© 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 - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 13037
    Prev
    Next
    INDEX
    Explanations

    technical terms and concepts related to detector design and evaluation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ruptedException
    -0.69
    ApplicationTests
    -0.69
    ."</
    -0.66
     endregion
    -0.65
    NUMX
    -0.65
    onViewCreated
    -0.65
     }}"></
    -0.64
    Grüße
    -0.64
    ]='\
    -0.64
    }*/
    
    -0.63
    POSITIVE LOGITS
     Although
    0.69
    Although
    0.67
    ↵↵
    0.64
    There
    0.64
     There
    0.62
     externi
    0.62
    :
    0.61
     Though
    0.60
    Though
    0.58
     Aunque
    0.57
    Activations Density 0.909%

    No Known Activations