© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 44088
    Prev
    Next
    INDEX
    Explanations

    punctuation and special characters used in programming or data representation

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

    multilingual "others"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    <unused68>
    -0.81
    <unused41>
    -0.81
    <unused74>
    -0.81
    <unused79>
    -0.81
    <unused14>
    -0.81
    <unused8>
    -0.81
    <unused16>
    -0.81
    <unused23>
    -0.81
    [@BOS@]
    -0.81
    <unused1>
    -0.81
    POSITIVE LOGITS
     felicitación
    0.35
     dalších
    0.35
     asesinado
    0.33
     všetko
    0.31
     realización
    0.31
     ménage
    0.30
    васто
    0.30
    îtra
    0.30
     diğer
    0.29
     demás
    0.29
    Activations Density 0.022%

    No Known Activations