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. 25-GEMMASCOPE-TRANSCODER-16K
    4. 9594
    Prev
    Next
    INDEX
    Explanations

    words associated with possibility and personal opinions, especially using modal verbs like "would," "could," and "may," and first-person pronouns such as "I," "we," and "myself," often within the context of religious or moral discussions.

    oai_token-act-pair · gemini-2.0-flash

    hedging statements

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.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
     Италијани
    -0.60
    #+#
    -0.59
     Italijani
    -0.54
    EndTag
    -0.53
     ***!
    -0.52
    gnore
    -0.51
     Chwiliwch
    -0.50
    ्यालय
    -0.49
    WindowConstants
    -0.48
    AddHtmlAttribute
    -0.48
    POSITIVE LOGITS
    اں
    0.54
    edding
    0.50
    bl
    0.49
    INF
    0.48
     Embedding
    0.47
    Bap
    0.47
    บค
    0.46
    Biographie
    0.46
    Історія
    0.46
    Indexing
    0.45
    Activations Density 6.446%

    No Known Activations