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. 15-GEMMASCOPE-RES-16K
    6. 2440
    Prev
    Next
    INDEX
    Explanations

    sentences that describe conclusions or assessments about someone's health or well-being

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ValueStyle
    -0.69
     Cina
    -0.68
    EndProject
    -0.67
    Chinese
    -0.59
    +:+
    -0.58
     china
    -0.58
    Égypte
    -0.58
    ьаж
    -0.57
    PyExc
    -0.56
    //});
    -0.54
    POSITIVE LOGITS
    IUrlHelper
    0.61
    Hentet
    0.49
     FontFamily
    0.48
     conferred
    0.47
    TileEntity
    0.46
    Бахар
    0.46
     CallOverrides
    0.46
    NamedQueries
    0.45
     roundabout
    0.44
    ⃣
    0.44
    Activations Density 0.033%

    No Known Activations