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. 8561
    Prev
    Next
    INDEX
    Explanations

    the presence of significant punctuation and formatting within the text

    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
     â
    -0.57
    DataAnnotations
    -0.49
    脚注の使い方
    -0.49
    iotic
    -0.48
     des
    -0.47
     Ã
    -0.47
    lnd
    -0.46
     "
    -0.46
    rxjs
    -0.45
    MockMvc
    -0.45
    POSITIVE LOGITS
    الحياه
    0.93
     Италијани
    0.91
    titleMargin
    0.85
    RegressionTest
    0.82
    LookAnd
    0.81
    SharedDtor
    0.78
    MigrationBuilder
    0.78
     CreateTagHelper
    0.77
    richTextPanel
    0.77
     disambiguazione
    0.76
    Activations Density 0.049%

    No Known Activations