© 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. 41310
    Prev
    Next
    INDEX
    Explanations

    references to financial stakeholders

    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
    WithOptions
    -0.17
    ube
    -0.17
    aget
    -0.16
    etto
    -0.16
    icus
    -0.16
    aybe
    -0.15
    ffen
    -0.15
    inx
    -0.15
    ignet
    -0.14
    endid
    -0.14
    POSITIVE LOGITS
     expect
    0.71
     expectations
    0.65
     Expect
    0.65
     expects
    0.65
     expectation
    0.63
     EXPECT
    0.61
     expecting
    0.60
    expect
    0.60
    Expect
    0.58
     demand
    0.57
    Activations Density 0.042%

    No Known Activations