© 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 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. 14055
    Prev
    Next
    INDEX
    Explanations

    references to deception or subterfuge, particularly in the form of "Trojan horse" concepts

    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
    coop
    -0.16
    Dispatch
    -0.15
    SGlobal
    -0.15
    uth
    -0.15
    Hashtable
    -0.15
     FetchType
    -0.14
    /GPL
    -0.14
    nty
    -0.14
    antha
    -0.14
    decorators
    -0.14
    POSITIVE LOGITS
     horse
    0.84
     horses
    0.76
     Horse
    0.72
    horse
    0.72
    orse
    0.43
     horsepower
    0.41
     pony
    0.40
     HOR
    0.39
     hors
    0.39
    orses
    0.37
    Activations Density 0.010%

    No Known Activations