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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 3628
    Prev
    Next
    INDEX
    Explanations

    locations, especially those related to state agencies and events

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

    Locations (office, bedroom)

    np_acts-logits-general · gemini-2.0-flash

    locations of places or people

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     يتيمه
    -0.56
     فريبيس
    -0.55
    PointerException
    -0.55
     męska
    -0.52
    adaptiveStyles
    -0.52
     ModelExpression
    -0.51
     kasarigan
    -0.50
    setViewportView
    -0.49
     arşivlendi
    -0.49
     zimowa
    -0.48
    POSITIVE LOGITS
    Kaynakça
    0.39
     SME
    0.38
     Nani
    0.37
    locene
    0.36
     DCM
    0.36
     DPM
    0.36
     Jun
    0.36
     lenker
    0.35
    staun
    0.35
     chambers
    0.35
    Activations Density 0.644%

    No Known Activations