© 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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 369
    Prev
    Next
    INDEX
    Explanations

    phrases related to subjective experiences and opinions

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

    Sentences starting with coordinating conjunctions

    np_acts-logits-general · gemini-2.0-flash

    as she grew

    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
     Надо
    -1.02
    Надо
    -0.94
     надо
    -0.94
     muß
    -0.88
     Moslem
    -0.85
     daß
    -0.84
    :-)
    -0.83
     läßt
    -0.80
    けっこう
    -0.78
    ;-)
    -0.77
    POSITIVE LOGITS
     Alongside
    1.04
    Alongside
    0.96
     “[
    0.87
     Whilst
    0.86
    “[
    0.82
     Throughout
    0.81
     alongside
    0.81
     Additionally
    0.81
    Whilst
    0.80
    Upon
    0.80
    Activations Density 0.229%

    No Known Activations