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. 4-GEMMASCOPE-RES-16K
    6. 15961
    Prev
    Next
    INDEX
    Explanations

    scientific or technical terminology related to physics and chemistry

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    abestanden
    -1.08
    SpringRunner
    -1.05
    SequentialGroup
    -1.04
    nocześnie
    -1.01
     Exacts
    -1.01
     Yoruba
    -1.00
    TagMode
    -0.98
     MainAxisSize
    -0.97
     comigo
    -0.97
     Lolita
    -0.95
    POSITIVE LOGITS
     Schild
    0.73
     Cio
    0.72
    CLE
    0.67
     tio
    0.64
     Rog
    0.64
     Heil
    0.64
    AccessException
    0.64
     Mia
    0.63
     lle
    0.62
     Ellison
    0.60
    Activations Density 10.456%

    No Known Activations