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-2-2B
    4. MLP Out - 16k
    5. 1-GEMMASCOPE-MLP-16K
    6. 8827
    Prev
    Next
    INDEX
    Explanations

    concepts related to intermediate frequencies and their variations in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k
    Configuration
    google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.1.hook_mlp_out
    Hook Layer
    1
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     intermediate
    -0.77
    intermediate
    -0.76
    Intermediate
    -0.73
     Intermediate
    -0.72
     Mus
    -0.70
    Mus
    -0.68
     intermédiaire
    -0.65
     rafra
    -0.63
    Rüyada
    -0.61
     CURIAM
    -0.60
    POSITIVE LOGITS
    Redaktion
    0.52
     surla
    0.50
     Signalez
    0.50
    ..\..\
    0.49
    Mvc
    0.49
    তথ্যসূত্র
    0.49
    typeorm
    0.48
    tidase
    0.47
    ChildScrollView
    0.46
     الحره
    0.45
    Activations Density 0.466%

    No Known Activations