© 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. 7213
    Prev
    Next
    INDEX
    Explanations

    capitalized letters in the middle of a word preceded by lowercase letters and words related to youth

    oai_token-act-pair · gemini-2.0-flash

    Mix of languages/code

    np_max-act-logits · gemini-2.0-flash
    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
     Majefty
    -1.30
     Efq
    -1.27
     Chriftian
    -1.26
     itſelf
    -1.25
     myſelf
    -1.22
     Anſ
    -1.20
     Houſe
    -1.16
     Diſ
    -1.16
     Monfieur
    -1.13
     raiſ
    -1.13
    POSITIVE LOGITS
     die
    0.53
     door
    0.44
     op
    0.44
     identificación
    0.44
     zo
    0.43
     bu
    0.42
     k
    0.41
     z
    0.40
     v
    0.40
     ste
    0.40
    Activations Density 0.448%

    No Known Activations