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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 29029
    Prev
    Next
    INDEX
    Explanations

    expressions of love and support in social interactions

    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
    ForRow
    -0.16
    agua
    -0.15
    £p
    -0.15
    CustomAttributes
    -0.14
    幸ç¦ı
    -0.14
    uisse
    -0.14
    ÏĥοÏħ
    -0.14
    ofire
    -0.14
    isku
    -0.14
    dio
    -0.14
    POSITIVE LOGITS
     empathy
    0.30
     support
    0.30
     encouragement
    0.26
     Support
    0.25
     empath
    0.25
     compassion
    0.25
     gentle
    0.24
     understanding
    0.24
     caring
    0.23
    compass
    0.23
    Activations Density 0.108%

    No Known Activations