© 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. 22137
    Prev
    Next
    INDEX
    Explanations

    interactions and questions within comments or discussions

    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
    pron
    -0.18
    оÑī
    -0.16
     pron
    -0.15
     Pron
    -0.14
    vla
    -0.14
    άÏģÏĩ
    -0.14
    iani
    -0.14
    _Controller
    -0.14
    _DLL
    -0.13
     Wikimedia
    -0.13
    POSITIVE LOGITS
     user
    0.33
     viewer
    0.29
     reader
    0.26
    user
    0.26
    çĶ¨æĪ·
    0.24
     Anonymous
    0.24
     someone
    0.23
    олÑĮзоваÑĤ
    0.23
     anonymous
    0.22
    -user
    0.22
    Activations Density 0.043%

    No Known Activations