© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 26199
    Prev
    Next
    INDEX
    Explanations

    negative phrases or words suggesting disapproval

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    ondo
    -0.15
    idity
    -0.15
    iance
    -0.15
    onde
    -0.14
    æĵļ
    -0.14
    रत
    -0.13
    Å¡i
    -0.13
    ogl
    -0.13
    ãĥł
    -0.13
     dol
    -0.12
    POSITIVE LOGITS
     mention
    0.59
     Mention
    0.49
    mention
    0.49
     mentions
    0.46
     mentioning
    0.45
     mentioned
    0.40
    mentioned
    0.34
    mentions
    0.33
     menc
    0.32
     Ment
    0.30
    Activations Density 0.016%

    No Known Activations