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. 1423
    Prev
    Next
    INDEX
    Explanations

    terms related to "sin," particularly in medical or analytical contexts

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

    "Sin" or "sin" (related to "sinusitis")

    np_acts-logits-general · gemini-2.0-flash

    Sin prefix followed by varied words

    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
     שוליים
    -0.97
    tvguidetime
    -0.86
    脚注の使い方
    -0.86
    ]--;
    -0.75
    IntoConstraints
    -0.74
    ')")
    -0.74
    ,:);
    -0.73
    ]]);
    -0.72
    Sklici
    -0.71
    "]);
    
    -0.71
    POSITIVE LOGITS
     SIN
    0.75
    usitis
    0.70
     Sin
    0.69
     sins
    0.62
    Sin
    0.59
     sin
    0.58
     sinned
    0.57
     Sins
    0.56
     Sinha
    0.54
    SIN
    0.54
    Activations Density 0.079%

    No Known Activations