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. 10713
    Prev
    Next
    INDEX
    Explanations

    references to specific plant varieties and their characteristics

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

    Plant cultivars and varieties

    np_acts-logits-general · gemini-2.0-flash

    variety names

    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
    曖昧さ回避
    -0.73
    ConstraintMaker
    -0.72
    tagHelperRunner
    -0.66
     Yahya
    -0.66
    ագրություններ
    -0.65
    IonicModule
    -0.64
     مرئيه
    -0.63
    enumi
    -0.62
    homonymie
    -0.62
    ValueStyle
    -0.62
    POSITIVE LOGITS
     pregi
    0.52
     finest
    0.51
     variedades
    0.50
     brand
    0.47
     ’
    0.46
     coû
    0.44
    的好
    0.44
     terciopelo
    0.43
     amélior
    0.43
     Blue
    0.43
    Activations Density 0.304%

    No Known Activations