© 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-2-2B
    4. Residual Stream - 65k
    5. 13-GEMMASCOPE-RES-65K
    6. 57391
    Prev
    Next
    INDEX
    Explanations

    statements that challenge perceptions of reality or truth

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.13.hook_resid_post
    Hook Layer
    13
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MigrationBuilder
    -0.65
    LookAnd
    -0.64
     "..\..\..\
    -0.62
    #
    -0.60
     للاسماء
    -0.60
    AndEndTag
    -0.59
     समीक्षक
    -0.59
    interesting
    -0.56
     Hast
    -0.55
     Capit
    -0.55
    POSITIVE LOGITS
     pourtant
    0.54
    明明
    0.52
    Truth
    0.50
     mentira
    0.50
     fake
    0.49
     Wirklichkeit
    0.49
    ikin
    0.49
    contr
    0.49
    osin
    0.48
     untrue
    0.48
    Activations Density 0.100%

    No Known Activations