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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 28314
    Prev
    Next
    INDEX
    Explanations

    instances of manipulation or manipulation-related concepts in political 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
     Bounty
    -0.15
    arez
    -0.14
    NameValuePair
    -0.14
    ennon
    -0.14
    ModelState
    -0.14
    .vo
    -0.14
     Nach
    -0.14
    cke
    -0.13
     Bloss
    -0.13
    .editor
    -0.13
    POSITIVE LOGITS
     Mach
    0.26
     political
    0.21
     wheel
    0.20
     politics
    0.20
     deals
    0.20
     transaction
    0.20
    wheel
    0.19
     deal
    0.19
     survivor
    0.17
     elbows
    0.17
    Activations Density 0.065%

    No Known Activations