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. 8657
    Prev
    Next
    INDEX
    Explanations

    emotional expressions and reactions to art or film

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

    crying or sadness

    np_acts-logits-general · gemini-2.0-flash

    tears and crying

    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
    +#+#
    -0.98
    ⓘ
    -0.80
    UserScript
    -0.77
     समीक्षाओं
    -0.74
    IsContent
    -0.73
     مرئيه
    -0.69
    存于互联网档案馆
    -0.68
    Tikang
    -0.68
    PerformLayout
    -0.68
    rungsseite
    -0.68
    POSITIVE LOGITS
     tears
    2.02
     emotional
    1.73
     Tears
    1.70
     wept
    1.63
    Tears
    1.63
    tears
    1.61
     tear
    1.57
    emotional
    1.51
     sobbing
    1.48
     Emotional
    1.46
    Activations Density 0.218%

    No Known Activations