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. 1792
    Prev
    Next
    INDEX
    Explanations

    questions starting with "Is" or "Are"

    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
    CloseOperation
    -1.09
     itſelf
    -0.88
    AddTagHelper
    -0.84
     pleaſure
    -0.84
     Longfellow
    -0.83
     Ruman
    -0.83
     Româ
    -0.82
     themſelves
    -0.81
     ſever
    -0.81
    ucoup
    -0.81
    POSITIVE LOGITS
     Is
    1.12
    Is
    1.01
     IS
    0.95
     is
    0.95
    mIs
    0.84
     setIs
    0.83
     Was
    0.79
    bIs
    0.78
     Ishi
    0.78
    is
    0.78
    Activations Density 0.170%

    No Known Activations