© 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. 13071
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns followed by words indicating ownership or association

    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
    ville
    -0.16
    åħ¸
    -0.15
    quam
    -0.15
    zin
    -0.14
    erif
    -0.14
    PÅĻi
    -0.14
    familia
    -0.14
    leon
    -0.14
    ospace
    -0.13
    Rocket
    -0.13
    POSITIVE LOGITS
     colleagues
    0.45
     team
    0.44
     colleague
    0.41
     co
    0.34
    team
    0.32
     coworkers
    0.31
     teammates
    0.30
     associates
    0.30
     collaborators
    0.28
     teammate
    0.28
    Activations Density 0.027%

    No Known Activations