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-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 11997
    Prev
    Next
    INDEX
    Explanations

    specific scientific measurements or results related to cancer research

    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
     ***!
    -0.66
     Chwiliwch
    -0.54
     CreateTagHelper
    -0.54
     surla
    -0.53
    AddTagHelper
    -0.49
     typelib
    -0.47
    Tembelea
    -0.47
    ContentLoaded
    -0.47
     HEX
    -0.45
    
    -0.44
    POSITIVE LOGITS
     tavola
    0.42
     hinweg
    0.41
     wohl
    0.39
     eût
    0.39
     című
    0.38
     שוליים
    0.38
    hadapi
    0.38
     fallait
    0.38
    makeText
    0.36
    JvmStatic
    0.36
    Activations Density 0.227%

    No Known Activations