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. 12912
    Prev
    Next
    INDEX
    Explanations

    negative constructs and expressions of inability or difficulty

    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
    DoubleQuotes
    -0.70
     springfox
    -0.69
     eminently
    -0.56
    Disposition
    -0.55
    SnackBar
    -0.54
     Drapeau
    -0.52
    ativement
    -0.51
    JsonFormat
    -0.50
     Muffins
    -0.50
     cleanest
    -0.50
    POSITIVE LOGITS
    Spoljašnje
    0.71
    GEBURTSDATUM
    0.67
    FormTagHelper
    0.63
     समीक्षक
    0.61
    argout
    0.59
    Geplaatst
    0.59
     control
    0.58
    orgue
    0.54
    Externé
    0.52
     initComponents
    0.52
    Activations Density 0.128%

    No Known Activations