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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 714
    Prev
    Next
    INDEX
    Explanations

    emotions and reactions related to anger and fear

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

    Emotions or feelings

    np_acts-logits-general · gemini-2.0-flash

    emotions and sentiments

    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
    •jailbreak_evil_confidant

    No Comments

    Negative Logits
    brigens
    -0.56
    كذا
    -0.55
    MemoryWarning
    -0.51
     Référence
    -0.50
    :✨
    -0.49
    :+:
    -0.47
    ließlich
    -0.46
     S
    -0.46
    żyw
    -0.45
     AssemblyTitle
    -0.44
    POSITIVE LOGITS
     propOrder
    0.76
     pleaſure
    0.70
     manifested
    0.69
     itſelf
    0.68
     quæ
    0.67
    fulness
    0.67
     expressed
    0.67
    WriteBarrier
    0.67
     Efq
    0.66
     saites
    0.64
    Activations Density 0.363%

    No Known Activations