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. 4-GEMMASCOPE-RES-16K
    6. 16025
    Prev
    Next
    INDEX
    Explanations

    references to important notes or annotations in a document

    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
     Brito
    -0.67
     consum
    -0.63
    buttonBar
    -0.60
     Svenska
    -0.59
    archiviato
    -0.59
    yscy
    -0.59
     BorderRadius
    -0.58
    AsUp
    -0.58
    Pengertian
    -0.58
    ftance
    -0.58
    POSITIVE LOGITS
     note
    2.61
     Note
    2.39
    note
    2.32
     notes
    2.23
     NOTE
    2.20
    Note
    2.17
     Notes
    2.09
    notes
    1.92
     NOTES
    1.84
    Notes
    1.80
    Activations Density 0.082%

    No Known Activations