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 - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 13567
    Prev
    Next
    INDEX
    Explanations

    terms related to bomb and material specifications in a technical or regulatory context

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

    bomb, bombs, bomber, bombshell

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    huizen
    -0.43
    hånd
    -0.42
     wilayah
    -0.42
     față
    -0.41
     extiende
    -0.41
    orElse
    -0.41
     Pute
    -0.41
     écart
    -0.41
    ethane
    -0.40
     riqueza
    -0.40
    POSITIVE LOGITS
     Bom
    2.47
    Bom
    2.41
     bom
    2.20
    bom
    2.14
     BOM
    2.11
    BOM
    1.95
     bomb
    1.92
     Bomb
    1.88
     bombs
    1.80
     Bombs
    1.80
    Activations Density 0.005%

    No Known Activations