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-2-9B
    4. Residual Stream - 16k
    5. 16-GEMMASCOPE-RES-16K
    6. 12846
    Prev
    Next
    INDEX
    Explanations

    LaTeX formatting commands related to figures and images

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_16/width_16k/average_l0_75
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ist
    -0.33
    oltre
    -0.29
     experienced
    -0.28
     only
    -0.27
     лишь
    -0.27
     onely
    -0.27
     is
    -0.26
    NonQuery
    -0.26
     cap
    -0.25
     interested
    -0.25
    POSITIVE LOGITS
     EconPapers
    0.92
     estekak
    0.82
     autorytatywna
    0.78
    Hentet
    0.77
    Географиясе
    0.76
     betweenstory
    0.76
    expandindo
    0.75
    ThroughAttribute
    0.75
    tvguidetime
    0.73
     CreateTagHelper
    0.72
    Activations Density 0.073%

    No Known Activations