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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 11742
    Prev
    Next
    INDEX
    Explanations

    This neuron detects a mix of programming code snippets, portuguese language and the phrase "I can't help".

    oai_token-act-pair · gemini-2.0-flash

    non-english

    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
     but
    -3.89
    but
    -3.22
     But
    -3.08
    But
    -3.03
     pero
    -2.58
     BUT
    -2.56
     tetapi
    -2.47
     nhưng
    -2.39
    但
    -2.27
    BUT
    -2.25
    POSITIVE LOGITS
    findpost
    0.63
     referenties
    0.58
     disponibilités
    0.53
    Atsauces
    0.52
    SourceChecksum
    0.49
    FormTagHelper
    0.49
    eqn
    0.48
    γέν
    0.47
     iné
    0.47
     betweenstory
    0.47
    Activations Density 0.366%

    No Known Activations