© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 83062
    Prev
    Next
    INDEX
    Explanations

    steep ascent

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

    The neuron fires strongly on phrasing that describes climbing or ascending steep terrain (e.g. “climb,” “straight up,” “up … m to the ridge”).

    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
    tellingen
    -0.87
     zeigt
    -0.82
     drive
    -0.82
    叭
    -0.81
    wingen
    -0.79
    drive
    -0.78
    麓
    -0.78
     mowing
    -0.77
     transactions
    -0.77
     Ride
    -0.77
    POSITIVE LOGITS
     steep
    1.41
     stairs
    1.13
     strenuous
    1.12
     steeper
    1.10
     cardio
    1.04
    Reward
    1.01
     ascent
    1.01
    Cardio
    1.01
     climb
    1.00
     rewarded
    0.97
    Activations Density 0.014%

    No Known Activations