© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 8941
    Prev
    Next
    INDEX
    Explanations

    references to spatial relationships and actions that indicate movement or position

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

    Spatial relationships: above, over, nearest

    np_acts-logits-general · gemini-2.0-flash

    tall trees and buildings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ImageContext
    -0.54
     yhte
    -0.48
     suelos
    -0.47
     piele
    -0.44
     územ
    -0.42
     صوتيه
    -0.41
     girdle
    -0.41
     heridas
    -0.40
     embolism
    -0.39
     nessuno
    -0.38
    POSITIVE LOGITS
     trees
    1.28
     buildings
    1.26
    buildings
    1.05
     tree
    1.01
    trees
    1.00
     houses
    1.00
     Buildings
    0.99
     Trees
    0.99
     TREES
    0.94
    tall
    0.93
    Activations Density 0.247%

    No Known Activations