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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 29913
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and spaces in a text

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

    "IL, USA" or "terms"

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
     AssemblyCulture
    -0.88
    httphttps
    -0.88
    iſchen
    -0.85
    ſehen
    -0.84
    iſche
    -0.84
    ſchaft
    -0.84
     ब्रेकडाउन
    -0.84
    ſſung
    -0.84
    enablog
    -0.82
    ſicht
    -0.82
    POSITIVE LOGITS
    ↵↵
    0.37
    <strong>
    0.33
     gruesa
    0.33
    Còn
    0.32
     artistique
    0.32
     Keuangan
    0.32
     económico
    0.31
    2
    0.31
    出
    0.31
     entera
    0.30
    Activations Density 0.000%

    No Known Activations