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. 18-GEMMASCOPE-RES-16K
    6. 4148
    Prev
    Next
    INDEX
    Explanations

    references to changing or switching actions and their contexts

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

    after "switch" or "switched"

    np_acts-logits-general · gemini-2.0-flash

    switching to different states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ReusableCell
    -0.59
    IntoConstraints
    -0.54
    chitis
    -0.54
    ]))
    
    -0.51
     négociations
    -0.48
     simak
    -0.47
     wikipagina
    -0.47
    leyball
    -0.46
    //
    -0.46
     WebDriverWait
    -0.46
    POSITIVE LOGITS
     formats
    0.74
    SourceChecksum
    0.71
     regimes
    0.70
     switch
    0.70
     modes
    0.69
     styles
    0.68
    bootstrapcdn
    0.63
     switches
    0.61
     switching
    0.60
    ScopeManager
    0.60
    Activations Density 0.187%

    No Known Activations