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. 113943
    Prev
    Next
    INDEX
    Explanations

    terms related to specific animal prints and brands, particularly those associated with pumas and panthers

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

    big cats: Panther, Puma, Leopard, Lynx

    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
    []{"
    -0.41
    存于互联网档案馆
    -0.41
     Chio
    -0.40
     recal
    -0.39
    PYX
    -0.38
     وص
    -0.38
    mixed
    -0.38
     نص
    -0.37
     roots
    -0.37
    '],'
    -0.37
    POSITIVE LOGITS
     Panther
    1.09
    Panther
    1.01
     leopard
    0.94
     Leopard
    0.93
    Leopard
    0.91
    etah
    0.87
     panther
    0.86
     Puma
    0.83
     Jaguar
    0.82
     Panthers
    0.79
    Activations Density 0.006%

    No Known Activations