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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 11776
    Prev
    Next
    INDEX
    Explanations

    specific terms related to specialty or uniqueness across various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     gambe
    -0.47
     volledige
    -0.44
    resizingMask
    -0.43
     completa
    -0.43
     mengalir
    -0.41
     basura
    -0.41
     piatto
    -0.41
     montaż
    -0.40
     água
    -0.40
     plán
    -0.40
    POSITIVE LOGITS
     special
    1.12
     especial
    1.05
    special
    1.02
     Particular
    0.93
     especiales
    0.92
     specially
    0.91
    特殊的
    0.91
    特別な
    0.91
     particular
    0.91
     speciale
    0.89
    Activations Density 0.672%

    No Known Activations