© 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-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 13206
    Prev
    Next
    INDEX
    Explanations

    descriptions of people ending up on the ground

    oai_token-act-pair · gemini-2.0-flash

    action verbs and falling

    np_token-act-pair-logits · gpt-4o-mini

    falling or collapsing

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    sizeCache
    -0.90
     ExecuteAsync
    -0.90
    KURZBESCHREIBUNG
    -0.89
    webElementXpaths
    -0.88
    Personendaten
    -0.86
    WithIOException
    -0.85
     disambiguazione
    -0.85
     '\\;'
    -0.84
     المعيارى
    -0.82
     resourceCulture
    -0.81
    POSITIVE LOGITS
     kereszt
    0.45
     on
    0.43
    كمة
    0.40
    achts
    0.38
     llamo
    0.37
    꿈
    0.37
    tagli
    0.36
     link
    0.36
     over
    0.36
     complete
    0.36
    Activations Density 1.352%

    No Known Activations