© 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. 695
    Prev
    Next
    INDEX
    Explanations

    soil compaction

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

    content related to plant care conditions, soil properties, and environmental factors affecting plant growth.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron strongly activates on very common, high‐frequency “function” words (e.g. in, with, but, will, they) rather than on rarer content words.

    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
    رژی
    -0.88
    Gustav
    -0.83
     PROTECT
    -0.82
    delser
    -0.81
    כים
    -0.81
     PRIM
    -0.81
     Belles
    -0.80
    lüğ
    -0.80
     cori
    -0.79
     dirt
    -0.79
    POSITIVE LOGITS
     compaction
    1.44
     compacted
    1.42
     compact
    1.28
    compact
    1.13
     Compact
    1.10
    Compact
    0.99
     anaerobic
    0.98
     kompakt
    0.98
     compactness
    0.96
     excessive
    0.93
    Activations Density 0.014%

    No Known Activations