
    -0.14
     Ville
    -0.14
    884
    -0.14
    ãĥĪãĥª
    -0.13
     street
    -0.13
     cover
    -0.13
    POSITIVE LOGITS
     nich
    0.31
     carve
    0.27
     niche
    0.27
     carving
    0.24
     carved
    0.24
     out
    0.23
     slice
    0.18
    nic
    0.18
     path
    0.18
     nick
    0.18
    Activations Density 0.010%

    No Known Activations