© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 1274
    Prev
    Next
    INDEX
    Explanations

    phrases and terms related to novelty and the introduction of new concepts or products

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

    references to newly released content, particularly in adult entertainment

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

    something is new

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    csolódó
    -0.62
    OCCURRED
    -0.61
    anair
    -0.57
    ulose
    -0.54
    TypedDataSet
    -0.53
    jsonwebtoken
    -0.53
    vocato
    -0.52
    ReadKey
    -0.52
    oyu
    -0.52
    abinieri
    -0.51
    POSITIVE LOGITS
     newest
    0.76
     new
    0.76
    new
    0.73
     nouve
    0.72
     newer
    0.70
     NEW
    0.69
     novità
    0.65
     nuevas
    0.65
     arrival
    0.65
     nuevo
    0.65
    Activations Density 0.040%

    No Known Activations