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. 25-GEMMASCOPE-RES-16K
    6. 10444
    Prev
    Next
    INDEX
    Explanations

    elements of suspense and tension within a narrative

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    GenerationType
    -0.52
    wiring
    -0.49
    떻
    -0.48
    ignition
    -0.48
     laughed
    -0.47
    ktory
    -0.47
    DataAnnotations
    -0.47
     laugh
    -0.47
     biologique
    -0.47
    ρακ
    -0.46
    POSITIVE LOGITS
     somber
    0.95
     sombre
    0.92
     atmosphere
    0.92
     melancholy
    0.90
     melancho
    0.90
     gloom
    0.89
     brooding
    0.89
     mood
    0.89
     gloomy
    0.88
     moody
    0.87
    Activations Density 0.308%

    No Known Activations