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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 13033
    Prev
    Next
    INDEX
    Explanations

    quantifiable evaluations of groups or categories

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

    Preceding superlative adjectives

    np_acts-logits-general · gemini-2.0-flash

    most, largest, highest, newest, best

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    est
    -0.61
    MORE
    -0.53
    Best
    -0.52
    bootstrapcdn
    -0.49
    nya
    -0.48
     besonders
    -0.48
    er
    -0.47
     hosszú
    -0.47
    特别
    -0.46
     lieber
    -0.46
    POSITIVE LOGITS
    =$?
    1.13
    rungsseite
    0.94
     ujednoznacz
    0.94
     surla
    0.94
     StatefulWidget
    0.94
    ContentAsync
    0.88
    "])
    
    0.86
    ]){
    
    0.84
     StatelessWidget
    0.83
     leggi
    0.81
    Activations Density 0.192%

    No Known Activations