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. 22-GEMMASCOPE-RES-16K
    6. 3301
    Prev
    Next
    INDEX
    Explanations

    questions and termination marks in dialogues or testimony

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Appears after questions in transcripts

    np_acts-logits-general · gemini-2.0-flash

    question starters or endings

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    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>
    -0.77
    !")
    
    -0.68
    ↵↵↵
    -0.59
    :
    
    -0.58
    -0.57
    ,}
    -0.57
    <bos>
    -0.55
    ↵↵↵↵
    -0.54
    ,
    
    -0.53
    .*;
    
    -0.52
    POSITIVE LOGITS
    findpost
    1.10
    XmlAccessType
    1.02
    GraphicsUnit
    0.97
     '\\;'
    0.94
    AutoresizingMask
    0.92
    ftagPool
    0.90
    ISupport
    0.88
    Personensuche
    0.86
     GenerationType
    0.86
    ագրություններ
    0.85
    Activations Density 0.220%

    No Known Activations