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. 15-GEMMASCOPE-RES-16K
    6. 8682
    Prev
    Next
    INDEX
    Explanations

    patterns or structures in procedural descriptions and methodologies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    mybatisplus
    -0.65
    Diweddarwch
    -0.60
    obacterium
    -0.58
    __':
    -0.54
    TableAdapter
    -0.53
    gänger
    -0.53
     propOrder
    -0.53
     '|
    -0.53
    锈钢
    -0.52
    SqlClient
    -0.52
    POSITIVE LOGITS
     lastly
    1.95
     Lastly
    1.94
    Lastly
    1.90
     Finally
    1.90
    Finally
    1.89
     finally
    1.80
     enfin
    1.73
    finally
    1.69
     infine
    1.64
     Enfin
    1.61
    Activations Density 0.639%

    No Known Activations