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. 3-GEMMASCOPE-RES-16K
    6. 9998
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate diversity or variation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     nucléaire
    -0.79
     vectorielle
    -0.74
     yoksa
    -0.73
     dessutom
    -0.73
     fumée
    -0.73
     operazioni
    -0.70
     flèche
    -0.70
     sólidos
    -0.68
     poitrine
    -0.68
     larmes
    -0.68
    POSITIVE LOGITS
     different
    2.90
    Different
    2.64
    different
    2.62
     Different
    2.56
     DIFFERENT
    2.25
     diferentes
    1.93
     diferente
    1.82
    不同
    1.80
    不同的
    1.79
     différentes
    1.78
    Activations Density 0.164%

    No Known Activations