© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 3945
    Prev
    Next
    INDEX
    Explanations

    colons and the structure of argumentation in text

    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
    ValueStyle
    -0.72
     setCode
    -0.69
     zatem
    -0.68
     כה
    -0.67
    tvguidetime
    -0.66
    empêcher
    -0.64
    จึง
    -0.63
    athread
    -0.61
    deutsch
    -0.59
     noter
    -0.58
    POSITIVE LOGITS
     makeStyles
    0.61
     it
    0.58
     there
    0.56
    If
    0.55
    When
    0.55
     when
    0.55
     these
    0.54
     If
    0.53
     When
    0.52
     you
    0.52
    Activations Density 0.183%

    No Known Activations