© 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-2-9B-IT
    7. Residual Stream - 16k
    9. 9-GEMMASCOPE-RES-16K
    11. 5773
    Prev
    Next
    INDEX
    Explanations

    proper nouns, particularly names and titles

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

    ending in "ou" followed by a consonant

    np_acts-logits-general · gemini-2.0-flash

    Pou, Brou, Hou, douc- tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    wój
    -0.52
     confirmer
    -0.50
     régal
    -0.49
     mystère
    -0.45
     Conferencia
    -0.42
     néglig
    -0.40
     merveilleux
    -0.40
    *************
    -0.40
     enregistrer
    -0.40
     perfeito
    -0.40
    POSITIVE LOGITS
     Crouch
    0.70
     InputDecoration
    0.69
    ouss
    0.69
    rouch
    0.67
    gou
    0.66
    ougou
    0.65
    ou
    0.65
     Cavity
    0.65
    outi
    0.65
    Hou
    0.64
    Activations Density 1.974%

    No Known Activations