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. 9-GEMMASCOPE-RES-131K
    6. 14453
    Prev
    Next
    INDEX
    Explanations

    references to legal cases and entities, particularly involving the term "North" in relation to legal disputes

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

    true value

    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
    ConstraintMaker
    -0.36
     mucho
    -0.35
    ONSORED
    -0.35
     novedades
    -0.35
    ジュアル
    -0.34
     Autorisations
    -0.33
    misu
    -0.33
     varying
    -0.33
    rouvez
    -0.32
     Wiktionnaire
    -0.32
    POSITIVE LOGITS
     true
    1.17
    True
    1.17
     True
    1.16
    true
    1.15
     TRUE
    1.13
    TRUE
    1.00
     truer
    0.94
     wahren
    0.93
     truest
    0.90
     verdadeiro
    0.89
    Activations Density 0.014%

    No Known Activations