© 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. 12553
    Prev
    Next
    INDEX
    Explanations

    phrases that highlight significant threats or challenges

    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
    umi
    -0.19
    inecraft
    -0.14
    ÅĻet
    -0.14
    å°ij女
    -0.14
    vv
    -0.14
    ican
    -0.14
    ido
    -0.13
    ather
    -0.13
    athers
    -0.13
    erek
    -0.13
    POSITIVE LOGITS
     challenge
    0.23
     lesson
    0.22
     single
    0.21
     danger
    0.21
     enemy
    0.19
     Lesson
    0.19
    single
    0.19
     threat
    0.18
     thing
    0.18
     asset
    0.18
    Activations Density 0.048%

    No Known Activations