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. 24-GEMMASCOPE-RES-16K
    6. 2431
    Prev
    Next
    INDEX
    Explanations

    mentions of environmental concerns and civic engagement in discussions about local infrastructure projects

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

    Follows a time period or number

    np_acts-logits-general · gemini-2.0-flash

    year in context

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Whilst
    -1.71
    Whilst
    -1.62
     whilst
    -1.41
     poichè
    -1.19
     twenty
    -1.08
     enquiry
    -1.08
     fucking
    -1.06
     Schluß
    -1.06
     FUCKING
    -1.05
     läßt
    -1.04
    POSITIVE LOGITS
     —
    1.02
     
    0.90
     Sept
    0.85
    Fig
    0.83
     Fig
    0.82
    ,’’
    0.81
     Gov
    0.81
     
    0.80
     ‘‘
    0.79
     Aug
    0.79
    Activations Density 1.700%

    No Known Activations