© 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-2-2B
    4. Residual Stream - 16k
    5. 4-GEMMASCOPE-RES-16K
    6. 10822
    Prev
    Next
    INDEX
    Explanations

    statements emphasizing the word "fact" and its variations

    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
    SiO
    -0.54
     aimez
    -0.53
    DIS
    -0.47
    erez
    -0.45
     SwitchCompat
    -0.44
    itolo
    -0.44
    Uni
    -0.43
    BAM
    -0.43
     करें
    -0.43
     masing
    -0.43
    POSITIVE LOGITS
     indeed
    1.30
    indeed
    1.29
     fact
    1.23
    Indeed
    1.20
     Indeed
    1.18
     anzi
    1.13
    Bahkan
    1.03
     Bahkan
    1.03
     Infatti
    1.03
    事实上
    1.01
    Activations Density 0.258%

    No Known Activations