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. 15-GEMMASCOPE-RES-16K
    6. 1738
    Prev
    Next
    INDEX
    Explanations

    terms and structures associated with GitHub repositories and programming elements

    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
    <eos>
    -0.63
     
    -0.62
     “
    -0.61
     (
    -0.61
     "
    -0.57
    ContentAsync
    -0.55
    分ほど
    -0.53
    flashdata
    -0.52
     pre
    -0.51
      
    -0.51
    POSITIVE LOGITS
     itſelf
    1.24
     houſe
    1.19
     purpoſe
    1.17
     Efq
    1.15
     Reſ
    1.11
    ſelf
    1.09
     themſelves
    1.08
     Houſe
    1.08
     Chriſt
    1.07
     pleaſure
    1.07
    Activations Density 0.009%

    No Known Activations