Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 70639
    Prev
    Next
    INDEX
    Explanations

    numeric values and their associated statistics or probabilities

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

    Mathematical and statistical expressions containing numbers, percentages, or terms like "number of" and "proportion" that quantify amounts or relationships within scientific or technical contexts.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    __).
    -0.44
     parado
    -0.40
    MessageState
    -0.38
     Pins
    -0.36
     Pines
    -0.36
    ñoz
    -0.36
    ('.');
    -0.35
     sztu
    -0.34
     Focal
    -0.34
     Elin
    -0.34
    POSITIVE LOGITS
    errHandler
    0.66
     betweenstory
    0.58
     <>",
    0.55
     defaultstate
    0.54
    تقاوى
    0.49
    awtextra
    0.48
    AddTagHelper
    0.47
    WritableDatabase
    0.47
     ब्रेकडाउन
    0.47
     bleue
    0.46
    Activations Density 0.268%

    No Known Activations