© 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-2-2B
    3. 21-GEMMASCOPE-TRANSCODER-16K
    4. 12173
    Prev
    Next
    INDEX
    Explanations

    occurrences of "and" and various words from the start of sentences (e.g. "the", "i", "cade")

    oai_token-act-pair · gemini-2.0-flash

    terms related to technology and gaming

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

    Video games

    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
    RUnlock
    -0.74
     doInBackground
    -0.65
    +#+
    -0.65
    rrggbb
    -0.63
     communiquez
    -0.62
    stahl
    -0.61
    omock
    -0.61
    __(/*!
    -0.60
     الحره
    -0.58
     мәкал
    -0.57
    POSITIVE LOGITS
     Switch
    0.58
     Super
    0.54
     switch
    0.54
    Switch
    0.50
    amek
    0.49
     pipa
    0.47
     super
    0.47
    Super
    0.45
    AppMethodBeat
    0.45
    super
    0.44
    Activations Density 1.898%

    No Known Activations