© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-2B
    3. 22-GEMMASCOPE-TRANSCODER-16K
    4. 10410
    Prev
    Next
    INDEX
    Explanations

    references to Pokemon, evolution, and Kamen Rider, but also sees common parts of speech, like the and to, as somewhat relevant

    oai_token-act-pair · gemini-2.0-flash

    Pokémon gyms/trainers

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Obrázky
    -0.60
    kuuta
    -0.56
    例文帳に追加
    -0.56
    }{*}{
    -0.54
    jspx
    -0.52
     DRM
    -0.52
     gynhyrchwyd
    -0.49
    colage
    -0.49
    +:+
    -0.48
    IsPostBack
    -0.48
    POSITIVE LOGITS
     Ditto
    0.63
     gym
    0.63
     Gym
    0.60
     gyms
    0.60
     trainer
    0.60
     gimnasio
    0.59
     Trainer
    0.57
    gym
    0.57
     trainers
    0.57
     Joh
    0.57
    Activations Density 1.197%

    No Known Activations