© 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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 4386
    Prev
    Next
    INDEX
    Explanations

    historical figures and their titles or roles

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

    Follows "the," "iph," "princip," or "Cali"

    np_acts-logits-general · gemini-2.0-flash

    king, caliphate, principality

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     Assy
    -0.60
     Ovid
    -0.60
     Maced
    -0.59
     Pliny
    -0.59
     Nadal
    -0.57
     sensei
    -0.57
     plebe
    -0.54
     Huguen
    -0.54
     Brest
    -0.53
     Sensei
    -0.52
    POSITIVE LOGITS
     puissante
    0.53
     llorar
    0.50
     regeringen
    0.49
     treino
    0.48
     turístico
    0.48
     étrangère
    0.48
     protocolos
    0.48
     souverain
    0.47
     político
    0.47
     étranger
    0.47
    Activations Density 0.572%

    No Known Activations