© 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. 69431
    Prev
    Next
    INDEX
    Explanations

    phrases indicating approximation or estimation

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

    roughly followed by quantities or comparisons

    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
     graag
    -0.54
    Datuak
    -0.46
     vanskelig
    -0.46
     genoemd
    -0.45
    keyColumn
    -0.45
    paravant
    -0.44
    postmedia
    -0.44
     ApiException
    -0.44
    ⓧ
    -0.43
    setcounter
    -0.43
    POSITIVE LOGITS
     speaking
    0.65
    ingly
    0.54
     equivalent
    0.52
     halfway
    0.52
     idea
    0.48
     half
    0.47
     notion
    0.46
    speaking
    0.46
     apro
    0.45
    Apro
    0.45
    Activations Density 0.006%

    No Known Activations