© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 3278
    Prev
    Next
    INDEX
    Explanations

    references to legal claims and relationships between individuals involved in a legal case

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

    past tense verbs and events

    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
    COLN
    -0.84
    AutoresizingMask
    -0.80
    InstrumentedTest
    -0.75
    Personendaten
    -0.73
     оригіналу
    -0.69
    elemField
    -0.68
    rungsseite
    -0.68
    IntoConstraints
    -0.67
    })();
    -0.67
    GenerationType
    -0.67
    POSITIVE LOGITS
     hindsight
    0.66
     retrospect
    0.63
    當時
    0.58
    Did
    0.57
    当時の
    0.57
    did
    0.56
     did
    0.54
     Did
    0.53
    当时
    0.53
     happened
    0.52
    Activations Density 0.901%

    No Known Activations