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. Gemma-2-2B
    3. 22-GEMMASCOPE-TRANSCODER-16K
    4. 14354
    Prev
    Next
    INDEX
    Explanations

    references to the first person

    oai_token-act-pair · gemini-2.0-flash

    "I"

    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
     Efq
    -1.06
     Jefus
    -1.03
     Monfieur
    -1.01
     Houſe
    -0.94
     Cæsar
    -0.92
     Majefty
    -0.90
     raiſ
    -0.88
     myſelf
    -0.86
     Theſe
    -0.86
     himſelf
    -0.85
    POSITIVE LOGITS
     can
    0.60
     don
    0.46
      
    0.44
     di
    0.44
     
    0.43
    interopRequire
    0.43
     createSprite
    0.40
    тербург
    0.40
    V
    0.39
     gynhyrchwyd
    0.39
    Activations Density 2.316%

    No Known Activations