© 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. 10-GEMMASCOPE-RES-131K
    4. 42786
    Prev
    Next
    INDEX
    Explanations

    cell

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

    the word "cell" in programming contexts.

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

    The neuron selectively activates on the word "cell" (and its variants like “cells” or “cell_”) in code or configuration contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    镞
    -2.33
    愎
    -2.11
    verture
    -1.98
    silien
    -1.98
    无声
    -1.97
    AutoresizingMask
    -1.96
    WRENCE
    -1.96
     dritte
    -1.95
    𒊹
    -1.95
     Kombination
    -1.91
    POSITIVE LOGITS
    u
    2.58
    er
    2.55
    quíni
    2.52
     selepas
    2.42
    يديا
    2.39
    ၟ
    2.38
    नलोड
    2.38
    e
    2.31
    )
    2.31
    
    2.30
    Activations Density 0.005%

    No Known Activations