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. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 7965
    Prev
    Next
    INDEX
    Explanations

    variations of the verb "has" in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    verdi
    -0.83
    vode
    -0.83
    Morte
    -0.79
     Coolidge
    -0.77
     TPC
    -0.77
     olympique
    -0.77
    mitian
    -0.77
     Comer
    -0.76
     MOP
    -0.76
     Revere
    -0.76
    POSITIVE LOGITS
     has
    0.97
    ness
    0.95
     Has
    0.89
     HAS
    0.89
    nt
    0.82
    HandlerContext
    0.80
    Twas
    0.80
    has
    0.77
     έχει
    0.76
    expandindo
    0.74
    Activations Density 0.145%

    No Known Activations