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. 13162
    Prev
    Next
    INDEX
    Explanations

    expressions of surprise, acknowledgment, or emotional reactions

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

    あ followed by particles or nouns

    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
    WebVitals
    -0.44
     zdan
    -0.40
    ipedi
    -0.40
     sztuk
    -0.40
    cumin
    -0.39
    ựng
    -0.39
     Grüßen
    -0.39
    GRANTED
    -0.39
     TableColumn
    -0.38
     Pollo
    -0.37
    POSITIVE LOGITS
    あ
    1.98
     あ
    1.18
    をあ
    0.96
    があ
    0.93
    にあ
    0.86
    あれ
    0.83
    あの
    0.66
    ああ
    0.62
    あー
    0.60
    あい
    0.56
    Activations Density 0.014%

    No Known Activations