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. 18-GEMMASCOPE-RES-16K
    6. 14549
    Prev
    Next
    INDEX
    Explanations

    phrases indicating exceptions or qualifications in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    followed by "except" or "exception"

    np_acts-logits-general · gemini-2.0-flash

    introducing an exception

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    any
    -0.55
    BooleanField
    -0.51
    UnusedPrivate
    -0.50
     bson
    -0.49
    Any
    -0.48
    ANY
    -0.48
     andern
    -0.45
    dotenv
    -0.45
    持
    -0.45
    listo
    -0.44
    POSITIVE LOGITS
     exception
    1.08
     maybe
    1.04
     perhaps
    0.92
    maybe
    0.90
     exceptions
    0.86
    perhaps
    0.82
     excepción
    0.78
     Maybe
    0.78
     للاسماء
    0.77
     except
    0.76
    Activations Density 0.141%

    No Known Activations