© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 25988
    Prev
    Next
    INDEX
    Explanations

    references to feelings and interpersonal dynamics in political contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    endif
    -0.15
    cmp
    -0.14
    ãģĿãĤĮãģ¯
    -0.14
     Ø¥ÙĦا
    -0.13
    èµĸ
    -0.13
    ertest
    -0.13
    olem
    -0.13
    Ø´ÙĪر
    -0.13
    окон
    -0.13
    IfNeeded
    -0.12
    POSITIVE LOGITS
     rather
    1.46
    rather
    1.18
     Rather
    1.11
     instead
    1.03
    Rather
    1.01
    instead
    0.90
     Instead
    0.79
     plutôt
    0.76
    Instead
    0.73
     chứ
    0.72
    Activations Density 0.219%

    No Known Activations