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. 18-GEMMASCOPE-RES-16K
    6. 742
    Prev
    Next
    INDEX
    Explanations

    specific scientific terms related to experimental methods and substances

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

    before "spor", "thy", and "Dun"

    np_acts-logits-general · gemini-2.0-flash

    Crypto, onion, soap, names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    +#+#
    -1.03
     ThemeData
    -0.82
     /\.
    -0.82
    CppMethod
    -0.77
    runApp
    -0.75
    usermodel
    -0.74
    UnifiedTopology
    -0.73
     MainAxisSize
    -0.73
    routeProvider
    -0.73
     Scribd
    -0.72
    POSITIVE LOGITS
    sz
    0.63
    YPTO
    0.61
     Sz
    0.59
     Barton
    0.59
     Nicho
    0.58
     Ceci
    0.57
     Holl
    0.57
    மி
    0.56
    XXXXX
    0.56
    ائم
    0.55
    Activations Density 0.716%

    No Known Activations