© 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. 12257
    Prev
    Next
    INDEX
    Explanations

    words relating to water and dryness

    oai_token-act-pair · gemini-2.0-flash

    Wetness/moisture

    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
     wet
    -1.88
    Wet
    -1.80
     Wet
    -1.76
    wet
    -1.73
     WET
    -1.70
     dry
    -1.66
     moist
    -1.59
    WET
    -1.55
     Dry
    -1.52
    dry
    -1.47
    POSITIVE LOGITS
     Sambo
    0.48
     Schmuck
    0.48
     boks
    0.46
     pacemaker
    0.43
     Benito
    0.42
    Charm
    0.42
    ujudkan
    0.42
     Mackie
    0.41
    ComponentName
    0.41
     Cavalier
    0.41
    Activations Density 2.103%

    No Known Activations