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. 25-GEMMASCOPE-RES-16K
    6. 4420
    Prev
    Next
    INDEX
    Explanations

    technical terminology related to electrical components and their specifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    findpost
    -0.74
    HasForeignKey
    -0.67
    antd
    -0.62
     initComponents
    -0.60
    CloseOperation
    -0.58
    onActivityResult
    -0.57
    TEntity
    -0.56
     pinulongan
    -0.55
     klap
    -0.55
     acert
    -0.53
    POSITIVE LOGITS
     strands
    0.83
     stranded
    0.81
    stranded
    0.81
     strand
    0.77
     core
    0.74
     braided
    0.72
     conductor
    0.72
     wire
    0.71
     cores
    0.71
     wound
    0.70
    Activations Density 0.557%

    No Known Activations