© 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. 4-GEMMASCOPE-RES-16K
    11. 15342
    Prev
    Next
    INDEX
    Explanations

    instances of the word "hold" and related phrases implying accountability or responsibility

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    genossen
    -0.69
    caso
    -0.69
    ucs
    -0.67
    nyez
    -0.66
     bác
    -0.65
    🆚
    -0.64
     Bakker
    -0.64
     Acker
    -0.63
    PyExc
    -0.63
     Hyman
    -0.61
    POSITIVE LOGITS
     hold
    2.53
     Hold
    2.45
    hold
    2.37
    Hold
    2.35
     Holds
    2.33
     holds
    2.32
     HOLD
    2.28
    holds
    2.28
     holding
    2.27
    HOLD
    2.22
    Activations Density 0.064%

    No Known Activations