© 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. 21560
    Prev
    Next
    INDEX
    Explanations

    terms related to the assessment of health and its discrepancies between patients and physicians

    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
     divisible
    -0.16
    leta
    -0.16
     lul
    -0.14
    ãģĵãĤĵãģ«ãģ¡ãģ¯
    -0.14
    Å¡etÅĻenÃŃ
    -0.14
    VML
    -0.14
    ä»ģ
    -0.13
    ubby
    -0.13
    ëįĶëĭĪ
    -0.13
    ç§ĭ
    -0.13
    POSITIVE LOGITS
     reported
    0.29
     ratings
    0.27
     rated
    0.25
    -reported
    0.25
    reported
    0.24
     perceived
    0.24
    -rated
    0.22
     satisfaction
    0.22
     Satisfaction
    0.22
     feelings
    0.22
    Activations Density 0.132%

    No Known Activations