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. 5415
    Prev
    Next
    INDEX
    Explanations

    narratives that emphasize the importance of thorough research and informed decision-making

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

    research and study

    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
     MessageBoxIcon
    -0.61
     ProtoMessage
    -0.60
     ब्रेकडाउन
    -0.55
     يتيمه
    -0.55
    tinyos
    -0.54
    portlet
    -0.51
     betweenstory
    -0.50
     الرياضيه
    -0.49
     setCode
    -0.48
    parsedMessage
    -0.46
    POSITIVE LOGITS
     researching
    0.78
     research
    0.74
     researched
    0.73
     investigación
    0.68
    research
    0.63
     researches
    0.62
     investigate
    0.61
     RESEARCH
    0.59
    RESEARCH
    0.56
     investigation
    0.55
    Activations Density 0.022%

    No Known Activations