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 - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53460
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and certain programming symbols

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

    code-related structure after punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AdapterView
    -0.54
     termica
    -0.53
    scriptId
    -0.52
     enví
    -0.51
    openSession
    -0.50
     compless
    -0.49
     présidenti
    -0.48
     coppia
    -0.47
     derfor
    -0.47
     polaire
    -0.47
    POSITIVE LOGITS
    RegressionTest
    0.88
     '\\;'
    0.84
     مشين
    0.82
    webElementGuid
    0.82
    /*
    0.81
    CloseOperation
    0.79
     שוליים
    0.78
    OGND
    0.77
    +#+#
    0.75
    MLLoader
    0.74
    Activations Density 0.070%

    No Known Activations