© 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. 91833
    Prev
    Next
    INDEX
    Explanations

    initial followed by a process or state

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

    The neuron is detecting occurrences of the word “initial” (in any casing).

    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
    recently
    -1.40
    currently
    -1.38
    considering
    -1.33
    becomes
    -1.32
    eventually
    -1.30
    previously
    -1.29
    鱚
    -1.28
    allowing
    -1.27
     quatrième
    -1.27
    becoming
    -1.25
    POSITIVE LOGITS
     initial
    1.59
     glance
    1.47
     Initial
    1.42
    Initial
    1.38
    initial
    1.36
     assessment
    1.35
     stages
    1.30
     setup
    1.23
     set
    1.20
     burst
    1.20
    Activations Density 0.034%

    No Known Activations