© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-RES-MATRYOSHKA-DC
    7. 2025
    Prev
    Next
    INDEX
    Explanations

    the use of Keras for saving models

    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/blocks.22.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    chanind/pile-uncopyrighted-gemma-1024-abbrv-1B
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tagHelperRunner
    -0.85
    +#+#
    -0.59
    //
    -0.58
     kaarangay
    -0.58
    脚注の使い方
    -0.55
     ostavi
    -0.50
    ThroughAttribute
    -0.50
    entibus
    -0.50
    .*")]
    -0.50
     proie
    -0.50
    POSITIVE LOGITS
     training
    1.24
     Training
    1.05
     trained
    1.04
     train
    1.04
    Training
    0.99
    train
    0.98
    training
    0.98
     trainer
    0.94
     TRAIN
    0.91
    Train
    0.90
    Activations Density 0.840%

    No Known Activations