© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 2879
    Prev
    Next
    INDEX
    Explanations

    technical terms related to hardware connections and power management

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

    unplugging, disabling, or disconnecting

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     Signalez
    -0.56
    AndEndTag
    -0.50
     plegable
    -0.49
    <unused47>
    -0.49
    <unused8>
    -0.48
    <unused41>
    -0.48
    <unused80>
    -0.48
    <unused43>
    -0.48
    <unused28>
    -0.48
    <unused16>
    -0.48
    POSITIVE LOGITS
    NegativeButton
    0.42
     ویکی‌پدی
    0.39
    WriteTagHelper
    0.36
    oneofs
    0.36
    IsContent
    0.36
    seamnă
    0.35
    RTLI
    0.34
    SourceChecksum
    0.33
    findpost
    0.33
     fieldNum
    0.32
    Activations Density 0.020%

    No Known Activations