© 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
    3. Gemma-2-2B
    5. 25-GEMMASCOPE-TRANSCODER-16K
    7. 3327
    Prev
    Next
    INDEX
    Explanations

    parts of JSON files, specifically colons, quotation marks, brackets and keys that seem to denote values in JSON objects

    oai_token-act-pair · gemini-2.0-flash

    bracket

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.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
    TagMode
    -0.48
    MLLoader
    -0.47
    ät
    -0.44
     Où
    -0.43
    ه
    -0.40
     religieuses
    -0.40
     olmayan
    -0.39
    an
    -0.39
     médicale
    -0.39
     admirably
    -0.37
    POSITIVE LOGITS
    itename
    0.66
    colorPrimary
    0.65
    KEYCODE
    0.60
    InvalidProtocol
    0.59
    itieren
    0.58
    OwnProperty
    0.58
    NSMutable
    0.58
    ngdoc
    0.58
    parsedMessage
    0.58
    ngOnInit
    0.57
    Activations Density 14.371%

    No Known Activations