© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 2944
    Prev
    Next
    INDEX
    Explanations

    references to the season of Fall

    oai_token-act-pair · gemini-2.0-flash

    say seasons

    np_max-act-logits · gemini-2.0-flash

    say "seasons" Method used: 2 — tokens after max all indicate seasons

    np_max-act-logits · gpt-5-mini
    Triggered by @chenshw0109
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     spring
    -1.77
     winter
    -1.76
    Spring
    -1.75
     Spring
    -1.74
     summer
    -1.72
     Summer
    -1.71
     Winter
    -1.70
    Winter
    -1.67
    Summer
    -1.67
    winter
    -1.66
    POSITIVE LOGITS
     late
    0.53
    UIControlState
    0.52
    mybatisplus
    0.49
    
    0.48
    pektive
    0.48
    Cuáles
    0.47
    AsUp
    0.47
    FormState
    0.44
     early
    0.43
    Early
    0.43
    Activations Density 1.197%

    No Known Activations