© 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. 13184
    Prev
    Next
    INDEX
    Explanations

    mentions of studies, results, or other related information.

    oai_token-act-pair · gemini-2.0-flash

    present tense verbs

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

    research results

    np_max-act-logits · gemini-2.0-flash

    plural nouns

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    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
    elemField
    -0.71
    gets
    -0.70
    goes
    -0.68
     включает
    -0.68
    Makes
    -0.68
     sings
    -0.68
     plays
    -0.66
     itſelf
    -0.66
     делает
    -0.65
     listens
    -0.65
    POSITIVE LOGITS
     have
    1.25
     are
    1.12
     remain
    1.10
     continue
    1.09
     require
    1.09
     happen
    1.09
     occur
    1.04
     come
    1.02
     apply
    1.02
     become
    1.02
    Activations Density 3.559%

    No Known Activations