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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 2035
    Prev
    Next
    INDEX
    Explanations

    phrases related to handling and storing items effectively

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    phrasal verbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ]));
    
    -0.77
    ----</
    -0.72
    ]))
    
    -0.70
    })));
    -0.69
    "]);
    
    -0.68
    "));
    
    -0.64
    "]));
    -0.62
    ])));
    -0.62
    })),
    -0.60
    </tfoot>
    -0.60
    POSITIVE LOGITS
     purpoſe
    0.71
     myſelf
    0.67
     himſelf
    0.66
     houſe
    0.64
     laſt
    0.64
     ſame
    0.63
     becauſe
    0.60
     ailleurs
    0.58
     beſt
    0.56
     themſelves
    0.56
    Activations Density 0.135%

    No Known Activations