© 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 - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 0
    Prev
    Next
    INDEX
    Explanations

    terms and phrases related to legal disclaimers and conditions

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

    legal disclaimers regarding warranties and liabilities

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

    KIND

    np_acts-logits-general · gemini-2.5-flash-lite

    The marked tokens appear in technical and scientific documentation across diverse domains (software licenses, source code, academic papers, HTML/web content). The marked tokens represent various linguistic elements: common noun phrases ("KIND"), URL protocol prefixes ("://"), LaTeX commands and symbols ("begin", "!"), HTML entities ("amp", "lementary"), and mathematical notation markers ("!"). These are content-specific tokens that serve structural or semantic purposes within their respective contexts rather than expressing a single unified pattern.

    eleuther_acts_top20 · claude-4-5-haikuTriggered by @kparkhamchuk
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    /*
    -0.68
     doInBackground
    -0.64
    OGND
    -0.62
    //
    -0.60
    ως
    -0.59
     nahilalakip
    -0.59
    Slf
    -0.58
    featureID
    -0.57
    prüche
    -0.57
    oneofs
    -0.56
    POSITIVE LOGITS
    mml
    0.71
     للمعارف
    0.65
     embora
    0.59
    VideoCapture
    0.57
     CanadaChoose
    0.54
     biogas
    0.52
     پرد
    0.51
    httphttps
    0.51
    kowym
    0.51
     σχέ
    0.50
    Activations Density 0.003%

    No Known Activations