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. 56472
    Prev
    Next
    INDEX
    Explanations

    scientific terminology related to RNA sequencing and library preparation

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

    Illumina sequencing platform technology

    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
     po
    -0.38
     Bel
    -0.35
     Py
    -0.35
    spaceBetween
    -0.35
    Filmografie
    -0.34
    րջ
    -0.34
     peace
    -0.33
    EDEFAULT
    -0.33
     Be
    -0.32
     leg
    -0.32
    POSITIVE LOGITS
     otomatig
    0.79
     sequencing
    0.66
    Бахар
    0.60
     المعيارى
    0.59
     Sequencing
    0.58
    NameInMap
    0.57
     Efq
    0.57
     Picchu
    0.56
    umina
    0.56
    SharedCtor
    0.56
    Activations Density 0.266%

    No Known Activations