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. 22-GEMMASCOPE-RES-16K
    6. 11153
    Prev
    Next
    INDEX
    Explanations

    phrases and relative positions within sentences

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

    "Of" followed by articles or pronouns

    np_acts-logits-general · gemini-2.0-flash

    end of

    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
     préalable
    -0.66
     wireType
    -0.62
    الإنجليزية
    -0.62
    المكان
    -0.59
    Климат
    -0.59
     különböző
    -0.58
     parts
    -0.57
     místa
    -0.57
    istoitu
    -0.57
    Aspects
    -0.57
    POSITIVE LOGITS
     ostavi
    0.58
     spectrum
    0.57
     RoutedEventArgs
    0.56
     nowhere
    0.56
     MonoBehaviour
    0.54
     town
    0.53
    cipar
    0.52
     الاطلاع
    0.52
     typelib
    0.52
     story
    0.51
    Activations Density 0.296%

    No Known Activations