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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 15772
    Prev
    Next
    INDEX
    Explanations

    clauses with speculative or conditional wording, often involving 'could', 'have', and words related to imagination or possibilities

    oai_token-act-pair · gemini-2.0-flash

    imagination and fictional scenarios

    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
     imaginary
    -2.03
     fictional
    -1.80
     imagine
    -1.79
    imaginary
    -1.78
     imagined
    -1.77
     immagin
    -1.75
     fictitious
    -1.71
    Imaginary
    -1.70
     hypothetical
    -1.70
     Imagin
    -1.69
    POSITIVE LOGITS
    riwal
    0.70
     disambiguazione
    0.60
    +#+
    0.57
     facie
    0.54
    PyExc
    0.54
    󠁢
    0.51
    }]
    0.51
    ecore
    0.50
    ⓘ
    0.50
    tuuri
    0.49
    Activations Density 1.401%

    No Known Activations