© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 30-GEMMASCOPE-RES-16K
    6. 12252
    Prev
    Next
    INDEX
    Explanations

    phrases expressing uncertainty or dependence on others' actions

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

    Contractions with "not" or "'ve"

    np_acts-logits-general · gemini-2.0-flash

    negative contractions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DrawerToggle
    -0.62
     clichés
    -0.47
     myſelf
    -0.46
    英語版
    -0.46
     naïve
    -0.45
     fiancée
    -0.43
    íncia
    -0.42
    niająca
    -0.41
     proxies
    -0.41
    iomanip
    -0.41
    POSITIVE LOGITS
     wasnt
    1.18
     doesnt
    1.18
     theyre
    1.16
     isnt
    1.16
     didnt
    1.13
     youll
    1.12
     youre
    1.10
     Thats
    1.09
     couldnt
    1.09
     arent
    1.08
    Activations Density 0.296%

    No Known Activations