© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 41044
    Prev
    Next
    INDEX
    Explanations

    mentions of specific character names or figures in a narrative 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
    ivers
    -0.16
     thresholds
    -0.15
    xia
    -0.14
    æ²»
    -0.14
    threshold
    -0.14
     Invent
    -0.14
    åĶ®
    -0.14
     threshold
    -0.14
    æ¡IJ
    -0.13
    اÙĨÙĩ
    -0.13
    POSITIVE LOGITS
    man
    0.30
    stein
    0.28
    berg
    0.28
    baum
    0.24
    thal
    0.23
    ман
    0.22
    mann
    0.22
    handler
    0.22
    haft
    0.20
    feld
    0.20
    Activations Density 0.067%

    No Known Activations