© 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. 10287
    Prev
    Next
    INDEX
    Explanations

    URLs and image references

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

    URL or image source

    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
    ValueStyle
    -0.51
     Infórmanos
    -0.48
    BufferedWriter
    -0.45
     createSlice
    -0.42
     дописавши
    -0.42
    createStatement
    -0.41
    urnum
    -0.39
    marle
    -0.39
    FunctionFlags
    -0.37
     overclock
    -0.36
    POSITIVE LOGITS
     images
    0.82
     image
    0.82
    URL
    0.80
     url
    0.77
     URL
    0.76
     imageUrl
    0.75
    ImageUrl
    0.75
    image
    0.73
    url
    0.73
     Images
    0.71
    Activations Density 0.015%

    No Known Activations