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. 0-GEMMASCOPE-TRANSCODER-16K
    4. 7037
    Prev
    Next
    INDEX
    Explanations

    the verb 'pick' or its variations

    oai_token-act-pair · gemini-2.0-flash

    picked

    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
     itſelf
    -1.57
     myſelf
    -1.52
     themſelves
    -1.34
     pleaſure
    -1.34
     ―――――
    -1.29
     iſt
    -1.28
     purpoſe
    -1.28
     Monfieur
    -1.26
     Efq
    -1.25
     ſta
    -1.24
    POSITIVE LOGITS
    ,
    0.92
     in
    0.88
     a
    0.88
     an
    0.84
     as
    0.81
     and
    0.78
     D
    0.73
     B
    0.72
     I
    0.71
     
    0.70
    Activations Density 0.862%

    No Known Activations