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. 20-GEMMASCOPE-RES-131K
    6. 8480
    Prev
    Next
    INDEX
    Explanations

    instances of the word "wander" and its variations

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

    wandering or rambling

    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
    Guns
    -0.44
     виправивши
    -0.41
    FIX
    -0.41
    MENAFN
    -0.40
    ########.
    -0.40
     endblock
    -0.40
    setupUi
    -0.39
     SqlParameter
    -0.39
    dataclass
    -0.39
     estekak
    -0.38
    POSITIVE LOGITS
     wandering
    1.18
     wander
    1.16
     Wandering
    1.06
     wanders
    1.05
     Wander
    1.02
     wandered
    0.99
    Wander
    0.96
     wanderer
    0.93
     Wanderer
    0.91
    wander
    0.88
    Activations Density 0.013%

    No Known Activations