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. 11-GEMMASCOPE-RES-16K
    6. 5635
    Prev
    Next
    INDEX
    Explanations

    specifications and attributes of graphical elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Datuak
    -0.81
    featureID
    -0.67
    lgari
    -0.66
     Normdatei
    -0.63
     насељу
    -0.56
    IonicModule
    -0.55
     kasarigan
    -0.53
     tartalomajánló
    -0.52
    PasswordField
    -0.52
    __':
    
    -0.51
    POSITIVE LOGITS
    英語版
    0.63
     ſame
    0.63
     venons
    0.60
    </u>
    0.58
    arithmic
    0.57
    šanai
    0.56
     koe
    0.56
    ̈́
    0.56
     vertes
    0.56
    lägen
    0.56
    Activations Density 0.054%

    No Known Activations