© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 13742
    Prev
    Next
    INDEX
    Explanations

    French language text and also conversational turns, as indicated by the presence of text like "bonjour" and angled brackets with names

    oai_token-act-pair · gemini-2.0-flash

    French text

    np_max-act-logits · gemini-2.0-flash

    French function words Method used: 1. Reason: MAX_ACTIVATING_TOKENS (▁un, ▁mon, ▁le, ▁au, ▁de, ▁son) are French articles/pronouns

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    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
    GEBURTSDATUM
    -0.65
     AssemblyVersion
    -0.52
    racene
    -0.51
    anin
    -0.48
    SharedDtor
    -0.47
    цина
    -0.46
    verifyException
    -0.45
    chkin
    -0.45
    coc
    -0.44
    craper
    -0.44
    POSITIVE LOGITS
     contextLoads
    0.59
     propOrder
    0.56
    UserScript
    0.53
     raiſ
    0.52
     myſelf
    0.51
     ſhall
    0.49
    igshid
    0.49
     Diction
    0.48
     fhall
    0.48
    penia
    0.48
    Activations Density 0.517%

    No Known Activations