© 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. 41996
    Prev
    Next
    INDEX
    Explanations

    symbols and punctuation marks, primarily brackets, quotes, and asterisks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     desmotivaciones
    -0.84
     للاسماء
    -0.79
     zijne
    -0.79
     zelve
    -0.78
    SequentialGroup
    -0.77
     zoude
    -0.76
     mijne
    -0.72
     suaminya
    -0.69
    pingente
    -0.67
    majánló
    -0.67
    POSITIVE LOGITS
    ()")
    0.39
    ']))
    
    0.39
     pat
    0.38
     aca
    0.37
     “
    0.37
     mu
    0.37
     ped
    0.35
     ”
    0.35
     cs
    0.34
     «
    0.34
    Activations Density 0.052%

    No Known Activations