© 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 - 131k
    9. 9-GEMMASCOPE-RES-131K
    11. 120397
    Prev
    Next
    INDEX
    Explanations

    specific suffixes that are prevalent in certain languages or names, particularly those with 'kh' and 'akh'

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

    Kazakh

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    SpringBootTest
    -0.42
    LocalizedString
    -0.41
    able
    -0.41
    erman
    -0.39
     kema
    -0.39
     dommages
    -0.39
     Hardin
    -0.39
    absolute
    -0.38
    beans
    -0.38
     Hasan
    -0.37
    POSITIVE LOGITS
    kh
    2.13
    akh
    1.64
    KH
    1.38
    ukh
    1.33
    ikh
    1.28
    kha
    1.28
    ख
    1.18
     kh
    1.15
    Kh
    1.07
     KH
    1.02
    Activations Density 0.007%

    No Known Activations