© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 6264
    Prev
    Next
    INDEX
    Explanations

    language associated with statistical meta-analysis

    oai_token-act-pair · gemini-2.0-flash

    tokens beginning with 'ma' or 'un'

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

    percent

    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
     nakalista
    -0.90
    WebElementEntity
    -0.89
     Roskov
    -0.86
    SourceChecksum
    -0.84
     Efq
    -0.81
     ſeveral
    -0.77
    FieldBuilder
    -0.77
    endphp
    -0.77
     الحره
    -0.76
    fjspx
    -0.76
    POSITIVE LOGITS
    ma
    0.42
     
    0.42
     un
    0.39
    1
    0.37
    tapan
    0.36
     unc
    0.36
    MAI
    0.35
    oboda
    0.35
     mun
    0.35
     ma
    0.35
    Activations Density 3.000%

    No Known Activations