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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 23-GEMMASCOPE-RES-16K
    11. 9493
    Prev
    Next
    INDEX
    Explanations

    exaggerations and qualifying words that express excessive or diminished amounts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.23.hook_resid_post
    Hook Layer
    23
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     lepší
    -0.57
     făcut
    -0.56
     dobré
    -0.54
     dobrý
    -0.54
    nější
    -0.53
    jší
    -0.52
     sacré
    -0.52
     soumis
    -0.51
     szép
    -0.48
    可怕
    -0.48
    POSITIVE LOGITS
     quickly
    1.15
     smoothly
    1.11
     differently
    1.08
    quickly
    1.06
     gracefully
    1.05
     efficiently
    1.04
     successfully
    1.02
    ConstraintMaker
    1.01
     loudly
    1.01
     aggressively
    1.00
    Activations Density 0.214%

    No Known Activations