Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 12753
    Prev
    Next
    INDEX
    Explanations

    terms associated with musical instruments and their components

    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
     olympiques
    -0.68
     dysplasia
    -0.66
     noires
    -0.62
     éché
    -0.62
     silica
    -0.62
    raujo
    -0.61
     eggs
    -0.60
     Beek
    -0.60
    îtra
    -0.60
     Turch
    -0.60
    POSITIVE LOGITS
     instruments
    1.60
     Instruments
    1.47
     instrument
    1.44
     Instrument
    1.43
     INSTRUMENTS
    1.33
     INSTRUMENT
    1.32
    instrument
    1.28
    Instrument
    1.28
    instruments
    1.24
    Instruments
    1.24
    Activations Density 0.235%

    No Known Activations