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. 22-GEMMASCOPE-RES-16K
    6. 13822
    Prev
    Next
    INDEX
    Explanations

    references to superhero films, particularly those related to Marvel characters and their success in the entertainment industry

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

    comic books or superheroes

    np_acts-logits-general · gemini-2.0-flash

    Superhero and comic book characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    spotify
    -0.51
     Chart
    -0.51
     rumen
    -0.50
    Chart
    -0.49
    RenderAtEndOf
    -0.47
    fao
    -0.46
    նակ
    -0.46
    一首
    -0.45
    SocketChannel
    -0.45
     concorso
    -0.44
    POSITIVE LOGITS
     superhero
    1.68
     superheroes
    1.53
     Superhero
    1.48
     Batman
    1.41
     Superman
    1.26
     Marvel
    1.25
    Batman
    1.22
     comic
    1.19
     hero
    1.19
     batman
    1.14
    Activations Density 0.338%

    No Known Activations