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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 5842
    Prev
    Next
    INDEX
    Explanations

    This neuron is hard to characterize, but it seems to activate on a mix of seemingly unrelated words, sometimes highlighting acronyms, numbers, and words related to research studies and airline companies.

    oai_token-act-pair · gemini-2.0-flash

    French words

    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
     AK
    -0.84
    af
    -0.79
     AH
    -0.79
     ak
    -0.78
     APS
    -0.77
     AF
    -0.76
     AP
    -0.76
     ag
    -0.76
    ap
    -0.75
     ACP
    -0.75
    POSITIVE LOGITS
    émission
    0.60
    investissement
    0.59
    étranger
    0.58
    humanité
    0.58
    intérieur
    0.57
    extérieur
    0.56
    échelle
    0.56
    essentiel
    0.56
    enfance
    0.56
    entraînement
    0.55
    Activations Density 1.852%

    No Known Activations