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. 20-GEMMASCOPE-RES-16K
    6. 3508
    Prev
    Next
    INDEX
    Explanations

    references to data structures or code snippets related to account management and messaging systems

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

    Non-English words and characters

    np_acts-logits-general · gemini-2.0-flash

    Slavic words and phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
     ciudadana
    -0.61
    ഏ
    -0.58
     colombiana
    -0.57
    MLLoader
    -0.56
    ओं
    -0.56
     agricoles
    -0.55
    mė
    -0.55
     AttributeSet
    -0.54
     amorosa
    -0.54
    InteropServices
    -0.54
    POSITIVE LOGITS
    })));
    0.62
    }}],
    0.57
     himself
    0.56
    uls
    0.55
    ]]:
    0.55
    nod
    0.55
    GraphicsUnit
    0.55
     ligh
    0.54
    TypedDataSet
    0.54
    =$?
    0.53
    Activations Density 0.036%

    No Known Activations