© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 2995
    Prev
    Next
    INDEX
    Explanations

    conjunctions and their usage in connecting phrases

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

    conjunctions or separators

    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
    HideFlags
    -0.61
    berdayakan
    -0.52
     الحره
    -0.51
    principalColumn
    -0.51
    msgTypes
    -0.51
    原始内容存档于
    -0.48
    chyma
    -0.48
     يتيمه
    -0.48
    IBOutlet
    -0.48
     propOrder
    -0.47
    POSITIVE LOGITS
     Wiktionnaire
    0.44
     ExecuteAsync
    0.38
     autorytatywna
    0.35
     nbr
    0.33
     ddl
    0.33
    叶凌天
    0.32
     allied
    0.32
     dna
    0.32
    ſelves
    0.31
     mum
    0.31
    Activations Density 0.091%

    No Known Activations