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. 11096
    Prev
    Next
    INDEX
    Explanations

    Arabic letters and phrases that express preference or selection

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

    Non-English abbreviations and symbols

    np_acts-logits-general · gemini-2.0-flash

    Arabic, English, and language codes

    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
     Efq
    -1.05
    ImageContext
    -1.01
    ########.
    -0.99
     Eſ
    -0.98
     تضيفلها
    -0.98
     faſt
    -0.93
     purpoſe
    -0.92
     whoſe
    -0.92
     iſt
    -0.92
     विश्वसनीयता
    -0.91
    POSITIVE LOGITS
     se
    0.49
     re
    0.49
     נ
    0.45
     is
    0.44
     can
    0.43
     έχει
    0.42
     
    0.42
     ansiedade
    0.41
     does
    0.41
     van
    0.40
    Activations Density 0.001%

    No Known Activations