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. 128952
    Prev
    Next
    INDEX
    Explanations

    mentions of the car manufacturer Peugeot and its associations or partnerships, particularly with Dongfeng

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

    car brands (Peugeot, Honda, Volkswagen)

    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
    awaiter
    -0.45
    localctx
    -0.40
    MessageState
    -0.40
    MLLoader
    -0.39
    bukkit
    -0.39
    ActionPerformed
    -0.35
    modb
    -0.35
    исленность
    -0.34
     EconPapers
    -0.34
    dataclass
    -0.32
    POSITIVE LOGITS
    WriteTagHelper
    0.64
    roën
    0.57
    ConstraintMaker
    0.57
    DIPSETTING
    0.56
    новниш
    0.55
     विश्वसनीयता
    0.52
     Citroën
    0.52
     فريبيس
    0.52
    SequentialGroup
    0.50
    iecie
    0.49
    Activations Density 0.027%

    No Known Activations