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. 80462
    Prev
    Next
    INDEX
    Explanations

    programming structures and keywords particularly related to object-oriented programming and Android development

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

    code structure and attribute assignment

    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
    UserScript
    -0.51
    RTEE
    -0.48
    complexType
    -0.42
     chi̍t
    -0.40
    <bos>
    -0.40
    awtextra
    -0.40
    }-*/;
    -0.38
    etheless
    -0.38
    scaleY
    -0.37
    ngdoc
    -0.36
    POSITIVE LOGITS
     externi
    0.48
     jsPsych
    0.47
     gunpowder
    0.44
    errHandler
    0.43
    ์
    0.42
    IUrlHelper
    0.41
     ब्रेकडाउन
    0.41
     moglie
    0.41
    utilisons
    0.41
     InputDecoration
    0.40
    Activations Density 0.001%

    No Known Activations