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. 25-GEMMASCOPE-RES-16K
    6. 3393
    Prev
    Next
    INDEX
    Explanations

    references to commands or actions related to software or settings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ch
    -0.93
     W
    -0.93
     Ne
    -0.92
     Ro
    -0.90
     Du
    -0.88
     Se
    -0.87
     Am
    -0.86
     Su
    -0.86
     Le
    -0.85
     Har
    -0.85
    POSITIVE LOGITS
     disambiguazione
    0.81
    Autoritní
    0.81
     supérieurs
    0.78
     Paving
    0.77
     découver
    0.77
     Wreath
    0.77
     cérami
    0.77
     Fittings
    0.77
     Restra
    0.76
     réfrig
    0.76
    Activations Density 0.057%

    No Known Activations