Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 12804
    Prev
    Next
    INDEX
    Explanations

    conversations and personal exchanges between individuals

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

    French subject pronouns followed by verbs

    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
     raiſ
    -0.78
     uſed
    -0.69
     chofe
    -0.68
    SequentialGroup
    -0.66
     gethan
    -0.64
     paſſ
    -0.63
     ſte
    -0.62
     recently
    -0.62
     deſt
    -0.61
     ſta
    -0.60
    POSITIVE LOGITS
    webElementGuid
    0.33
    当时
    0.33
    WebControls
    0.32
    當時
    0.32
     était
    0.31
     سكانية
    0.29
    lihatan
    0.28
    endphp
    0.27
    Defs
    0.26
     пить
    0.25
    Activations Density 0.025%

    No Known Activations