© 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-2-9B
    7. Residual Stream - 16k
    9. 37-GEMMASCOPE-RES-16K
    11. 5349
    Prev
    Next
    INDEX
    Explanations

    references to data structure representations and visualizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.37.hook_resid_post
    Hook Layer
    37
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    +#+
    -0.46
     محفوظة
    -0.33
     contextLoads
    -0.32
    Abitanti
    -0.32
    OSSARY
    -0.30
    сылкі
    -0.30
     patata
    -0.29
     jagung
    -0.28
    stdafx
    -0.28
    usermodel
    -0.28
    POSITIVE LOGITS
     yellow
    2.31
     purple
    2.30
     blue
    2.23
     pink
    2.13
     orange
    2.11
     green
    2.00
     gray
    1.98
     black
    1.98
     grey
    1.95
     white
    1.94
    Activations Density 1.992%

    No Known Activations