Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 8-RES-MATRYOSHKA-DC
    4. 31180
    Prev
    Next
    INDEX
    Explanations

    the word "innocent" followed by a noun describing a person

    oai_token-act-pair · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard
    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
     tendance
    -0.75
     tendenza
    -0.68
     oração
    -0.64
     useRouter
    -0.64
     reducers
    -0.63
     τά
    -0.63
     casket
    -0.60
     reda
    -0.58
     Kear
    -0.57
     quæ
    -0.57
    POSITIVE LOGITS
     Innocence
    1.48
    innoc
    1.40
    Innoc
    1.38
    innocent
    1.36
     Innoc
    1.32
     Innocent
    1.32
     innocent
    1.29
     innocence
    1.24
     inocente
    1.13
     innoc
    1.13
    Activations Density 0.005%

    No Known Activations