Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 22-GEMMASCOPE-TRANSCODER-16K
    4. 14465
    Prev
    Next
    INDEX
    Explanations

    grammatical constructs such as "with respect to", "not yet", "as well as", and location names

    oai_token-act-pair · gemini-2.0-flash

    Non-English words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.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
     
    -0.92
    ,
    -0.84
     the
    -0.84
     (
    -0.84
     in
    -0.80
     to
    -0.75
     a
    -0.75
     T
    -0.73
    ↵↵
    -0.73
    <eos>
    -0.73
    POSITIVE LOGITS
     Theſe
    1.41
     Efq
    1.26
     auffi
    1.25
     Monfieur
    1.21
    parsedMessage
    1.19
    aarrggbb
    1.17
     pleaſure
    1.16
     ſche
    1.16
     ―――――
    1.11
     للمعارف
    1.10
    Activations Density 4.453%

    No Known Activations