© 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. 21-GEMMASCOPE-TRANSCODER-16K
    7. 12202
    Prev
    Next
    INDEX
    Explanations

    mentions of a dog named Spike behaving badly, and the speaker's reactions.

    oai_token-act-pair · gemini-2.0-flash

    words related to facial expressions

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

    facial expressions

    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
     Italijanski
    -0.88
    Portail
    -0.83
    Portale
    -0.79
    awaiter
    -0.76
    aarrggbb
    -0.75
    AddTagHelper
    -0.75
    PRNewswire
    -0.75
    mybatisplus
    -0.74
    tagHelperRunner
    -0.73
    الحياه
    -0.71
    POSITIVE LOGITS
     face
    0.95
     faces
    0.84
     tone
    0.74
     facial
    0.71
     Face
    0.69
    顔
    0.68
    face
    0.67
     expression
    0.66
    faces
    0.63
     expressions
    0.62
    Activations Density 1.883%

    No Known Activations