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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 8189
    Prev
    Next
    INDEX
    Explanations

    phrases indicating successful operations or outcomes

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

    completion or success

    np_acts-logits-general · gemini-2.0-flash

    successful completion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     déclaration
    -0.32
     connais
    -0.31
    一些
    -0.29
     cierta
    -0.29
     asamblea
    -0.28
    ķ
    -0.28
     Asamblea
    -0.28
    ốp
    -0.27
     conocidas
    -0.27
    WindowConstants
    -0.27
    POSITIVE LOGITS
    sizeCache
    0.82
    MessageTagHelper
    0.81
    RTCF
    0.76
     '\\;'
    0.73
    ChildScrollView
    0.71
    IndentedString
    0.71
     ब्रेकडाउन
    0.69
     HasFactory
    0.69
     <<<<<<<<<<<<<<
    0.68
     contextLoads
    0.68
    Activations Density 0.329%

    No Known Activations