© 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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 10932
    Prev
    Next
    INDEX
    Explanations

    instances of the word "friendly" and related positive attributes or traits

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

    being gentle or friendly

    np_acts-logits-general · gemini-2.0-flash

    friendly, gentle, or benign descriptors

    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
    TemporalType
    -0.69
    <h5>
    -0.62
    IsContent
    -0.60
    ็จ
    -0.56
     NSCoder
    -0.52
    makeConstraints
    -0.51
    /***/
    -0.51
     nacionais
    -0.50
    кисло
    -0.49
     NgModule
    -0.48
    POSITIVE LOGITS
     reminder
    0.72
     genti
    0.70
     Genti
    0.69
     Gentle
    0.69
     nhàng
    0.67
     reminders
    0.66
    emang
    0.66
     GENTLE
    0.65
     gentle
    0.65
    ništvo
    0.64
    Activations Density 0.153%

    No Known Activations