© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 97789
    Prev
    Next
    INDEX
    Explanations

    full-length

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires whenever it sees the word “full,” for example as part of compounds like “full-time,” “full-length,” etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     opinión
    -0.93
     observa
    -0.93
    个
    -0.78
     seinen
    -0.75
    /")
    -0.75
    NEE
    -0.74
     deciso
    -0.73
     marca
    -0.72
     forma
    -0.71
    Policy
    -0.71
    POSITIVE LOGITS
    scale
    0.84
     scale
    0.78
    EMPO
    0.76
    lecteur
    0.76
    teens
    0.73
     fully
    0.73
     treatment
    0.73
     đầy
    0.72
    reatment
    0.71
    ifiquement
    0.71
    Activations Density 0.039%

    No Known Activations