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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 54347
    Prev
    Next
    INDEX
    Explanations

    references to military conflicts and related actions in geopolitical contexts

    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
    inton
    -0.14
     respectively
    -0.13
     belie
    -0.12
     Persons
    -0.12
     jointly
    -0.12
     arbitrarily
    -0.12
     solely
    -0.11
     strictly
    -0.11
    ince
    -0.11
    .volley
    -0.11
    POSITIVE LOGITS
    OffsetTable
    0.16
    GuidId
    0.14
    tml
    0.14
     addCriterion
    0.13
    ubi
    0.13
    ШÐIJ
    0.12
     гла
    0.12
     escorte
    0.12
     meanwhile
    0.12
    uld
    0.12
    Activations Density 0.013%

    No Known Activations