© 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 - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 3563
    Prev
    Next
    INDEX
    Explanations

    specific technical attributes and specifications related to products and components

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

    Replacing parts or components

    np_acts-logits-general · gemini-2.0-flash

    replacement and spare parts

    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
     Roskov
    -0.83
     étoit
    -0.71
    bootstrapcdn
    -0.69
     voorbij
    -0.68
     pouvoit
    -0.64
     métallique
    -0.63
     doInBackground
    -0.63
    GEBURTSDATUM
    -0.62
     BrowserModule
    -0.62
     onOptions
    -0.61
    POSITIVE LOGITS
     replacement
    1.27
     spare
    1.21
     Replacement
    1.13
     replacements
    1.05
    spare
    1.02
    Replacement
    1.01
     Spare
    1.01
     spares
    1.01
    replacement
    1.01
    Spare
    0.99
    Activations Density 0.139%

    No Known Activations