© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 6070
    Prev
    Next
    INDEX
    Explanations

    emotional responses, particularly those related to feelings of hurt and personal impact

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Emotional harm or offense

    np_acts-logits-general · gemini-2.0-flash

    hurt feelings and ego

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     تضيفلها
    -0.67
    évaluateur
    -0.66
    awaiter
    -0.64
     transférez
    -0.58
    ագրություններ
    -0.57
    //
    -0.56
    windowFixed
    -0.56
    WebElementEntity
    -0.54
    Xna
    -0.52
    ExtendWith
    -0.52
    POSITIVE LOGITS
     pride
    1.33
     ego
    1.14
     dignity
    1.11
     feelings
    1.09
     egos
    1.07
    pride
    1.06
     hurt
    1.05
     offended
    0.99
     sensibilities
    0.93
     sensitivities
    0.91
    Activations Density 0.161%

    No Known Activations