Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 2063
    Prev
    Next
    INDEX
    Explanations

    numerals or dates mentioned in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ↓,
    -0.58
     abgerufen
    -0.46
    voici
    -0.46
     thậm
    -0.45
     %@",
    -0.45
     Mitter
    -0.42
    ingar
    -0.40
     Ibero
    -0.39
     Tsing
    -0.39
     directions
    -0.38
    POSITIVE LOGITS
    ⓧ
    1.02
     ligiloj
    0.97
    expandindo
    0.94
    writeFieldEnd
    0.85
    parsedMessage
    0.84
    ConstraintMaker
    0.81
    XmlAccessType
    0.80
    ValueStyle
    0.79
    LookAnd
    0.77
    webElementXpaths
    0.77
    Activations Density 0.167%

    No Known Activations