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-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 7780
    Prev
    Next
    INDEX
    Explanations

    fax and attention

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     MatDialog
    -0.89
    っていない
    -0.88
     должны
    -0.83
     ficar
    -0.82
     autorità
    -0.79
    楽しみ
    -0.79
     staying
    -0.79
     devait
    -0.79
     Ayah
    -0.77
     enjoying
    -0.77
    POSITIVE LOGITS
     fax
    1.31
    Fax
    1.20
    Attn
    1.02
    fax
    1.01
    ATTN
    0.97
     Fax
    0.96
    FAX
    0.95
     doc
    0.91
     attention
    0.90
    attention
    0.88
    Activations Density 0.004%

    No Known Activations