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. 15-GEMMASCOPE-RES-16K
    6. 8844
    Prev
    Next
    INDEX
    Explanations

    emotional expressions and the struggle to maintain composure

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    SizeMode
    -0.60
     Ours
    -0.55
     jScrollPane
    -0.54
    Hauptartikel
    -0.54
     nakalista
    -0.53
    takt
    -0.50
     Whitfield
    -0.50
    DeleteBehavior
    -0.49
     débuts
    -0.49
     dAtA
    -0.49
    POSITIVE LOGITS
     feelings
    0.84
     feeling
    0.73
     feel
    0.72
     thoughts
    0.68
     feels
    0.66
     Feelings
    0.65
     thinking
    0.64
     perasaan
    0.63
     FEEL
    0.62
    feelings
    0.61
    Activations Density 0.296%

    No Known Activations