© 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 EvalsExportsinterp-engineNEWAPI 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. 12792
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the attribution of information to anonymous sources or officials

    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
    alin
    -0.18
     himself
    -0.16
    ubbo
    -0.16
    ritch
    -0.15
    265
    -0.14
     aut
    -0.14
    a
    -0.14
    ìłĢ
    -0.14
    elp
    -0.14
    mi
    -0.14
    POSITIVE LOGITS
     anonymity
    0.32
     anonymous
    0.27
     anonymously
    0.27
    onym
    0.26
    anonymous
    0.25
     Anonymous
    0.24
    onymous
    0.24
     anonym
    0.24
    Anonymous
    0.22
     request
    0.21
    Activations Density 0.018%

    No Known Activations