© 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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 55942
    Prev
    Next
    INDEX
    Explanations

    actions related to evacuation and structural changes

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

    states of being absent or non-functional

    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
    parsedMessage
    -0.70
    帖最后由
    -0.67
    
    -0.62
    alyptus
    -0.59
     Paglinawan
    -0.59
    ፩
    -0.58
    #
    -0.57
    ">//
    -0.57
    fromnode
    -0.56
    VIAF
    -0.54
    POSITIVE LOGITS
     Wilton
    0.32
    Vanjske
    0.32
     emergencia
    0.32
    ISupport
    0.31
     pilote
    0.31
     contactez
    0.31
     colazione
    0.31
     прока
    0.30
    MockBean
    0.30
     conséquences
    0.29
    Activations Density 0.073%

    No Known Activations