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. 18-GEMMASCOPE-RES-16K
    6. 1497
    Prev
    Next
    INDEX
    Explanations

    references to legislative or regulatory frameworks

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

    understanding, awareness, or knowledge

    np_acts-logits-general · gemini-2.0-flash

    knowing or realizing facts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MigrationBuilder
    -0.72
    RTDA
    -0.71
    uestamente
    -0.68
    RegressionTest
    -0.65
     Paglinawan
    -0.64
    artifactId
    -0.63
    localctx
    -0.62
    kháu
    -0.62
    
    -0.62
    AccessorTable
    -0.61
    POSITIVE LOGITS
     realize
    1.37
     realise
    1.33
     realizes
    1.31
     realized
    1.30
     realised
    1.28
     realises
    1.28
     knew
    1.13
     realizing
    1.11
     realization
    1.09
     realising
    1.08
    Activations Density 0.352%

    No Known Activations