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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14934
    Prev
    Next
    INDEX
    Explanations

    modal verbs indicating obligation or necessity

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

    "should" or similar advice/suggestions

    np_acts-logits-general · gemini-2.0-flash

    recommendations or predictions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    }$​
    -0.46
     poderosos
    -0.37
     Fluss
    -0.36
    getValues
    -0.36
     ​​
    -0.35
    fara
    -0.35
    $​
    -0.35
     Kome
    -0.34
    税込
    -0.34
    addValue
    -0.34
    POSITIVE LOGITS
    should
    1.25
     should
    1.24
     Should
    1.21
    Should
    1.20
     SHOULD
    1.15
    hould
    1.13
     ſhould
    0.98
     fhould
    0.88
     ought
    0.82
     Sollte
    0.82
    Activations Density 0.124%

    No Known Activations