© 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 - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 582
    Prev
    Next
    INDEX
    Explanations

    evidence of causation or correlation in scientific discussions

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

    phrases that suggest causation or indicate relationships between concepts

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

    introducing findings or conclusions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    everything
    -0.69
     precisione
    -0.68
     EVERYTHING
    -0.65
     drôle
    -0.65
     noires
    -0.64
    twimg
    -0.64
     étoient
    -0.62
     sanitario
    -0.61
     feroit
    -0.60
     auroit
    -0.58
    POSITIVE LOGITS
     possible
    0.63
     direct
    0.63
     possibility
    0.63
     potential
    0.62
    XmlAccessType
    0.60
     lack
    0.60
    numerusform
    0.59
     differences
    0.57
     role
    0.55
     возможности
    0.55
    Activations Density 0.498%

    No Known Activations