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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 114175
    Prev
    Next
    INDEX
    Explanations

    references to order and invoice information

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

    document ids and numbers

    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
    addPreferredGap
    -0.49
     suivantes
    -0.48
     vertes
    -0.44
    الحياه
    -0.44
    onded
    -0.43
    Становништво
    -0.42
     Comprometido
    -0.42
     schooling
    -0.41
    Carriera
    -0.41
    arthed
    -0.41
    POSITIVE LOGITS
    msgTypes
    0.46
    SimpleDateFormat
    0.40
     ProtoMessage
    0.40
    Hentet
    0.38
    MessageTagHelper
    0.37
    DropTable
    0.37
    RectangleBorder
    0.37
    randomUUID
    0.36
     AssemblyCulture
    0.36
     мәкал
    0.35
    Activations Density 0.018%

    No Known Activations