Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 24296
    Prev
    Next
    INDEX
    Explanations

    items related to mechanical objects and their functionalities

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

    words before punctuation or specific phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     škole
    -0.40
     quirúrg
    -0.39
     objav
    -0.35
     casada
    -0.35
     secretario
    -0.35
     wojsk
    -0.35
    exécution
    -0.34
     trekken
    -0.34
     vêtement
    -0.34
    agir
    -0.33
    POSITIVE LOGITS
    expandindo
    0.77
     disambiguazione
    0.73
     ujednoznacz
    0.71
     Roskov
    0.69
     AssemblyTitle
    0.64
    RTEE
    0.59
     propOrder
    0.56
    RetentionPolicy
    0.55
    MockBean
    0.54
    日閲覧
    0.54
    Activations Density 0.232%

    No Known Activations