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. 3124
    Prev
    Next
    INDEX
    Explanations

    concepts related to diversity and inclusion in educational settings

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

    Follows common words (era, that, and, book, our, are)

    np_acts-logits-general · gemini-2.0-flash

    time and future contexts

    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
     hierbei
    -0.67
    Utili
    -0.66
     downfall
    -0.64
     aforementioned
    -0.60
     intricacies
    -0.60
     utilizes
    -0.59
     somit
    -0.59
     uniquement
    -0.59
     Utilizing
    -0.58
     solucionar
    -0.58
    POSITIVE LOGITS
     (;;)
    0.76
     manifestly
    0.76
     jedenfalls
    0.71
    
    0.70
    ^(@)
    0.70
     ―――――
    0.68
    =[];
    
    0.67
     tens
    0.65
     insuffisamment
    0.63
     tolerably
    0.63
    Activations Density 1.017%

    No Known Activations