© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 27223
    Prev
    Next
    INDEX
    Explanations

    instances of the word "thorough" and its variations

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

    thoroughly examined

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    stateProvider
    -0.81
    ReusableCell
    -0.79
    Angelo
    -0.70
    Aza
    -0.69
    anlage
    -0.68
     spesa
    -0.65
     Cardona
    -0.65
    etzt
    -0.65
    anlagen
    -0.64
    djangoproject
    -0.64
    POSITIVE LOGITS
    oughly
    1.27
     Thorough
    1.02
     thoroughly
    0.98
     thorough
    0.94
     THOR
    0.91
    bred
    0.90
    thor
    0.87
    Thor
    0.85
    HOUT
    0.82
     Hout
    0.79
    Activations Density 0.014%

    No Known Activations