© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 41706
    Prev
    Next
    INDEX
    Explanations

    references to economic policies and government interventions

    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
    alta
    -0.14
     Imper
    -0.14
    yne
    -0.14
     Pocket
    -0.14
    ť
    -0.14
     Sonuç
    -0.13
     Wass
    -0.13
     Retrofit
    -0.13
    609
    -0.13
    lednÃŃ
    -0.13
    POSITIVE LOGITS
     Interest
    0.23
     monetary
    0.23
     stim
    0.23
     easing
    0.22
     stimulus
    0.22
     interest
    0.22
     accommod
    0.21
     Monetary
    0.20
     quantitative
    0.20
     Quant
    0.20
    Activations Density 0.051%

    No Known Activations