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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 1-GEMMASCOPE-RES-16K
    6. 13442
    Prev
    Next
    INDEX
    Explanations

    words that imply strangeness or peculiarity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     yoksa
    -0.66
     genomen
    -0.63
    Geplaatst
    -0.62
    وردار
    -0.61
    Datuak
    -0.61
    стоин
    -0.60
     respectivas
    -0.58
    ोंने
    -0.57
    iesis
    -0.56
    mortar
    -0.55
    POSITIVE LOGITS
     strange
    3.77
     weird
    3.39
    strange
    3.33
    Strange
    3.01
    weird
    2.94
     Strange
    2.92
     strangest
    2.87
     Weird
    2.84
    Weird
    2.81
     wierd
    2.80
    Activations Density 0.109%

    No Known Activations