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

    Neuronpedia

    APICircuit 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. 21-GEMMASCOPE-RES-16K
    6. 754
    Prev
    Next
    INDEX
    Explanations

    themes of kindness, compassion, and the treatment of others

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    SourceChecksum
    -0.72
    makeConstraints
    -0.68
    aarrggbb
    -0.64
     externes
    -0.53
    transQ
    -0.52
    WithIOException
    -0.51
    oredCriteria
    -0.51
    rungsseite
    -0.49
    AutoScaleMode
    -0.49
    DoubleQuotes
    -0.47
    POSITIVE LOGITS
     compassion
    1.73
     kindness
    1.72
     compassionate
    1.52
     empathy
    1.46
     Compassion
    1.43
     Kindness
    1.42
    Compassion
    1.38
     sympathy
    1.33
    kindness
    1.33
     caring
    1.32
    Activations Density 0.308%

    No Known Activations