© 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
    3. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 8095
    Prev
    Next
    INDEX
    Explanations

    code, specifically C++ code for the Mangos game engine, identifying spells and units

    oai_token-act-pair · gemini-2.0-flash

    say "short capitalized words"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     parlour
    -0.81
     prostitute
    -0.77
     anthrop
    -0.74
     pleaſure
    -0.74
     barbarians
    -0.73
     fountains
    -0.73
     labours
    -0.72
     fairies
    -0.71
     crests
    -0.71
     photolibrary
    -0.71
    POSITIVE LOGITS
     He
    0.49
     ng
    0.49
    LikeLiked
    0.47
     µ
    0.47
    uchter
    0.46
    AddHtmlAttribute
    0.46
     If
    0.44
    ParallelGroup
    0.44
     total
    0.44
    VersionUID
    0.44
    Activations Density 10.753%

    No Known Activations