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. 15-GEMMASCOPE-RES-16K
    6. 12716
    Prev
    Next
    INDEX
    Explanations

    expressions related to surprise or unexpected occurrences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    StructEnd
    -0.51
     ProtoMessage
    -0.48
    onViewCreated
    -0.47
    teborg
    -0.46
    crossed
    -0.45
    utat
    -0.45
     dit
    -0.44
    Unless
    -0.44
     referenties
    -0.43
     constamment
    -0.43
    POSITIVE LOGITS
     deserved
    0.75
     logical
    0.75
     unsur
    0.72
     understandable
    0.70
     logique
    0.69
    Appropriate
    0.68
     logically
    0.68
     Appropriate
    0.68
     appropriate
    0.65
     appropriateness
    0.65
    Activations Density 0.306%

    No Known Activations