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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 22708
    Prev
    Next
    INDEX
    Explanations

    detailed descriptions of gardens and parks

    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
    usra
    -0.16
     ticking
    -0.15
    isor
    -0.15
     köy
    -0.14
    IFY
    -0.14
    okit
    -0.14
    anggal
    -0.14
    estar
    -0.14
    ForObject
    -0.14
    çĶº
    -0.14
    POSITIVE LOGITS
     benches
    0.29
     fountain
    0.29
     paths
    0.27
     bench
    0.26
     ponds
    0.25
     pond
    0.25
     pathways
    0.25
     path
    0.24
     Fountain
    0.24
    bench
    0.23
    Activations Density 0.020%

    No Known Activations