© 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. 18-GEMMASCOPE-RES-16K
    6. 6774
    Prev
    Next
    INDEX
    Explanations

    references to statements or claims made by individuals in a professional context

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

    Tokens before names or locations

    np_acts-logits-general · gemini-2.0-flash

    Riyaz and place 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
     resistance
    -0.58
     resistenza
    -0.57
     resistances
    -0.55
    Omega
    -0.51
    Ω
    -0.51
     résistance
    -0.50
    racha
    -0.49
     resistencia
    -0.48
     Resistance
    -0.48
    kien
    -0.47
    POSITIVE LOGITS
     Kashmir
    0.96
     propOrder
    0.94
     Jammu
    0.94
     Srinagar
    0.92
    Srin
    0.81
    Kash
    0.81
    Sucesor
    0.75
    aarrggbb
    0.73
    SpringBootTest
    0.71
    WithIOException
    0.71
    Activations Density 0.038%

    No Known Activations