© 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-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 58063
    Prev
    Next
    INDEX
    Explanations

    phrases related to conditions, requirements, or actions in procedural contexts

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

    states of being or conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     convaincre
    -0.43
     Davis
    -0.38
     Barnett
    -0.38
     gelingt
    -0.36
     sekitar
    -0.35
    Every
    -0.35
     Wise
    -0.35
     pregunto
    -0.35
     Davies
    -0.34
    áneo
    -0.34
    POSITIVE LOGITS
    Personendaten
    0.84
     للمعارف
    0.72
     cherchés
    0.68
    
    0.66
     Administrativna
    0.64
    MLLoader
    0.60
    ItemBackground
    0.60
     незавершена
    0.59
     Chwiliwch
    0.58
     comuniques
    0.57
    Activations Density 0.036%

    No Known Activations