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. 22-GEMMASCOPE-RES-16K
    6. 8146
    Prev
    Next
    INDEX
    Explanations

    terms related to alerting or notifying about error messages and warnings

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

    alerts, warnings, and notifications

    np_acts-logits-general · gemini-2.0-flash

    alert messages and warnings

    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
    DockStyle
    -0.58
     undersigned
    -0.56
    transmitters
    -0.51
    loroethane
    -0.51
    itability
    -0.51
     Unsc
    -0.50
    getColumnIndex
    -0.50
    ğimiz
    -0.50
    batore
    -0.49
    脚注の使い方
    -0.49
    POSITIVE LOGITS
     alert
    1.57
    alert
    1.41
     warning
    1.39
     Alert
    1.39
    Alert
    1.28
     notification
    1.24
     alerts
    1.23
     Warning
    1.19
     message
    1.18
     ALERT
    1.18
    Activations Density 0.257%

    No Known Activations