© 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. 84720
    Prev
    Next
    INDEX
    Explanations

    placeholder for inputs

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

    The neuron is primarily activated by “placeholder=” attributes in HTML (and similar placeholder markup).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the word "placeholder" in HTML and code contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    mentions of placeholders in technical text or code, especially UI/input placeholder attributes and placeholder images.

    oai_token-act-pair · gpt-5Triggered 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
     COMERC
    -1.01
     versioni
    -0.92
     péché
    -0.92
    uña
    -0.91
    tubers
    -0.90
     affari
    -0.86
     皮肤
    -0.86
     perfetta
    -0.86
    asantry
    -0.84
    probab
    -0.82
    POSITIVE LOGITS
    Enter
    1.59
     Enter
    1.55
     enter
    1.43
    enter
    1.41
     your
    1.26
    TextColor
    1.13
    ENTER
    1.13
    请输入
    1.08
     ENTER
    1.06
    ütfen
    0.91
    Activations Density 0.001%

    No Known Activations