© 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. 0-GEMMASCOPE-TRANSCODER-16K
    7. 6752
    Prev
    Next
    INDEX
    Explanations

    snippets of text from a variety of formal contexts such as product reviews, user data, court cases and movie cast lists

    oai_token-act-pair · gemini-2.0-flash

    Software/legal text snippets

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.0.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
    <eos>
    -1.28
    ↵↵
    -1.16
    ↵
    -1.09
    ↵↵↵
    -1.00
    ↵↵↵↵
    -0.94
     I
    -0.78
     my
    -0.73
    ↵↵↵↵↵↵
    -0.71
     ...
    -0.71
    ↵↵↵↵↵
    -0.69
    POSITIVE LOGITS
     itſelf
    1.51
     myſelf
    1.29
     Jefus
    1.24
     Reſ
    1.23
     houſe
    1.23
     Monfieur
    1.21
     raiſ
    1.20
     pleaſure
    1.20
     himſelf
    1.20
     Majefty
    1.20
    Activations Density 3.760%

    No Known Activations