© 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. 24-GEMMASCOPE-RES-16K
    6. 1988
    Prev
    Next
    INDEX
    Explanations

    phrases related to photography and image quality specifications

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

    Image resizing or manipulation

    np_acts-logits-general · gemini-2.0-flash

    image cropping and resizing

    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
     visuelle
    -0.63
     visuals
    -0.53
     audiovisual
    -0.53
     visuales
    -0.53
    映像
    -0.52
    modelBuilder
    -0.51
     visual
    -0.51
    iric
    -0.51
    Dichloropropane
    -0.50
    Visual
    -0.50
    POSITIVE LOGITS
     cropped
    1.13
     crop
    1.03
     cropping
    0.97
     Crop
    0.95
     resized
    0.95
    crop
    0.88
    Crop
    0.87
     resize
    0.86
     scaled
    0.84
     resizing
    0.83
    Activations Density 0.289%

    No Known Activations