© 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. 43093
    Prev
    Next
    INDEX
    Explanations

    phrases indicating personal opinions and subjective experiences

    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
    apon
    -0.15
    éŀ
    -0.15
    unger
    -0.15
    otomy
    -0.15
    uali
    -0.14
    outil
    -0.14
    ordova
    -0.14
    iland
    -0.14
    .hasNext
    -0.14
    bject
    -0.14
    POSITIVE LOGITS
     fond
    0.39
     partial
    0.35
     big
    0.30
     Fond
    0.29
     known
    0.26
    Partial
    0.26
    partial
    0.25
     Partial
    0.25
    big
    0.23
     fans
    0.22
    Activations Density 0.072%

    No Known Activations