© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 2-GEMMASCOPE-RES-16K
    6. 12833
    Prev
    Next
    INDEX
    Explanations

    references to measurement units, specifically kilograms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 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
     llorar
    -0.44
     unknownFields
    -0.42
     Jaffe
    -0.38
     matéria
    -0.35
     llorando
    -0.34
    createQuery
    -0.33
     Dokumente
    -0.32
    RequiresApi
    -0.32
     sendiri
    -0.31
     femenino
    -0.31
    POSITIVE LOGITS
    Convention
    0.81
     Convention
    0.80
     convention
    0.79
     Conventions
    0.74
     ſei
    0.71
     beſch
    0.70
     geſch
    0.69
    iſchen
    0.69
    convention
    0.69
     ſans
    0.68
    Activations Density 0.064%

    No Known Activations