© 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. 28999
    Prev
    Next
    INDEX
    Explanations

    phrases indicating criticism or scrutiny towards media and political figures

    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
    canf
    -0.18
    ãĥŃãĥ³
    -0.15
    ken
    -0.14
    imen
    -0.14
    ispers
    -0.14
     pride
    -0.14
     reserve
    -0.13
     multiples
    -0.13
     unnatural
    -0.13
    bia
    -0.13
    POSITIVE LOGITS
     about
    0.35
    about
    0.26
     ABOUT
    0.26
    _about
    0.25
     About
    0.24
    -about
    0.24
    About
    0.23
     tentang
    0.23
    .about
    0.23
    åħ³äºİ
    0.22
    Activations Density 0.038%

    No Known Activations