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-2-2B
    4. Residual Stream - 16k
    5. 2-GEMMASCOPE-RES-16K
    6. 6330
    Prev
    Next
    INDEX
    Explanations

    structured mathematical expressions, particularly arrays and their manipulations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     băr
    -0.51
    chiato
    -0.51
    Carriera
    -0.51
     للاسماء
    -0.50
    sfw
    -0.50
     stretched
    -0.48
    риях
    -0.48
    География
    -0.48
     ModelExpression
    -0.47
    ieszka
    -0.47
    POSITIVE LOGITS
    basicConfig
    0.79
    parsedMessage
    0.75
    出版年
    0.73
    bonucle
    0.67
    \]
    0.67
     OMITBAD
    0.67
    сылкі
    0.67
    crollView
    0.66
    BackStack
    0.64
    }\]
    0.64
    Activations Density 0.092%

    No Known Activations