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. 632
    Prev
    Next
    INDEX
    Explanations

    phrases and words related to customer service and interpersonal interactions

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

    Positive adjectives related to people and services

    np_acts-logits-general · gemini-2.0-flash

    friendly and helpful staff

    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
    hanced
    -0.55
    thâu
    -0.55
     ImGui
    -0.54
     referrerpolicy
    -0.53
    IActionResult
    -0.53
    WaitGroup
    -0.52
    はじめに
    -0.52
    enterOuterAlt
    -0.51
    DoubleQuotes
    -0.51
    }],
    
    -0.50
    POSITIVE LOGITS
     friendly
    1.19
     responsive
    1.03
     receptive
    0.93
     willing
    0.93
     Friendly
    0.92
     courteous
    0.91
    friendly
    0.90
     supportive
    0.89
     helpful
    0.89
    Friendly
    0.88
    Activations Density 0.112%

    No Known Activations