© 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
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 10411
    Prev
    Next
    INDEX
    Explanations

    US city and state names, especially those in California

    oai_token-act-pair · gemini-2.0-flash

    Location context

    np_max-act-logits · gemini-2.0-flash

    say "US states" Method used: 2 — Tokens after max token are state abbreviations/names; say "US states"

    np_max-act-logits · gpt-5-miniTriggered by @chenshw0109
    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
     correctes
    -0.56
     produisons
    -0.55
    الدراسه
    -0.54
    帖最后由
    -0.52
    envolvimento
    -0.49
    
    -0.48
     cardiaque
    -0.48
    imedes
    -0.46
     Infórmanos
    -0.46
    adget
    -0.46
    POSITIVE LOGITS
     Span
    0.72
     Fla
    0.69
    FLA
    0.64
     Filip
    0.60
    Span
    0.60
     Ca
    0.60
     Kas
    0.59
     esp
    0.58
     FLA
    0.58
    Spa
    0.58
    Activations Density 2.256%

    No Known Activations