© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 96609
    Prev
    Next
    INDEX
    Explanations

    specific technical details or specifications related to various products and components

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

    parts or locations

    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
     للاسماء
    -0.89
    RenderAtEndOf
    -0.88
    ьаж
    -0.81
     OFDb
    -0.79
     <<<<<<<<<<<<<<
    -0.77
    цездатний
    -0.77
     noDo
    -0.73
    évaluateur
    -0.72
     BoxFit
    -0.71
     ujednoznacz
    -0.70
    POSITIVE LOGITS
     encant
    0.35
     cloud
    0.29
    KindOfClass
    0.29
    él
    0.28
    led
    0.27
    chutz
    0.27
    cloud
    0.26
     sahiptir
    0.26
    Join
    0.26
    é
    0.26
    Activations Density 0.074%

    No Known Activations