© 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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10746
    Prev
    Next
    INDEX
    Explanations

    phrases related to financial transactions and account management

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

    Follows periods or end-of-turn tokens

    np_acts-logits-general · gemini-2.0-flash

    XML and code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     iprot
    -0.48
    يكب
    -0.48
    TagMode
    -0.47
    下载附件
    -0.45
     muualla
    -0.45
    GenerationType
    -0.44
    UnusedPrivate
    -0.44
     ligiloj
    -0.44
    MigrationBuilder
    -0.43
    Życiorys
    -0.41
    POSITIVE LOGITS
    webElementXpaths
    0.48
     AssemblyCulture
    0.43
     RELATIONS
    0.39
    ClientRect
    0.38
    ="#"><
    0.38
    GOTREF
    0.37
     transfieras
    0.36
     comprim
    0.36
    relations
    0.35
     khuy
    0.35
    Activations Density 0.811%

    No Known Activations