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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. MLP Out - 131k
    5. 6-GEMMASCOPE-MLP-131K
    6. 109717
    Prev
    Next
    INDEX
    Explanations

    instances of the word "che" and its variations, indicating a focus on products related to "cheese" or something similar

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.6.hook_mlp_out
    Hook Layer
    6
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    che
    -2.41
    CHE
    -1.70
    Che
    -1.66
     Che
    -1.64
     che
    -1.61
     CHE
    -1.46
    chel
    -1.15
    chee
    -1.08
     chemo
    -1.07
    chea
    -1.04
    POSITIVE LOGITS
     cim
    0.40
    RTLI
    0.40
     Fetal
    0.38
    kiksi
    0.38
    Rüyada
    0.38
     serv
    0.37
    arpa
    0.36
    ひ
    0.35
     brille
    0.35
    erati
    0.35
    Activations Density 0.003%

    No Known Activations