© 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. 23-GEMMASCOPE-TRANSCODER-16K
    7. 14287
    Prev
    Next
    INDEX
    Explanations

    phrases like "a very", "The [Adjective] Ones", "First [Word]", "So [Adjective]", "coming very", and some brand names or company positions, possibly personal writing.

    oai_token-act-pair · gemini-2.0-flash

    personal writing

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_23/width_16k/average_l0_25
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.23.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     purpoſe
    -0.96
    AsUp
    -0.93
     myſelf
    -0.93
     itſelf
    -0.93
     uſed
    -0.90
     deſt
    -0.90
     uſe
    -0.89
     themſelves
    -0.88
     poffible
    -0.86
     Roskov
    -0.86
    POSITIVE LOGITS
     B
    0.42
    博
    0.41
     Lo
    0.40
     Almost
    0.40
     lo
    0.39
     G
    0.38
     b
    0.38
     Next
    0.37
     Ol
    0.37
    ände
    0.36
    Activations Density 6.990%

    No Known Activations