© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 4-GEMMASCOPE-RES-16K
    6. 2858
    Prev
    Next
    INDEX
    Explanations

    phrases related to reasoning or argumentation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    contentLoaded
    -0.79
    multer
    -0.78
     GenerationType
    -0.74
    HasAnnotation
    -0.72
     BorderRadius
    -0.70
    AppMethodBeat
    -0.66
    MigrationBuilder
    -0.65
    دانشنامهٔ
    -0.63
     ویکی‌پدی
    -0.63
    CompleteListener
    -0.63
    POSITIVE LOGITS
     logic
    2.50
     Logic
    2.21
     logical
    2.00
    Logic
    2.00
     LOGIC
    1.98
    logic
    1.97
    LOGIC
    1.81
     lógica
    1.79
    逻辑
    1.72
     Logical
    1.72
    Activations Density 0.112%

    No Known Activations