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. 1025
    Prev
    Next
    INDEX
    Explanations

    terms related to software licensing and compliance

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

    Code-related licenses/permissions and software methods

    np_acts-logits-general · gemini-2.0-flash

    obtain, use, or delete

    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
    ,
    -0.61
     off
    -0.54
    .
    -0.52
    -
    -0.52
    ;
    -0.51
    <eos>
    -0.51
     so
    -0.48
     just
    -0.44
    off
    -0.44
    ?
    -0.44
    POSITIVE LOGITS
     Савезне
    1.16
     متعلقه
    1.13
    Personensuche
    1.11
    contentLoaded
    1.09
    GEBURTSDATUM
    1.07
     unknownFields
    1.06
     الحره
    1.05
     AssemblyProduct
    1.02
     ſind
    1.01
    ///</
    1.01
    Activations Density 0.166%

    No Known Activations