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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 3532
    Prev
    Next
    INDEX
    Explanations

    Greek words related to etymology or historical context

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

    Following words describing scientific or technical concepts

    np_acts-logits-general · gemini-2.0-flash

    words ending in -tron, -fili, -cli, OLED, hash, frac

    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
    amsa
    -0.53
    WaitGroup
    -0.52
    tivism
    -0.51
    IsContent
    -0.50
    מבר
    -0.50
    DECREF
    -0.50
     bezeichneter
    -0.49
    QMetaType
    -0.48
    MENAFN
    -0.48
    AccessorTable
    -0.48
    POSITIVE LOGITS
     Kai
    0.52
     kai
    0.52
     Wiese
    0.50
     paleta
    0.48
    Xr
    0.47
     biblia
    0.46
    Kai
    0.45
     ledig
    0.45
     ECONOM
    0.45
     Aero
    0.44
    Activations Density 0.106%

    No Known Activations