© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 49399
    Prev
    Next
    INDEX
    Explanations

    falling into or for something

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

    The neuron fires on occurrences of the verb “fall” in its various forms (fell, fallen, falling).

    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
     solidaridad
    -0.82
     Reisedaten
    -0.80
     articula
    -0.79
    を持った
    -0.76
     escolares
    -0.75
     अब
    -0.75
     Medik
    -0.75
    btn
    -0.75
     Rosales
    -0.74
    作品の
    -0.74
    POSITIVE LOGITS
     victim
    1.73
     into
    1.69
     asleep
    1.64
     prey
    1.51
    victim
    1.34
    Victim
    1.17
     Victim
    1.14
     succumbed
    1.02
    fell
    1.00
     back
    0.99
    Activations Density 0.008%

    No Known Activations