© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 6733
    Prev
    Next
    INDEX
    Explanations

    references to well-known characters and franchises in popular culture

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

    fictional characters in franchises

    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
    IVEREF
    -0.56
    enterOuterAlt
    -0.54
    numerusform
    -0.53
    CloseOperation
    -0.52
     GenerationType
    -0.48
    exitRule
    -0.47
    parsedMessage
    -0.47
    styleType
    -0.46
    Попис
    -0.46
    findpost
    -0.46
    POSITIVE LOGITS
     characters
    1.33
    characters
    1.07
     Characters
    1.05
     character
    1.00
    Characters
    0.98
     CHARACTERS
    0.92
     iconic
    0.89
     캐릭터
    0.86
     karakter
    0.82
     Character
    0.81
    Activations Density 0.027%

    No Known Activations