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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 7906
    Prev
    Next
    INDEX
    Explanations

    expressions of uncertainty and conjecture

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Comes before modal verbs ("would", "might", "may", "could")

    np_acts-logits-general · gemini-2.0-flash

    modal verb followed by 'have'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    were
    -0.88
     been
    -0.86
    been
    -0.78
    was
    -0.77
     Was
    -0.72
    Were
    -0.71
    Was
    -0.70
    Been
    -0.69
     were
    -0.69
     Did
    -0.66
    POSITIVE LOGITS
     للاسماء
    1.12
     itſelf
    1.04
     Efq
    1.02
     leaſt
    0.98
     myſelf
    0.97
     chofe
    0.97
     Houſe
    0.96
    ########.
    0.94
     houſe
    0.91
     Monfieur
    0.90
    Activations Density 0.097%

    No Known Activations