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. 18-GEMMASCOPE-RES-16K
    6. 778
    Prev
    Next
    INDEX
    Explanations

    terms related to legal frameworks and compliance

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

    purpose, intent, or design

    np_acts-logits-general · gemini-2.0-flash

    intended purpose or design

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    verwijspagina
    -0.90
    IUrlHelper
    -0.80
     nahilalakip
    -0.69
     صوتيه
    -0.68
    ьаж
    -0.67
     disponibilités
    -0.65
     disambiguazione
    -0.63
    Népesség
    -0.62
    nodoc
    -0.61
     fubject
    -0.59
    POSITIVE LOGITS
     intended
    2.16
    intended
    1.91
     Intended
    1.77
     designed
    1.60
    Intended
    1.54
     intent
    1.52
     intention
    1.48
     purpose
    1.43
     Designed
    1.38
    designed
    1.37
    Activations Density 0.481%

    No Known Activations