Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 14727
    Prev
    Next
    INDEX
    Explanations

    terms related to measurement and quantification in scientific contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    enumi
    -0.50
     dụ
    -0.38
     esterni
    -0.37
    DALE
    -0.35
     Jack
    -0.35
     rijden
    -0.35
    terase
    -0.34
     createSlice
    -0.34
     akár
    -0.34
    BeginContext
    -0.34
    POSITIVE LOGITS
    featureID
    0.61
     frequency
    0.60
     intensity
    0.59
     Umfang
    0.56
    исленность
    0.54
    IVEREF
    0.52
     level
    0.52
     severity
    0.52
     ſever
    0.52
     length
    0.51
    Activations Density 1.115%

    No Known Activations