Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 23326
    Prev
    Next
    INDEX
    Explanations

    statements challenging religious morality

    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
    411
    -0.16
    ometr
    -0.15
     Trot
    -0.15
    ubes
    -0.15
    ühl
    -0.15
     ëĭ¤ìļ´ë°Ľê¸°
    -0.14
    è°Ī
    -0.14
    "label
    -0.14
    onder
    -0.14
    .fd
    -0.14
    POSITIVE LOGITS
     forcing
    0.21
     serve
    0.21
     serves
    0.20
     forces
    0.20
     opens
    0.20
    force
    0.20
     contribution
    0.20
     Force
    0.20
     Forced
    0.20
     Forces
    0.19
    Activations Density 0.068%

    No Known Activations