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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 16-GEMMASCOPE-RES-65K
    11. 27602
    Prev
    Next
    INDEX
    Explanations

    error messages related to element identification

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

    constitutional" and "2 :" followed by numbers.- It also seems to be drawn to non-English words or unusual sequences based on `TOP_POSITIVE_LOGITS`.- The `TOP_ACTIVATING_TEXTS` show specific instances of these patterns, including references to legal/scientific contexts ('punishment', 'Characterization', 'roots'). constitutional punishment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     שוליים
    -0.88
    brainly
    -0.84
     חיצוניים
    -0.81
    #+#
    -0.72
     Waray
    -0.68
    InSection
    -0.66
    expandindo
    -0.66
    الإنجليزية
    -0.65
    __':
    -0.65
     مشين
    -0.64
    POSITIVE LOGITS
     slutt
    0.47
    enumi
    0.46
     ...
    0.44
     Grüßen
    0.43
    ...
    0.42
     bety
    0.41
     oog
    0.40
     transférez
    0.40
     Ainda
    0.40
     повіт
    0.39
    Activations Density 0.048%

    No Known Activations