© 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. 12191
    Prev
    Next
    INDEX
    Explanations

    instances of the word "hustle" and its variations, indicating a focus on industriousness or determination

    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
    omik
    -0.16
    룸
    -0.15
    ureka
    -0.14
    urette
    -0.14
    AZE
    -0.14
    ety
    -0.14
    irket
    -0.14
    ately
    -0.14
     Chief
    -0.14
    438
    -0.14
    POSITIVE LOGITS
    lers
    0.24
    ler
    0.22
    wit
    0.20
    ling
    0.19
    oldt
    0.17
    led
    0.17
     Hust
    0.16
    LER
    0.16
    wife
    0.15
    les
    0.15
    Activations Density 0.002%

    No Known Activations