© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 14639
    Prev
    Next
    INDEX
    Explanations

    javascript code accessing request parameters

    oai_token-act-pair · gemini-2.0-flash

    programming language constructs or keywords

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

    Code snippets

    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
    InstrumentedTest
    -0.44
    ブリッド
    -0.40
     législ
    -0.40
    خاذ
    -0.40
     يتيمه
    -0.39
     OnInit
    -0.39
    atre
    -0.38
     vie
    -0.37
    äte
    -0.37
     useContext
    -0.36
    POSITIVE LOGITS
    はじめに
    0.79
    Personendaten
    0.78
    AddTagHelper
    0.75
    }")]
    0.74
    migrationBuilder
    0.71
     للاسماء
    0.71
     архивлан
    0.69
     kaarangay
    0.65
     endwhile
    0.65
     Spisak
    0.63
    Activations Density 0.436%

    No Known Activations