Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 28118
    Prev
    Next
    INDEX
    Explanations

    the presence of introductory elements or markers in text

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

    verbs describing actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     HasFactory
    -0.72
    enumi
    -0.70
    fillType
    -0.64
     vzduchu
    -0.64
    Gtk
    -0.61
    antMatchers
    -0.60
    fact
    -0.60
    hljs
    -0.59
    teza
    -0.58
     AQ
    -0.57
    POSITIVE LOGITS
     nakalista
    0.81
    =$?
    0.75
     onCancelled
    0.72
     began
    0.72
     appeared
    0.71
     went
    0.70
     wrote
    0.70
    Personendaten
    0.69
    }}}
    
    0.69
     drove
    0.68
    Activations Density 0.585%

    No Known Activations