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. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 13959
    Prev
    Next
    INDEX
    Explanations

    details related to football player's transfers and contracts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    
    -0.57
     Kaka
    -0.52
     singlet
    -0.51
     unknownFields
    -0.49
     Serena
    -0.49
     Neymar
    -0.49
    ContentAsync
    -0.48
    NewGuid
    -0.47
    lename
    -0.46
     krzy
    -0.46
    POSITIVE LOGITS
    }))
    
    0.79
     betweenstory
    0.72
     promotion
    0.65
    انيف
    0.62
     TextInputType
    0.60
    MLLoader
    0.59
    TemporalType
    0.56
    ''');
    0.55
    Automatic
    0.55
    phrine
    0.54
    Activations Density 0.142%

    No Known Activations