© 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-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 12766
    Prev
    Next
    INDEX
    Explanations

    technical terms and acronyms related to scientific processes or systems

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

    "Di", "Diag", or "di" followed by a letter

    np_acts-logits-general · gemini-2.0-flash

    DI words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Krak
    -0.76
    MarshalTo
    -0.61
     Ogre
    -0.59
     CCR
    -0.56
     Pott
    -0.56
     Moo
    -0.56
     Krakowie
    -0.55
    AnchorStyles
    -0.55
     Grafton
    -0.54
    Krak
    -0.54
    POSITIVE LOGITS
     Di
    1.72
     DI
    1.71
     Dia
    1.63
    DI
    1.61
    Di
    1.56
    di
    1.53
    Dia
    1.52
     Dial
    1.49
     Dietz
    1.43
     Dip
    1.39
    Activations Density 0.325%

    No Known Activations