© 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
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 9092
    Prev
    Next
    INDEX
    Explanations

    references to the term "Po" in various contexts, particularly related to characters, positions, and artistic roles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bootstrapcdn
    -0.64
     polished
    -0.63
     poor
    -0.59
    reportWebVitals
    -0.59
    pooled
    -0.58
    Poor
    -0.57
    population
    -0.56
    poor
    -0.56
    populated
    -0.56
     Polished
    -0.56
    POSITIVE LOGITS
     зрения
    0.40
     MainAxisSize
    0.37
    ulating
    0.34
     الرخصة
    0.33
     negru
    0.32
    ルター
    0.32
    SharedDtor
    0.32
    thenReturn
    0.32
    ExecuteReader
    0.31
    HttpDelete
    0.31
    Activations Density 0.542%

    No Known Activations