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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 105435
    Prev
    Next
    INDEX
    Explanations

    symbols and formatting elements often used in data representation or coding contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     intervención
    -0.42
     amenazas
    -0.40
     sorpresas
    -0.39
     explosión
    -0.38
     wykonania
    -0.37
     candidatura
    -0.37
     licencias
    -0.36
     exposiciones
    -0.36
     discapacidad
    -0.35
     posteriormente
    -0.35
    POSITIVE LOGITS
    
    0.71
    ftagPool
    0.68
    ۜ
    0.61
     nakalista
    0.61
     ſind
    0.60
    <unused52>
    0.60
    addCriterion
    0.60
    <unused8>
    0.60
    <pad>
    0.60
    [@BOS@]
    0.60
    Activations Density 0.001%

    No Known Activations