© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 40773
    Prev
    Next
    INDEX
    Explanations

    descriptions of women, particularly those with long hair and attractive features

    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
    cki
    -0.16
    andles
    -0.15
    andle
    -0.15
    andler
    -0.15
    olf
    -0.14
    âĸ¡âĸ¡
    -0.14
    ecal
    -0.14
    arov
    -0.14
    ANDLE
    -0.14
    üf
    -0.14
    POSITIVE LOGITS
     hair
    0.68
     Hair
    0.58
    Hair
    0.56
    hair
    0.53
     волоÑģ
    0.44
     locks
    0.42
     hairstyle
    0.40
     hairs
    0.40
    髪
    0.39
     hairstyles
    0.38
    Activations Density 0.035%

    No Known Activations